Market orientation andbusiness economic

performanceA mediated model

Albert Maydeu-OlivaresMarketing Department Instituto de Empresa Madrid Spain and

Department of Psychology University of Barcelona BarcelonaSpain and

Nora LadoDepartment of Business Economics Universidad Carlos III de Madrid

Madrid Spain and Economics DepartmentEuropean University Institute

Keywords Market orientation Innovation Marketing research

Abstract Previous studies have found that market orientation signiregcantly predicts economicperformance The present study attempts to provide a necessarily partial model for how this impacttakes place using innovation degree innovation performance and customer loyalty as intermediatevariables The study targets the insurance industry in the European Union The sample accountedfor 22 percent of the companies and 17 percent of the insurance premiums in this market Theresults suggest that the addition of these variables improves predictions of objective economicperformance 52 percent over what is explained by market orientation alone Furthermore thestudy found that the effects of market orientation on economic performance are completelychanneled (mediated) through these variables particularly through innovation degree andinnovation performance Based on the results the paper provide guidelines for improving themarket share premium growth and proregtability of European Union insurance regrms

In a time characterized by increasingly rapid change in consumer preferenceseven faster technological progress and growing competitive rivalry it becomesessential for companies to develop mechanisms within their organizations togenerate market information analyze it and respond accordingly The set ofactivities developed by companies permanently to monitor analyze andrespond to these market changes is referred to in the Marketing literature asordfmarket orientationordm Over the last decade there has been a growing interest inthe construct of market orientation (Webster 1994 Day 1992) and itsusefulness in increasing companiesrsquo economic performance (Narver and Slater1990 Ruekert 1992 Jaworski and Kohli 1993 Lambin 1996 Deng and Dart1994) However it is still not well understood why there is such an effect and plusmn

The Emerald Research Register for this journal is available at The current issue and full text archive of this journal is available at

httpwwwemeraldinsightcomresearchregister httpwwwemeraldinsightcom0956-4233htm

This research has been partially funded by the Spanish Ministry of Education and CultureDireccioAcircn General de EnsenAumlanza Superior with project SEC2001-1169

IJSIM143

284

International Journal of ServiceIndustry ManagementVol 14 No 3 2003pp 284-309

q MCB UP Limited0956-4233DOI 10110809564230310478837

particularly plusmn how it operates More research is clearly needed in this area(DeshpandeAcirc et al 1993 DeshpandeAcirc and Farley 1998)

Two streams of research have developed aimed at reglling this gap Onestream of research has focused on the effects of market orientation oninnovation The other has investigated the links between market orientationand relationship marketing

Within the regrst stream we regnd a line of research that analyzes the effects ofmarket orientation on innovation performance For instance Atuahene-Gima(1996) and Gatignon and Xuered (1997) found a signiregcant relationshipbetween market orientation and several measures of new product performanceYet another line of research within this regrst stream focuses on the effects ofmarket orientation on the degree of business innovation For instance Lukasand Ferrell (2000) investigated how the different components of marketorientation are linked to an increase in business innovations (imitations linesextensions and new-to-the world) Also Han et al (1998) provide empiricalevidence concerning the market orientation-organizational innovativeness-performance chain In their substantial contribution to the advancement on theunderstanding of the variables that intermediate and make possible the marketorientation-performance relationship these authors concluded that marketorientation is conducive to facilitating both technical innovations (involvingeither products or processes) and administrative-organizational innovationsInterestingly Calantone et al (1994) found a non-signiregcant empiricalrelationship between degree of innovation and degree of innovation successconcluding that these two phenomena appear to be distinct An increase involume of innovations plusmn degree of innovation plusmn does not necessarily imply anincrease in new productsrsquo success rate plusmn degree of innovation success

As for the second stream of research three studies have investigated thelinks between market orientation and relationship marketing Webb et al(2000) and Baker et al (1999) report the impact of market orientation on keyrelationship constructs As with market orientation the focal point in therelationship marketing literature is satisfying customersrsquo needs moreeffectively than the competition while looking at customer relations from along-term perspective (Steinman et al 2000) Since the regnal objective ofrelationship marketing is to enhance customer loyalty it is necessary toincorporate the construct of market orientation in relationship marketingmodels (Webb et al 2000)

Clearly a broader model that integrates both streams of research innovationand relationship marketing while distinguishing between regrmsrsquo degree ofinnovation and their innovation success is needed to deepen our understandingof the mechanisms that lead the more market-oriented regrms to a bettereconomic performance To regll in this gap we postulate and empirical test amodel that hypothesizes that innovation degree innovation performance andcustomer loyalty actually mediate the effects of market orientation ontobusiness economic performance

Businesseconomic

performance

285

The rest of this paper is organized as follows First we review the constructof market orientation and we discuss existing research evidence on its effectson regrmsrsquo economic performance Then the role of innovation degreeinnovation performance and customer loyalty on this market orientation-businessperformance relationship is discussed followed by the formulation of a set ofhypotheses to be tested Next we present our empirical study with dueattention being given to a description of the sample and the validity andreliability issues associated with the measurement instruments This isfollowed by analysis of the research results which reveal the linkages betweenmarket orientation innovation performance and business performance Finallywe summarize the regndings of the study and we provide directions for furtherresearch

Theoretical frameworkMarket orientationMarket orientation was deregned by Narver and Slater (1990) as the competitivestrategy that most efregciently generates the right kinds of behavior to createenhanced value for the consumer and therefore assures better long-term resultsfor corporations According to these authors market orientation is based onorientation towards the customer orientation towards competitors andinter-functional coordination Kohli and Jaworski (1990) identify threestructural components of market orientation

(1) generation and analysis of all relevant information about the market

(2) dissemination of this information among the various departments of theorganization in order to coordinate and arrange strategic planning and

(3) implementation of strategic initiatives designed to satisfy the market

In reviewing this construct Lado et al (1998a) have provided a broaderderegnition of market orientation which they deregne as a competitive strategy thatinvolves all functional areas and levels of the organization and embraces thedifferent market participants These market participants or market forces are

the regnal customer

the intermediate customer (distributor)

the competitors and

environmental factors

To create and hold on to a competitive advantage companies must analyze andact on every one of these market forces with proper coordination between theirfunctions As a result in this theoretical framework market orientation can beconceptualized as consisting of nine facets

(1) analysis of the regnal customers

(2) analysis of intermediate customers (distributors)

IJSIM143

286

(3) analysis of the competitors

(4) analysis of the market environment

(5) strategic actions on the regnal customers

(6) strategic actions on intermediate customers (distributors)

(7) strategic actions on the competitors

(8) strategic actions on the market environment and

(9) inter-functional coordination

That market orientation is conceptualized as consisting of nine facets shouldnot be taken to imply that market orientation is a multidimensional conceptLado et al (1998a) have shown that these facets are well accounted for by aone-factor model Therefore these nine facets should be taken as the conceptualcomponents of a unidimensional construct of market orientation and aunidimensional measure of market orientation is called for

Market orientation as predictor of regrmsrsquo economic performanceSeveral studies have found a consistent positive relationship betweenbusinessesrsquo degree of market orientation and their economic performance(Deng and Dart 1994 Fritz 1996 Greenley 1995 Greenley and Foxall 19971998 Jaworski and Kohli 1993 Narver and Slater 1990 Pelham and Wilson1996 Pitt et al 1996 Ruekert 1992 Selnes et al 1996 Slater and Narver1994) Yet in most of these studies (eg Deng and Dart 1994 Fritz 1996Greenley 1995 Greenley and Foxall 1997 1998 Pelham and Wilson 1996Ruekert 1992 Selnes et al 1996) a wide cross-section of industries wasemployed as target population In so doing the observed co-variation betweenmarket orientation and economic performance confounds within-industry andbetween-industry market orientation variability It is important to separatethese two sources of variability since from an applied perspective interest liesin assessing increments in regrmsrsquo economic performance due to within-industrymarket orientation variability

The role of innovation degreeIn as much as the concept of market orientation subsumes knowledge aboutclientsrsquo present and future needs competitors trailing and a control ofenvironmental factors market orientation generates market intelligence and itmay be an important source of ideas for new products and services In thissense Cooper (1994) reports that a quality relationship with customersprovides valuable information to new productsrsquo development in the servicesector Also Subramanian (1997) reports a positive signiregcant associationbetween a multidimensional measure of innovation and organizationalperformance in the banking industry while DeshpandeAcirc et al (1993) report apositive association between degree of innovation and economic performancein a sample of Japanese corporations As Gatignon and Xuered (1997 p 77)

Businesseconomic

performance

287

afregrm in a recent article ordfit is possible that the strategic orientation of the regrmleads to at least in part superior performance because of the innovation thatare brought to market Although being market-oriented may lead to generalbeneregts for the regrmrsquos marketing activities the ability to bring to market newproducts which present the characteristics necessary to be successful may becriticalordm

Market orientation may also be an important determinant of innovation inthe services sector According to Atuahene-Gima (1996) in services like theinsurance and banking industries innovation success depends on the regrmrsquosmarket orientation especially on its customer orientation Being in touch withyour clients wants and needs and being able to respond appropriately to themis a key to innovation success in the service sector Furthermore the marketenvironment in the service sector is likely to be more competitive in terms ofproduct innovation than in other industries Innovation in services is moreeasily and quickly imitated (Tufano 1992) and more difregcult to protect bymeans of patenting Thus it may be than in this sector the relationshipbetween market orientation innovation and business performance isparticularly strong

The role of innovation performanceIn many instances new products arise from the coordination betweenmarketing and other business units such as RampD Also competitorsrsquomonitoring and a close relationship with distributors are key elements to thegeneration of new concepts for new product development As these areremacrected in the market orientation facets of regnal client analysis andenvironmental analysis one should expect a direct link from market orientationto new product performance

We regnd support for this hypothesis in the literature (eg Ottum and Moore1997 Slater and Narver 1994) Also in a meta-analysis on the determinantsof new product success Montoya-Weiss and Calantone (1994) identifymarket-related activities as one of the four more important factors thatdiscriminate between a new product success or failure Successful regrmsdevelop superior products that are attuned to customer wants and needs andthey also have strong marketing knowledge and skills to develop and launchthe product (Calantone et al 1996) As Cooper (1994 p 64) concluded insummarizing the results of new products research ordfa strong Market orientationis critical both to success and cycle time reductionordm

Innovation degree innovation performance and business performance areall linked together Calantone et al (1994) have investigated whether the sheervolume of innovation engaged in by the regrm determines the level of newproduct success Their regndings suggest that the degree of innovation of a regrmsis related to its new product performance Hence regrms that attempt to bringout more innovations may be more likely to succeed Similarly recent research

IJSIM143

288

shows that increased levels of innovation are associated to superiorperformance (Robinson et al 1992 DeshpandeAcirc et al 1993)

Market orientation and customer loyaltyDeshpandeAcirc et al (1993 p 24) point out that ordfthe canons of the marketingconcept assert that proregt is a reward for customer orientation which creates asatisreged customer but we have only the beginning of systematic empiricaldocumentation of the presumed relationshipordm In the present competitivemarket environment characterized by globalization with rapid market entry ofnew products and maturity conditions in many products and servicesattaining a high level of customer loyalty has emerged as a central managerialconcern Clearly customer loyalty constitutes an important objective forstrategic marketing planning (Kotler 1984) and represents an important basisfor developing a sustainable competitive advantage plusmn an advantage that canbe realized through market orientation A high degree of market orientationleads to customer loyalty which in the long run contributes to better economicperformance In the service sector the intangible nature of services gives rise toinformationrsquos asymmetry between buyers and sellers This results in higherrisk perceptions and greater difregculty in customerrsquos quality evaluation(Nayyar 1990) As a result market orientation becomes a crucial instrument toestablish long-term relations with customers in service regrms Kohli andJaworski (1990) posited a positive relationship between a regrmrsquos marketorientation level and customer satisfaction Webb et al (2000) and Lai (2003)provide further empirical support to this relationship Jones et al (2002)empirical research on business-to business buyer-seller relationship suggeststhat a strong salespersonrsquos customer orientation tends to reduce the customerrsquospropensity to switch suppliers Harrison-Walker (2001) found a positiverelationship between market orientation and customer retention customerwillingness to pay a price premium customer propensity to spread positiveword-of-mouth communication and customer propensity to not-alternateamong brandsservice providers All these variables are widely used measuresof the customer loyalty concept (Jacoby and Chestnut 1978 Odin et al 2001Dick and Basu 1994)

On the other hand customer loyalty is expected to have a positive impact onbusiness economic performance since market-oriented regrms have a largenumber of satisreged customer and therefore a higher rate of repeated purchases(Dick and Basu 1994 McCullough et al 1986 Loveman 1998 Kamakura et al2002)

The arguments put forth in the preceding sections can be summarized into aseries of hypotheses Our initial hypotheses is

H1 Within an industry the more market oriented regrms are the better theirobjective economic performance

Businesseconomic

performance

289

If this hypotheses is tenable then we shall investigate what is role ofinnovation degree innovation performance and customer loyalty in relation tothe hypothesized impact of market orientation on economic performance Wehypothesize that each of these variables taken separately is an intermediatemediational variable According to Baron and Kenny (1986) an intermediatevariable is said to be a mediator if when introduced within a directedrelationship the directed relationship vanishes (complete mediational effect) orat least it signiregcantly decreases (partial mediational effect)

Thus our hypotheses are

H2a Within an industry the impact of market orientation on economicperformance is at least partially mediated through innovation degreeThat is the more market oriented regrms are the higher their innovationdegree This higher innovation degree results in better economicperformance

H2b Within an industry the impact of market orientation on economicperformance is at least partially mediated through innovationperformance That is the more market oriented regrms are the highertheir innovation performance This higher innovation performanceresults in better economic performance

H2c Within an industry the impact of market orientation on economicperformance is at least partially mediated through customer loyaltyThat is the more market oriented regrms are the higher their customerloyalty This higher customer loyalty results in better economicperformance

If in turn these hypotheses are tenable then we need to specify a model thatintegrates these three intermediate variables within the hypothesized directedrelationship of market orientation on economic performance We hypothesizedthat

H3 Taken jointly innovation degree innovation performance andcustomer loyalty completely mediate the impact of marketorientation on economic performance Furthermore the relationshipbetween innovation degree and economic performance is all conveyedthrough innovation performance

This last hypothesis is graphically depicted in Figure 1

Empirical studyWith the objective of isolating within-industry variation we shall adopt asingle-industry approach focusing on the insurance industry This clearlyprevents the generalization of the results outside the scope of the industryconsidered On the other side we can meaningfully assess the impact of unitincrements in market orientation on regrmsrsquo economic performance and sound

IJSIM143

290

inferences can be drawn on the target population based on therepresentativeness of the sample used

The confounding of within-industry and between industry variation is notthe only threat to the validity of inferences drawn on the relationship betweenmarket orientation and economic performance A second threat is the noiseintroduced by environmental variables such as market turbulence marketgrowth rate buyer and supplier power and competitive intensity on businessperformance A standard approach to minimize this threat is to focus theresearch on a single market The drawback of this approach is that we are notable to capture regrmsrsquo behavior in facing increasing globalization and marketintegration As a compromise between these two ends the present studytargets the European Union market In this market the key characteristics of asingle market are preserved but it is also an environment in which we canpresently observe how regrms struggle in meeting the challenges ofinternationalization and market integration

A third threat to the validity of inferences drawn on the relationship betweenmarket orientation and economic performance lies in the use of subjectivemeasures of economic performance (ie managersrsquo evaluations of theircompaniesrsquo performance) Positive effects of market orientation on economicperformance have been reported when subjective assessments of performancewere used However when objective measures of economic performance havebeen used mixed results emerged For instance Ruekert (1992) and Lambin(1996) report a positive relationship between market orientation and objectivelymeasured economic performance However Bhuian (1997) Jaworski and Kohli(1993) and Selnes et al (1996) failed to regnd any signiregcant relationshipClearly when market orientation and economic performance are concurrentlyassessed by the regrmsrsquo managers a perceptual bias may be introduced A casein point Van Bruggen and Smidts (1995) found within one single company

Figure 1Hypothetical model

Businesseconomic

performance

291

(which has only one performance) a substantial degree of variation insubjective performance assessments In fact they report a positive relationshipbetween market orientation and judgments about the company performancewithin a single company As they point out ordfit might be that managers have amore positive view of their companyrsquos market orientation when they perceivetheir company to be performing wellordm (Van Bruggen and Smidts 1995 p 13)Hence it is important to employ objective measures of economic performanceand we shall do so in the present study

Market orientation in the services sector the European insurance industryThe insurance sector is of particular interest from a market orientationviewpoint as it works with intangible commodities in which service qualityand customer orientation are crucial elements The competitive characteristicsgenerated by the European Union provide an additional interest in studyingmarket orientation in this area The insurance sector in Europe has traditionallyoperated subject to strict regulations and strong protection from internationalcompetition However for some years now the European Commission has beenworking on the liberalization of this sector Effective implementation of this hasbrought about a major increase in competition within the sector and hasprovoked a major restructuring of insurance companies and groups Thecompetitive climate in Europe has also been inmacruenced by a downside in theeconomic cycle and changes in consumer behavior European customers nowshow greater service expectations and less loyalty As a result rivalry amongcompetitors is increasing as is the importance of competitive strategies adaptedto this sectorrsquos needs In this background the degree of orientation toward thecustomer distributors competition and the general socio-economicenvironment is becoming an increasingly important area of study not onlyfor academics but also for the business world

Lado et al (1998ab) have investigated quite extensively the marketorientation of insurance regrms within the European Union These authors havenot found signiregcant mean differences in market orientation by countryFurthermore they report substantial agreement between the factor structuresof market orientation across countries Thus it seems that the Europeaninsurance sector can be considered a homogenous population with respect tomarket orientation

DataThe population universe considered in this article is deregned as the set ofinsurance companies operating in the European Union which meet thefollowing conditions

they operate in private insurance or ordfmass insuranceordm

they have a market share of more than 005 percent and

their management is independent

IJSIM143

292

The list of European insurance companies was taken from the Financial TimesYearbook for 1996

It was assumed that senior executives were the people best qualireged to assessthe companyrsquos market orientation as well as their innovation degree innovationperformance and customersrsquo loyalty Therefore information from these variableswas gathered via a postal questionnaire submitted to the senior executive in eachof the 554 companies comprising the target population

We obtained 122 valid questionnaires giving a response rate of 22 percentThis sample accounts for over 17 percent of total insurance premiums in theEuropean Union

In order to assess response bias the questionnaires were divided intoquartiles on the basis of reception date (Armstrong and Overton 1977) Ananalysis of early and late responses did not indicate any signiregcant differencein terms of means and covariances

MeasuresBusiness economic performance is a complex construct with multiple possibleobserved indicators Here we measure this construct using three remacrectiveindicators[1]

(1) domestic market share (Jaworski and Kohli 1993 DeshpandeAcirc et al 1993Selnes et al 1996 Greenley and Foxall 1997 1998)

(2) premium growth (which is equivalent to sales growth for the insurancecompanies business (Slater and Narver 1994 Narver and Slater 1990Ruekert 1992 Greenley and Foxall 1997 1998) and

(3) proregtability per year averaged over the last three years(similar to thereturn on investment (ROI) rate (Greenley and Foxall 1997 1998))

All three indicators were expressed as percentages These data were obtainedfrom the managers responding the questionnaire Their responses were carefullycontrasted with published regnancial information (eg Reuters Insurance Brieregng)

Market orientation was measured using the Market Orientation Scale-Revised(MOS-R) This scale is a shortened version of the MOS validated by Lado et al(1998a) in the population of insurance companies of Belgium and Spain Ladoet al (1998a) shortened the original MOS scale while extending the previousvalidation study to target all insurance companies operating in the EuropeanUnion In the Appendix we provide the 30 items composing the MOS-R Eachitem is to be rated on a ten-point Likert-type scale ranging from 0 (completedisagreement) to 10 (complete agreement)

Innovation degree and innovation performance were assessed by means ofmulti-item questionnaires akin to Miller and Friesenrsquos (1982) Innovationperformance was measured by a four-item questionnaire regarding the successof a new productservice (deregned as an improved product a product extensionor a new product line) introduced by the company The questions involved

Businesseconomic

performance

293

whether the new productservice had succeeded in meeting the sales growthmarket share and proregt objectives set up by the company

Innovation degree was assessed by a three-item questionnaire that inquiredthe rate of new productsservices introduced by the company relative tocompetitors the amount of new productsservices marketed by the companyover the past three years and the nature of change of the new productsservices

Finally we used a four-item questionnaire based on existing literature (egDick and Basu 1994 Javalgi and Moberg 1997) to evaluate managersrsquoperceptions of their customersrsquo loyalty The questionnaire taps on theproportion of their customersrsquo insurance premiums taken on by the companythe average time a customer remains in the companyrsquos portfolio the probabilityof a customer renewing a premium and the overall perception of the companycustomersrsquo loyalty

Scale scores for innovation degree innovation performance and customerloyalty were obtained as an unweighted sum of the corresponding items Sincein all three cases Likert-type items on a 0-7 scale were used scale scores forthese variables range from 0-27 0-27 and 0-28 respectively For marketorientation we computed a score for each of its facets as an unweighted sum ofthe corresponding items Then a global market orientation score was obtainedas a sum of the facetsrsquo scores inversely weighted by their number of itemsHence this market orientation score assigns equal weights to each its facetsand ranges from 0-90

The scalesrsquo reliability (as assessed by coefregcient alpha) in this sample were088 (market orientation) 070 (innovation degree) 091 (innovationperformance) and 076 (customer loyalty) The means standard deviationsand correlations among all variables considered in this study are presented inTable I As can be seen in this table the three indicators of business economicperformance are signiregcantly but not largely correlated (the correlations rangefrom 020 to 029) The correlations among the hypothesized intermediatevariables (innovation degree innovation performance and customer loyalty)are not high except for innovation degree and innovation performance whichshare 36 percent of their variance The correlations of market orientation withthe intermediate variables appear signiregcantly larger (they range from 055 to058) than with the dependent variables (they range from 023 to 036) We alsoobserve in Table I that managers report on average a high degree of innovationin their businesses not so high a level of customer loyalty and a level ofinnovation performance just at the scale mean The average self-reporteddegree of market orientation is 56 on a 0-90 scale

MethodAll hypotheses were contrasted using covariance structure analysis asimplemented in LISREL 850 (see Joreskog et al 1999) Since all three indicatorsof business performance are highly positively skewed and present a high

IJSIM143

294

degree of kurtosis throughout this paper rather than attempting to transformthese variables to near-normality we shall employ an estimation approach thatis robust to non-normality of the observed variables The parameter estimateswere obtained using maximum likelihood estimation with standard errorssuitable for non-normal data (Joreskog et al 1999 Equation A24) and twotest statistics were used to assess the goodness of regt (GFI) of the model theSatorra-Bentler scaled chi-squared statistic (Satorra and Bentler 1988Equation 41) and Brownersquos (1984 Equation 220a) chi-squared statisticcorrected for non-normality To evaluate better the goodness of regt of thismodel several additional indices will also provided the root mean squarederror of approximation (RMSEA (Steiger 1990)) the standardized root meansquared residual (SRMSR (Joreskog et al 1999)) the GFI (Tanaka and Huba1985) and the Comparative Fit Index (CFI) using the independence model asbaseline (Bentler 1990 see also McDonald and Marsh 1990) Adequate to goodregt is suggested by RMSEA and SRMSR values approaching 005 For the GFIand the CFI indices values between 090 and 100 indicate adequate to excellentregt (but see Hu and Bentler 1999)

ResultsH1The model used to estimate the effects of market orientation on insurancebusinessesrsquo performance consists of a latent variable representing economicperformance with three indicators (market share premium growth andproregtability) and a single exogenous variable (market orientation) This model

MST PG PROF INNODR INNPERF LOYAL MO

MST 100PG 020 100PROF 023 029 100INNODR 030 034 035 1000INNPERF 027 032 041 062 100LOYAL 025 019 038 040 038 100MO 024 030 036 055 058 057 100x 455 788 626 1825 1422 1996 5628Std 535 749 571 356 292 520 1331

NotesN = 122 all correlations are signiregcant (a = 001) except those marked by which are onlysigniregcant at an a = 005MS = market share PG = premium growth PROF = proregtability INNODR = innovation degreeINNPERF = innovation performance LOYAL = customer loyalty MO = market orientation

Table IMeans standard

deviations andinter-correlations

Businesseconomic

performance

295

is depicted in Figure 2 The parameter estimates and GFIs for this model aregiven in Table II The model shows a good regt although note that it only hastwo degrees of freedom According to the model the best objective indicator ofbusiness economic performance is proregtability per year over 34 percent of itsvariance is accounted for by the model The standardized regressioncoefregcients reveal that proregtability per year is the best objective indicator of

Parameter estimates Goodness of regt R 2

Parameter Value Index Value Variable Value

b1 006 [061] (002) MFF X 2001 ( p = 099) Market share 015

b2 166 [039] (049) S-B X 2001 ( p = 099) Premium growth 024

b3 293 [049] (078) B X 2001 ( p = 099) Proregtability 034

b4 267 [059] (084) RMSEA 001 Business performance 037u1 17713 [100] (2241) df 2u2 2427 [085] (713) SRMSR 001u3 4248 [076] (1364) GFI 100u4 2141 [066] (515) CFI 100

NotesRobust asymptotic standard errors are provided in parentheses standardized parameterestimates are provided in square bracketsMFF X 2 = Minimum regt function chi-square SB X 2 = Satorra-Bentler scaled chi-squareB X 2 = Brownersquos chi-square corrected for non-normality RMSEA = root mean squared error ofapproximation SRMSR = standardized root mean squared residual GFI = goodness of regt indexCFI = comparative regt index R 2 = squared multiple correlations for endogenous variables

Table IIEstimation resultsfor the modeldepicted in Figure 2

Figure 2Market orientation aspredictor of economicperformance

IJSIM143

296

overall business performance Finally according to the model almost 37percent of overall business economic performance is accounted for by thedegree of market orientation

An inspection of the total effects of market orientation on the indicators ofeconomic performance suggests that unit increments of market orientation asmeasured by the MOS-R are associated with 0095 0168 and 0153 incrementsin domestic market share premium growth and proregtability per year averagedover the last three years respectively

H2A mediated model for the relationship between market orientation andbusiness performance is depicted in Figure 3 In this context a mediating effectis said to exist when

both mediating paths b5 b6 are signiregcant and

the direct effect of the exogenous variable on the outcome variablevanishes (complete mediational effect) or is signiregcantly lower (partialmediational effect) when a mediator variable is introduced in the model

Condition (2) amounts to b1 in Figure 3 becoming zero or signiregcantly less thatthan value reported for Figure 2

We used the mediated model depicted in Figure 3 to test for mediatingeffects of innovation degree innovation performance and customer loyaltyseparately on the impact of market orientation on business economicperformance We found that when either innovation performance or innovation

Figure 3Mediated model

Businesseconomic

performance

297

degree were used as mediating variable all the mediating paths weresigniregcant and that direct path from market orientation to businessperformance was not signiregcantly different from zero b1 = 004 t = 162for innovation performance b1 = 004 t = 186 for innovation degree Hencetaken separately both innovation degree and innovation performancecompletely mediate the impact of market orientation on businessperformance After regxing b1 at zero we re-estimated these two mediationalmodels The resulting parameter estimates and GFIs for these two models areshown in Table III On the other hand customer loyalty was found not to havea mediational effect between market orientation and business performanceThe parameter estimates and GFIs for this model are also given in Table III

As can be seen in Table III the mediating paths are signiregcant but thedirect path b1 is signiregcantly different from zero at a = 001 Furthermore a 99percent conregdence interval for the value for b1 reported in Table II (003 009)includes the value of b1 estimated in the mediational model using customerloyalty 004 Hence this variable does not even partially mediate on the impactof market orientation on business economic performance The standardizeddirect impact of market orientation on business performance (0408) is morethan twice the standardized impact of market orientation conveyed throughcustomer loyalty (0191)

The percentage of variance of business economic performance explained bythe model when innovation performance innovation degree or customer loyaltyare used as mediators is very similar (465 percent 453 percent and 437percent respectively)

H3The full model to be regtted corresponding to the hypothesis depicted in Figure 1is presented in Figure 3 The parameter estimates and goodness of regt testcorresponding to this model are given in Table IV (see also Figure 4) As can beseen in this Table the model regts these data very well

All the postulated relationships were found to be signiregcant at an a = 001Lagrange multiplier tests indicated that the regt of the model would notsigniregcantly improve by

adding a direct effect of market orientation to business performance nor

adding a direct effect of innovation degree on economic performance

Result (1) is in accordance with the results discussed above where we saw thatinnovation degree and innovation performance even when taken separatelycompletely mediate the impact of market orientation on business performanceResult (2) conregrms our hypothesis that innovation performance completelymediates the impact of innovation degree on business performance

Given that all effects of market orientation on business performance gothrough either innovation degree-innovation performance or through customer

IJSIM143

298

Par

amet

eres

tim

ates

Goo

dnes

sof

regt

Par

amet

erV

alue

Par

amet

erV

alue

Index

Val

ue

Innova

tion

per

form

ance

as

med

iato

rb 1

0(reg

xed

)u

117

713

[10

0](2

241

)M

FF

X2

560

(p=

035

)b 2

155

[04

0](0

42)

u2

241

4[0

84]

(71

1)S-B

X2

380

(p=

058

)b 3

263

[04

8](0

67)

u3

431

6[0

77]

(45

5)B

X2

429

(p=

051

)b 4

248

[05

9](o

81)

u4

211

3[0

65]

(45

5)df

5R

MSE

A0

03b 5

023

[05

8](0

03)

u5

180

3[0

67]

(28

0)SR

MSR

005

GF

I0

98b 6

018

[06

8](0

05)

CF

I0

99

Innova

tion

deg

ree

as

med

iato

rb

10

(regxed

)u

117

713

[10

0](2

241

)M

FF

X2

717

(p=

021

)b 2

169

[05

0](0

36)

u2

234

0[0

82]

(73

8)S-B

X2

497

(p=

043

)b 3

282

[05

1](0

82)

u3

415

6[0

74]

(45

0)B

X2

613

(p=

029

)b 4

226

[05

4](0

79)

u4

232

9[0

71]

(46

0)df

5R

MSE

A0

06b 5

012

[05

5](0

02)

u5

590

[06

9](0

78)

SR

MSR

005

GF

I0

98b 6

031

[06

7](0

09)

CF

I0

98

Cust

om

erlo

yalty

as

med

iato

rb 1

004

[04

1](0

02)

u1

177

13[1

00]

(22

41)

MF

FX

22

30(p

=0

68)

b 21

61[0

40]

(05

1)u

224

05

[08

4](7

11)

S-B

X2

155

(p=

082

)b 3

252

[04

5](0

74)

u3

447

5[0

80]

(45

5)B

X2

309

(p=

054

)b 4

265

[06

2](0

92)

u4

201

3[0

62]

(45

5)R

MSE

A0

0df

4b 5

015

[05

7](0

02)

u5

863

[06

8](1

04)

SR

MSR

002

GF

I0

99b 6

013

[03

4](0

05)

CF

I1

00

Note

s

Rob

ust

asym

pto

tic

stan

dar

der

rors

are

pro

vid

edin

par

enth

eses

st

andar

diz

edpar

amet

eres

tim

ates

are

pro

vid

edin

squar

ebra

cket

sM

FF

X2=

Min

imum

regt

funct

ion

chi-sq

uar

eSB

X2=

Sat

orra

-Ben

tler

scal

edch

i-sq

uar

eB

X2=

Bro

wnersquo

sch

i-sq

uar

eco

rrec

ted

for

non

-nor

mal

ity

RM

SE

A=

root

mea

nsq

uar

eder

ror

ofap

pro

xim

atio

n

SR

MSR

=st

andar

diz

edro

otm

ean

squar

edre

sidual

G

FI

=goo

dnes

sof

regt

index

C

FI

=co

mpar

ativ

ereg

tin

dex

Table IIIEstimation results

for the modeldepicted in Figure 3

Businesseconomic

performance

299

Par

amet

eres

tim

ates

Goo

dnes

sof

regt

R2

Par

amet

erV

alue

Par

amet

erV

alue

Index

Val

ue

Var

iable

Val

ue

b 10

12[0

55]

(22

41)

u1

177

13[1

00]

(00

2)M

FF

X2

105

9(p

=0

56)

Mar

ket

shar

e0

16b 2

013

[03

4](0

04)

u2

590

[06

9](0

78)

S-B

X2

785

(p=

080

)P

rem

ium

gro

wth

019

b 30

15[0

57]

(00

2)u

314

48

[05

4](2

59)

BX

212

67

(p=

039

)P

roreg

tabilit

y0

38b 4

078

[04

4](0

15)

u4

863

[06

8](1

04)

df

12B

usi

nes

sper

form

ance

056

RM

SE

A0

0b 5

016

[05

4](0

05)

u5

239

1[0

84]

(72

3)SR

MSR

005

Innov

atio

nper

form

ance

046

b 60

16[0

37]

(00

7)u

645

09

[08

1](1

271

)G

FI

098

Innov

atio

ndeg

ree

031

b 71

42[0

40]

(04

5)u

720

12

[06

2](4

60)

CF

I1

00C

ust

omer

loyal

ty0

32b 8

217

[04

4](0

64)

b 92

32[0

62]

(08

5)

Note

sR

obust

asym

pto

tic

stan

dar

der

rors

are

pro

vid

edin

par

enth

eses

st

andar

diz

edpar

amet

eres

tim

ates

are

pro

vid

edin

squar

ebra

cket

sM

FF

X2=

Min

imum

regtfu

nct

ion

chi-sq

uar

eSB

X2=

Sat

orra

-Ben

tler

scal

edch

i-sq

uar

eB

X2=

Bro

wnersquo

sch

i-sq

uar

eco

rrec

ted

for

non

-nor

mal

ity

RM

SE

A=

root

mea

nsq

uar

eder

ror

ofap

pro

xim

atio

n

SR

MSR

=st

andar

diz

edro

otm

ean

squar

edre

sidual

G

FI

=goo

dnes

sof

regt

index

C

FI

=co

mpar

ativ

ereg

tin

dex

R

2=

squar

edm

ult

iple

corr

elat

ions

for

endog

enou

svar

iable

s

Table IVEstimation resultsfor the modeldepicted in Figure 4

IJSIM143

300

loyalty a question arises as to the relative importance of the speciregc effectsgoing through these variables The standardized speciregc effect (computed as inBollen 1987) going through innovation degree and innovation performance is031 and 021 going through customer loyalty Hence the impact of marketorientation going through innovation is 50 percent more than that goingthrough customer loyalty

We can also see in Table IV that over 30 percent of the variance of theintermediate variables (innovation degree innovation performance andcustomer loyalty) are explained by market orientation In fact almost 50percent (464 percent to be exact) of innovation performance is explained bymarket orientation Furthermore note that the percentage of variance ofbusiness performance explained by the model is 561 percent a 52 percentincrement over what is explained by market orientation alone (see Table II) andover a 20 percent increment over what is explained by the mediational modelsconsidered previously Hence the inclusion of all three intermediate variables inthe model improves considerably our prediction of business performance

Furthermore we observe in Table IV that the direct effect of marketorientation on all three intermediate variables appear to be equal Also thedirect effects of customer loyalty and of innovation performance on economicperformance appear to be equal We re-estimated the model to test theseconstraints obtaining b1 = b2 = b3 = 0133 b5 = b6 = 0157 Satorra-BentlerX 2(15) = 8849 p = 0885 CFI = 100 GFI = 0975 SRMR = 006

ConclusionsMarket orientation can be deregned as a strategy used to reach a sustainablecompetitive advantage based on the generation and use of information within

Figure 4Full model

Businesseconomic

performance

301

organizations and on the selection of markets to be satisreged In thisframework we believe that competitive advantage results from the use ofresources and capabilities to generate differential satisfaction in proregtablemarkets Sustainability is achieved because the performance of the marketorientationrsquos behaviors requires complex organizational knowledge that cannoteasily be imitated by competitors Thus we hypothesize that the satisfaction ofproregtable markets permits the regrm to achieve a psychologically differentialposition that leads to brand loyalty and thus to higher proregts Previous studieshave

found a clear impact of market orientation on economic performance

assessed the effects of market orientation on innovation and

investigated the relationship between market orientation and relationshipmarketing variables

The present study proposes and tests a model that integrates both streams ofresearch innovation and relationship marketing More research is needed toinvestigate the role of customer satisfaction within this framework

In our necessarily partial model innovation degree innovation performanceand customer loyalty are used as intermediate variables on the effect of marketorientation on business performance Our results suggest that the addition ofthese variables help improve our predictions of business economic performance52 percent over what is explained by market orientation alone Also we foundthat innovation degree and innovation performance each taken separatelycompletely mediate the effect of market orientation on economic performanceFurthermore the impact of innovation degree on economic performance iscompletely channeled through innovation performance Customer loyalty byitself does not meditate the impact of market orientation on economicperformance but when considered along with innovation degree andinnovation performance it conveys some of the effects of market orientationon business performance This seemingly contradictory result arises from thefact that all three intermediate variables are interrelated

Our results should not be taken to imply that there are no other variablesmediate the effect of market orientation on economic performance We believethat other variables that have not been taken into account in this study such asproduct quality and customer satisfaction may also be signiregcant mediatorsHowever our results do suggest that whenever innovation degree andinnovation performance are included in the model as intermediate variablesthe effects of market orientation on business performance will mostly beconveyed through these variables Also in this study we have adopted thecurrently most widely accepted approach to measuring customer loyalty whichis based on behavioral loyalty measures (eg share of category requirementsrenew the policy probability the average customer last in the companyportfolio) It would be worth extending the present study by including not only

IJSIM143

302

behavioral measures of customer loyalty but also attitudinal measures alongthe lines of Dick and Basu (1994)

In our opinion two important contributions of the present research are outuse of objective measures of business performance and our focusing oninternational markets Despite the growing role of globalization and marketintegration and despite the increasing internationalization of corporationsmost studies on market orientation have focused on domestic markets (withnotable exceptions such as Selnes et al 1996 Webster 1994) Similarly moststudies on product innovation have also focused on domestic markets There isa lack of research yielding empirical support to the validity in an internationalsetting to research results obtained in domestic markets To regll this gap wetargeted the European Union market

Our study focused on a single industry the insurance sector Our sampleaccounted for 22 percent of the companies and 17 percent of the insurancepremiums in the targeted market An advantage of our single-industryapproach is that (with obvious reservations arising from the non-experimentalnature of our study and the fact that our sample should not be considered tohave been obtained at random) we can draw tentative predictions from ourmodel concerning the impact of market orientation on economic performance ininsurance companies operating in the European Union market An evidentdrawback of the single-industry approach adopted here is that it is not clearhow the present results extrapolate to other industries even when operating inthe same market

We have found that within the European Union insurance regrms thatconstantly monitor the evolution of current and potential customerrsquos needsmodify the attributes of the products to adapt them to the distributorsrequirements analyze competitorsrsquo marketing policy and products and knowbest of environmental trends especially technological and legal changes aremore likely to develop more new products have their new products accepted bythe market and obtain more loyal customers In turn increased customerloyalty and increased new product success will result in improved economicperformance However these conclusions must be taken with some caution aswe have used overall measures of innovation and innovation performanceFurther research is needed that takes into account the various aspects thatconstitute innovation

Previous studies have concluded that insurance companies still seethemselves as being product-focused and the industry as a whole is generallydistribution driven (Sodano 2000 Lambin 1996) This lack of closeness withcustomers has in turn contributed to the industry lack of product innovationsand product portfolios of commodity products that only compete on price Nowthe most successful regrms are redirecting their focus to the market needs andthey are beginning to exploit customer data and use market research togenerate ideas for designing new products In this context these companies

Businesseconomic

performance

303

now face choosing among too many new service options The managerialimplication of our study is that by enhancing their market orientation regrmswill know and service its customers better Thus they will generate moreinnovations by adopting a market-based product development process Alsoincreasing levels of market orientation enable regrms to discriminate more easilywhich new products have a higher success probability thus enhancing both theefregcacy and efregciency of new product development

Note

1 We considered employing the total volume of premiums for each insurance company as anadditional indicator although it seemed to us to be a better indicator of a companyrsquos sizerather than of its performance As we suspected the total volume of premiums wasuncorrelated with any of the variables considered in this study except for the companyrsquosmarket share (r = 024 p 001) Hence it is clear that volume of premiums should not beused as an indicator of insurance companiesrsquo performance

References

Armstrong JS and Overton T (1977) ordfEstimating nonresponse bias in mail surveysordm Journalof Marketing Research Vol 14 pp 396-402

Atuahene-Gima K (1996) ordfMarket orientation and innovationordm Journal of Business ResearchVol 35 pp 93-103

Baker TL Simpson PM and Sigauw JA (1999) ordfThe impact of suppliersrsquo perceptions ofreseller market orientation on key relationship constructsordm Journal of the Academy ofMarketing Science Vol 27 No 1 pp 50-7

Baron RM and Kenny DA (1986) ordfThe moderator-mediator variable distinction in socialpsychological research conceptual strategic and statistical considerationsordm Journal ofPersonality and Social Psychology Vol 51 pp 1173-82

Bentler PM (1990) ordfComparative regt indexes in structural modelsordm Psychological BulletinVol 107 pp 238-46

Bhuian SN (1997) ordfExploring market orientation in banks an empirical examination in SaudiArabiaordm The Journal of Service Marketing Vol 11 No 5 pp 317-28

Bollen KA (1987) ordfTotal direct and indirect effects in structural equation modelsordm inClogg CC (Ed) Sociological Methodology American Sociological AssociationWashington DC pp 37-69

Browne MW (1984) ordfAsymptotically distribution free methods for the analysis of covariancestructuresordm British Journal of Mathematical and Statistical Psychology Vol 37 pp 62-83

Calantone RJ di Benedetto AC and Bhoovaraghavan S (1994) ordfExamining the relationshipbetween degree of innovation and new product successordm Journal of Business ResearchVol 30 pp 143-8

Calantone RJ Schmidt JB and Song MX (1996) ordfControllable factors of new productssuccess a cross-national comparisonordm Marketing Science Vol 15 No 4 pp 341-58

Cooper RC (1994) ordfNew product the factors that drive successordm International MarketingReview Vol 11 No 1 pp 60-76

Day G (1992) ordfMarketingrsquos contribution to the strategy dialogueordm Journal of the Academy ofMarketing Science Vol 20 Fall pp 323-9

IJSIM143

304

The rest of this paper is organized as follows First we review the constructof market orientation and we discuss existing research evidence on its effectson regrmsrsquo economic performance Then the role of innovation degreeinnovation performance and customer loyalty on this market orientation-businessperformance relationship is discussed followed by the formulation of a set ofhypotheses to be tested Next we present our empirical study with dueattention being given to a description of the sample and the validity andreliability issues associated with the measurement instruments This isfollowed by analysis of the research results which reveal the linkages betweenmarket orientation innovation performance and business performance Finallywe summarize the regndings of the study and we provide directions for furtherresearch

Theoretical frameworkMarket orientationMarket orientation was deregned by Narver and Slater (1990) as the competitivestrategy that most efregciently generates the right kinds of behavior to createenhanced value for the consumer and therefore assures better long-term resultsfor corporations According to these authors market orientation is based onorientation towards the customer orientation towards competitors andinter-functional coordination Kohli and Jaworski (1990) identify threestructural components of market orientation

(1) generation and analysis of all relevant information about the market

(2) dissemination of this information among the various departments of theorganization in order to coordinate and arrange strategic planning and

(3) implementation of strategic initiatives designed to satisfy the market

In reviewing this construct Lado et al (1998a) have provided a broaderderegnition of market orientation which they deregne as a competitive strategy thatinvolves all functional areas and levels of the organization and embraces thedifferent market participants These market participants or market forces are

the regnal customer

the intermediate customer (distributor)

the competitors and

environmental factors

To create and hold on to a competitive advantage companies must analyze andact on every one of these market forces with proper coordination between theirfunctions As a result in this theoretical framework market orientation can beconceptualized as consisting of nine facets

(1) analysis of the regnal customers

(2) analysis of intermediate customers (distributors)

IJSIM143

286

(3) analysis of the competitors

(4) analysis of the market environment

(5) strategic actions on the regnal customers

(6) strategic actions on intermediate customers (distributors)

(7) strategic actions on the competitors

(8) strategic actions on the market environment and

(9) inter-functional coordination

That market orientation is conceptualized as consisting of nine facets shouldnot be taken to imply that market orientation is a multidimensional conceptLado et al (1998a) have shown that these facets are well accounted for by aone-factor model Therefore these nine facets should be taken as the conceptualcomponents of a unidimensional construct of market orientation and aunidimensional measure of market orientation is called for

Market orientation as predictor of regrmsrsquo economic performanceSeveral studies have found a consistent positive relationship betweenbusinessesrsquo degree of market orientation and their economic performance(Deng and Dart 1994 Fritz 1996 Greenley 1995 Greenley and Foxall 19971998 Jaworski and Kohli 1993 Narver and Slater 1990 Pelham and Wilson1996 Pitt et al 1996 Ruekert 1992 Selnes et al 1996 Slater and Narver1994) Yet in most of these studies (eg Deng and Dart 1994 Fritz 1996Greenley 1995 Greenley and Foxall 1997 1998 Pelham and Wilson 1996Ruekert 1992 Selnes et al 1996) a wide cross-section of industries wasemployed as target population In so doing the observed co-variation betweenmarket orientation and economic performance confounds within-industry andbetween-industry market orientation variability It is important to separatethese two sources of variability since from an applied perspective interest liesin assessing increments in regrmsrsquo economic performance due to within-industrymarket orientation variability

The role of innovation degreeIn as much as the concept of market orientation subsumes knowledge aboutclientsrsquo present and future needs competitors trailing and a control ofenvironmental factors market orientation generates market intelligence and itmay be an important source of ideas for new products and services In thissense Cooper (1994) reports that a quality relationship with customersprovides valuable information to new productsrsquo development in the servicesector Also Subramanian (1997) reports a positive signiregcant associationbetween a multidimensional measure of innovation and organizationalperformance in the banking industry while DeshpandeAcirc et al (1993) report apositive association between degree of innovation and economic performancein a sample of Japanese corporations As Gatignon and Xuered (1997 p 77)

Businesseconomic

performance

287

afregrm in a recent article ordfit is possible that the strategic orientation of the regrmleads to at least in part superior performance because of the innovation thatare brought to market Although being market-oriented may lead to generalbeneregts for the regrmrsquos marketing activities the ability to bring to market newproducts which present the characteristics necessary to be successful may becriticalordm

Market orientation may also be an important determinant of innovation inthe services sector According to Atuahene-Gima (1996) in services like theinsurance and banking industries innovation success depends on the regrmrsquosmarket orientation especially on its customer orientation Being in touch withyour clients wants and needs and being able to respond appropriately to themis a key to innovation success in the service sector Furthermore the marketenvironment in the service sector is likely to be more competitive in terms ofproduct innovation than in other industries Innovation in services is moreeasily and quickly imitated (Tufano 1992) and more difregcult to protect bymeans of patenting Thus it may be than in this sector the relationshipbetween market orientation innovation and business performance isparticularly strong

The role of innovation performanceIn many instances new products arise from the coordination betweenmarketing and other business units such as RampD Also competitorsrsquomonitoring and a close relationship with distributors are key elements to thegeneration of new concepts for new product development As these areremacrected in the market orientation facets of regnal client analysis andenvironmental analysis one should expect a direct link from market orientationto new product performance

We regnd support for this hypothesis in the literature (eg Ottum and Moore1997 Slater and Narver 1994) Also in a meta-analysis on the determinantsof new product success Montoya-Weiss and Calantone (1994) identifymarket-related activities as one of the four more important factors thatdiscriminate between a new product success or failure Successful regrmsdevelop superior products that are attuned to customer wants and needs andthey also have strong marketing knowledge and skills to develop and launchthe product (Calantone et al 1996) As Cooper (1994 p 64) concluded insummarizing the results of new products research ordfa strong Market orientationis critical both to success and cycle time reductionordm

Innovation degree innovation performance and business performance areall linked together Calantone et al (1994) have investigated whether the sheervolume of innovation engaged in by the regrm determines the level of newproduct success Their regndings suggest that the degree of innovation of a regrmsis related to its new product performance Hence regrms that attempt to bringout more innovations may be more likely to succeed Similarly recent research

IJSIM143

288

shows that increased levels of innovation are associated to superiorperformance (Robinson et al 1992 DeshpandeAcirc et al 1993)

Market orientation and customer loyaltyDeshpandeAcirc et al (1993 p 24) point out that ordfthe canons of the marketingconcept assert that proregt is a reward for customer orientation which creates asatisreged customer but we have only the beginning of systematic empiricaldocumentation of the presumed relationshipordm In the present competitivemarket environment characterized by globalization with rapid market entry ofnew products and maturity conditions in many products and servicesattaining a high level of customer loyalty has emerged as a central managerialconcern Clearly customer loyalty constitutes an important objective forstrategic marketing planning (Kotler 1984) and represents an important basisfor developing a sustainable competitive advantage plusmn an advantage that canbe realized through market orientation A high degree of market orientationleads to customer loyalty which in the long run contributes to better economicperformance In the service sector the intangible nature of services gives rise toinformationrsquos asymmetry between buyers and sellers This results in higherrisk perceptions and greater difregculty in customerrsquos quality evaluation(Nayyar 1990) As a result market orientation becomes a crucial instrument toestablish long-term relations with customers in service regrms Kohli andJaworski (1990) posited a positive relationship between a regrmrsquos marketorientation level and customer satisfaction Webb et al (2000) and Lai (2003)provide further empirical support to this relationship Jones et al (2002)empirical research on business-to business buyer-seller relationship suggeststhat a strong salespersonrsquos customer orientation tends to reduce the customerrsquospropensity to switch suppliers Harrison-Walker (2001) found a positiverelationship between market orientation and customer retention customerwillingness to pay a price premium customer propensity to spread positiveword-of-mouth communication and customer propensity to not-alternateamong brandsservice providers All these variables are widely used measuresof the customer loyalty concept (Jacoby and Chestnut 1978 Odin et al 2001Dick and Basu 1994)

On the other hand customer loyalty is expected to have a positive impact onbusiness economic performance since market-oriented regrms have a largenumber of satisreged customer and therefore a higher rate of repeated purchases(Dick and Basu 1994 McCullough et al 1986 Loveman 1998 Kamakura et al2002)

The arguments put forth in the preceding sections can be summarized into aseries of hypotheses Our initial hypotheses is

H1 Within an industry the more market oriented regrms are the better theirobjective economic performance

Businesseconomic

performance

289

If this hypotheses is tenable then we shall investigate what is role ofinnovation degree innovation performance and customer loyalty in relation tothe hypothesized impact of market orientation on economic performance Wehypothesize that each of these variables taken separately is an intermediatemediational variable According to Baron and Kenny (1986) an intermediatevariable is said to be a mediator if when introduced within a directedrelationship the directed relationship vanishes (complete mediational effect) orat least it signiregcantly decreases (partial mediational effect)

Thus our hypotheses are

H2a Within an industry the impact of market orientation on economicperformance is at least partially mediated through innovation degreeThat is the more market oriented regrms are the higher their innovationdegree This higher innovation degree results in better economicperformance

H2b Within an industry the impact of market orientation on economicperformance is at least partially mediated through innovationperformance That is the more market oriented regrms are the highertheir innovation performance This higher innovation performanceresults in better economic performance

H2c Within an industry the impact of market orientation on economicperformance is at least partially mediated through customer loyaltyThat is the more market oriented regrms are the higher their customerloyalty This higher customer loyalty results in better economicperformance

If in turn these hypotheses are tenable then we need to specify a model thatintegrates these three intermediate variables within the hypothesized directedrelationship of market orientation on economic performance We hypothesizedthat

H3 Taken jointly innovation degree innovation performance andcustomer loyalty completely mediate the impact of marketorientation on economic performance Furthermore the relationshipbetween innovation degree and economic performance is all conveyedthrough innovation performance

This last hypothesis is graphically depicted in Figure 1

Empirical studyWith the objective of isolating within-industry variation we shall adopt asingle-industry approach focusing on the insurance industry This clearlyprevents the generalization of the results outside the scope of the industryconsidered On the other side we can meaningfully assess the impact of unitincrements in market orientation on regrmsrsquo economic performance and sound

IJSIM143

290

inferences can be drawn on the target population based on therepresentativeness of the sample used

The confounding of within-industry and between industry variation is notthe only threat to the validity of inferences drawn on the relationship betweenmarket orientation and economic performance A second threat is the noiseintroduced by environmental variables such as market turbulence marketgrowth rate buyer and supplier power and competitive intensity on businessperformance A standard approach to minimize this threat is to focus theresearch on a single market The drawback of this approach is that we are notable to capture regrmsrsquo behavior in facing increasing globalization and marketintegration As a compromise between these two ends the present studytargets the European Union market In this market the key characteristics of asingle market are preserved but it is also an environment in which we canpresently observe how regrms struggle in meeting the challenges ofinternationalization and market integration

A third threat to the validity of inferences drawn on the relationship betweenmarket orientation and economic performance lies in the use of subjectivemeasures of economic performance (ie managersrsquo evaluations of theircompaniesrsquo performance) Positive effects of market orientation on economicperformance have been reported when subjective assessments of performancewere used However when objective measures of economic performance havebeen used mixed results emerged For instance Ruekert (1992) and Lambin(1996) report a positive relationship between market orientation and objectivelymeasured economic performance However Bhuian (1997) Jaworski and Kohli(1993) and Selnes et al (1996) failed to regnd any signiregcant relationshipClearly when market orientation and economic performance are concurrentlyassessed by the regrmsrsquo managers a perceptual bias may be introduced A casein point Van Bruggen and Smidts (1995) found within one single company

Figure 1Hypothetical model

Businesseconomic

performance

291

(which has only one performance) a substantial degree of variation insubjective performance assessments In fact they report a positive relationshipbetween market orientation and judgments about the company performancewithin a single company As they point out ordfit might be that managers have amore positive view of their companyrsquos market orientation when they perceivetheir company to be performing wellordm (Van Bruggen and Smidts 1995 p 13)Hence it is important to employ objective measures of economic performanceand we shall do so in the present study

Market orientation in the services sector the European insurance industryThe insurance sector is of particular interest from a market orientationviewpoint as it works with intangible commodities in which service qualityand customer orientation are crucial elements The competitive characteristicsgenerated by the European Union provide an additional interest in studyingmarket orientation in this area The insurance sector in Europe has traditionallyoperated subject to strict regulations and strong protection from internationalcompetition However for some years now the European Commission has beenworking on the liberalization of this sector Effective implementation of this hasbrought about a major increase in competition within the sector and hasprovoked a major restructuring of insurance companies and groups Thecompetitive climate in Europe has also been inmacruenced by a downside in theeconomic cycle and changes in consumer behavior European customers nowshow greater service expectations and less loyalty As a result rivalry amongcompetitors is increasing as is the importance of competitive strategies adaptedto this sectorrsquos needs In this background the degree of orientation toward thecustomer distributors competition and the general socio-economicenvironment is becoming an increasingly important area of study not onlyfor academics but also for the business world

Lado et al (1998ab) have investigated quite extensively the marketorientation of insurance regrms within the European Union These authors havenot found signiregcant mean differences in market orientation by countryFurthermore they report substantial agreement between the factor structuresof market orientation across countries Thus it seems that the Europeaninsurance sector can be considered a homogenous population with respect tomarket orientation

DataThe population universe considered in this article is deregned as the set ofinsurance companies operating in the European Union which meet thefollowing conditions

they operate in private insurance or ordfmass insuranceordm

they have a market share of more than 005 percent and

their management is independent

IJSIM143

292

The list of European insurance companies was taken from the Financial TimesYearbook for 1996

It was assumed that senior executives were the people best qualireged to assessthe companyrsquos market orientation as well as their innovation degree innovationperformance and customersrsquo loyalty Therefore information from these variableswas gathered via a postal questionnaire submitted to the senior executive in eachof the 554 companies comprising the target population

We obtained 122 valid questionnaires giving a response rate of 22 percentThis sample accounts for over 17 percent of total insurance premiums in theEuropean Union

In order to assess response bias the questionnaires were divided intoquartiles on the basis of reception date (Armstrong and Overton 1977) Ananalysis of early and late responses did not indicate any signiregcant differencein terms of means and covariances

MeasuresBusiness economic performance is a complex construct with multiple possibleobserved indicators Here we measure this construct using three remacrectiveindicators[1]

(1) domestic market share (Jaworski and Kohli 1993 DeshpandeAcirc et al 1993Selnes et al 1996 Greenley and Foxall 1997 1998)

(2) premium growth (which is equivalent to sales growth for the insurancecompanies business (Slater and Narver 1994 Narver and Slater 1990Ruekert 1992 Greenley and Foxall 1997 1998) and

(3) proregtability per year averaged over the last three years(similar to thereturn on investment (ROI) rate (Greenley and Foxall 1997 1998))

All three indicators were expressed as percentages These data were obtainedfrom the managers responding the questionnaire Their responses were carefullycontrasted with published regnancial information (eg Reuters Insurance Brieregng)

Market orientation was measured using the Market Orientation Scale-Revised(MOS-R) This scale is a shortened version of the MOS validated by Lado et al(1998a) in the population of insurance companies of Belgium and Spain Ladoet al (1998a) shortened the original MOS scale while extending the previousvalidation study to target all insurance companies operating in the EuropeanUnion In the Appendix we provide the 30 items composing the MOS-R Eachitem is to be rated on a ten-point Likert-type scale ranging from 0 (completedisagreement) to 10 (complete agreement)

Innovation degree and innovation performance were assessed by means ofmulti-item questionnaires akin to Miller and Friesenrsquos (1982) Innovationperformance was measured by a four-item questionnaire regarding the successof a new productservice (deregned as an improved product a product extensionor a new product line) introduced by the company The questions involved

Businesseconomic

performance

293

whether the new productservice had succeeded in meeting the sales growthmarket share and proregt objectives set up by the company

Innovation degree was assessed by a three-item questionnaire that inquiredthe rate of new productsservices introduced by the company relative tocompetitors the amount of new productsservices marketed by the companyover the past three years and the nature of change of the new productsservices

Finally we used a four-item questionnaire based on existing literature (egDick and Basu 1994 Javalgi and Moberg 1997) to evaluate managersrsquoperceptions of their customersrsquo loyalty The questionnaire taps on theproportion of their customersrsquo insurance premiums taken on by the companythe average time a customer remains in the companyrsquos portfolio the probabilityof a customer renewing a premium and the overall perception of the companycustomersrsquo loyalty

Scale scores for innovation degree innovation performance and customerloyalty were obtained as an unweighted sum of the corresponding items Sincein all three cases Likert-type items on a 0-7 scale were used scale scores forthese variables range from 0-27 0-27 and 0-28 respectively For marketorientation we computed a score for each of its facets as an unweighted sum ofthe corresponding items Then a global market orientation score was obtainedas a sum of the facetsrsquo scores inversely weighted by their number of itemsHence this market orientation score assigns equal weights to each its facetsand ranges from 0-90

The scalesrsquo reliability (as assessed by coefregcient alpha) in this sample were088 (market orientation) 070 (innovation degree) 091 (innovationperformance) and 076 (customer loyalty) The means standard deviationsand correlations among all variables considered in this study are presented inTable I As can be seen in this table the three indicators of business economicperformance are signiregcantly but not largely correlated (the correlations rangefrom 020 to 029) The correlations among the hypothesized intermediatevariables (innovation degree innovation performance and customer loyalty)are not high except for innovation degree and innovation performance whichshare 36 percent of their variance The correlations of market orientation withthe intermediate variables appear signiregcantly larger (they range from 055 to058) than with the dependent variables (they range from 023 to 036) We alsoobserve in Table I that managers report on average a high degree of innovationin their businesses not so high a level of customer loyalty and a level ofinnovation performance just at the scale mean The average self-reporteddegree of market orientation is 56 on a 0-90 scale

MethodAll hypotheses were contrasted using covariance structure analysis asimplemented in LISREL 850 (see Joreskog et al 1999) Since all three indicatorsof business performance are highly positively skewed and present a high

IJSIM143

294

degree of kurtosis throughout this paper rather than attempting to transformthese variables to near-normality we shall employ an estimation approach thatis robust to non-normality of the observed variables The parameter estimateswere obtained using maximum likelihood estimation with standard errorssuitable for non-normal data (Joreskog et al 1999 Equation A24) and twotest statistics were used to assess the goodness of regt (GFI) of the model theSatorra-Bentler scaled chi-squared statistic (Satorra and Bentler 1988Equation 41) and Brownersquos (1984 Equation 220a) chi-squared statisticcorrected for non-normality To evaluate better the goodness of regt of thismodel several additional indices will also provided the root mean squarederror of approximation (RMSEA (Steiger 1990)) the standardized root meansquared residual (SRMSR (Joreskog et al 1999)) the GFI (Tanaka and Huba1985) and the Comparative Fit Index (CFI) using the independence model asbaseline (Bentler 1990 see also McDonald and Marsh 1990) Adequate to goodregt is suggested by RMSEA and SRMSR values approaching 005 For the GFIand the CFI indices values between 090 and 100 indicate adequate to excellentregt (but see Hu and Bentler 1999)

ResultsH1The model used to estimate the effects of market orientation on insurancebusinessesrsquo performance consists of a latent variable representing economicperformance with three indicators (market share premium growth andproregtability) and a single exogenous variable (market orientation) This model

MST PG PROF INNODR INNPERF LOYAL MO

MST 100PG 020 100PROF 023 029 100INNODR 030 034 035 1000INNPERF 027 032 041 062 100LOYAL 025 019 038 040 038 100MO 024 030 036 055 058 057 100x 455 788 626 1825 1422 1996 5628Std 535 749 571 356 292 520 1331

NotesN = 122 all correlations are signiregcant (a = 001) except those marked by which are onlysigniregcant at an a = 005MS = market share PG = premium growth PROF = proregtability INNODR = innovation degreeINNPERF = innovation performance LOYAL = customer loyalty MO = market orientation

Table IMeans standard

deviations andinter-correlations

Businesseconomic

performance

295

is depicted in Figure 2 The parameter estimates and GFIs for this model aregiven in Table II The model shows a good regt although note that it only hastwo degrees of freedom According to the model the best objective indicator ofbusiness economic performance is proregtability per year over 34 percent of itsvariance is accounted for by the model The standardized regressioncoefregcients reveal that proregtability per year is the best objective indicator of

Parameter estimates Goodness of regt R 2

Parameter Value Index Value Variable Value

b1 006 [061] (002) MFF X 2001 ( p = 099) Market share 015

b2 166 [039] (049) S-B X 2001 ( p = 099) Premium growth 024

b3 293 [049] (078) B X 2001 ( p = 099) Proregtability 034

b4 267 [059] (084) RMSEA 001 Business performance 037u1 17713 [100] (2241) df 2u2 2427 [085] (713) SRMSR 001u3 4248 [076] (1364) GFI 100u4 2141 [066] (515) CFI 100

NotesRobust asymptotic standard errors are provided in parentheses standardized parameterestimates are provided in square bracketsMFF X 2 = Minimum regt function chi-square SB X 2 = Satorra-Bentler scaled chi-squareB X 2 = Brownersquos chi-square corrected for non-normality RMSEA = root mean squared error ofapproximation SRMSR = standardized root mean squared residual GFI = goodness of regt indexCFI = comparative regt index R 2 = squared multiple correlations for endogenous variables

Table IIEstimation resultsfor the modeldepicted in Figure 2

Figure 2Market orientation aspredictor of economicperformance

IJSIM143

296

overall business performance Finally according to the model almost 37percent of overall business economic performance is accounted for by thedegree of market orientation

An inspection of the total effects of market orientation on the indicators ofeconomic performance suggests that unit increments of market orientation asmeasured by the MOS-R are associated with 0095 0168 and 0153 incrementsin domestic market share premium growth and proregtability per year averagedover the last three years respectively

H2A mediated model for the relationship between market orientation andbusiness performance is depicted in Figure 3 In this context a mediating effectis said to exist when

both mediating paths b5 b6 are signiregcant and

the direct effect of the exogenous variable on the outcome variablevanishes (complete mediational effect) or is signiregcantly lower (partialmediational effect) when a mediator variable is introduced in the model

Condition (2) amounts to b1 in Figure 3 becoming zero or signiregcantly less thatthan value reported for Figure 2

We used the mediated model depicted in Figure 3 to test for mediatingeffects of innovation degree innovation performance and customer loyaltyseparately on the impact of market orientation on business economicperformance We found that when either innovation performance or innovation

Figure 3Mediated model

Businesseconomic

performance

297

degree were used as mediating variable all the mediating paths weresigniregcant and that direct path from market orientation to businessperformance was not signiregcantly different from zero b1 = 004 t = 162for innovation performance b1 = 004 t = 186 for innovation degree Hencetaken separately both innovation degree and innovation performancecompletely mediate the impact of market orientation on businessperformance After regxing b1 at zero we re-estimated these two mediationalmodels The resulting parameter estimates and GFIs for these two models areshown in Table III On the other hand customer loyalty was found not to havea mediational effect between market orientation and business performanceThe parameter estimates and GFIs for this model are also given in Table III

As can be seen in Table III the mediating paths are signiregcant but thedirect path b1 is signiregcantly different from zero at a = 001 Furthermore a 99percent conregdence interval for the value for b1 reported in Table II (003 009)includes the value of b1 estimated in the mediational model using customerloyalty 004 Hence this variable does not even partially mediate on the impactof market orientation on business economic performance The standardizeddirect impact of market orientation on business performance (0408) is morethan twice the standardized impact of market orientation conveyed throughcustomer loyalty (0191)

The percentage of variance of business economic performance explained bythe model when innovation performance innovation degree or customer loyaltyare used as mediators is very similar (465 percent 453 percent and 437percent respectively)

H3The full model to be regtted corresponding to the hypothesis depicted in Figure 1is presented in Figure 3 The parameter estimates and goodness of regt testcorresponding to this model are given in Table IV (see also Figure 4) As can beseen in this Table the model regts these data very well

All the postulated relationships were found to be signiregcant at an a = 001Lagrange multiplier tests indicated that the regt of the model would notsigniregcantly improve by

adding a direct effect of market orientation to business performance nor

adding a direct effect of innovation degree on economic performance

Result (1) is in accordance with the results discussed above where we saw thatinnovation degree and innovation performance even when taken separatelycompletely mediate the impact of market orientation on business performanceResult (2) conregrms our hypothesis that innovation performance completelymediates the impact of innovation degree on business performance

Given that all effects of market orientation on business performance gothrough either innovation degree-innovation performance or through customer

IJSIM143

298

Par

amet

eres

tim

ates

Goo

dnes

sof

regt

Par

amet

erV

alue

Par

amet

erV

alue

Index

Val

ue

Innova

tion

per

form

ance

as

med

iato

rb 1

0(reg

xed

)u

117

713

[10

0](2

241

)M

FF

X2

560

(p=

035

)b 2

155

[04

0](0

42)

u2

241

4[0

84]

(71

1)S-B

X2

380

(p=

058

)b 3

263

[04

8](0

67)

u3

431

6[0

77]

(45

5)B

X2

429

(p=

051

)b 4

248

[05

9](o

81)

u4

211

3[0

65]

(45

5)df

5R

MSE

A0

03b 5

023

[05

8](0

03)

u5

180

3[0

67]

(28

0)SR

MSR

005

GF

I0

98b 6

018

[06

8](0

05)

CF

I0

99

Innova

tion

deg

ree

as

med

iato

rb

10

(regxed

)u

117

713

[10

0](2

241

)M

FF

X2

717

(p=

021

)b 2

169

[05

0](0

36)

u2

234

0[0

82]

(73

8)S-B

X2

497

(p=

043

)b 3

282

[05

1](0

82)

u3

415

6[0

74]

(45

0)B

X2

613

(p=

029

)b 4

226

[05

4](0

79)

u4

232

9[0

71]

(46

0)df

5R

MSE

A0

06b 5

012

[05

5](0

02)

u5

590

[06

9](0

78)

SR

MSR

005

GF

I0

98b 6

031

[06

7](0

09)

CF

I0

98

Cust

om

erlo

yalty

as

med

iato

rb 1

004

[04

1](0

02)

u1

177

13[1

00]

(22

41)

MF

FX

22

30(p

=0

68)

b 21

61[0

40]

(05

1)u

224

05

[08

4](7

11)

S-B

X2

155

(p=

082

)b 3

252

[04

5](0

74)

u3

447

5[0

80]

(45

5)B

X2

309

(p=

054

)b 4

265

[06

2](0

92)

u4

201

3[0

62]

(45

5)R

MSE

A0

0df

4b 5

015

[05

7](0

02)

u5

863

[06

8](1

04)

SR

MSR

002

GF

I0

99b 6

013

[03

4](0

05)

CF

I1

00

Note

s

Rob

ust

asym

pto

tic

stan

dar

der

rors

are

pro

vid

edin

par

enth

eses

st

andar

diz

edpar

amet

eres

tim

ates

are

pro

vid

edin

squar

ebra

cket

sM

FF

X2=

Min

imum

regt

funct

ion

chi-sq

uar

eSB

X2=

Sat

orra

-Ben

tler

scal

edch

i-sq

uar

eB

X2=

Bro

wnersquo

sch

i-sq

uar

eco

rrec

ted

for

non

-nor

mal

ity

RM

SE

A=

root

mea

nsq

uar

eder

ror

ofap

pro

xim

atio

n

SR

MSR

=st

andar

diz

edro

otm

ean

squar

edre

sidual

G

FI

=goo

dnes

sof

regt

index

C

FI

=co

mpar

ativ

ereg

tin

dex

Table IIIEstimation results

for the modeldepicted in Figure 3

Businesseconomic

performance

299

Par

amet

eres

tim

ates

Goo

dnes

sof

regt

R2

Par

amet

erV

alue

Par

amet

erV

alue

Index

Val

ue

Var

iable

Val

ue

b 10

12[0

55]

(22

41)

u1

177

13[1

00]

(00

2)M

FF

X2

105

9(p

=0

56)

Mar

ket

shar

e0

16b 2

013

[03

4](0

04)

u2

590

[06

9](0

78)

S-B

X2

785

(p=

080

)P

rem

ium

gro

wth

019

b 30

15[0

57]

(00

2)u

314

48

[05

4](2

59)

BX

212

67

(p=

039

)P

roreg

tabilit

y0

38b 4

078

[04

4](0

15)

u4

863

[06

8](1

04)

df

12B

usi

nes

sper

form

ance

056

RM

SE

A0

0b 5

016

[05

4](0

05)

u5

239

1[0

84]

(72

3)SR

MSR

005

Innov

atio

nper

form

ance

046

b 60

16[0

37]

(00

7)u

645

09

[08

1](1

271

)G

FI

098

Innov

atio

ndeg

ree

031

b 71

42[0

40]

(04

5)u

720

12

[06

2](4

60)

CF

I1

00C

ust

omer

loyal

ty0

32b 8

217

[04

4](0

64)

b 92

32[0

62]

(08

5)

Note

sR

obust

asym

pto

tic

stan

dar

der

rors

are

pro

vid

edin

par

enth

eses

st

andar

diz

edpar

amet

eres

tim

ates

are

pro

vid

edin

squar

ebra

cket

sM

FF

X2=

Min

imum

regtfu

nct

ion

chi-sq

uar

eSB

X2=

Sat

orra

-Ben

tler

scal

edch

i-sq

uar

eB

X2=

Bro

wnersquo

sch

i-sq

uar

eco

rrec

ted

for

non

-nor

mal

ity

RM

SE

A=

root

mea

nsq

uar

eder

ror

ofap

pro

xim

atio

n

SR

MSR

=st

andar

diz

edro

otm

ean

squar

edre

sidual

G

FI

=goo

dnes

sof

regt

index

C

FI

=co

mpar

ativ

ereg

tin

dex

R

2=

squar

edm

ult

iple

corr

elat

ions

for

endog

enou

svar

iable

s

Table IVEstimation resultsfor the modeldepicted in Figure 4

IJSIM143

300

loyalty a question arises as to the relative importance of the speciregc effectsgoing through these variables The standardized speciregc effect (computed as inBollen 1987) going through innovation degree and innovation performance is031 and 021 going through customer loyalty Hence the impact of marketorientation going through innovation is 50 percent more than that goingthrough customer loyalty

We can also see in Table IV that over 30 percent of the variance of theintermediate variables (innovation degree innovation performance andcustomer loyalty) are explained by market orientation In fact almost 50percent (464 percent to be exact) of innovation performance is explained bymarket orientation Furthermore note that the percentage of variance ofbusiness performance explained by the model is 561 percent a 52 percentincrement over what is explained by market orientation alone (see Table II) andover a 20 percent increment over what is explained by the mediational modelsconsidered previously Hence the inclusion of all three intermediate variables inthe model improves considerably our prediction of business performance

Furthermore we observe in Table IV that the direct effect of marketorientation on all three intermediate variables appear to be equal Also thedirect effects of customer loyalty and of innovation performance on economicperformance appear to be equal We re-estimated the model to test theseconstraints obtaining b1 = b2 = b3 = 0133 b5 = b6 = 0157 Satorra-BentlerX 2(15) = 8849 p = 0885 CFI = 100 GFI = 0975 SRMR = 006

ConclusionsMarket orientation can be deregned as a strategy used to reach a sustainablecompetitive advantage based on the generation and use of information within

Figure 4Full model

Businesseconomic

performance

301

organizations and on the selection of markets to be satisreged In thisframework we believe that competitive advantage results from the use ofresources and capabilities to generate differential satisfaction in proregtablemarkets Sustainability is achieved because the performance of the marketorientationrsquos behaviors requires complex organizational knowledge that cannoteasily be imitated by competitors Thus we hypothesize that the satisfaction ofproregtable markets permits the regrm to achieve a psychologically differentialposition that leads to brand loyalty and thus to higher proregts Previous studieshave

found a clear impact of market orientation on economic performance

assessed the effects of market orientation on innovation and

investigated the relationship between market orientation and relationshipmarketing variables

The present study proposes and tests a model that integrates both streams ofresearch innovation and relationship marketing More research is needed toinvestigate the role of customer satisfaction within this framework

In our necessarily partial model innovation degree innovation performanceand customer loyalty are used as intermediate variables on the effect of marketorientation on business performance Our results suggest that the addition ofthese variables help improve our predictions of business economic performance52 percent over what is explained by market orientation alone Also we foundthat innovation degree and innovation performance each taken separatelycompletely mediate the effect of market orientation on economic performanceFurthermore the impact of innovation degree on economic performance iscompletely channeled through innovation performance Customer loyalty byitself does not meditate the impact of market orientation on economicperformance but when considered along with innovation degree andinnovation performance it conveys some of the effects of market orientationon business performance This seemingly contradictory result arises from thefact that all three intermediate variables are interrelated

Our results should not be taken to imply that there are no other variablesmediate the effect of market orientation on economic performance We believethat other variables that have not been taken into account in this study such asproduct quality and customer satisfaction may also be signiregcant mediatorsHowever our results do suggest that whenever innovation degree andinnovation performance are included in the model as intermediate variablesthe effects of market orientation on business performance will mostly beconveyed through these variables Also in this study we have adopted thecurrently most widely accepted approach to measuring customer loyalty whichis based on behavioral loyalty measures (eg share of category requirementsrenew the policy probability the average customer last in the companyportfolio) It would be worth extending the present study by including not only

IJSIM143

302

behavioral measures of customer loyalty but also attitudinal measures alongthe lines of Dick and Basu (1994)

In our opinion two important contributions of the present research are outuse of objective measures of business performance and our focusing oninternational markets Despite the growing role of globalization and marketintegration and despite the increasing internationalization of corporationsmost studies on market orientation have focused on domestic markets (withnotable exceptions such as Selnes et al 1996 Webster 1994) Similarly moststudies on product innovation have also focused on domestic markets There isa lack of research yielding empirical support to the validity in an internationalsetting to research results obtained in domestic markets To regll this gap wetargeted the European Union market

Our study focused on a single industry the insurance sector Our sampleaccounted for 22 percent of the companies and 17 percent of the insurancepremiums in the targeted market An advantage of our single-industryapproach is that (with obvious reservations arising from the non-experimentalnature of our study and the fact that our sample should not be considered tohave been obtained at random) we can draw tentative predictions from ourmodel concerning the impact of market orientation on economic performance ininsurance companies operating in the European Union market An evidentdrawback of the single-industry approach adopted here is that it is not clearhow the present results extrapolate to other industries even when operating inthe same market

We have found that within the European Union insurance regrms thatconstantly monitor the evolution of current and potential customerrsquos needsmodify the attributes of the products to adapt them to the distributorsrequirements analyze competitorsrsquo marketing policy and products and knowbest of environmental trends especially technological and legal changes aremore likely to develop more new products have their new products accepted bythe market and obtain more loyal customers In turn increased customerloyalty and increased new product success will result in improved economicperformance However these conclusions must be taken with some caution aswe have used overall measures of innovation and innovation performanceFurther research is needed that takes into account the various aspects thatconstitute innovation

Previous studies have concluded that insurance companies still seethemselves as being product-focused and the industry as a whole is generallydistribution driven (Sodano 2000 Lambin 1996) This lack of closeness withcustomers has in turn contributed to the industry lack of product innovationsand product portfolios of commodity products that only compete on price Nowthe most successful regrms are redirecting their focus to the market needs andthey are beginning to exploit customer data and use market research togenerate ideas for designing new products In this context these companies

Businesseconomic

performance

303

now face choosing among too many new service options The managerialimplication of our study is that by enhancing their market orientation regrmswill know and service its customers better Thus they will generate moreinnovations by adopting a market-based product development process Alsoincreasing levels of market orientation enable regrms to discriminate more easilywhich new products have a higher success probability thus enhancing both theefregcacy and efregciency of new product development

Note

1 We considered employing the total volume of premiums for each insurance company as anadditional indicator although it seemed to us to be a better indicator of a companyrsquos sizerather than of its performance As we suspected the total volume of premiums wasuncorrelated with any of the variables considered in this study except for the companyrsquosmarket share (r = 024 p 001) Hence it is clear that volume of premiums should not beused as an indicator of insurance companiesrsquo performance

References

Armstrong JS and Overton T (1977) ordfEstimating nonresponse bias in mail surveysordm Journalof Marketing Research Vol 14 pp 396-402

Atuahene-Gima K (1996) ordfMarket orientation and innovationordm Journal of Business ResearchVol 35 pp 93-103

Baker TL Simpson PM and Sigauw JA (1999) ordfThe impact of suppliersrsquo perceptions ofreseller market orientation on key relationship constructsordm Journal of the Academy ofMarketing Science Vol 27 No 1 pp 50-7

Baron RM and Kenny DA (1986) ordfThe moderator-mediator variable distinction in socialpsychological research conceptual strategic and statistical considerationsordm Journal ofPersonality and Social Psychology Vol 51 pp 1173-82

Bentler PM (1990) ordfComparative regt indexes in structural modelsordm Psychological BulletinVol 107 pp 238-46

Bhuian SN (1997) ordfExploring market orientation in banks an empirical examination in SaudiArabiaordm The Journal of Service Marketing Vol 11 No 5 pp 317-28

Bollen KA (1987) ordfTotal direct and indirect effects in structural equation modelsordm inClogg CC (Ed) Sociological Methodology American Sociological AssociationWashington DC pp 37-69

Browne MW (1984) ordfAsymptotically distribution free methods for the analysis of covariancestructuresordm British Journal of Mathematical and Statistical Psychology Vol 37 pp 62-83

Calantone RJ di Benedetto AC and Bhoovaraghavan S (1994) ordfExamining the relationshipbetween degree of innovation and new product successordm Journal of Business ResearchVol 30 pp 143-8

Calantone RJ Schmidt JB and Song MX (1996) ordfControllable factors of new productssuccess a cross-national comparisonordm Marketing Science Vol 15 No 4 pp 341-58

Cooper RC (1994) ordfNew product the factors that drive successordm International MarketingReview Vol 11 No 1 pp 60-76

Day G (1992) ordfMarketingrsquos contribution to the strategy dialogueordm Journal of the Academy ofMarketing Science Vol 20 Fall pp 323-9

IJSIM143

304

(3) analysis of the competitors

(4) analysis of the market environment

(5) strategic actions on the regnal customers

(6) strategic actions on intermediate customers (distributors)

(7) strategic actions on the competitors

(8) strategic actions on the market environment and

(9) inter-functional coordination

That market orientation is conceptualized as consisting of nine facets shouldnot be taken to imply that market orientation is a multidimensional conceptLado et al (1998a) have shown that these facets are well accounted for by aone-factor model Therefore these nine facets should be taken as the conceptualcomponents of a unidimensional construct of market orientation and aunidimensional measure of market orientation is called for

Market orientation as predictor of regrmsrsquo economic performanceSeveral studies have found a consistent positive relationship betweenbusinessesrsquo degree of market orientation and their economic performance(Deng and Dart 1994 Fritz 1996 Greenley 1995 Greenley and Foxall 19971998 Jaworski and Kohli 1993 Narver and Slater 1990 Pelham and Wilson1996 Pitt et al 1996 Ruekert 1992 Selnes et al 1996 Slater and Narver1994) Yet in most of these studies (eg Deng and Dart 1994 Fritz 1996Greenley 1995 Greenley and Foxall 1997 1998 Pelham and Wilson 1996Ruekert 1992 Selnes et al 1996) a wide cross-section of industries wasemployed as target population In so doing the observed co-variation betweenmarket orientation and economic performance confounds within-industry andbetween-industry market orientation variability It is important to separatethese two sources of variability since from an applied perspective interest liesin assessing increments in regrmsrsquo economic performance due to within-industrymarket orientation variability

The role of innovation degreeIn as much as the concept of market orientation subsumes knowledge aboutclientsrsquo present and future needs competitors trailing and a control ofenvironmental factors market orientation generates market intelligence and itmay be an important source of ideas for new products and services In thissense Cooper (1994) reports that a quality relationship with customersprovides valuable information to new productsrsquo development in the servicesector Also Subramanian (1997) reports a positive signiregcant associationbetween a multidimensional measure of innovation and organizationalperformance in the banking industry while DeshpandeAcirc et al (1993) report apositive association between degree of innovation and economic performancein a sample of Japanese corporations As Gatignon and Xuered (1997 p 77)

Businesseconomic

performance

287

afregrm in a recent article ordfit is possible that the strategic orientation of the regrmleads to at least in part superior performance because of the innovation thatare brought to market Although being market-oriented may lead to generalbeneregts for the regrmrsquos marketing activities the ability to bring to market newproducts which present the characteristics necessary to be successful may becriticalordm

Market orientation may also be an important determinant of innovation inthe services sector According to Atuahene-Gima (1996) in services like theinsurance and banking industries innovation success depends on the regrmrsquosmarket orientation especially on its customer orientation Being in touch withyour clients wants and needs and being able to respond appropriately to themis a key to innovation success in the service sector Furthermore the marketenvironment in the service sector is likely to be more competitive in terms ofproduct innovation than in other industries Innovation in services is moreeasily and quickly imitated (Tufano 1992) and more difregcult to protect bymeans of patenting Thus it may be than in this sector the relationshipbetween market orientation innovation and business performance isparticularly strong

The role of innovation performanceIn many instances new products arise from the coordination betweenmarketing and other business units such as RampD Also competitorsrsquomonitoring and a close relationship with distributors are key elements to thegeneration of new concepts for new product development As these areremacrected in the market orientation facets of regnal client analysis andenvironmental analysis one should expect a direct link from market orientationto new product performance

We regnd support for this hypothesis in the literature (eg Ottum and Moore1997 Slater and Narver 1994) Also in a meta-analysis on the determinantsof new product success Montoya-Weiss and Calantone (1994) identifymarket-related activities as one of the four more important factors thatdiscriminate between a new product success or failure Successful regrmsdevelop superior products that are attuned to customer wants and needs andthey also have strong marketing knowledge and skills to develop and launchthe product (Calantone et al 1996) As Cooper (1994 p 64) concluded insummarizing the results of new products research ordfa strong Market orientationis critical both to success and cycle time reductionordm

Innovation degree innovation performance and business performance areall linked together Calantone et al (1994) have investigated whether the sheervolume of innovation engaged in by the regrm determines the level of newproduct success Their regndings suggest that the degree of innovation of a regrmsis related to its new product performance Hence regrms that attempt to bringout more innovations may be more likely to succeed Similarly recent research

IJSIM143

288

shows that increased levels of innovation are associated to superiorperformance (Robinson et al 1992 DeshpandeAcirc et al 1993)

Market orientation and customer loyaltyDeshpandeAcirc et al (1993 p 24) point out that ordfthe canons of the marketingconcept assert that proregt is a reward for customer orientation which creates asatisreged customer but we have only the beginning of systematic empiricaldocumentation of the presumed relationshipordm In the present competitivemarket environment characterized by globalization with rapid market entry ofnew products and maturity conditions in many products and servicesattaining a high level of customer loyalty has emerged as a central managerialconcern Clearly customer loyalty constitutes an important objective forstrategic marketing planning (Kotler 1984) and represents an important basisfor developing a sustainable competitive advantage plusmn an advantage that canbe realized through market orientation A high degree of market orientationleads to customer loyalty which in the long run contributes to better economicperformance In the service sector the intangible nature of services gives rise toinformationrsquos asymmetry between buyers and sellers This results in higherrisk perceptions and greater difregculty in customerrsquos quality evaluation(Nayyar 1990) As a result market orientation becomes a crucial instrument toestablish long-term relations with customers in service regrms Kohli andJaworski (1990) posited a positive relationship between a regrmrsquos marketorientation level and customer satisfaction Webb et al (2000) and Lai (2003)provide further empirical support to this relationship Jones et al (2002)empirical research on business-to business buyer-seller relationship suggeststhat a strong salespersonrsquos customer orientation tends to reduce the customerrsquospropensity to switch suppliers Harrison-Walker (2001) found a positiverelationship between market orientation and customer retention customerwillingness to pay a price premium customer propensity to spread positiveword-of-mouth communication and customer propensity to not-alternateamong brandsservice providers All these variables are widely used measuresof the customer loyalty concept (Jacoby and Chestnut 1978 Odin et al 2001Dick and Basu 1994)

On the other hand customer loyalty is expected to have a positive impact onbusiness economic performance since market-oriented regrms have a largenumber of satisreged customer and therefore a higher rate of repeated purchases(Dick and Basu 1994 McCullough et al 1986 Loveman 1998 Kamakura et al2002)

The arguments put forth in the preceding sections can be summarized into aseries of hypotheses Our initial hypotheses is

H1 Within an industry the more market oriented regrms are the better theirobjective economic performance

Businesseconomic

performance

289

If this hypotheses is tenable then we shall investigate what is role ofinnovation degree innovation performance and customer loyalty in relation tothe hypothesized impact of market orientation on economic performance Wehypothesize that each of these variables taken separately is an intermediatemediational variable According to Baron and Kenny (1986) an intermediatevariable is said to be a mediator if when introduced within a directedrelationship the directed relationship vanishes (complete mediational effect) orat least it signiregcantly decreases (partial mediational effect)

Thus our hypotheses are

H2a Within an industry the impact of market orientation on economicperformance is at least partially mediated through innovation degreeThat is the more market oriented regrms are the higher their innovationdegree This higher innovation degree results in better economicperformance

H2b Within an industry the impact of market orientation on economicperformance is at least partially mediated through innovationperformance That is the more market oriented regrms are the highertheir innovation performance This higher innovation performanceresults in better economic performance

H2c Within an industry the impact of market orientation on economicperformance is at least partially mediated through customer loyaltyThat is the more market oriented regrms are the higher their customerloyalty This higher customer loyalty results in better economicperformance

If in turn these hypotheses are tenable then we need to specify a model thatintegrates these three intermediate variables within the hypothesized directedrelationship of market orientation on economic performance We hypothesizedthat

H3 Taken jointly innovation degree innovation performance andcustomer loyalty completely mediate the impact of marketorientation on economic performance Furthermore the relationshipbetween innovation degree and economic performance is all conveyedthrough innovation performance

This last hypothesis is graphically depicted in Figure 1

Empirical studyWith the objective of isolating within-industry variation we shall adopt asingle-industry approach focusing on the insurance industry This clearlyprevents the generalization of the results outside the scope of the industryconsidered On the other side we can meaningfully assess the impact of unitincrements in market orientation on regrmsrsquo economic performance and sound

IJSIM143

290

inferences can be drawn on the target population based on therepresentativeness of the sample used

The confounding of within-industry and between industry variation is notthe only threat to the validity of inferences drawn on the relationship betweenmarket orientation and economic performance A second threat is the noiseintroduced by environmental variables such as market turbulence marketgrowth rate buyer and supplier power and competitive intensity on businessperformance A standard approach to minimize this threat is to focus theresearch on a single market The drawback of this approach is that we are notable to capture regrmsrsquo behavior in facing increasing globalization and marketintegration As a compromise between these two ends the present studytargets the European Union market In this market the key characteristics of asingle market are preserved but it is also an environment in which we canpresently observe how regrms struggle in meeting the challenges ofinternationalization and market integration

A third threat to the validity of inferences drawn on the relationship betweenmarket orientation and economic performance lies in the use of subjectivemeasures of economic performance (ie managersrsquo evaluations of theircompaniesrsquo performance) Positive effects of market orientation on economicperformance have been reported when subjective assessments of performancewere used However when objective measures of economic performance havebeen used mixed results emerged For instance Ruekert (1992) and Lambin(1996) report a positive relationship between market orientation and objectivelymeasured economic performance However Bhuian (1997) Jaworski and Kohli(1993) and Selnes et al (1996) failed to regnd any signiregcant relationshipClearly when market orientation and economic performance are concurrentlyassessed by the regrmsrsquo managers a perceptual bias may be introduced A casein point Van Bruggen and Smidts (1995) found within one single company

Figure 1Hypothetical model

Businesseconomic

performance

291

(which has only one performance) a substantial degree of variation insubjective performance assessments In fact they report a positive relationshipbetween market orientation and judgments about the company performancewithin a single company As they point out ordfit might be that managers have amore positive view of their companyrsquos market orientation when they perceivetheir company to be performing wellordm (Van Bruggen and Smidts 1995 p 13)Hence it is important to employ objective measures of economic performanceand we shall do so in the present study

Market orientation in the services sector the European insurance industryThe insurance sector is of particular interest from a market orientationviewpoint as it works with intangible commodities in which service qualityand customer orientation are crucial elements The competitive characteristicsgenerated by the European Union provide an additional interest in studyingmarket orientation in this area The insurance sector in Europe has traditionallyoperated subject to strict regulations and strong protection from internationalcompetition However for some years now the European Commission has beenworking on the liberalization of this sector Effective implementation of this hasbrought about a major increase in competition within the sector and hasprovoked a major restructuring of insurance companies and groups Thecompetitive climate in Europe has also been inmacruenced by a downside in theeconomic cycle and changes in consumer behavior European customers nowshow greater service expectations and less loyalty As a result rivalry amongcompetitors is increasing as is the importance of competitive strategies adaptedto this sectorrsquos needs In this background the degree of orientation toward thecustomer distributors competition and the general socio-economicenvironment is becoming an increasingly important area of study not onlyfor academics but also for the business world

Lado et al (1998ab) have investigated quite extensively the marketorientation of insurance regrms within the European Union These authors havenot found signiregcant mean differences in market orientation by countryFurthermore they report substantial agreement between the factor structuresof market orientation across countries Thus it seems that the Europeaninsurance sector can be considered a homogenous population with respect tomarket orientation

DataThe population universe considered in this article is deregned as the set ofinsurance companies operating in the European Union which meet thefollowing conditions

they operate in private insurance or ordfmass insuranceordm

they have a market share of more than 005 percent and

their management is independent

IJSIM143

292

The list of European insurance companies was taken from the Financial TimesYearbook for 1996

It was assumed that senior executives were the people best qualireged to assessthe companyrsquos market orientation as well as their innovation degree innovationperformance and customersrsquo loyalty Therefore information from these variableswas gathered via a postal questionnaire submitted to the senior executive in eachof the 554 companies comprising the target population

We obtained 122 valid questionnaires giving a response rate of 22 percentThis sample accounts for over 17 percent of total insurance premiums in theEuropean Union

In order to assess response bias the questionnaires were divided intoquartiles on the basis of reception date (Armstrong and Overton 1977) Ananalysis of early and late responses did not indicate any signiregcant differencein terms of means and covariances

MeasuresBusiness economic performance is a complex construct with multiple possibleobserved indicators Here we measure this construct using three remacrectiveindicators[1]

(1) domestic market share (Jaworski and Kohli 1993 DeshpandeAcirc et al 1993Selnes et al 1996 Greenley and Foxall 1997 1998)

(2) premium growth (which is equivalent to sales growth for the insurancecompanies business (Slater and Narver 1994 Narver and Slater 1990Ruekert 1992 Greenley and Foxall 1997 1998) and

(3) proregtability per year averaged over the last three years(similar to thereturn on investment (ROI) rate (Greenley and Foxall 1997 1998))

All three indicators were expressed as percentages These data were obtainedfrom the managers responding the questionnaire Their responses were carefullycontrasted with published regnancial information (eg Reuters Insurance Brieregng)

Market orientation was measured using the Market Orientation Scale-Revised(MOS-R) This scale is a shortened version of the MOS validated by Lado et al(1998a) in the population of insurance companies of Belgium and Spain Ladoet al (1998a) shortened the original MOS scale while extending the previousvalidation study to target all insurance companies operating in the EuropeanUnion In the Appendix we provide the 30 items composing the MOS-R Eachitem is to be rated on a ten-point Likert-type scale ranging from 0 (completedisagreement) to 10 (complete agreement)

Innovation degree and innovation performance were assessed by means ofmulti-item questionnaires akin to Miller and Friesenrsquos (1982) Innovationperformance was measured by a four-item questionnaire regarding the successof a new productservice (deregned as an improved product a product extensionor a new product line) introduced by the company The questions involved

Businesseconomic

performance

293

whether the new productservice had succeeded in meeting the sales growthmarket share and proregt objectives set up by the company

Innovation degree was assessed by a three-item questionnaire that inquiredthe rate of new productsservices introduced by the company relative tocompetitors the amount of new productsservices marketed by the companyover the past three years and the nature of change of the new productsservices

Finally we used a four-item questionnaire based on existing literature (egDick and Basu 1994 Javalgi and Moberg 1997) to evaluate managersrsquoperceptions of their customersrsquo loyalty The questionnaire taps on theproportion of their customersrsquo insurance premiums taken on by the companythe average time a customer remains in the companyrsquos portfolio the probabilityof a customer renewing a premium and the overall perception of the companycustomersrsquo loyalty

Scale scores for innovation degree innovation performance and customerloyalty were obtained as an unweighted sum of the corresponding items Sincein all three cases Likert-type items on a 0-7 scale were used scale scores forthese variables range from 0-27 0-27 and 0-28 respectively For marketorientation we computed a score for each of its facets as an unweighted sum ofthe corresponding items Then a global market orientation score was obtainedas a sum of the facetsrsquo scores inversely weighted by their number of itemsHence this market orientation score assigns equal weights to each its facetsand ranges from 0-90

The scalesrsquo reliability (as assessed by coefregcient alpha) in this sample were088 (market orientation) 070 (innovation degree) 091 (innovationperformance) and 076 (customer loyalty) The means standard deviationsand correlations among all variables considered in this study are presented inTable I As can be seen in this table the three indicators of business economicperformance are signiregcantly but not largely correlated (the correlations rangefrom 020 to 029) The correlations among the hypothesized intermediatevariables (innovation degree innovation performance and customer loyalty)are not high except for innovation degree and innovation performance whichshare 36 percent of their variance The correlations of market orientation withthe intermediate variables appear signiregcantly larger (they range from 055 to058) than with the dependent variables (they range from 023 to 036) We alsoobserve in Table I that managers report on average a high degree of innovationin their businesses not so high a level of customer loyalty and a level ofinnovation performance just at the scale mean The average self-reporteddegree of market orientation is 56 on a 0-90 scale

MethodAll hypotheses were contrasted using covariance structure analysis asimplemented in LISREL 850 (see Joreskog et al 1999) Since all three indicatorsof business performance are highly positively skewed and present a high

IJSIM143

294

degree of kurtosis throughout this paper rather than attempting to transformthese variables to near-normality we shall employ an estimation approach thatis robust to non-normality of the observed variables The parameter estimateswere obtained using maximum likelihood estimation with standard errorssuitable for non-normal data (Joreskog et al 1999 Equation A24) and twotest statistics were used to assess the goodness of regt (GFI) of the model theSatorra-Bentler scaled chi-squared statistic (Satorra and Bentler 1988Equation 41) and Brownersquos (1984 Equation 220a) chi-squared statisticcorrected for non-normality To evaluate better the goodness of regt of thismodel several additional indices will also provided the root mean squarederror of approximation (RMSEA (Steiger 1990)) the standardized root meansquared residual (SRMSR (Joreskog et al 1999)) the GFI (Tanaka and Huba1985) and the Comparative Fit Index (CFI) using the independence model asbaseline (Bentler 1990 see also McDonald and Marsh 1990) Adequate to goodregt is suggested by RMSEA and SRMSR values approaching 005 For the GFIand the CFI indices values between 090 and 100 indicate adequate to excellentregt (but see Hu and Bentler 1999)

ResultsH1The model used to estimate the effects of market orientation on insurancebusinessesrsquo performance consists of a latent variable representing economicperformance with three indicators (market share premium growth andproregtability) and a single exogenous variable (market orientation) This model

MST PG PROF INNODR INNPERF LOYAL MO

MST 100PG 020 100PROF 023 029 100INNODR 030 034 035 1000INNPERF 027 032 041 062 100LOYAL 025 019 038 040 038 100MO 024 030 036 055 058 057 100x 455 788 626 1825 1422 1996 5628Std 535 749 571 356 292 520 1331

NotesN = 122 all correlations are signiregcant (a = 001) except those marked by which are onlysigniregcant at an a = 005MS = market share PG = premium growth PROF = proregtability INNODR = innovation degreeINNPERF = innovation performance LOYAL = customer loyalty MO = market orientation

Table IMeans standard

deviations andinter-correlations

Businesseconomic

performance

295

is depicted in Figure 2 The parameter estimates and GFIs for this model aregiven in Table II The model shows a good regt although note that it only hastwo degrees of freedom According to the model the best objective indicator ofbusiness economic performance is proregtability per year over 34 percent of itsvariance is accounted for by the model The standardized regressioncoefregcients reveal that proregtability per year is the best objective indicator of

Parameter estimates Goodness of regt R 2

Parameter Value Index Value Variable Value

b1 006 [061] (002) MFF X 2001 ( p = 099) Market share 015

b2 166 [039] (049) S-B X 2001 ( p = 099) Premium growth 024

b3 293 [049] (078) B X 2001 ( p = 099) Proregtability 034

b4 267 [059] (084) RMSEA 001 Business performance 037u1 17713 [100] (2241) df 2u2 2427 [085] (713) SRMSR 001u3 4248 [076] (1364) GFI 100u4 2141 [066] (515) CFI 100

NotesRobust asymptotic standard errors are provided in parentheses standardized parameterestimates are provided in square bracketsMFF X 2 = Minimum regt function chi-square SB X 2 = Satorra-Bentler scaled chi-squareB X 2 = Brownersquos chi-square corrected for non-normality RMSEA = root mean squared error ofapproximation SRMSR = standardized root mean squared residual GFI = goodness of regt indexCFI = comparative regt index R 2 = squared multiple correlations for endogenous variables

Table IIEstimation resultsfor the modeldepicted in Figure 2

Figure 2Market orientation aspredictor of economicperformance

IJSIM143

296

overall business performance Finally according to the model almost 37percent of overall business economic performance is accounted for by thedegree of market orientation

An inspection of the total effects of market orientation on the indicators ofeconomic performance suggests that unit increments of market orientation asmeasured by the MOS-R are associated with 0095 0168 and 0153 incrementsin domestic market share premium growth and proregtability per year averagedover the last three years respectively

H2A mediated model for the relationship between market orientation andbusiness performance is depicted in Figure 3 In this context a mediating effectis said to exist when

both mediating paths b5 b6 are signiregcant and

the direct effect of the exogenous variable on the outcome variablevanishes (complete mediational effect) or is signiregcantly lower (partialmediational effect) when a mediator variable is introduced in the model

Condition (2) amounts to b1 in Figure 3 becoming zero or signiregcantly less thatthan value reported for Figure 2

We used the mediated model depicted in Figure 3 to test for mediatingeffects of innovation degree innovation performance and customer loyaltyseparately on the impact of market orientation on business economicperformance We found that when either innovation performance or innovation

Figure 3Mediated model

Businesseconomic

performance

297

degree were used as mediating variable all the mediating paths weresigniregcant and that direct path from market orientation to businessperformance was not signiregcantly different from zero b1 = 004 t = 162for innovation performance b1 = 004 t = 186 for innovation degree Hencetaken separately both innovation degree and innovation performancecompletely mediate the impact of market orientation on businessperformance After regxing b1 at zero we re-estimated these two mediationalmodels The resulting parameter estimates and GFIs for these two models areshown in Table III On the other hand customer loyalty was found not to havea mediational effect between market orientation and business performanceThe parameter estimates and GFIs for this model are also given in Table III

As can be seen in Table III the mediating paths are signiregcant but thedirect path b1 is signiregcantly different from zero at a = 001 Furthermore a 99percent conregdence interval for the value for b1 reported in Table II (003 009)includes the value of b1 estimated in the mediational model using customerloyalty 004 Hence this variable does not even partially mediate on the impactof market orientation on business economic performance The standardizeddirect impact of market orientation on business performance (0408) is morethan twice the standardized impact of market orientation conveyed throughcustomer loyalty (0191)

The percentage of variance of business economic performance explained bythe model when innovation performance innovation degree or customer loyaltyare used as mediators is very similar (465 percent 453 percent and 437percent respectively)

H3The full model to be regtted corresponding to the hypothesis depicted in Figure 1is presented in Figure 3 The parameter estimates and goodness of regt testcorresponding to this model are given in Table IV (see also Figure 4) As can beseen in this Table the model regts these data very well

All the postulated relationships were found to be signiregcant at an a = 001Lagrange multiplier tests indicated that the regt of the model would notsigniregcantly improve by

adding a direct effect of market orientation to business performance nor

adding a direct effect of innovation degree on economic performance

Result (1) is in accordance with the results discussed above where we saw thatinnovation degree and innovation performance even when taken separatelycompletely mediate the impact of market orientation on business performanceResult (2) conregrms our hypothesis that innovation performance completelymediates the impact of innovation degree on business performance

Given that all effects of market orientation on business performance gothrough either innovation degree-innovation performance or through customer

IJSIM143

298

Par

amet

eres

tim

ates

Goo

dnes

sof

regt

Par

amet

erV

alue

Par

amet

erV

alue

Index

Val

ue

Innova

tion

per

form

ance

as

med

iato

rb 1

0(reg

xed

)u

117

713

[10

0](2

241

)M

FF

X2

560

(p=

035

)b 2

155

[04

0](0

42)

u2

241

4[0

84]

(71

1)S-B

X2

380

(p=

058

)b 3

263

[04

8](0

67)

u3

431

6[0

77]

(45

5)B

X2

429

(p=

051

)b 4

248

[05

9](o

81)

u4

211

3[0

65]

(45

5)df

5R

MSE

A0

03b 5

023

[05

8](0

03)

u5

180

3[0

67]

(28

0)SR

MSR

005

GF

I0

98b 6

018

[06

8](0

05)

CF

I0

99

Innova

tion

deg

ree

as

med

iato

rb

10

(regxed

)u

117

713

[10

0](2

241

)M

FF

X2

717

(p=

021

)b 2

169

[05

0](0

36)

u2

234

0[0

82]

(73

8)S-B

X2

497

(p=

043

)b 3

282

[05

1](0

82)

u3

415

6[0

74]

(45

0)B

X2

613

(p=

029

)b 4

226

[05

4](0

79)

u4

232

9[0

71]

(46

0)df

5R

MSE

A0

06b 5

012

[05

5](0

02)

u5

590

[06

9](0

78)

SR

MSR

005

GF

I0

98b 6

031

[06

7](0

09)

CF

I0

98

Cust

om

erlo

yalty

as

med

iato

rb 1

004

[04

1](0

02)

u1

177

13[1

00]

(22

41)

MF

FX

22

30(p

=0

68)

b 21

61[0

40]

(05

1)u

224

05

[08

4](7

11)

S-B

X2

155

(p=

082

)b 3

252

[04

5](0

74)

u3

447

5[0

80]

(45

5)B

X2

309

(p=

054

)b 4

265

[06

2](0

92)

u4

201

3[0

62]

(45

5)R

MSE

A0

0df

4b 5

015

[05

7](0

02)

u5

863

[06

8](1

04)

SR

MSR

002

GF

I0

99b 6

013

[03

4](0

05)

CF

I1

00

Note

s

Rob

ust

asym

pto

tic

stan

dar

der

rors

are

pro

vid

edin

par

enth

eses

st

andar

diz

edpar

amet

eres

tim

ates

are

pro

vid

edin

squar

ebra

cket

sM

FF

X2=

Min

imum

regt

funct

ion

chi-sq

uar

eSB

X2=

Sat

orra

-Ben

tler

scal

edch

i-sq

uar

eB

X2=

Bro

wnersquo

sch

i-sq

uar

eco

rrec

ted

for

non

-nor

mal

ity

RM

SE

A=

root

mea

nsq

uar

eder

ror

ofap

pro

xim

atio

n

SR

MSR

=st

andar

diz

edro

otm

ean

squar

edre

sidual

G

FI

=goo

dnes

sof

regt

index

C

FI

=co

mpar

ativ

ereg

tin

dex

Table IIIEstimation results

for the modeldepicted in Figure 3

Businesseconomic

performance

299

Par

amet

eres

tim

ates

Goo

dnes

sof

regt

R2

Par

amet

erV

alue

Par

amet

erV

alue

Index

Val

ue

Var

iable

Val

ue

b 10

12[0

55]

(22

41)

u1

177

13[1

00]

(00

2)M

FF

X2

105

9(p

=0

56)

Mar

ket

shar

e0

16b 2

013

[03

4](0

04)

u2

590

[06

9](0

78)

S-B

X2

785

(p=

080

)P

rem

ium

gro

wth

019

b 30

15[0

57]

(00

2)u

314

48

[05

4](2

59)

BX

212

67

(p=

039

)P

roreg

tabilit

y0

38b 4

078

[04

4](0

15)

u4

863

[06

8](1

04)

df

12B

usi

nes

sper

form

ance

056

RM

SE

A0

0b 5

016

[05

4](0

05)

u5

239

1[0

84]

(72

3)SR

MSR

005

Innov

atio

nper

form

ance

046

b 60

16[0

37]

(00

7)u

645

09

[08

1](1

271

)G

FI

098

Innov

atio

ndeg

ree

031

b 71

42[0

40]

(04

5)u

720

12

[06

2](4

60)

CF

I1

00C

ust

omer

loyal

ty0

32b 8

217

[04

4](0

64)

b 92

32[0

62]

(08

5)

Note

sR

obust

asym

pto

tic

stan

dar

der

rors

are

pro

vid

edin

par

enth

eses

st

andar

diz

edpar

amet

eres

tim

ates

are

pro

vid

edin

squar

ebra

cket

sM

FF

X2=

Min

imum

regtfu

nct

ion

chi-sq

uar

eSB

X2=

Sat

orra

-Ben

tler

scal

edch

i-sq

uar

eB

X2=

Bro

wnersquo

sch

i-sq

uar

eco

rrec

ted

for

non

-nor

mal

ity

RM

SE

A=

root

mea

nsq

uar

eder

ror

ofap

pro

xim

atio

n

SR

MSR

=st

andar

diz

edro

otm

ean

squar

edre

sidual

G

FI

=goo

dnes

sof

regt

index

C

FI

=co

mpar

ativ

ereg

tin

dex

R

2=

squar

edm

ult

iple

corr

elat

ions

for

endog

enou

svar

iable

s

Table IVEstimation resultsfor the modeldepicted in Figure 4

IJSIM143

300

loyalty a question arises as to the relative importance of the speciregc effectsgoing through these variables The standardized speciregc effect (computed as inBollen 1987) going through innovation degree and innovation performance is031 and 021 going through customer loyalty Hence the impact of marketorientation going through innovation is 50 percent more than that goingthrough customer loyalty

We can also see in Table IV that over 30 percent of the variance of theintermediate variables (innovation degree innovation performance andcustomer loyalty) are explained by market orientation In fact almost 50percent (464 percent to be exact) of innovation performance is explained bymarket orientation Furthermore note that the percentage of variance ofbusiness performance explained by the model is 561 percent a 52 percentincrement over what is explained by market orientation alone (see Table II) andover a 20 percent increment over what is explained by the mediational modelsconsidered previously Hence the inclusion of all three intermediate variables inthe model improves considerably our prediction of business performance

Furthermore we observe in Table IV that the direct effect of marketorientation on all three intermediate variables appear to be equal Also thedirect effects of customer loyalty and of innovation performance on economicperformance appear to be equal We re-estimated the model to test theseconstraints obtaining b1 = b2 = b3 = 0133 b5 = b6 = 0157 Satorra-BentlerX 2(15) = 8849 p = 0885 CFI = 100 GFI = 0975 SRMR = 006

ConclusionsMarket orientation can be deregned as a strategy used to reach a sustainablecompetitive advantage based on the generation and use of information within

Figure 4Full model

Businesseconomic

performance

301

organizations and on the selection of markets to be satisreged In thisframework we believe that competitive advantage results from the use ofresources and capabilities to generate differential satisfaction in proregtablemarkets Sustainability is achieved because the performance of the marketorientationrsquos behaviors requires complex organizational knowledge that cannoteasily be imitated by competitors Thus we hypothesize that the satisfaction ofproregtable markets permits the regrm to achieve a psychologically differentialposition that leads to brand loyalty and thus to higher proregts Previous studieshave

found a clear impact of market orientation on economic performance

assessed the effects of market orientation on innovation and

investigated the relationship between market orientation and relationshipmarketing variables

The present study proposes and tests a model that integrates both streams ofresearch innovation and relationship marketing More research is needed toinvestigate the role of customer satisfaction within this framework

In our necessarily partial model innovation degree innovation performanceand customer loyalty are used as intermediate variables on the effect of marketorientation on business performance Our results suggest that the addition ofthese variables help improve our predictions of business economic performance52 percent over what is explained by market orientation alone Also we foundthat innovation degree and innovation performance each taken separatelycompletely mediate the effect of market orientation on economic performanceFurthermore the impact of innovation degree on economic performance iscompletely channeled through innovation performance Customer loyalty byitself does not meditate the impact of market orientation on economicperformance but when considered along with innovation degree andinnovation performance it conveys some of the effects of market orientationon business performance This seemingly contradictory result arises from thefact that all three intermediate variables are interrelated

Our results should not be taken to imply that there are no other variablesmediate the effect of market orientation on economic performance We believethat other variables that have not been taken into account in this study such asproduct quality and customer satisfaction may also be signiregcant mediatorsHowever our results do suggest that whenever innovation degree andinnovation performance are included in the model as intermediate variablesthe effects of market orientation on business performance will mostly beconveyed through these variables Also in this study we have adopted thecurrently most widely accepted approach to measuring customer loyalty whichis based on behavioral loyalty measures (eg share of category requirementsrenew the policy probability the average customer last in the companyportfolio) It would be worth extending the present study by including not only

IJSIM143

302

behavioral measures of customer loyalty but also attitudinal measures alongthe lines of Dick and Basu (1994)

In our opinion two important contributions of the present research are outuse of objective measures of business performance and our focusing oninternational markets Despite the growing role of globalization and marketintegration and despite the increasing internationalization of corporationsmost studies on market orientation have focused on domestic markets (withnotable exceptions such as Selnes et al 1996 Webster 1994) Similarly moststudies on product innovation have also focused on domestic markets There isa lack of research yielding empirical support to the validity in an internationalsetting to research results obtained in domestic markets To regll this gap wetargeted the European Union market

Our study focused on a single industry the insurance sector Our sampleaccounted for 22 percent of the companies and 17 percent of the insurancepremiums in the targeted market An advantage of our single-industryapproach is that (with obvious reservations arising from the non-experimentalnature of our study and the fact that our sample should not be considered tohave been obtained at random) we can draw tentative predictions from ourmodel concerning the impact of market orientation on economic performance ininsurance companies operating in the European Union market An evidentdrawback of the single-industry approach adopted here is that it is not clearhow the present results extrapolate to other industries even when operating inthe same market

We have found that within the European Union insurance regrms thatconstantly monitor the evolution of current and potential customerrsquos needsmodify the attributes of the products to adapt them to the distributorsrequirements analyze competitorsrsquo marketing policy and products and knowbest of environmental trends especially technological and legal changes aremore likely to develop more new products have their new products accepted bythe market and obtain more loyal customers In turn increased customerloyalty and increased new product success will result in improved economicperformance However these conclusions must be taken with some caution aswe have used overall measures of innovation and innovation performanceFurther research is needed that takes into account the various aspects thatconstitute innovation

Previous studies have concluded that insurance companies still seethemselves as being product-focused and the industry as a whole is generallydistribution driven (Sodano 2000 Lambin 1996) This lack of closeness withcustomers has in turn contributed to the industry lack of product innovationsand product portfolios of commodity products that only compete on price Nowthe most successful regrms are redirecting their focus to the market needs andthey are beginning to exploit customer data and use market research togenerate ideas for designing new products In this context these companies

Businesseconomic

performance

303

now face choosing among too many new service options The managerialimplication of our study is that by enhancing their market orientation regrmswill know and service its customers better Thus they will generate moreinnovations by adopting a market-based product development process Alsoincreasing levels of market orientation enable regrms to discriminate more easilywhich new products have a higher success probability thus enhancing both theefregcacy and efregciency of new product development

Note

1 We considered employing the total volume of premiums for each insurance company as anadditional indicator although it seemed to us to be a better indicator of a companyrsquos sizerather than of its performance As we suspected the total volume of premiums wasuncorrelated with any of the variables considered in this study except for the companyrsquosmarket share (r = 024 p 001) Hence it is clear that volume of premiums should not beused as an indicator of insurance companiesrsquo performance

References

Armstrong JS and Overton T (1977) ordfEstimating nonresponse bias in mail surveysordm Journalof Marketing Research Vol 14 pp 396-402

Atuahene-Gima K (1996) ordfMarket orientation and innovationordm Journal of Business ResearchVol 35 pp 93-103

Baker TL Simpson PM and Sigauw JA (1999) ordfThe impact of suppliersrsquo perceptions ofreseller market orientation on key relationship constructsordm Journal of the Academy ofMarketing Science Vol 27 No 1 pp 50-7

Baron RM and Kenny DA (1986) ordfThe moderator-mediator variable distinction in socialpsychological research conceptual strategic and statistical considerationsordm Journal ofPersonality and Social Psychology Vol 51 pp 1173-82

Bentler PM (1990) ordfComparative regt indexes in structural modelsordm Psychological BulletinVol 107 pp 238-46

Bhuian SN (1997) ordfExploring market orientation in banks an empirical examination in SaudiArabiaordm The Journal of Service Marketing Vol 11 No 5 pp 317-28

Bollen KA (1987) ordfTotal direct and indirect effects in structural equation modelsordm inClogg CC (Ed) Sociological Methodology American Sociological AssociationWashington DC pp 37-69

Browne MW (1984) ordfAsymptotically distribution free methods for the analysis of covariancestructuresordm British Journal of Mathematical and Statistical Psychology Vol 37 pp 62-83

Calantone RJ di Benedetto AC and Bhoovaraghavan S (1994) ordfExamining the relationshipbetween degree of innovation and new product successordm Journal of Business ResearchVol 30 pp 143-8

Calantone RJ Schmidt JB and Song MX (1996) ordfControllable factors of new productssuccess a cross-national comparisonordm Marketing Science Vol 15 No 4 pp 341-58

Cooper RC (1994) ordfNew product the factors that drive successordm International MarketingReview Vol 11 No 1 pp 60-76

Day G (1992) ordfMarketingrsquos contribution to the strategy dialogueordm Journal of the Academy ofMarketing Science Vol 20 Fall pp 323-9

IJSIM143

304

afregrm in a recent article ordfit is possible that the strategic orientation of the regrmleads to at least in part superior performance because of the innovation thatare brought to market Although being market-oriented may lead to generalbeneregts for the regrmrsquos marketing activities the ability to bring to market newproducts which present the characteristics necessary to be successful may becriticalordm

Market orientation may also be an important determinant of innovation inthe services sector According to Atuahene-Gima (1996) in services like theinsurance and banking industries innovation success depends on the regrmrsquosmarket orientation especially on its customer orientation Being in touch withyour clients wants and needs and being able to respond appropriately to themis a key to innovation success in the service sector Furthermore the marketenvironment in the service sector is likely to be more competitive in terms ofproduct innovation than in other industries Innovation in services is moreeasily and quickly imitated (Tufano 1992) and more difregcult to protect bymeans of patenting Thus it may be than in this sector the relationshipbetween market orientation innovation and business performance isparticularly strong

The role of innovation performanceIn many instances new products arise from the coordination betweenmarketing and other business units such as RampD Also competitorsrsquomonitoring and a close relationship with distributors are key elements to thegeneration of new concepts for new product development As these areremacrected in the market orientation facets of regnal client analysis andenvironmental analysis one should expect a direct link from market orientationto new product performance

We regnd support for this hypothesis in the literature (eg Ottum and Moore1997 Slater and Narver 1994) Also in a meta-analysis on the determinantsof new product success Montoya-Weiss and Calantone (1994) identifymarket-related activities as one of the four more important factors thatdiscriminate between a new product success or failure Successful regrmsdevelop superior products that are attuned to customer wants and needs andthey also have strong marketing knowledge and skills to develop and launchthe product (Calantone et al 1996) As Cooper (1994 p 64) concluded insummarizing the results of new products research ordfa strong Market orientationis critical both to success and cycle time reductionordm

Innovation degree innovation performance and business performance areall linked together Calantone et al (1994) have investigated whether the sheervolume of innovation engaged in by the regrm determines the level of newproduct success Their regndings suggest that the degree of innovation of a regrmsis related to its new product performance Hence regrms that attempt to bringout more innovations may be more likely to succeed Similarly recent research

IJSIM143

288

shows that increased levels of innovation are associated to superiorperformance (Robinson et al 1992 DeshpandeAcirc et al 1993)

Market orientation and customer loyaltyDeshpandeAcirc et al (1993 p 24) point out that ordfthe canons of the marketingconcept assert that proregt is a reward for customer orientation which creates asatisreged customer but we have only the beginning of systematic empiricaldocumentation of the presumed relationshipordm In the present competitivemarket environment characterized by globalization with rapid market entry ofnew products and maturity conditions in many products and servicesattaining a high level of customer loyalty has emerged as a central managerialconcern Clearly customer loyalty constitutes an important objective forstrategic marketing planning (Kotler 1984) and represents an important basisfor developing a sustainable competitive advantage plusmn an advantage that canbe realized through market orientation A high degree of market orientationleads to customer loyalty which in the long run contributes to better economicperformance In the service sector the intangible nature of services gives rise toinformationrsquos asymmetry between buyers and sellers This results in higherrisk perceptions and greater difregculty in customerrsquos quality evaluation(Nayyar 1990) As a result market orientation becomes a crucial instrument toestablish long-term relations with customers in service regrms Kohli andJaworski (1990) posited a positive relationship between a regrmrsquos marketorientation level and customer satisfaction Webb et al (2000) and Lai (2003)provide further empirical support to this relationship Jones et al (2002)empirical research on business-to business buyer-seller relationship suggeststhat a strong salespersonrsquos customer orientation tends to reduce the customerrsquospropensity to switch suppliers Harrison-Walker (2001) found a positiverelationship between market orientation and customer retention customerwillingness to pay a price premium customer propensity to spread positiveword-of-mouth communication and customer propensity to not-alternateamong brandsservice providers All these variables are widely used measuresof the customer loyalty concept (Jacoby and Chestnut 1978 Odin et al 2001Dick and Basu 1994)

On the other hand customer loyalty is expected to have a positive impact onbusiness economic performance since market-oriented regrms have a largenumber of satisreged customer and therefore a higher rate of repeated purchases(Dick and Basu 1994 McCullough et al 1986 Loveman 1998 Kamakura et al2002)

The arguments put forth in the preceding sections can be summarized into aseries of hypotheses Our initial hypotheses is

H1 Within an industry the more market oriented regrms are the better theirobjective economic performance

Businesseconomic

performance

289

If this hypotheses is tenable then we shall investigate what is role ofinnovation degree innovation performance and customer loyalty in relation tothe hypothesized impact of market orientation on economic performance Wehypothesize that each of these variables taken separately is an intermediatemediational variable According to Baron and Kenny (1986) an intermediatevariable is said to be a mediator if when introduced within a directedrelationship the directed relationship vanishes (complete mediational effect) orat least it signiregcantly decreases (partial mediational effect)

Thus our hypotheses are

H2a Within an industry the impact of market orientation on economicperformance is at least partially mediated through innovation degreeThat is the more market oriented regrms are the higher their innovationdegree This higher innovation degree results in better economicperformance

H2b Within an industry the impact of market orientation on economicperformance is at least partially mediated through innovationperformance That is the more market oriented regrms are the highertheir innovation performance This higher innovation performanceresults in better economic performance

H2c Within an industry the impact of market orientation on economicperformance is at least partially mediated through customer loyaltyThat is the more market oriented regrms are the higher their customerloyalty This higher customer loyalty results in better economicperformance

If in turn these hypotheses are tenable then we need to specify a model thatintegrates these three intermediate variables within the hypothesized directedrelationship of market orientation on economic performance We hypothesizedthat

H3 Taken jointly innovation degree innovation performance andcustomer loyalty completely mediate the impact of marketorientation on economic performance Furthermore the relationshipbetween innovation degree and economic performance is all conveyedthrough innovation performance

This last hypothesis is graphically depicted in Figure 1

Empirical studyWith the objective of isolating within-industry variation we shall adopt asingle-industry approach focusing on the insurance industry This clearlyprevents the generalization of the results outside the scope of the industryconsidered On the other side we can meaningfully assess the impact of unitincrements in market orientation on regrmsrsquo economic performance and sound

IJSIM143

290

inferences can be drawn on the target population based on therepresentativeness of the sample used

The confounding of within-industry and between industry variation is notthe only threat to the validity of inferences drawn on the relationship betweenmarket orientation and economic performance A second threat is the noiseintroduced by environmental variables such as market turbulence marketgrowth rate buyer and supplier power and competitive intensity on businessperformance A standard approach to minimize this threat is to focus theresearch on a single market The drawback of this approach is that we are notable to capture regrmsrsquo behavior in facing increasing globalization and marketintegration As a compromise between these two ends the present studytargets the European Union market In this market the key characteristics of asingle market are preserved but it is also an environment in which we canpresently observe how regrms struggle in meeting the challenges ofinternationalization and market integration

A third threat to the validity of inferences drawn on the relationship betweenmarket orientation and economic performance lies in the use of subjectivemeasures of economic performance (ie managersrsquo evaluations of theircompaniesrsquo performance) Positive effects of market orientation on economicperformance have been reported when subjective assessments of performancewere used However when objective measures of economic performance havebeen used mixed results emerged For instance Ruekert (1992) and Lambin(1996) report a positive relationship between market orientation and objectivelymeasured economic performance However Bhuian (1997) Jaworski and Kohli(1993) and Selnes et al (1996) failed to regnd any signiregcant relationshipClearly when market orientation and economic performance are concurrentlyassessed by the regrmsrsquo managers a perceptual bias may be introduced A casein point Van Bruggen and Smidts (1995) found within one single company

Figure 1Hypothetical model

Businesseconomic

performance

291

(which has only one performance) a substantial degree of variation insubjective performance assessments In fact they report a positive relationshipbetween market orientation and judgments about the company performancewithin a single company As they point out ordfit might be that managers have amore positive view of their companyrsquos market orientation when they perceivetheir company to be performing wellordm (Van Bruggen and Smidts 1995 p 13)Hence it is important to employ objective measures of economic performanceand we shall do so in the present study

Market orientation in the services sector the European insurance industryThe insurance sector is of particular interest from a market orientationviewpoint as it works with intangible commodities in which service qualityand customer orientation are crucial elements The competitive characteristicsgenerated by the European Union provide an additional interest in studyingmarket orientation in this area The insurance sector in Europe has traditionallyoperated subject to strict regulations and strong protection from internationalcompetition However for some years now the European Commission has beenworking on the liberalization of this sector Effective implementation of this hasbrought about a major increase in competition within the sector and hasprovoked a major restructuring of insurance companies and groups Thecompetitive climate in Europe has also been inmacruenced by a downside in theeconomic cycle and changes in consumer behavior European customers nowshow greater service expectations and less loyalty As a result rivalry amongcompetitors is increasing as is the importance of competitive strategies adaptedto this sectorrsquos needs In this background the degree of orientation toward thecustomer distributors competition and the general socio-economicenvironment is becoming an increasingly important area of study not onlyfor academics but also for the business world

Lado et al (1998ab) have investigated quite extensively the marketorientation of insurance regrms within the European Union These authors havenot found signiregcant mean differences in market orientation by countryFurthermore they report substantial agreement between the factor structuresof market orientation across countries Thus it seems that the Europeaninsurance sector can be considered a homogenous population with respect tomarket orientation

DataThe population universe considered in this article is deregned as the set ofinsurance companies operating in the European Union which meet thefollowing conditions

they operate in private insurance or ordfmass insuranceordm

they have a market share of more than 005 percent and

their management is independent

IJSIM143

292

The list of European insurance companies was taken from the Financial TimesYearbook for 1996

It was assumed that senior executives were the people best qualireged to assessthe companyrsquos market orientation as well as their innovation degree innovationperformance and customersrsquo loyalty Therefore information from these variableswas gathered via a postal questionnaire submitted to the senior executive in eachof the 554 companies comprising the target population

We obtained 122 valid questionnaires giving a response rate of 22 percentThis sample accounts for over 17 percent of total insurance premiums in theEuropean Union

In order to assess response bias the questionnaires were divided intoquartiles on the basis of reception date (Armstrong and Overton 1977) Ananalysis of early and late responses did not indicate any signiregcant differencein terms of means and covariances

MeasuresBusiness economic performance is a complex construct with multiple possibleobserved indicators Here we measure this construct using three remacrectiveindicators[1]

(1) domestic market share (Jaworski and Kohli 1993 DeshpandeAcirc et al 1993Selnes et al 1996 Greenley and Foxall 1997 1998)

(2) premium growth (which is equivalent to sales growth for the insurancecompanies business (Slater and Narver 1994 Narver and Slater 1990Ruekert 1992 Greenley and Foxall 1997 1998) and

(3) proregtability per year averaged over the last three years(similar to thereturn on investment (ROI) rate (Greenley and Foxall 1997 1998))

All three indicators were expressed as percentages These data were obtainedfrom the managers responding the questionnaire Their responses were carefullycontrasted with published regnancial information (eg Reuters Insurance Brieregng)

Market orientation was measured using the Market Orientation Scale-Revised(MOS-R) This scale is a shortened version of the MOS validated by Lado et al(1998a) in the population of insurance companies of Belgium and Spain Ladoet al (1998a) shortened the original MOS scale while extending the previousvalidation study to target all insurance companies operating in the EuropeanUnion In the Appendix we provide the 30 items composing the MOS-R Eachitem is to be rated on a ten-point Likert-type scale ranging from 0 (completedisagreement) to 10 (complete agreement)

Innovation degree and innovation performance were assessed by means ofmulti-item questionnaires akin to Miller and Friesenrsquos (1982) Innovationperformance was measured by a four-item questionnaire regarding the successof a new productservice (deregned as an improved product a product extensionor a new product line) introduced by the company The questions involved

Businesseconomic

performance

293

whether the new productservice had succeeded in meeting the sales growthmarket share and proregt objectives set up by the company

Innovation degree was assessed by a three-item questionnaire that inquiredthe rate of new productsservices introduced by the company relative tocompetitors the amount of new productsservices marketed by the companyover the past three years and the nature of change of the new productsservices

Finally we used a four-item questionnaire based on existing literature (egDick and Basu 1994 Javalgi and Moberg 1997) to evaluate managersrsquoperceptions of their customersrsquo loyalty The questionnaire taps on theproportion of their customersrsquo insurance premiums taken on by the companythe average time a customer remains in the companyrsquos portfolio the probabilityof a customer renewing a premium and the overall perception of the companycustomersrsquo loyalty

Scale scores for innovation degree innovation performance and customerloyalty were obtained as an unweighted sum of the corresponding items Sincein all three cases Likert-type items on a 0-7 scale were used scale scores forthese variables range from 0-27 0-27 and 0-28 respectively For marketorientation we computed a score for each of its facets as an unweighted sum ofthe corresponding items Then a global market orientation score was obtainedas a sum of the facetsrsquo scores inversely weighted by their number of itemsHence this market orientation score assigns equal weights to each its facetsand ranges from 0-90

The scalesrsquo reliability (as assessed by coefregcient alpha) in this sample were088 (market orientation) 070 (innovation degree) 091 (innovationperformance) and 076 (customer loyalty) The means standard deviationsand correlations among all variables considered in this study are presented inTable I As can be seen in this table the three indicators of business economicperformance are signiregcantly but not largely correlated (the correlations rangefrom 020 to 029) The correlations among the hypothesized intermediatevariables (innovation degree innovation performance and customer loyalty)are not high except for innovation degree and innovation performance whichshare 36 percent of their variance The correlations of market orientation withthe intermediate variables appear signiregcantly larger (they range from 055 to058) than with the dependent variables (they range from 023 to 036) We alsoobserve in Table I that managers report on average a high degree of innovationin their businesses not so high a level of customer loyalty and a level ofinnovation performance just at the scale mean The average self-reporteddegree of market orientation is 56 on a 0-90 scale

MethodAll hypotheses were contrasted using covariance structure analysis asimplemented in LISREL 850 (see Joreskog et al 1999) Since all three indicatorsof business performance are highly positively skewed and present a high

IJSIM143

294

degree of kurtosis throughout this paper rather than attempting to transformthese variables to near-normality we shall employ an estimation approach thatis robust to non-normality of the observed variables The parameter estimateswere obtained using maximum likelihood estimation with standard errorssuitable for non-normal data (Joreskog et al 1999 Equation A24) and twotest statistics were used to assess the goodness of regt (GFI) of the model theSatorra-Bentler scaled chi-squared statistic (Satorra and Bentler 1988Equation 41) and Brownersquos (1984 Equation 220a) chi-squared statisticcorrected for non-normality To evaluate better the goodness of regt of thismodel several additional indices will also provided the root mean squarederror of approximation (RMSEA (Steiger 1990)) the standardized root meansquared residual (SRMSR (Joreskog et al 1999)) the GFI (Tanaka and Huba1985) and the Comparative Fit Index (CFI) using the independence model asbaseline (Bentler 1990 see also McDonald and Marsh 1990) Adequate to goodregt is suggested by RMSEA and SRMSR values approaching 005 For the GFIand the CFI indices values between 090 and 100 indicate adequate to excellentregt (but see Hu and Bentler 1999)

ResultsH1The model used to estimate the effects of market orientation on insurancebusinessesrsquo performance consists of a latent variable representing economicperformance with three indicators (market share premium growth andproregtability) and a single exogenous variable (market orientation) This model

MST PG PROF INNODR INNPERF LOYAL MO

MST 100PG 020 100PROF 023 029 100INNODR 030 034 035 1000INNPERF 027 032 041 062 100LOYAL 025 019 038 040 038 100MO 024 030 036 055 058 057 100x 455 788 626 1825 1422 1996 5628Std 535 749 571 356 292 520 1331

NotesN = 122 all correlations are signiregcant (a = 001) except those marked by which are onlysigniregcant at an a = 005MS = market share PG = premium growth PROF = proregtability INNODR = innovation degreeINNPERF = innovation performance LOYAL = customer loyalty MO = market orientation

Table IMeans standard

deviations andinter-correlations

Businesseconomic

performance

295

is depicted in Figure 2 The parameter estimates and GFIs for this model aregiven in Table II The model shows a good regt although note that it only hastwo degrees of freedom According to the model the best objective indicator ofbusiness economic performance is proregtability per year over 34 percent of itsvariance is accounted for by the model The standardized regressioncoefregcients reveal that proregtability per year is the best objective indicator of

Parameter estimates Goodness of regt R 2

Parameter Value Index Value Variable Value

b1 006 [061] (002) MFF X 2001 ( p = 099) Market share 015

b2 166 [039] (049) S-B X 2001 ( p = 099) Premium growth 024

b3 293 [049] (078) B X 2001 ( p = 099) Proregtability 034

b4 267 [059] (084) RMSEA 001 Business performance 037u1 17713 [100] (2241) df 2u2 2427 [085] (713) SRMSR 001u3 4248 [076] (1364) GFI 100u4 2141 [066] (515) CFI 100

NotesRobust asymptotic standard errors are provided in parentheses standardized parameterestimates are provided in square bracketsMFF X 2 = Minimum regt function chi-square SB X 2 = Satorra-Bentler scaled chi-squareB X 2 = Brownersquos chi-square corrected for non-normality RMSEA = root mean squared error ofapproximation SRMSR = standardized root mean squared residual GFI = goodness of regt indexCFI = comparative regt index R 2 = squared multiple correlations for endogenous variables

Table IIEstimation resultsfor the modeldepicted in Figure 2

Figure 2Market orientation aspredictor of economicperformance

IJSIM143

296

overall business performance Finally according to the model almost 37percent of overall business economic performance is accounted for by thedegree of market orientation

An inspection of the total effects of market orientation on the indicators ofeconomic performance suggests that unit increments of market orientation asmeasured by the MOS-R are associated with 0095 0168 and 0153 incrementsin domestic market share premium growth and proregtability per year averagedover the last three years respectively

H2A mediated model for the relationship between market orientation andbusiness performance is depicted in Figure 3 In this context a mediating effectis said to exist when

both mediating paths b5 b6 are signiregcant and

the direct effect of the exogenous variable on the outcome variablevanishes (complete mediational effect) or is signiregcantly lower (partialmediational effect) when a mediator variable is introduced in the model

Condition (2) amounts to b1 in Figure 3 becoming zero or signiregcantly less thatthan value reported for Figure 2

We used the mediated model depicted in Figure 3 to test for mediatingeffects of innovation degree innovation performance and customer loyaltyseparately on the impact of market orientation on business economicperformance We found that when either innovation performance or innovation

Figure 3Mediated model

Businesseconomic

performance

297

degree were used as mediating variable all the mediating paths weresigniregcant and that direct path from market orientation to businessperformance was not signiregcantly different from zero b1 = 004 t = 162for innovation performance b1 = 004 t = 186 for innovation degree Hencetaken separately both innovation degree and innovation performancecompletely mediate the impact of market orientation on businessperformance After regxing b1 at zero we re-estimated these two mediationalmodels The resulting parameter estimates and GFIs for these two models areshown in Table III On the other hand customer loyalty was found not to havea mediational effect between market orientation and business performanceThe parameter estimates and GFIs for this model are also given in Table III

As can be seen in Table III the mediating paths are signiregcant but thedirect path b1 is signiregcantly different from zero at a = 001 Furthermore a 99percent conregdence interval for the value for b1 reported in Table II (003 009)includes the value of b1 estimated in the mediational model using customerloyalty 004 Hence this variable does not even partially mediate on the impactof market orientation on business economic performance The standardizeddirect impact of market orientation on business performance (0408) is morethan twice the standardized impact of market orientation conveyed throughcustomer loyalty (0191)

The percentage of variance of business economic performance explained bythe model when innovation performance innovation degree or customer loyaltyare used as mediators is very similar (465 percent 453 percent and 437percent respectively)

H3The full model to be regtted corresponding to the hypothesis depicted in Figure 1is presented in Figure 3 The parameter estimates and goodness of regt testcorresponding to this model are given in Table IV (see also Figure 4) As can beseen in this Table the model regts these data very well

All the postulated relationships were found to be signiregcant at an a = 001Lagrange multiplier tests indicated that the regt of the model would notsigniregcantly improve by

adding a direct effect of market orientation to business performance nor

adding a direct effect of innovation degree on economic performance

Result (1) is in accordance with the results discussed above where we saw thatinnovation degree and innovation performance even when taken separatelycompletely mediate the impact of market orientation on business performanceResult (2) conregrms our hypothesis that innovation performance completelymediates the impact of innovation degree on business performance

Given that all effects of market orientation on business performance gothrough either innovation degree-innovation performance or through customer

IJSIM143

298

Par

amet

eres

tim

ates

Goo

dnes

sof

regt

Par

amet

erV

alue

Par

amet

erV

alue

Index

Val

ue

Innova

tion

per

form

ance

as

med

iato

rb 1

0(reg

xed

)u

117

713

[10

0](2

241

)M

FF

X2

560

(p=

035

)b 2

155

[04

0](0

42)

u2

241

4[0

84]

(71

1)S-B

X2

380

(p=

058

)b 3

263

[04

8](0

67)

u3

431

6[0

77]

(45

5)B

X2

429

(p=

051

)b 4

248

[05

9](o

81)

u4

211

3[0

65]

(45

5)df

5R

MSE

A0

03b 5

023

[05

8](0

03)

u5

180

3[0

67]

(28

0)SR

MSR

005

GF

I0

98b 6

018

[06

8](0

05)

CF

I0

99

Innova

tion

deg

ree

as

med

iato

rb

10

(regxed

)u

117

713

[10

0](2

241

)M

FF

X2

717

(p=

021

)b 2

169

[05

0](0

36)

u2

234

0[0

82]

(73

8)S-B

X2

497

(p=

043

)b 3

282

[05

1](0

82)

u3

415

6[0

74]

(45

0)B

X2

613

(p=

029

)b 4

226

[05

4](0

79)

u4

232

9[0

71]

(46

0)df

5R

MSE

A0

06b 5

012

[05

5](0

02)

u5

590

[06

9](0

78)

SR

MSR

005

GF

I0

98b 6

031

[06

7](0

09)

CF

I0

98

Cust

om

erlo

yalty

as

med

iato

rb 1

004

[04

1](0

02)

u1

177

13[1

00]

(22

41)

MF

FX

22

30(p

=0

68)

b 21

61[0

40]

(05

1)u

224

05

[08

4](7

11)

S-B

X2

155

(p=

082

)b 3

252

[04

5](0

74)

u3

447

5[0

80]

(45

5)B

X2

309

(p=

054

)b 4

265

[06

2](0

92)

u4

201

3[0

62]

(45

5)R

MSE

A0

0df

4b 5

015

[05

7](0

02)

u5

863

[06

8](1

04)

SR

MSR

002

GF

I0

99b 6

013

[03

4](0

05)

CF

I1

00

Note

s

Rob

ust

asym

pto

tic

stan

dar

der

rors

are

pro

vid

edin

par

enth

eses

st

andar

diz

edpar

amet

eres

tim

ates

are

pro

vid

edin

squar

ebra

cket

sM

FF

X2=

Min

imum

regt

funct

ion

chi-sq

uar

eSB

X2=

Sat

orra

-Ben

tler

scal

edch

i-sq

uar

eB

X2=

Bro

wnersquo

sch

i-sq

uar

eco

rrec

ted

for

non

-nor

mal

ity

RM

SE

A=

root

mea

nsq

uar

eder

ror

ofap

pro

xim

atio

n

SR

MSR

=st

andar

diz

edro

otm

ean

squar

edre

sidual

G

FI

=goo

dnes

sof

regt

index

C

FI

=co

mpar

ativ

ereg

tin

dex

Table IIIEstimation results

for the modeldepicted in Figure 3

Businesseconomic

performance

299

Par

amet

eres

tim

ates

Goo

dnes

sof

regt

R2

Par

amet

erV

alue

Par

amet

erV

alue

Index

Val

ue

Var

iable

Val

ue

b 10

12[0

55]

(22

41)

u1

177

13[1

00]

(00

2)M

FF

X2

105

9(p

=0

56)

Mar

ket

shar

e0

16b 2

013

[03

4](0

04)

u2

590

[06

9](0

78)

S-B

X2

785

(p=

080

)P

rem

ium

gro

wth

019

b 30

15[0

57]

(00

2)u

314

48

[05

4](2

59)

BX

212

67

(p=

039

)P

roreg

tabilit

y0

38b 4

078

[04

4](0

15)

u4

863

[06

8](1

04)

df

12B

usi

nes

sper

form

ance

056

RM

SE

A0

0b 5

016

[05

4](0

05)

u5

239

1[0

84]

(72

3)SR

MSR

005

Innov

atio

nper

form

ance

046

b 60

16[0

37]

(00

7)u

645

09

[08

1](1

271

)G

FI

098

Innov

atio

ndeg

ree

031

b 71

42[0

40]

(04

5)u

720

12

[06

2](4

60)

CF

I1

00C

ust

omer

loyal

ty0

32b 8

217

[04

4](0

64)

b 92

32[0

62]

(08

5)

Note

sR

obust

asym

pto

tic

stan

dar

der

rors

are

pro

vid

edin

par

enth

eses

st

andar

diz

edpar

amet

eres

tim

ates

are

pro

vid

edin

squar

ebra

cket

sM

FF

X2=

Min

imum

regtfu

nct

ion

chi-sq

uar

eSB

X2=

Sat

orra

-Ben

tler

scal

edch

i-sq

uar

eB

X2=

Bro

wnersquo

sch

i-sq

uar

eco

rrec

ted

for

non

-nor

mal

ity

RM

SE

A=

root

mea

nsq

uar

eder

ror

ofap

pro

xim

atio

n

SR

MSR

=st

andar

diz

edro

otm

ean

squar

edre

sidual

G

FI

=goo

dnes

sof

regt

index

C

FI

=co

mpar

ativ

ereg

tin

dex

R

2=

squar

edm

ult

iple

corr

elat

ions

for

endog

enou

svar

iable

s

Table IVEstimation resultsfor the modeldepicted in Figure 4

IJSIM143

300

loyalty a question arises as to the relative importance of the speciregc effectsgoing through these variables The standardized speciregc effect (computed as inBollen 1987) going through innovation degree and innovation performance is031 and 021 going through customer loyalty Hence the impact of marketorientation going through innovation is 50 percent more than that goingthrough customer loyalty

We can also see in Table IV that over 30 percent of the variance of theintermediate variables (innovation degree innovation performance andcustomer loyalty) are explained by market orientation In fact almost 50percent (464 percent to be exact) of innovation performance is explained bymarket orientation Furthermore note that the percentage of variance ofbusiness performance explained by the model is 561 percent a 52 percentincrement over what is explained by market orientation alone (see Table II) andover a 20 percent increment over what is explained by the mediational modelsconsidered previously Hence the inclusion of all three intermediate variables inthe model improves considerably our prediction of business performance

Furthermore we observe in Table IV that the direct effect of marketorientation on all three intermediate variables appear to be equal Also thedirect effects of customer loyalty and of innovation performance on economicperformance appear to be equal We re-estimated the model to test theseconstraints obtaining b1 = b2 = b3 = 0133 b5 = b6 = 0157 Satorra-BentlerX 2(15) = 8849 p = 0885 CFI = 100 GFI = 0975 SRMR = 006

ConclusionsMarket orientation can be deregned as a strategy used to reach a sustainablecompetitive advantage based on the generation and use of information within

Figure 4Full model

Businesseconomic

performance

301

organizations and on the selection of markets to be satisreged In thisframework we believe that competitive advantage results from the use ofresources and capabilities to generate differential satisfaction in proregtablemarkets Sustainability is achieved because the performance of the marketorientationrsquos behaviors requires complex organizational knowledge that cannoteasily be imitated by competitors Thus we hypothesize that the satisfaction ofproregtable markets permits the regrm to achieve a psychologically differentialposition that leads to brand loyalty and thus to higher proregts Previous studieshave

found a clear impact of market orientation on economic performance

assessed the effects of market orientation on innovation and

investigated the relationship between market orientation and relationshipmarketing variables

The present study proposes and tests a model that integrates both streams ofresearch innovation and relationship marketing More research is needed toinvestigate the role of customer satisfaction within this framework

In our necessarily partial model innovation degree innovation performanceand customer loyalty are used as intermediate variables on the effect of marketorientation on business performance Our results suggest that the addition ofthese variables help improve our predictions of business economic performance52 percent over what is explained by market orientation alone Also we foundthat innovation degree and innovation performance each taken separatelycompletely mediate the effect of market orientation on economic performanceFurthermore the impact of innovation degree on economic performance iscompletely channeled through innovation performance Customer loyalty byitself does not meditate the impact of market orientation on economicperformance but when considered along with innovation degree andinnovation performance it conveys some of the effects of market orientationon business performance This seemingly contradictory result arises from thefact that all three intermediate variables are interrelated

Our results should not be taken to imply that there are no other variablesmediate the effect of market orientation on economic performance We believethat other variables that have not been taken into account in this study such asproduct quality and customer satisfaction may also be signiregcant mediatorsHowever our results do suggest that whenever innovation degree andinnovation performance are included in the model as intermediate variablesthe effects of market orientation on business performance will mostly beconveyed through these variables Also in this study we have adopted thecurrently most widely accepted approach to measuring customer loyalty whichis based on behavioral loyalty measures (eg share of category requirementsrenew the policy probability the average customer last in the companyportfolio) It would be worth extending the present study by including not only

IJSIM143

302

behavioral measures of customer loyalty but also attitudinal measures alongthe lines of Dick and Basu (1994)

In our opinion two important contributions of the present research are outuse of objective measures of business performance and our focusing oninternational markets Despite the growing role of globalization and marketintegration and despite the increasing internationalization of corporationsmost studies on market orientation have focused on domestic markets (withnotable exceptions such as Selnes et al 1996 Webster 1994) Similarly moststudies on product innovation have also focused on domestic markets There isa lack of research yielding empirical support to the validity in an internationalsetting to research results obtained in domestic markets To regll this gap wetargeted the European Union market

Our study focused on a single industry the insurance sector Our sampleaccounted for 22 percent of the companies and 17 percent of the insurancepremiums in the targeted market An advantage of our single-industryapproach is that (with obvious reservations arising from the non-experimentalnature of our study and the fact that our sample should not be considered tohave been obtained at random) we can draw tentative predictions from ourmodel concerning the impact of market orientation on economic performance ininsurance companies operating in the European Union market An evidentdrawback of the single-industry approach adopted here is that it is not clearhow the present results extrapolate to other industries even when operating inthe same market

We have found that within the European Union insurance regrms thatconstantly monitor the evolution of current and potential customerrsquos needsmodify the attributes of the products to adapt them to the distributorsrequirements analyze competitorsrsquo marketing policy and products and knowbest of environmental trends especially technological and legal changes aremore likely to develop more new products have their new products accepted bythe market and obtain more loyal customers In turn increased customerloyalty and increased new product success will result in improved economicperformance However these conclusions must be taken with some caution aswe have used overall measures of innovation and innovation performanceFurther research is needed that takes into account the various aspects thatconstitute innovation

Previous studies have concluded that insurance companies still seethemselves as being product-focused and the industry as a whole is generallydistribution driven (Sodano 2000 Lambin 1996) This lack of closeness withcustomers has in turn contributed to the industry lack of product innovationsand product portfolios of commodity products that only compete on price Nowthe most successful regrms are redirecting their focus to the market needs andthey are beginning to exploit customer data and use market research togenerate ideas for designing new products In this context these companies

Businesseconomic

performance

303

now face choosing among too many new service options The managerialimplication of our study is that by enhancing their market orientation regrmswill know and service its customers better Thus they will generate moreinnovations by adopting a market-based product development process Alsoincreasing levels of market orientation enable regrms to discriminate more easilywhich new products have a higher success probability thus enhancing both theefregcacy and efregciency of new product development

Note

1 We considered employing the total volume of premiums for each insurance company as anadditional indicator although it seemed to us to be a better indicator of a companyrsquos sizerather than of its performance As we suspected the total volume of premiums wasuncorrelated with any of the variables considered in this study except for the companyrsquosmarket share (r = 024 p 001) Hence it is clear that volume of premiums should not beused as an indicator of insurance companiesrsquo performance

References

Armstrong JS and Overton T (1977) ordfEstimating nonresponse bias in mail surveysordm Journalof Marketing Research Vol 14 pp 396-402

Atuahene-Gima K (1996) ordfMarket orientation and innovationordm Journal of Business ResearchVol 35 pp 93-103

Baker TL Simpson PM and Sigauw JA (1999) ordfThe impact of suppliersrsquo perceptions ofreseller market orientation on key relationship constructsordm Journal of the Academy ofMarketing Science Vol 27 No 1 pp 50-7

Baron RM and Kenny DA (1986) ordfThe moderator-mediator variable distinction in socialpsychological research conceptual strategic and statistical considerationsordm Journal ofPersonality and Social Psychology Vol 51 pp 1173-82

Bentler PM (1990) ordfComparative regt indexes in structural modelsordm Psychological BulletinVol 107 pp 238-46

Bhuian SN (1997) ordfExploring market orientation in banks an empirical examination in SaudiArabiaordm The Journal of Service Marketing Vol 11 No 5 pp 317-28

Bollen KA (1987) ordfTotal direct and indirect effects in structural equation modelsordm inClogg CC (Ed) Sociological Methodology American Sociological AssociationWashington DC pp 37-69

Browne MW (1984) ordfAsymptotically distribution free methods for the analysis of covariancestructuresordm British Journal of Mathematical and Statistical Psychology Vol 37 pp 62-83

Calantone RJ di Benedetto AC and Bhoovaraghavan S (1994) ordfExamining the relationshipbetween degree of innovation and new product successordm Journal of Business ResearchVol 30 pp 143-8

Calantone RJ Schmidt JB and Song MX (1996) ordfControllable factors of new productssuccess a cross-national comparisonordm Marketing Science Vol 15 No 4 pp 341-58

Cooper RC (1994) ordfNew product the factors that drive successordm International MarketingReview Vol 11 No 1 pp 60-76

Day G (1992) ordfMarketingrsquos contribution to the strategy dialogueordm Journal of the Academy ofMarketing Science Vol 20 Fall pp 323-9

IJSIM143

304

shows that increased levels of innovation are associated to superiorperformance (Robinson et al 1992 DeshpandeAcirc et al 1993)

Market orientation and customer loyaltyDeshpandeAcirc et al (1993 p 24) point out that ordfthe canons of the marketingconcept assert that proregt is a reward for customer orientation which creates asatisreged customer but we have only the beginning of systematic empiricaldocumentation of the presumed relationshipordm In the present competitivemarket environment characterized by globalization with rapid market entry ofnew products and maturity conditions in many products and servicesattaining a high level of customer loyalty has emerged as a central managerialconcern Clearly customer loyalty constitutes an important objective forstrategic marketing planning (Kotler 1984) and represents an important basisfor developing a sustainable competitive advantage plusmn an advantage that canbe realized through market orientation A high degree of market orientationleads to customer loyalty which in the long run contributes to better economicperformance In the service sector the intangible nature of services gives rise toinformationrsquos asymmetry between buyers and sellers This results in higherrisk perceptions and greater difregculty in customerrsquos quality evaluation(Nayyar 1990) As a result market orientation becomes a crucial instrument toestablish long-term relations with customers in service regrms Kohli andJaworski (1990) posited a positive relationship between a regrmrsquos marketorientation level and customer satisfaction Webb et al (2000) and Lai (2003)provide further empirical support to this relationship Jones et al (2002)empirical research on business-to business buyer-seller relationship suggeststhat a strong salespersonrsquos customer orientation tends to reduce the customerrsquospropensity to switch suppliers Harrison-Walker (2001) found a positiverelationship between market orientation and customer retention customerwillingness to pay a price premium customer propensity to spread positiveword-of-mouth communication and customer propensity to not-alternateamong brandsservice providers All these variables are widely used measuresof the customer loyalty concept (Jacoby and Chestnut 1978 Odin et al 2001Dick and Basu 1994)

On the other hand customer loyalty is expected to have a positive impact onbusiness economic performance since market-oriented regrms have a largenumber of satisreged customer and therefore a higher rate of repeated purchases(Dick and Basu 1994 McCullough et al 1986 Loveman 1998 Kamakura et al2002)

The arguments put forth in the preceding sections can be summarized into aseries of hypotheses Our initial hypotheses is

H1 Within an industry the more market oriented regrms are the better theirobjective economic performance

Businesseconomic

performance

289

If this hypotheses is tenable then we shall investigate what is role ofinnovation degree innovation performance and customer loyalty in relation tothe hypothesized impact of market orientation on economic performance Wehypothesize that each of these variables taken separately is an intermediatemediational variable According to Baron and Kenny (1986) an intermediatevariable is said to be a mediator if when introduced within a directedrelationship the directed relationship vanishes (complete mediational effect) orat least it signiregcantly decreases (partial mediational effect)

Thus our hypotheses are

H2a Within an industry the impact of market orientation on economicperformance is at least partially mediated through innovation degreeThat is the more market oriented regrms are the higher their innovationdegree This higher innovation degree results in better economicperformance

H2b Within an industry the impact of market orientation on economicperformance is at least partially mediated through innovationperformance That is the more market oriented regrms are the highertheir innovation performance This higher innovation performanceresults in better economic performance

H2c Within an industry the impact of market orientation on economicperformance is at least partially mediated through customer loyaltyThat is the more market oriented regrms are the higher their customerloyalty This higher customer loyalty results in better economicperformance

If in turn these hypotheses are tenable then we need to specify a model thatintegrates these three intermediate variables within the hypothesized directedrelationship of market orientation on economic performance We hypothesizedthat

H3 Taken jointly innovation degree innovation performance andcustomer loyalty completely mediate the impact of marketorientation on economic performance Furthermore the relationshipbetween innovation degree and economic performance is all conveyedthrough innovation performance

This last hypothesis is graphically depicted in Figure 1

Empirical studyWith the objective of isolating within-industry variation we shall adopt asingle-industry approach focusing on the insurance industry This clearlyprevents the generalization of the results outside the scope of the industryconsidered On the other side we can meaningfully assess the impact of unitincrements in market orientation on regrmsrsquo economic performance and sound

IJSIM143

290

inferences can be drawn on the target population based on therepresentativeness of the sample used

The confounding of within-industry and between industry variation is notthe only threat to the validity of inferences drawn on the relationship betweenmarket orientation and economic performance A second threat is the noiseintroduced by environmental variables such as market turbulence marketgrowth rate buyer and supplier power and competitive intensity on businessperformance A standard approach to minimize this threat is to focus theresearch on a single market The drawback of this approach is that we are notable to capture regrmsrsquo behavior in facing increasing globalization and marketintegration As a compromise between these two ends the present studytargets the European Union market In this market the key characteristics of asingle market are preserved but it is also an environment in which we canpresently observe how regrms struggle in meeting the challenges ofinternationalization and market integration

A third threat to the validity of inferences drawn on the relationship betweenmarket orientation and economic performance lies in the use of subjectivemeasures of economic performance (ie managersrsquo evaluations of theircompaniesrsquo performance) Positive effects of market orientation on economicperformance have been reported when subjective assessments of performancewere used However when objective measures of economic performance havebeen used mixed results emerged For instance Ruekert (1992) and Lambin(1996) report a positive relationship between market orientation and objectivelymeasured economic performance However Bhuian (1997) Jaworski and Kohli(1993) and Selnes et al (1996) failed to regnd any signiregcant relationshipClearly when market orientation and economic performance are concurrentlyassessed by the regrmsrsquo managers a perceptual bias may be introduced A casein point Van Bruggen and Smidts (1995) found within one single company

Figure 1Hypothetical model

Businesseconomic

performance

291

(which has only one performance) a substantial degree of variation insubjective performance assessments In fact they report a positive relationshipbetween market orientation and judgments about the company performancewithin a single company As they point out ordfit might be that managers have amore positive view of their companyrsquos market orientation when they perceivetheir company to be performing wellordm (Van Bruggen and Smidts 1995 p 13)Hence it is important to employ objective measures of economic performanceand we shall do so in the present study

Market orientation in the services sector the European insurance industryThe insurance sector is of particular interest from a market orientationviewpoint as it works with intangible commodities in which service qualityand customer orientation are crucial elements The competitive characteristicsgenerated by the European Union provide an additional interest in studyingmarket orientation in this area The insurance sector in Europe has traditionallyoperated subject to strict regulations and strong protection from internationalcompetition However for some years now the European Commission has beenworking on the liberalization of this sector Effective implementation of this hasbrought about a major increase in competition within the sector and hasprovoked a major restructuring of insurance companies and groups Thecompetitive climate in Europe has also been inmacruenced by a downside in theeconomic cycle and changes in consumer behavior European customers nowshow greater service expectations and less loyalty As a result rivalry amongcompetitors is increasing as is the importance of competitive strategies adaptedto this sectorrsquos needs In this background the degree of orientation toward thecustomer distributors competition and the general socio-economicenvironment is becoming an increasingly important area of study not onlyfor academics but also for the business world

Lado et al (1998ab) have investigated quite extensively the marketorientation of insurance regrms within the European Union These authors havenot found signiregcant mean differences in market orientation by countryFurthermore they report substantial agreement between the factor structuresof market orientation across countries Thus it seems that the Europeaninsurance sector can be considered a homogenous population with respect tomarket orientation

DataThe population universe considered in this article is deregned as the set ofinsurance companies operating in the European Union which meet thefollowing conditions

they operate in private insurance or ordfmass insuranceordm

they have a market share of more than 005 percent and

their management is independent

IJSIM143

292

The list of European insurance companies was taken from the Financial TimesYearbook for 1996

It was assumed that senior executives were the people best qualireged to assessthe companyrsquos market orientation as well as their innovation degree innovationperformance and customersrsquo loyalty Therefore information from these variableswas gathered via a postal questionnaire submitted to the senior executive in eachof the 554 companies comprising the target population

We obtained 122 valid questionnaires giving a response rate of 22 percentThis sample accounts for over 17 percent of total insurance premiums in theEuropean Union

In order to assess response bias the questionnaires were divided intoquartiles on the basis of reception date (Armstrong and Overton 1977) Ananalysis of early and late responses did not indicate any signiregcant differencein terms of means and covariances

MeasuresBusiness economic performance is a complex construct with multiple possibleobserved indicators Here we measure this construct using three remacrectiveindicators[1]

(1) domestic market share (Jaworski and Kohli 1993 DeshpandeAcirc et al 1993Selnes et al 1996 Greenley and Foxall 1997 1998)

(2) premium growth (which is equivalent to sales growth for the insurancecompanies business (Slater and Narver 1994 Narver and Slater 1990Ruekert 1992 Greenley and Foxall 1997 1998) and

(3) proregtability per year averaged over the last three years(similar to thereturn on investment (ROI) rate (Greenley and Foxall 1997 1998))

All three indicators were expressed as percentages These data were obtainedfrom the managers responding the questionnaire Their responses were carefullycontrasted with published regnancial information (eg Reuters Insurance Brieregng)

Market orientation was measured using the Market Orientation Scale-Revised(MOS-R) This scale is a shortened version of the MOS validated by Lado et al(1998a) in the population of insurance companies of Belgium and Spain Ladoet al (1998a) shortened the original MOS scale while extending the previousvalidation study to target all insurance companies operating in the EuropeanUnion In the Appendix we provide the 30 items composing the MOS-R Eachitem is to be rated on a ten-point Likert-type scale ranging from 0 (completedisagreement) to 10 (complete agreement)

Innovation degree and innovation performance were assessed by means ofmulti-item questionnaires akin to Miller and Friesenrsquos (1982) Innovationperformance was measured by a four-item questionnaire regarding the successof a new productservice (deregned as an improved product a product extensionor a new product line) introduced by the company The questions involved

Businesseconomic

performance

293

whether the new productservice had succeeded in meeting the sales growthmarket share and proregt objectives set up by the company

Innovation degree was assessed by a three-item questionnaire that inquiredthe rate of new productsservices introduced by the company relative tocompetitors the amount of new productsservices marketed by the companyover the past three years and the nature of change of the new productsservices

Finally we used a four-item questionnaire based on existing literature (egDick and Basu 1994 Javalgi and Moberg 1997) to evaluate managersrsquoperceptions of their customersrsquo loyalty The questionnaire taps on theproportion of their customersrsquo insurance premiums taken on by the companythe average time a customer remains in the companyrsquos portfolio the probabilityof a customer renewing a premium and the overall perception of the companycustomersrsquo loyalty

Scale scores for innovation degree innovation performance and customerloyalty were obtained as an unweighted sum of the corresponding items Sincein all three cases Likert-type items on a 0-7 scale were used scale scores forthese variables range from 0-27 0-27 and 0-28 respectively For marketorientation we computed a score for each of its facets as an unweighted sum ofthe corresponding items Then a global market orientation score was obtainedas a sum of the facetsrsquo scores inversely weighted by their number of itemsHence this market orientation score assigns equal weights to each its facetsand ranges from 0-90

The scalesrsquo reliability (as assessed by coefregcient alpha) in this sample were088 (market orientation) 070 (innovation degree) 091 (innovationperformance) and 076 (customer loyalty) The means standard deviationsand correlations among all variables considered in this study are presented inTable I As can be seen in this table the three indicators of business economicperformance are signiregcantly but not largely correlated (the correlations rangefrom 020 to 029) The correlations among the hypothesized intermediatevariables (innovation degree innovation performance and customer loyalty)are not high except for innovation degree and innovation performance whichshare 36 percent of their variance The correlations of market orientation withthe intermediate variables appear signiregcantly larger (they range from 055 to058) than with the dependent variables (they range from 023 to 036) We alsoobserve in Table I that managers report on average a high degree of innovationin their businesses not so high a level of customer loyalty and a level ofinnovation performance just at the scale mean The average self-reporteddegree of market orientation is 56 on a 0-90 scale

MethodAll hypotheses were contrasted using covariance structure analysis asimplemented in LISREL 850 (see Joreskog et al 1999) Since all three indicatorsof business performance are highly positively skewed and present a high

IJSIM143

294

degree of kurtosis throughout this paper rather than attempting to transformthese variables to near-normality we shall employ an estimation approach thatis robust to non-normality of the observed variables The parameter estimateswere obtained using maximum likelihood estimation with standard errorssuitable for non-normal data (Joreskog et al 1999 Equation A24) and twotest statistics were used to assess the goodness of regt (GFI) of the model theSatorra-Bentler scaled chi-squared statistic (Satorra and Bentler 1988Equation 41) and Brownersquos (1984 Equation 220a) chi-squared statisticcorrected for non-normality To evaluate better the goodness of regt of thismodel several additional indices will also provided the root mean squarederror of approximation (RMSEA (Steiger 1990)) the standardized root meansquared residual (SRMSR (Joreskog et al 1999)) the GFI (Tanaka and Huba1985) and the Comparative Fit Index (CFI) using the independence model asbaseline (Bentler 1990 see also McDonald and Marsh 1990) Adequate to goodregt is suggested by RMSEA and SRMSR values approaching 005 For the GFIand the CFI indices values between 090 and 100 indicate adequate to excellentregt (but see Hu and Bentler 1999)

ResultsH1The model used to estimate the effects of market orientation on insurancebusinessesrsquo performance consists of a latent variable representing economicperformance with three indicators (market share premium growth andproregtability) and a single exogenous variable (market orientation) This model

MST PG PROF INNODR INNPERF LOYAL MO

MST 100PG 020 100PROF 023 029 100INNODR 030 034 035 1000INNPERF 027 032 041 062 100LOYAL 025 019 038 040 038 100MO 024 030 036 055 058 057 100x 455 788 626 1825 1422 1996 5628Std 535 749 571 356 292 520 1331

NotesN = 122 all correlations are signiregcant (a = 001) except those marked by which are onlysigniregcant at an a = 005MS = market share PG = premium growth PROF = proregtability INNODR = innovation degreeINNPERF = innovation performance LOYAL = customer loyalty MO = market orientation

Table IMeans standard

deviations andinter-correlations

Businesseconomic

performance

295

is depicted in Figure 2 The parameter estimates and GFIs for this model aregiven in Table II The model shows a good regt although note that it only hastwo degrees of freedom According to the model the best objective indicator ofbusiness economic performance is proregtability per year over 34 percent of itsvariance is accounted for by the model The standardized regressioncoefregcients reveal that proregtability per year is the best objective indicator of

Parameter estimates Goodness of regt R 2

Parameter Value Index Value Variable Value

b1 006 [061] (002) MFF X 2001 ( p = 099) Market share 015

b2 166 [039] (049) S-B X 2001 ( p = 099) Premium growth 024

b3 293 [049] (078) B X 2001 ( p = 099) Proregtability 034

b4 267 [059] (084) RMSEA 001 Business performance 037u1 17713 [100] (2241) df 2u2 2427 [085] (713) SRMSR 001u3 4248 [076] (1364) GFI 100u4 2141 [066] (515) CFI 100

NotesRobust asymptotic standard errors are provided in parentheses standardized parameterestimates are provided in square bracketsMFF X 2 = Minimum regt function chi-square SB X 2 = Satorra-Bentler scaled chi-squareB X 2 = Brownersquos chi-square corrected for non-normality RMSEA = root mean squared error ofapproximation SRMSR = standardized root mean squared residual GFI = goodness of regt indexCFI = comparative regt index R 2 = squared multiple correlations for endogenous variables

Table IIEstimation resultsfor the modeldepicted in Figure 2

Figure 2Market orientation aspredictor of economicperformance

IJSIM143

296

overall business performance Finally according to the model almost 37percent of overall business economic performance is accounted for by thedegree of market orientation

An inspection of the total effects of market orientation on the indicators ofeconomic performance suggests that unit increments of market orientation asmeasured by the MOS-R are associated with 0095 0168 and 0153 incrementsin domestic market share premium growth and proregtability per year averagedover the last three years respectively

H2A mediated model for the relationship between market orientation andbusiness performance is depicted in Figure 3 In this context a mediating effectis said to exist when

both mediating paths b5 b6 are signiregcant and

the direct effect of the exogenous variable on the outcome variablevanishes (complete mediational effect) or is signiregcantly lower (partialmediational effect) when a mediator variable is introduced in the model

Condition (2) amounts to b1 in Figure 3 becoming zero or signiregcantly less thatthan value reported for Figure 2

We used the mediated model depicted in Figure 3 to test for mediatingeffects of innovation degree innovation performance and customer loyaltyseparately on the impact of market orientation on business economicperformance We found that when either innovation performance or innovation

Figure 3Mediated model

Businesseconomic

performance

297

degree were used as mediating variable all the mediating paths weresigniregcant and that direct path from market orientation to businessperformance was not signiregcantly different from zero b1 = 004 t = 162for innovation performance b1 = 004 t = 186 for innovation degree Hencetaken separately both innovation degree and innovation performancecompletely mediate the impact of market orientation on businessperformance After regxing b1 at zero we re-estimated these two mediationalmodels The resulting parameter estimates and GFIs for these two models areshown in Table III On the other hand customer loyalty was found not to havea mediational effect between market orientation and business performanceThe parameter estimates and GFIs for this model are also given in Table III

As can be seen in Table III the mediating paths are signiregcant but thedirect path b1 is signiregcantly different from zero at a = 001 Furthermore a 99percent conregdence interval for the value for b1 reported in Table II (003 009)includes the value of b1 estimated in the mediational model using customerloyalty 004 Hence this variable does not even partially mediate on the impactof market orientation on business economic performance The standardizeddirect impact of market orientation on business performance (0408) is morethan twice the standardized impact of market orientation conveyed throughcustomer loyalty (0191)

The percentage of variance of business economic performance explained bythe model when innovation performance innovation degree or customer loyaltyare used as mediators is very similar (465 percent 453 percent and 437percent respectively)

H3The full model to be regtted corresponding to the hypothesis depicted in Figure 1is presented in Figure 3 The parameter estimates and goodness of regt testcorresponding to this model are given in Table IV (see also Figure 4) As can beseen in this Table the model regts these data very well

All the postulated relationships were found to be signiregcant at an a = 001Lagrange multiplier tests indicated that the regt of the model would notsigniregcantly improve by

adding a direct effect of market orientation to business performance nor

adding a direct effect of innovation degree on economic performance

Result (1) is in accordance with the results discussed above where we saw thatinnovation degree and innovation performance even when taken separatelycompletely mediate the impact of market orientation on business performanceResult (2) conregrms our hypothesis that innovation performance completelymediates the impact of innovation degree on business performance

Given that all effects of market orientation on business performance gothrough either innovation degree-innovation performance or through customer

IJSIM143

298

Par

amet

eres

tim

ates

Goo

dnes

sof

regt

Par

amet

erV

alue

Par

amet

erV

alue

Index

Val

ue

Innova

tion

per

form

ance

as

med

iato

rb 1

0(reg

xed

)u

117

713

[10

0](2

241

)M

FF

X2

560

(p=

035

)b 2

155

[04

0](0

42)

u2

241

4[0

84]

(71

1)S-B

X2

380

(p=

058

)b 3

263

[04

8](0

67)

u3

431

6[0

77]

(45

5)B

X2

429

(p=

051

)b 4

248

[05

9](o

81)

u4

211

3[0

65]

(45

5)df

5R

MSE

A0

03b 5

023

[05

8](0

03)

u5

180

3[0

67]

(28

0)SR

MSR

005

GF

I0

98b 6

018

[06

8](0

05)

CF

I0

99

Innova

tion

deg

ree

as

med

iato

rb

10

(regxed

)u

117

713

[10

0](2

241

)M

FF

X2

717

(p=

021

)b 2

169

[05

0](0

36)

u2

234

0[0

82]

(73

8)S-B

X2

497

(p=

043

)b 3

282

[05

1](0

82)

u3

415

6[0

74]

(45

0)B

X2

613

(p=

029

)b 4

226

[05

4](0

79)

u4

232

9[0

71]

(46

0)df

5R

MSE

A0

06b 5

012

[05

5](0

02)

u5

590

[06

9](0

78)

SR

MSR

005

GF

I0

98b 6

031

[06

7](0

09)

CF

I0

98

Cust

om

erlo

yalty

as

med

iato

rb 1

004

[04

1](0

02)

u1

177

13[1

00]

(22

41)

MF

FX

22

30(p

=0

68)

b 21

61[0

40]

(05

1)u

224

05

[08

4](7

11)

S-B

X2

155

(p=

082

)b 3

252

[04

5](0

74)

u3

447

5[0

80]

(45

5)B

X2

309

(p=

054

)b 4

265

[06

2](0

92)

u4

201

3[0

62]

(45

5)R

MSE

A0

0df

4b 5

015

[05

7](0

02)

u5

863

[06

8](1

04)

SR

MSR

002

GF

I0

99b 6

013

[03

4](0

05)

CF

I1

00

Note

s

Rob

ust

asym

pto

tic

stan

dar

der

rors

are

pro

vid

edin

par

enth

eses

st

andar

diz

edpar

amet

eres

tim

ates

are

pro

vid

edin

squar

ebra

cket

sM

FF

X2=

Min

imum

regt

funct

ion

chi-sq

uar

eSB

X2=

Sat

orra

-Ben

tler

scal

edch

i-sq

uar

eB

X2=

Bro

wnersquo

sch

i-sq

uar

eco

rrec

ted

for

non

-nor

mal

ity

RM

SE

A=

root

mea

nsq

uar

eder

ror

ofap

pro

xim

atio

n

SR

MSR

=st

andar

diz

edro

otm

ean

squar

edre

sidual

G

FI

=goo

dnes

sof

regt

index

C

FI

=co

mpar

ativ

ereg

tin

dex

Table IIIEstimation results

for the modeldepicted in Figure 3

Businesseconomic

performance

299

Par

amet

eres

tim

ates

Goo

dnes

sof

regt

R2

Par

amet

erV

alue

Par

amet

erV

alue

Index

Val

ue

Var

iable

Val

ue

b 10

12[0

55]

(22

41)

u1

177

13[1

00]

(00

2)M

FF

X2

105

9(p

=0

56)

Mar

ket

shar

e0

16b 2

013

[03

4](0

04)

u2

590

[06

9](0

78)

S-B

X2

785

(p=

080

)P

rem

ium

gro

wth

019

b 30

15[0

57]

(00

2)u

314

48

[05

4](2

59)

BX

212

67

(p=

039

)P

roreg

tabilit

y0

38b 4

078

[04

4](0

15)

u4

863

[06

8](1

04)

df

12B

usi

nes

sper

form

ance

056

RM

SE

A0

0b 5

016

[05

4](0

05)

u5

239

1[0

84]

(72

3)SR

MSR

005

Innov

atio

nper

form

ance

046

b 60

16[0

37]

(00

7)u

645

09

[08

1](1

271

)G

FI

098

Innov

atio

ndeg

ree

031

b 71

42[0

40]

(04

5)u

720

12

[06

2](4

60)

CF

I1

00C

ust

omer

loyal

ty0

32b 8

217

[04

4](0

64)

b 92

32[0

62]

(08

5)

Note

sR

obust

asym

pto

tic

stan

dar

der

rors

are

pro

vid

edin

par

enth

eses

st

andar

diz

edpar

amet

eres

tim

ates

are

pro

vid

edin

squar

ebra

cket

sM

FF

X2=

Min

imum

regtfu

nct

ion

chi-sq

uar

eSB

X2=

Sat

orra

-Ben

tler

scal

edch

i-sq

uar

eB

X2=

Bro

wnersquo

sch

i-sq

uar

eco

rrec

ted

for

non

-nor

mal

ity

RM

SE

A=

root

mea

nsq

uar

eder

ror

ofap

pro

xim

atio

n

SR

MSR

=st

andar

diz

edro

otm

ean

squar

edre

sidual

G

FI

=goo

dnes

sof

regt

index

C

FI

=co

mpar

ativ

ereg

tin

dex

R

2=

squar

edm

ult

iple

corr

elat

ions

for

endog

enou

svar

iable

s

Table IVEstimation resultsfor the modeldepicted in Figure 4

IJSIM143

300

loyalty a question arises as to the relative importance of the speciregc effectsgoing through these variables The standardized speciregc effect (computed as inBollen 1987) going through innovation degree and innovation performance is031 and 021 going through customer loyalty Hence the impact of marketorientation going through innovation is 50 percent more than that goingthrough customer loyalty

We can also see in Table IV that over 30 percent of the variance of theintermediate variables (innovation degree innovation performance andcustomer loyalty) are explained by market orientation In fact almost 50percent (464 percent to be exact) of innovation performance is explained bymarket orientation Furthermore note that the percentage of variance ofbusiness performance explained by the model is 561 percent a 52 percentincrement over what is explained by market orientation alone (see Table II) andover a 20 percent increment over what is explained by the mediational modelsconsidered previously Hence the inclusion of all three intermediate variables inthe model improves considerably our prediction of business performance

Furthermore we observe in Table IV that the direct effect of marketorientation on all three intermediate variables appear to be equal Also thedirect effects of customer loyalty and of innovation performance on economicperformance appear to be equal We re-estimated the model to test theseconstraints obtaining b1 = b2 = b3 = 0133 b5 = b6 = 0157 Satorra-BentlerX 2(15) = 8849 p = 0885 CFI = 100 GFI = 0975 SRMR = 006

ConclusionsMarket orientation can be deregned as a strategy used to reach a sustainablecompetitive advantage based on the generation and use of information within

Figure 4Full model

Businesseconomic

performance

301

organizations and on the selection of markets to be satisreged In thisframework we believe that competitive advantage results from the use ofresources and capabilities to generate differential satisfaction in proregtablemarkets Sustainability is achieved because the performance of the marketorientationrsquos behaviors requires complex organizational knowledge that cannoteasily be imitated by competitors Thus we hypothesize that the satisfaction ofproregtable markets permits the regrm to achieve a psychologically differentialposition that leads to brand loyalty and thus to higher proregts Previous studieshave

found a clear impact of market orientation on economic performance

assessed the effects of market orientation on innovation and

investigated the relationship between market orientation and relationshipmarketing variables

The present study proposes and tests a model that integrates both streams ofresearch innovation and relationship marketing More research is needed toinvestigate the role of customer satisfaction within this framework

In our necessarily partial model innovation degree innovation performanceand customer loyalty are used as intermediate variables on the effect of marketorientation on business performance Our results suggest that the addition ofthese variables help improve our predictions of business economic performance52 percent over what is explained by market orientation alone Also we foundthat innovation degree and innovation performance each taken separatelycompletely mediate the effect of market orientation on economic performanceFurthermore the impact of innovation degree on economic performance iscompletely channeled through innovation performance Customer loyalty byitself does not meditate the impact of market orientation on economicperformance but when considered along with innovation degree andinnovation performance it conveys some of the effects of market orientationon business performance This seemingly contradictory result arises from thefact that all three intermediate variables are interrelated

Our results should not be taken to imply that there are no other variablesmediate the effect of market orientation on economic performance We believethat other variables that have not been taken into account in this study such asproduct quality and customer satisfaction may also be signiregcant mediatorsHowever our results do suggest that whenever innovation degree andinnovation performance are included in the model as intermediate variablesthe effects of market orientation on business performance will mostly beconveyed through these variables Also in this study we have adopted thecurrently most widely accepted approach to measuring customer loyalty whichis based on behavioral loyalty measures (eg share of category requirementsrenew the policy probability the average customer last in the companyportfolio) It would be worth extending the present study by including not only

IJSIM143

302

behavioral measures of customer loyalty but also attitudinal measures alongthe lines of Dick and Basu (1994)

In our opinion two important contributions of the present research are outuse of objective measures of business performance and our focusing oninternational markets Despite the growing role of globalization and marketintegration and despite the increasing internationalization of corporationsmost studies on market orientation have focused on domestic markets (withnotable exceptions such as Selnes et al 1996 Webster 1994) Similarly moststudies on product innovation have also focused on domestic markets There isa lack of research yielding empirical support to the validity in an internationalsetting to research results obtained in domestic markets To regll this gap wetargeted the European Union market

Our study focused on a single industry the insurance sector Our sampleaccounted for 22 percent of the companies and 17 percent of the insurancepremiums in the targeted market An advantage of our single-industryapproach is that (with obvious reservations arising from the non-experimentalnature of our study and the fact that our sample should not be considered tohave been obtained at random) we can draw tentative predictions from ourmodel concerning the impact of market orientation on economic performance ininsurance companies operating in the European Union market An evidentdrawback of the single-industry approach adopted here is that it is not clearhow the present results extrapolate to other industries even when operating inthe same market

We have found that within the European Union insurance regrms thatconstantly monitor the evolution of current and potential customerrsquos needsmodify the attributes of the products to adapt them to the distributorsrequirements analyze competitorsrsquo marketing policy and products and knowbest of environmental trends especially technological and legal changes aremore likely to develop more new products have their new products accepted bythe market and obtain more loyal customers In turn increased customerloyalty and increased new product success will result in improved economicperformance However these conclusions must be taken with some caution aswe have used overall measures of innovation and innovation performanceFurther research is needed that takes into account the various aspects thatconstitute innovation

Previous studies have concluded that insurance companies still seethemselves as being product-focused and the industry as a whole is generallydistribution driven (Sodano 2000 Lambin 1996) This lack of closeness withcustomers has in turn contributed to the industry lack of product innovationsand product portfolios of commodity products that only compete on price Nowthe most successful regrms are redirecting their focus to the market needs andthey are beginning to exploit customer data and use market research togenerate ideas for designing new products In this context these companies

Businesseconomic

performance

303

now face choosing among too many new service options The managerialimplication of our study is that by enhancing their market orientation regrmswill know and service its customers better Thus they will generate moreinnovations by adopting a market-based product development process Alsoincreasing levels of market orientation enable regrms to discriminate more easilywhich new products have a higher success probability thus enhancing both theefregcacy and efregciency of new product development

Note

1 We considered employing the total volume of premiums for each insurance company as anadditional indicator although it seemed to us to be a better indicator of a companyrsquos sizerather than of its performance As we suspected the total volume of premiums wasuncorrelated with any of the variables considered in this study except for the companyrsquosmarket share (r = 024 p 001) Hence it is clear that volume of premiums should not beused as an indicator of insurance companiesrsquo performance

References

Armstrong JS and Overton T (1977) ordfEstimating nonresponse bias in mail surveysordm Journalof Marketing Research Vol 14 pp 396-402

Atuahene-Gima K (1996) ordfMarket orientation and innovationordm Journal of Business ResearchVol 35 pp 93-103

Baker TL Simpson PM and Sigauw JA (1999) ordfThe impact of suppliersrsquo perceptions ofreseller market orientation on key relationship constructsordm Journal of the Academy ofMarketing Science Vol 27 No 1 pp 50-7

Baron RM and Kenny DA (1986) ordfThe moderator-mediator variable distinction in socialpsychological research conceptual strategic and statistical considerationsordm Journal ofPersonality and Social Psychology Vol 51 pp 1173-82

Bentler PM (1990) ordfComparative regt indexes in structural modelsordm Psychological BulletinVol 107 pp 238-46

Bhuian SN (1997) ordfExploring market orientation in banks an empirical examination in SaudiArabiaordm The Journal of Service Marketing Vol 11 No 5 pp 317-28

Bollen KA (1987) ordfTotal direct and indirect effects in structural equation modelsordm inClogg CC (Ed) Sociological Methodology American Sociological AssociationWashington DC pp 37-69

Browne MW (1984) ordfAsymptotically distribution free methods for the analysis of covariancestructuresordm British Journal of Mathematical and Statistical Psychology Vol 37 pp 62-83

Calantone RJ di Benedetto AC and Bhoovaraghavan S (1994) ordfExamining the relationshipbetween degree of innovation and new product successordm Journal of Business ResearchVol 30 pp 143-8

Calantone RJ Schmidt JB and Song MX (1996) ordfControllable factors of new productssuccess a cross-national comparisonordm Marketing Science Vol 15 No 4 pp 341-58

Cooper RC (1994) ordfNew product the factors that drive successordm International MarketingReview Vol 11 No 1 pp 60-76

Day G (1992) ordfMarketingrsquos contribution to the strategy dialogueordm Journal of the Academy ofMarketing Science Vol 20 Fall pp 323-9

IJSIM143

304

If this hypotheses is tenable then we shall investigate what is role ofinnovation degree innovation performance and customer loyalty in relation tothe hypothesized impact of market orientation on economic performance Wehypothesize that each of these variables taken separately is an intermediatemediational variable According to Baron and Kenny (1986) an intermediatevariable is said to be a mediator if when introduced within a directedrelationship the directed relationship vanishes (complete mediational effect) orat least it signiregcantly decreases (partial mediational effect)

Thus our hypotheses are

H2a Within an industry the impact of market orientation on economicperformance is at least partially mediated through innovation degreeThat is the more market oriented regrms are the higher their innovationdegree This higher innovation degree results in better economicperformance

H2b Within an industry the impact of market orientation on economicperformance is at least partially mediated through innovationperformance That is the more market oriented regrms are the highertheir innovation performance This higher innovation performanceresults in better economic performance

H2c Within an industry the impact of market orientation on economicperformance is at least partially mediated through customer loyaltyThat is the more market oriented regrms are the higher their customerloyalty This higher customer loyalty results in better economicperformance

If in turn these hypotheses are tenable then we need to specify a model thatintegrates these three intermediate variables within the hypothesized directedrelationship of market orientation on economic performance We hypothesizedthat

H3 Taken jointly innovation degree innovation performance andcustomer loyalty completely mediate the impact of marketorientation on economic performance Furthermore the relationshipbetween innovation degree and economic performance is all conveyedthrough innovation performance

This last hypothesis is graphically depicted in Figure 1

Empirical studyWith the objective of isolating within-industry variation we shall adopt asingle-industry approach focusing on the insurance industry This clearlyprevents the generalization of the results outside the scope of the industryconsidered On the other side we can meaningfully assess the impact of unitincrements in market orientation on regrmsrsquo economic performance and sound

IJSIM143

290

inferences can be drawn on the target population based on therepresentativeness of the sample used

The confounding of within-industry and between industry variation is notthe only threat to the validity of inferences drawn on the relationship betweenmarket orientation and economic performance A second threat is the noiseintroduced by environmental variables such as market turbulence marketgrowth rate buyer and supplier power and competitive intensity on businessperformance A standard approach to minimize this threat is to focus theresearch on a single market The drawback of this approach is that we are notable to capture regrmsrsquo behavior in facing increasing globalization and marketintegration As a compromise between these two ends the present studytargets the European Union market In this market the key characteristics of asingle market are preserved but it is also an environment in which we canpresently observe how regrms struggle in meeting the challenges ofinternationalization and market integration

A third threat to the validity of inferences drawn on the relationship betweenmarket orientation and economic performance lies in the use of subjectivemeasures of economic performance (ie managersrsquo evaluations of theircompaniesrsquo performance) Positive effects of market orientation on economicperformance have been reported when subjective assessments of performancewere used However when objective measures of economic performance havebeen used mixed results emerged For instance Ruekert (1992) and Lambin(1996) report a positive relationship between market orientation and objectivelymeasured economic performance However Bhuian (1997) Jaworski and Kohli(1993) and Selnes et al (1996) failed to regnd any signiregcant relationshipClearly when market orientation and economic performance are concurrentlyassessed by the regrmsrsquo managers a perceptual bias may be introduced A casein point Van Bruggen and Smidts (1995) found within one single company

Figure 1Hypothetical model

Businesseconomic

performance

291

(which has only one performance) a substantial degree of variation insubjective performance assessments In fact they report a positive relationshipbetween market orientation and judgments about the company performancewithin a single company As they point out ordfit might be that managers have amore positive view of their companyrsquos market orientation when they perceivetheir company to be performing wellordm (Van Bruggen and Smidts 1995 p 13)Hence it is important to employ objective measures of economic performanceand we shall do so in the present study

Market orientation in the services sector the European insurance industryThe insurance sector is of particular interest from a market orientationviewpoint as it works with intangible commodities in which service qualityand customer orientation are crucial elements The competitive characteristicsgenerated by the European Union provide an additional interest in studyingmarket orientation in this area The insurance sector in Europe has traditionallyoperated subject to strict regulations and strong protection from internationalcompetition However for some years now the European Commission has beenworking on the liberalization of this sector Effective implementation of this hasbrought about a major increase in competition within the sector and hasprovoked a major restructuring of insurance companies and groups Thecompetitive climate in Europe has also been inmacruenced by a downside in theeconomic cycle and changes in consumer behavior European customers nowshow greater service expectations and less loyalty As a result rivalry amongcompetitors is increasing as is the importance of competitive strategies adaptedto this sectorrsquos needs In this background the degree of orientation toward thecustomer distributors competition and the general socio-economicenvironment is becoming an increasingly important area of study not onlyfor academics but also for the business world

Lado et al (1998ab) have investigated quite extensively the marketorientation of insurance regrms within the European Union These authors havenot found signiregcant mean differences in market orientation by countryFurthermore they report substantial agreement between the factor structuresof market orientation across countries Thus it seems that the Europeaninsurance sector can be considered a homogenous population with respect tomarket orientation

DataThe population universe considered in this article is deregned as the set ofinsurance companies operating in the European Union which meet thefollowing conditions

they operate in private insurance or ordfmass insuranceordm

they have a market share of more than 005 percent and

their management is independent

IJSIM143

292

The list of European insurance companies was taken from the Financial TimesYearbook for 1996

It was assumed that senior executives were the people best qualireged to assessthe companyrsquos market orientation as well as their innovation degree innovationperformance and customersrsquo loyalty Therefore information from these variableswas gathered via a postal questionnaire submitted to the senior executive in eachof the 554 companies comprising the target population

We obtained 122 valid questionnaires giving a response rate of 22 percentThis sample accounts for over 17 percent of total insurance premiums in theEuropean Union

In order to assess response bias the questionnaires were divided intoquartiles on the basis of reception date (Armstrong and Overton 1977) Ananalysis of early and late responses did not indicate any signiregcant differencein terms of means and covariances

MeasuresBusiness economic performance is a complex construct with multiple possibleobserved indicators Here we measure this construct using three remacrectiveindicators[1]

(1) domestic market share (Jaworski and Kohli 1993 DeshpandeAcirc et al 1993Selnes et al 1996 Greenley and Foxall 1997 1998)

(2) premium growth (which is equivalent to sales growth for the insurancecompanies business (Slater and Narver 1994 Narver and Slater 1990Ruekert 1992 Greenley and Foxall 1997 1998) and

(3) proregtability per year averaged over the last three years(similar to thereturn on investment (ROI) rate (Greenley and Foxall 1997 1998))

All three indicators were expressed as percentages These data were obtainedfrom the managers responding the questionnaire Their responses were carefullycontrasted with published regnancial information (eg Reuters Insurance Brieregng)

Market orientation was measured using the Market Orientation Scale-Revised(MOS-R) This scale is a shortened version of the MOS validated by Lado et al(1998a) in the population of insurance companies of Belgium and Spain Ladoet al (1998a) shortened the original MOS scale while extending the previousvalidation study to target all insurance companies operating in the EuropeanUnion In the Appendix we provide the 30 items composing the MOS-R Eachitem is to be rated on a ten-point Likert-type scale ranging from 0 (completedisagreement) to 10 (complete agreement)

Innovation degree and innovation performance were assessed by means ofmulti-item questionnaires akin to Miller and Friesenrsquos (1982) Innovationperformance was measured by a four-item questionnaire regarding the successof a new productservice (deregned as an improved product a product extensionor a new product line) introduced by the company The questions involved

Businesseconomic

performance

293

whether the new productservice had succeeded in meeting the sales growthmarket share and proregt objectives set up by the company

Innovation degree was assessed by a three-item questionnaire that inquiredthe rate of new productsservices introduced by the company relative tocompetitors the amount of new productsservices marketed by the companyover the past three years and the nature of change of the new productsservices

Finally we used a four-item questionnaire based on existing literature (egDick and Basu 1994 Javalgi and Moberg 1997) to evaluate managersrsquoperceptions of their customersrsquo loyalty The questionnaire taps on theproportion of their customersrsquo insurance premiums taken on by the companythe average time a customer remains in the companyrsquos portfolio the probabilityof a customer renewing a premium and the overall perception of the companycustomersrsquo loyalty

Scale scores for innovation degree innovation performance and customerloyalty were obtained as an unweighted sum of the corresponding items Sincein all three cases Likert-type items on a 0-7 scale were used scale scores forthese variables range from 0-27 0-27 and 0-28 respectively For marketorientation we computed a score for each of its facets as an unweighted sum ofthe corresponding items Then a global market orientation score was obtainedas a sum of the facetsrsquo scores inversely weighted by their number of itemsHence this market orientation score assigns equal weights to each its facetsand ranges from 0-90

The scalesrsquo reliability (as assessed by coefregcient alpha) in this sample were088 (market orientation) 070 (innovation degree) 091 (innovationperformance) and 076 (customer loyalty) The means standard deviationsand correlations among all variables considered in this study are presented inTable I As can be seen in this table the three indicators of business economicperformance are signiregcantly but not largely correlated (the correlations rangefrom 020 to 029) The correlations among the hypothesized intermediatevariables (innovation degree innovation performance and customer loyalty)are not high except for innovation degree and innovation performance whichshare 36 percent of their variance The correlations of market orientation withthe intermediate variables appear signiregcantly larger (they range from 055 to058) than with the dependent variables (they range from 023 to 036) We alsoobserve in Table I that managers report on average a high degree of innovationin their businesses not so high a level of customer loyalty and a level ofinnovation performance just at the scale mean The average self-reporteddegree of market orientation is 56 on a 0-90 scale

MethodAll hypotheses were contrasted using covariance structure analysis asimplemented in LISREL 850 (see Joreskog et al 1999) Since all three indicatorsof business performance are highly positively skewed and present a high

IJSIM143

294

degree of kurtosis throughout this paper rather than attempting to transformthese variables to near-normality we shall employ an estimation approach thatis robust to non-normality of the observed variables The parameter estimateswere obtained using maximum likelihood estimation with standard errorssuitable for non-normal data (Joreskog et al 1999 Equation A24) and twotest statistics were used to assess the goodness of regt (GFI) of the model theSatorra-Bentler scaled chi-squared statistic (Satorra and Bentler 1988Equation 41) and Brownersquos (1984 Equation 220a) chi-squared statisticcorrected for non-normality To evaluate better the goodness of regt of thismodel several additional indices will also provided the root mean squarederror of approximation (RMSEA (Steiger 1990)) the standardized root meansquared residual (SRMSR (Joreskog et al 1999)) the GFI (Tanaka and Huba1985) and the Comparative Fit Index (CFI) using the independence model asbaseline (Bentler 1990 see also McDonald and Marsh 1990) Adequate to goodregt is suggested by RMSEA and SRMSR values approaching 005 For the GFIand the CFI indices values between 090 and 100 indicate adequate to excellentregt (but see Hu and Bentler 1999)

ResultsH1The model used to estimate the effects of market orientation on insurancebusinessesrsquo performance consists of a latent variable representing economicperformance with three indicators (market share premium growth andproregtability) and a single exogenous variable (market orientation) This model

MST PG PROF INNODR INNPERF LOYAL MO

MST 100PG 020 100PROF 023 029 100INNODR 030 034 035 1000INNPERF 027 032 041 062 100LOYAL 025 019 038 040 038 100MO 024 030 036 055 058 057 100x 455 788 626 1825 1422 1996 5628Std 535 749 571 356 292 520 1331

NotesN = 122 all correlations are signiregcant (a = 001) except those marked by which are onlysigniregcant at an a = 005MS = market share PG = premium growth PROF = proregtability INNODR = innovation degreeINNPERF = innovation performance LOYAL = customer loyalty MO = market orientation

Table IMeans standard

deviations andinter-correlations

Businesseconomic

performance

295

is depicted in Figure 2 The parameter estimates and GFIs for this model aregiven in Table II The model shows a good regt although note that it only hastwo degrees of freedom According to the model the best objective indicator ofbusiness economic performance is proregtability per year over 34 percent of itsvariance is accounted for by the model The standardized regressioncoefregcients reveal that proregtability per year is the best objective indicator of

Parameter estimates Goodness of regt R 2

Parameter Value Index Value Variable Value

b1 006 [061] (002) MFF X 2001 ( p = 099) Market share 015

b2 166 [039] (049) S-B X 2001 ( p = 099) Premium growth 024

b3 293 [049] (078) B X 2001 ( p = 099) Proregtability 034

b4 267 [059] (084) RMSEA 001 Business performance 037u1 17713 [100] (2241) df 2u2 2427 [085] (713) SRMSR 001u3 4248 [076] (1364) GFI 100u4 2141 [066] (515) CFI 100

NotesRobust asymptotic standard errors are provided in parentheses standardized parameterestimates are provided in square bracketsMFF X 2 = Minimum regt function chi-square SB X 2 = Satorra-Bentler scaled chi-squareB X 2 = Brownersquos chi-square corrected for non-normality RMSEA = root mean squared error ofapproximation SRMSR = standardized root mean squared residual GFI = goodness of regt indexCFI = comparative regt index R 2 = squared multiple correlations for endogenous variables

Table IIEstimation resultsfor the modeldepicted in Figure 2

Figure 2Market orientation aspredictor of economicperformance

IJSIM143

296

overall business performance Finally according to the model almost 37percent of overall business economic performance is accounted for by thedegree of market orientation

An inspection of the total effects of market orientation on the indicators ofeconomic performance suggests that unit increments of market orientation asmeasured by the MOS-R are associated with 0095 0168 and 0153 incrementsin domestic market share premium growth and proregtability per year averagedover the last three years respectively

H2A mediated model for the relationship between market orientation andbusiness performance is depicted in Figure 3 In this context a mediating effectis said to exist when

both mediating paths b5 b6 are signiregcant and

the direct effect of the exogenous variable on the outcome variablevanishes (complete mediational effect) or is signiregcantly lower (partialmediational effect) when a mediator variable is introduced in the model

Condition (2) amounts to b1 in Figure 3 becoming zero or signiregcantly less thatthan value reported for Figure 2

We used the mediated model depicted in Figure 3 to test for mediatingeffects of innovation degree innovation performance and customer loyaltyseparately on the impact of market orientation on business economicperformance We found that when either innovation performance or innovation

Figure 3Mediated model

Businesseconomic

performance

297

degree were used as mediating variable all the mediating paths weresigniregcant and that direct path from market orientation to businessperformance was not signiregcantly different from zero b1 = 004 t = 162for innovation performance b1 = 004 t = 186 for innovation degree Hencetaken separately both innovation degree and innovation performancecompletely mediate the impact of market orientation on businessperformance After regxing b1 at zero we re-estimated these two mediationalmodels The resulting parameter estimates and GFIs for these two models areshown in Table III On the other hand customer loyalty was found not to havea mediational effect between market orientation and business performanceThe parameter estimates and GFIs for this model are also given in Table III

As can be seen in Table III the mediating paths are signiregcant but thedirect path b1 is signiregcantly different from zero at a = 001 Furthermore a 99percent conregdence interval for the value for b1 reported in Table II (003 009)includes the value of b1 estimated in the mediational model using customerloyalty 004 Hence this variable does not even partially mediate on the impactof market orientation on business economic performance The standardizeddirect impact of market orientation on business performance (0408) is morethan twice the standardized impact of market orientation conveyed throughcustomer loyalty (0191)

The percentage of variance of business economic performance explained bythe model when innovation performance innovation degree or customer loyaltyare used as mediators is very similar (465 percent 453 percent and 437percent respectively)

H3The full model to be regtted corresponding to the hypothesis depicted in Figure 1is presented in Figure 3 The parameter estimates and goodness of regt testcorresponding to this model are given in Table IV (see also Figure 4) As can beseen in this Table the model regts these data very well

All the postulated relationships were found to be signiregcant at an a = 001Lagrange multiplier tests indicated that the regt of the model would notsigniregcantly improve by

adding a direct effect of market orientation to business performance nor

adding a direct effect of innovation degree on economic performance

Result (1) is in accordance with the results discussed above where we saw thatinnovation degree and innovation performance even when taken separatelycompletely mediate the impact of market orientation on business performanceResult (2) conregrms our hypothesis that innovation performance completelymediates the impact of innovation degree on business performance

Given that all effects of market orientation on business performance gothrough either innovation degree-innovation performance or through customer

IJSIM143

298

Par

amet

eres

tim

ates

Goo

dnes

sof

regt

Par

amet

erV

alue

Par

amet

erV

alue

Index

Val

ue

Innova

tion

per

form

ance

as

med

iato

rb 1

0(reg

xed

)u

117

713

[10

0](2

241

)M

FF

X2

560

(p=

035

)b 2

155

[04

0](0

42)

u2

241

4[0

84]

(71

1)S-B

X2

380

(p=

058

)b 3

263

[04

8](0

67)

u3

431

6[0

77]

(45

5)B

X2

429

(p=

051

)b 4

248

[05

9](o

81)

u4

211

3[0

65]

(45

5)df

5R

MSE

A0

03b 5

023

[05

8](0

03)

u5

180

3[0

67]

(28

0)SR

MSR

005

GF

I0

98b 6

018

[06

8](0

05)

CF

I0

99

Innova

tion

deg

ree

as

med

iato

rb

10

(regxed

)u

117

713

[10

0](2

241

)M

FF

X2

717

(p=

021

)b 2

169

[05

0](0

36)

u2

234

0[0

82]

(73

8)S-B

X2

497

(p=

043

)b 3

282

[05

1](0

82)

u3

415

6[0

74]

(45

0)B

X2

613

(p=

029

)b 4

226

[05

4](0

79)

u4

232

9[0

71]

(46

0)df

5R

MSE

A0

06b 5

012

[05

5](0

02)

u5

590

[06

9](0

78)

SR

MSR

005

GF

I0

98b 6

031

[06

7](0

09)

CF

I0

98

Cust

om

erlo

yalty

as

med

iato

rb 1

004

[04

1](0

02)

u1

177

13[1

00]

(22

41)

MF

FX

22

30(p

=0

68)

b 21

61[0

40]

(05

1)u

224

05

[08

4](7

11)

S-B

X2

155

(p=

082

)b 3

252

[04

5](0

74)

u3

447

5[0

80]

(45

5)B

X2

309

(p=

054

)b 4

265

[06

2](0

92)

u4

201

3[0

62]

(45

5)R

MSE

A0

0df

4b 5

015

[05

7](0

02)

u5

863

[06

8](1

04)

SR

MSR

002

GF

I0

99b 6

013

[03

4](0

05)

CF

I1

00

Note

s

Rob

ust

asym

pto

tic

stan

dar

der

rors

are

pro

vid

edin

par

enth

eses

st

andar

diz

edpar

amet

eres

tim

ates

are

pro

vid

edin

squar

ebra

cket

sM

FF

X2=

Min

imum

regt

funct

ion

chi-sq

uar

eSB

X2=

Sat

orra

-Ben

tler

scal

edch

i-sq

uar

eB

X2=

Bro

wnersquo

sch

i-sq

uar

eco

rrec

ted

for

non

-nor

mal

ity

RM

SE

A=

root

mea

nsq

uar

eder

ror

ofap

pro

xim

atio

n

SR

MSR

=st

andar

diz

edro

otm

ean

squar

edre

sidual

G

FI

=goo

dnes

sof

regt

index

C

FI

=co

mpar

ativ

ereg

tin

dex

Table IIIEstimation results

for the modeldepicted in Figure 3

Businesseconomic

performance

299

Par

amet

eres

tim

ates

Goo

dnes

sof

regt

R2

Par

amet

erV

alue

Par

amet

erV

alue

Index

Val

ue

Var

iable

Val

ue

b 10

12[0

55]

(22

41)

u1

177

13[1

00]

(00

2)M

FF

X2

105

9(p

=0

56)

Mar

ket

shar

e0

16b 2

013

[03

4](0

04)

u2

590

[06

9](0

78)

S-B

X2

785

(p=

080

)P

rem

ium

gro

wth

019

b 30

15[0

57]

(00

2)u

314

48

[05

4](2

59)

BX

212

67

(p=

039

)P

roreg

tabilit

y0

38b 4

078

[04

4](0

15)

u4

863

[06

8](1

04)

df

12B

usi

nes

sper

form

ance

056

RM

SE

A0

0b 5

016

[05

4](0

05)

u5

239

1[0

84]

(72

3)SR

MSR

005

Innov

atio

nper

form

ance

046

b 60

16[0

37]

(00

7)u

645

09

[08

1](1

271

)G

FI

098

Innov

atio

ndeg

ree

031

b 71

42[0

40]

(04

5)u

720

12

[06

2](4

60)

CF

I1

00C

ust

omer

loyal

ty0

32b 8

217

[04

4](0

64)

b 92

32[0

62]

(08

5)

Note

sR

obust

asym

pto

tic

stan

dar

der

rors

are

pro

vid

edin

par

enth

eses

st

andar

diz

edpar

amet

eres

tim

ates

are

pro

vid

edin

squar

ebra

cket

sM

FF

X2=

Min

imum

regtfu

nct

ion

chi-sq

uar

eSB

X2=

Sat

orra

-Ben

tler

scal

edch

i-sq

uar

eB

X2=

Bro

wnersquo

sch

i-sq

uar

eco

rrec

ted

for

non

-nor

mal

ity

RM

SE

A=

root

mea

nsq

uar

eder

ror

ofap

pro

xim

atio

n

SR

MSR

=st

andar

diz

edro

otm

ean

squar

edre

sidual

G

FI

=goo

dnes

sof

regt

index

C

FI

=co

mpar

ativ

ereg

tin

dex

R

2=

squar

edm

ult

iple

corr

elat

ions

for

endog

enou

svar

iable

s

Table IVEstimation resultsfor the modeldepicted in Figure 4

IJSIM143

300

loyalty a question arises as to the relative importance of the speciregc effectsgoing through these variables The standardized speciregc effect (computed as inBollen 1987) going through innovation degree and innovation performance is031 and 021 going through customer loyalty Hence the impact of marketorientation going through innovation is 50 percent more than that goingthrough customer loyalty

We can also see in Table IV that over 30 percent of the variance of theintermediate variables (innovation degree innovation performance andcustomer loyalty) are explained by market orientation In fact almost 50percent (464 percent to be exact) of innovation performance is explained bymarket orientation Furthermore note that the percentage of variance ofbusiness performance explained by the model is 561 percent a 52 percentincrement over what is explained by market orientation alone (see Table II) andover a 20 percent increment over what is explained by the mediational modelsconsidered previously Hence the inclusion of all three intermediate variables inthe model improves considerably our prediction of business performance

Furthermore we observe in Table IV that the direct effect of marketorientation on all three intermediate variables appear to be equal Also thedirect effects of customer loyalty and of innovation performance on economicperformance appear to be equal We re-estimated the model to test theseconstraints obtaining b1 = b2 = b3 = 0133 b5 = b6 = 0157 Satorra-BentlerX 2(15) = 8849 p = 0885 CFI = 100 GFI = 0975 SRMR = 006

ConclusionsMarket orientation can be deregned as a strategy used to reach a sustainablecompetitive advantage based on the generation and use of information within

Figure 4Full model

Businesseconomic

performance

301

organizations and on the selection of markets to be satisreged In thisframework we believe that competitive advantage results from the use ofresources and capabilities to generate differential satisfaction in proregtablemarkets Sustainability is achieved because the performance of the marketorientationrsquos behaviors requires complex organizational knowledge that cannoteasily be imitated by competitors Thus we hypothesize that the satisfaction ofproregtable markets permits the regrm to achieve a psychologically differentialposition that leads to brand loyalty and thus to higher proregts Previous studieshave

found a clear impact of market orientation on economic performance

assessed the effects of market orientation on innovation and

investigated the relationship between market orientation and relationshipmarketing variables

The present study proposes and tests a model that integrates both streams ofresearch innovation and relationship marketing More research is needed toinvestigate the role of customer satisfaction within this framework

In our necessarily partial model innovation degree innovation performanceand customer loyalty are used as intermediate variables on the effect of marketorientation on business performance Our results suggest that the addition ofthese variables help improve our predictions of business economic performance52 percent over what is explained by market orientation alone Also we foundthat innovation degree and innovation performance each taken separatelycompletely mediate the effect of market orientation on economic performanceFurthermore the impact of innovation degree on economic performance iscompletely channeled through innovation performance Customer loyalty byitself does not meditate the impact of market orientation on economicperformance but when considered along with innovation degree andinnovation performance it conveys some of the effects of market orientationon business performance This seemingly contradictory result arises from thefact that all three intermediate variables are interrelated

Our results should not be taken to imply that there are no other variablesmediate the effect of market orientation on economic performance We believethat other variables that have not been taken into account in this study such asproduct quality and customer satisfaction may also be signiregcant mediatorsHowever our results do suggest that whenever innovation degree andinnovation performance are included in the model as intermediate variablesthe effects of market orientation on business performance will mostly beconveyed through these variables Also in this study we have adopted thecurrently most widely accepted approach to measuring customer loyalty whichis based on behavioral loyalty measures (eg share of category requirementsrenew the policy probability the average customer last in the companyportfolio) It would be worth extending the present study by including not only

IJSIM143

302

behavioral measures of customer loyalty but also attitudinal measures alongthe lines of Dick and Basu (1994)

In our opinion two important contributions of the present research are outuse of objective measures of business performance and our focusing oninternational markets Despite the growing role of globalization and marketintegration and despite the increasing internationalization of corporationsmost studies on market orientation have focused on domestic markets (withnotable exceptions such as Selnes et al 1996 Webster 1994) Similarly moststudies on product innovation have also focused on domestic markets There isa lack of research yielding empirical support to the validity in an internationalsetting to research results obtained in domestic markets To regll this gap wetargeted the European Union market

Our study focused on a single industry the insurance sector Our sampleaccounted for 22 percent of the companies and 17 percent of the insurancepremiums in the targeted market An advantage of our single-industryapproach is that (with obvious reservations arising from the non-experimentalnature of our study and the fact that our sample should not be considered tohave been obtained at random) we can draw tentative predictions from ourmodel concerning the impact of market orientation on economic performance ininsurance companies operating in the European Union market An evidentdrawback of the single-industry approach adopted here is that it is not clearhow the present results extrapolate to other industries even when operating inthe same market

We have found that within the European Union insurance regrms thatconstantly monitor the evolution of current and potential customerrsquos needsmodify the attributes of the products to adapt them to the distributorsrequirements analyze competitorsrsquo marketing policy and products and knowbest of environmental trends especially technological and legal changes aremore likely to develop more new products have their new products accepted bythe market and obtain more loyal customers In turn increased customerloyalty and increased new product success will result in improved economicperformance However these conclusions must be taken with some caution aswe have used overall measures of innovation and innovation performanceFurther research is needed that takes into account the various aspects thatconstitute innovation

Previous studies have concluded that insurance companies still seethemselves as being product-focused and the industry as a whole is generallydistribution driven (Sodano 2000 Lambin 1996) This lack of closeness withcustomers has in turn contributed to the industry lack of product innovationsand product portfolios of commodity products that only compete on price Nowthe most successful regrms are redirecting their focus to the market needs andthey are beginning to exploit customer data and use market research togenerate ideas for designing new products In this context these companies

Businesseconomic

performance

303

now face choosing among too many new service options The managerialimplication of our study is that by enhancing their market orientation regrmswill know and service its customers better Thus they will generate moreinnovations by adopting a market-based product development process Alsoincreasing levels of market orientation enable regrms to discriminate more easilywhich new products have a higher success probability thus enhancing both theefregcacy and efregciency of new product development

Note

1 We considered employing the total volume of premiums for each insurance company as anadditional indicator although it seemed to us to be a better indicator of a companyrsquos sizerather than of its performance As we suspected the total volume of premiums wasuncorrelated with any of the variables considered in this study except for the companyrsquosmarket share (r = 024 p 001) Hence it is clear that volume of premiums should not beused as an indicator of insurance companiesrsquo performance

References

Armstrong JS and Overton T (1977) ordfEstimating nonresponse bias in mail surveysordm Journalof Marketing Research Vol 14 pp 396-402

Atuahene-Gima K (1996) ordfMarket orientation and innovationordm Journal of Business ResearchVol 35 pp 93-103

Baker TL Simpson PM and Sigauw JA (1999) ordfThe impact of suppliersrsquo perceptions ofreseller market orientation on key relationship constructsordm Journal of the Academy ofMarketing Science Vol 27 No 1 pp 50-7

Baron RM and Kenny DA (1986) ordfThe moderator-mediator variable distinction in socialpsychological research conceptual strategic and statistical considerationsordm Journal ofPersonality and Social Psychology Vol 51 pp 1173-82

Bentler PM (1990) ordfComparative regt indexes in structural modelsordm Psychological BulletinVol 107 pp 238-46

Bhuian SN (1997) ordfExploring market orientation in banks an empirical examination in SaudiArabiaordm The Journal of Service Marketing Vol 11 No 5 pp 317-28

Bollen KA (1987) ordfTotal direct and indirect effects in structural equation modelsordm inClogg CC (Ed) Sociological Methodology American Sociological AssociationWashington DC pp 37-69

Browne MW (1984) ordfAsymptotically distribution free methods for the analysis of covariancestructuresordm British Journal of Mathematical and Statistical Psychology Vol 37 pp 62-83

Calantone RJ di Benedetto AC and Bhoovaraghavan S (1994) ordfExamining the relationshipbetween degree of innovation and new product successordm Journal of Business ResearchVol 30 pp 143-8

Calantone RJ Schmidt JB and Song MX (1996) ordfControllable factors of new productssuccess a cross-national comparisonordm Marketing Science Vol 15 No 4 pp 341-58

Cooper RC (1994) ordfNew product the factors that drive successordm International MarketingReview Vol 11 No 1 pp 60-76

Day G (1992) ordfMarketingrsquos contribution to the strategy dialogueordm Journal of the Academy ofMarketing Science Vol 20 Fall pp 323-9

IJSIM143

304

inferences can be drawn on the target population based on therepresentativeness of the sample used

The confounding of within-industry and between industry variation is notthe only threat to the validity of inferences drawn on the relationship betweenmarket orientation and economic performance A second threat is the noiseintroduced by environmental variables such as market turbulence marketgrowth rate buyer and supplier power and competitive intensity on businessperformance A standard approach to minimize this threat is to focus theresearch on a single market The drawback of this approach is that we are notable to capture regrmsrsquo behavior in facing increasing globalization and marketintegration As a compromise between these two ends the present studytargets the European Union market In this market the key characteristics of asingle market are preserved but it is also an environment in which we canpresently observe how regrms struggle in meeting the challenges ofinternationalization and market integration

A third threat to the validity of inferences drawn on the relationship betweenmarket orientation and economic performance lies in the use of subjectivemeasures of economic performance (ie managersrsquo evaluations of theircompaniesrsquo performance) Positive effects of market orientation on economicperformance have been reported when subjective assessments of performancewere used However when objective measures of economic performance havebeen used mixed results emerged For instance Ruekert (1992) and Lambin(1996) report a positive relationship between market orientation and objectivelymeasured economic performance However Bhuian (1997) Jaworski and Kohli(1993) and Selnes et al (1996) failed to regnd any signiregcant relationshipClearly when market orientation and economic performance are concurrentlyassessed by the regrmsrsquo managers a perceptual bias may be introduced A casein point Van Bruggen and Smidts (1995) found within one single company

Figure 1Hypothetical model

Businesseconomic

performance

291

(which has only one performance) a substantial degree of variation insubjective performance assessments In fact they report a positive relationshipbetween market orientation and judgments about the company performancewithin a single company As they point out ordfit might be that managers have amore positive view of their companyrsquos market orientation when they perceivetheir company to be performing wellordm (Van Bruggen and Smidts 1995 p 13)Hence it is important to employ objective measures of economic performanceand we shall do so in the present study

Market orientation in the services sector the European insurance industryThe insurance sector is of particular interest from a market orientationviewpoint as it works with intangible commodities in which service qualityand customer orientation are crucial elements The competitive characteristicsgenerated by the European Union provide an additional interest in studyingmarket orientation in this area The insurance sector in Europe has traditionallyoperated subject to strict regulations and strong protection from internationalcompetition However for some years now the European Commission has beenworking on the liberalization of this sector Effective implementation of this hasbrought about a major increase in competition within the sector and hasprovoked a major restructuring of insurance companies and groups Thecompetitive climate in Europe has also been inmacruenced by a downside in theeconomic cycle and changes in consumer behavior European customers nowshow greater service expectations and less loyalty As a result rivalry amongcompetitors is increasing as is the importance of competitive strategies adaptedto this sectorrsquos needs In this background the degree of orientation toward thecustomer distributors competition and the general socio-economicenvironment is becoming an increasingly important area of study not onlyfor academics but also for the business world

Lado et al (1998ab) have investigated quite extensively the marketorientation of insurance regrms within the European Union These authors havenot found signiregcant mean differences in market orientation by countryFurthermore they report substantial agreement between the factor structuresof market orientation across countries Thus it seems that the Europeaninsurance sector can be considered a homogenous population with respect tomarket orientation

DataThe population universe considered in this article is deregned as the set ofinsurance companies operating in the European Union which meet thefollowing conditions

they operate in private insurance or ordfmass insuranceordm

they have a market share of more than 005 percent and

their management is independent

IJSIM143

292

The list of European insurance companies was taken from the Financial TimesYearbook for 1996

It was assumed that senior executives were the people best qualireged to assessthe companyrsquos market orientation as well as their innovation degree innovationperformance and customersrsquo loyalty Therefore information from these variableswas gathered via a postal questionnaire submitted to the senior executive in eachof the 554 companies comprising the target population

We obtained 122 valid questionnaires giving a response rate of 22 percentThis sample accounts for over 17 percent of total insurance premiums in theEuropean Union

In order to assess response bias the questionnaires were divided intoquartiles on the basis of reception date (Armstrong and Overton 1977) Ananalysis of early and late responses did not indicate any signiregcant differencein terms of means and covariances

MeasuresBusiness economic performance is a complex construct with multiple possibleobserved indicators Here we measure this construct using three remacrectiveindicators[1]

(1) domestic market share (Jaworski and Kohli 1993 DeshpandeAcirc et al 1993Selnes et al 1996 Greenley and Foxall 1997 1998)

(2) premium growth (which is equivalent to sales growth for the insurancecompanies business (Slater and Narver 1994 Narver and Slater 1990Ruekert 1992 Greenley and Foxall 1997 1998) and

(3) proregtability per year averaged over the last three years(similar to thereturn on investment (ROI) rate (Greenley and Foxall 1997 1998))

All three indicators were expressed as percentages These data were obtainedfrom the managers responding the questionnaire Their responses were carefullycontrasted with published regnancial information (eg Reuters Insurance Brieregng)

Market orientation was measured using the Market Orientation Scale-Revised(MOS-R) This scale is a shortened version of the MOS validated by Lado et al(1998a) in the population of insurance companies of Belgium and Spain Ladoet al (1998a) shortened the original MOS scale while extending the previousvalidation study to target all insurance companies operating in the EuropeanUnion In the Appendix we provide the 30 items composing the MOS-R Eachitem is to be rated on a ten-point Likert-type scale ranging from 0 (completedisagreement) to 10 (complete agreement)

Innovation degree and innovation performance were assessed by means ofmulti-item questionnaires akin to Miller and Friesenrsquos (1982) Innovationperformance was measured by a four-item questionnaire regarding the successof a new productservice (deregned as an improved product a product extensionor a new product line) introduced by the company The questions involved

Businesseconomic

performance

293

whether the new productservice had succeeded in meeting the sales growthmarket share and proregt objectives set up by the company

Innovation degree was assessed by a three-item questionnaire that inquiredthe rate of new productsservices introduced by the company relative tocompetitors the amount of new productsservices marketed by the companyover the past three years and the nature of change of the new productsservices

Finally we used a four-item questionnaire based on existing literature (egDick and Basu 1994 Javalgi and Moberg 1997) to evaluate managersrsquoperceptions of their customersrsquo loyalty The questionnaire taps on theproportion of their customersrsquo insurance premiums taken on by the companythe average time a customer remains in the companyrsquos portfolio the probabilityof a customer renewing a premium and the overall perception of the companycustomersrsquo loyalty

Scale scores for innovation degree innovation performance and customerloyalty were obtained as an unweighted sum of the corresponding items Sincein all three cases Likert-type items on a 0-7 scale were used scale scores forthese variables range from 0-27 0-27 and 0-28 respectively For marketorientation we computed a score for each of its facets as an unweighted sum ofthe corresponding items Then a global market orientation score was obtainedas a sum of the facetsrsquo scores inversely weighted by their number of itemsHence this market orientation score assigns equal weights to each its facetsand ranges from 0-90

The scalesrsquo reliability (as assessed by coefregcient alpha) in this sample were088 (market orientation) 070 (innovation degree) 091 (innovationperformance) and 076 (customer loyalty) The means standard deviationsand correlations among all variables considered in this study are presented inTable I As can be seen in this table the three indicators of business economicperformance are signiregcantly but not largely correlated (the correlations rangefrom 020 to 029) The correlations among the hypothesized intermediatevariables (innovation degree innovation performance and customer loyalty)are not high except for innovation degree and innovation performance whichshare 36 percent of their variance The correlations of market orientation withthe intermediate variables appear signiregcantly larger (they range from 055 to058) than with the dependent variables (they range from 023 to 036) We alsoobserve in Table I that managers report on average a high degree of innovationin their businesses not so high a level of customer loyalty and a level ofinnovation performance just at the scale mean The average self-reporteddegree of market orientation is 56 on a 0-90 scale

MethodAll hypotheses were contrasted using covariance structure analysis asimplemented in LISREL 850 (see Joreskog et al 1999) Since all three indicatorsof business performance are highly positively skewed and present a high

IJSIM143

294

degree of kurtosis throughout this paper rather than attempting to transformthese variables to near-normality we shall employ an estimation approach thatis robust to non-normality of the observed variables The parameter estimateswere obtained using maximum likelihood estimation with standard errorssuitable for non-normal data (Joreskog et al 1999 Equation A24) and twotest statistics were used to assess the goodness of regt (GFI) of the model theSatorra-Bentler scaled chi-squared statistic (Satorra and Bentler 1988Equation 41) and Brownersquos (1984 Equation 220a) chi-squared statisticcorrected for non-normality To evaluate better the goodness of regt of thismodel several additional indices will also provided the root mean squarederror of approximation (RMSEA (Steiger 1990)) the standardized root meansquared residual (SRMSR (Joreskog et al 1999)) the GFI (Tanaka and Huba1985) and the Comparative Fit Index (CFI) using the independence model asbaseline (Bentler 1990 see also McDonald and Marsh 1990) Adequate to goodregt is suggested by RMSEA and SRMSR values approaching 005 For the GFIand the CFI indices values between 090 and 100 indicate adequate to excellentregt (but see Hu and Bentler 1999)

ResultsH1The model used to estimate the effects of market orientation on insurancebusinessesrsquo performance consists of a latent variable representing economicperformance with three indicators (market share premium growth andproregtability) and a single exogenous variable (market orientation) This model

MST PG PROF INNODR INNPERF LOYAL MO

MST 100PG 020 100PROF 023 029 100INNODR 030 034 035 1000INNPERF 027 032 041 062 100LOYAL 025 019 038 040 038 100MO 024 030 036 055 058 057 100x 455 788 626 1825 1422 1996 5628Std 535 749 571 356 292 520 1331

NotesN = 122 all correlations are signiregcant (a = 001) except those marked by which are onlysigniregcant at an a = 005MS = market share PG = premium growth PROF = proregtability INNODR = innovation degreeINNPERF = innovation performance LOYAL = customer loyalty MO = market orientation

Table IMeans standard

deviations andinter-correlations

Businesseconomic

performance

295

is depicted in Figure 2 The parameter estimates and GFIs for this model aregiven in Table II The model shows a good regt although note that it only hastwo degrees of freedom According to the model the best objective indicator ofbusiness economic performance is proregtability per year over 34 percent of itsvariance is accounted for by the model The standardized regressioncoefregcients reveal that proregtability per year is the best objective indicator of

Parameter estimates Goodness of regt R 2

Parameter Value Index Value Variable Value

b1 006 [061] (002) MFF X 2001 ( p = 099) Market share 015

b2 166 [039] (049) S-B X 2001 ( p = 099) Premium growth 024

b3 293 [049] (078) B X 2001 ( p = 099) Proregtability 034

b4 267 [059] (084) RMSEA 001 Business performance 037u1 17713 [100] (2241) df 2u2 2427 [085] (713) SRMSR 001u3 4248 [076] (1364) GFI 100u4 2141 [066] (515) CFI 100

NotesRobust asymptotic standard errors are provided in parentheses standardized parameterestimates are provided in square bracketsMFF X 2 = Minimum regt function chi-square SB X 2 = Satorra-Bentler scaled chi-squareB X 2 = Brownersquos chi-square corrected for non-normality RMSEA = root mean squared error ofapproximation SRMSR = standardized root mean squared residual GFI = goodness of regt indexCFI = comparative regt index R 2 = squared multiple correlations for endogenous variables

Table IIEstimation resultsfor the modeldepicted in Figure 2

Figure 2Market orientation aspredictor of economicperformance

IJSIM143

296

overall business performance Finally according to the model almost 37percent of overall business economic performance is accounted for by thedegree of market orientation

An inspection of the total effects of market orientation on the indicators ofeconomic performance suggests that unit increments of market orientation asmeasured by the MOS-R are associated with 0095 0168 and 0153 incrementsin domestic market share premium growth and proregtability per year averagedover the last three years respectively

H2A mediated model for the relationship between market orientation andbusiness performance is depicted in Figure 3 In this context a mediating effectis said to exist when

both mediating paths b5 b6 are signiregcant and

the direct effect of the exogenous variable on the outcome variablevanishes (complete mediational effect) or is signiregcantly lower (partialmediational effect) when a mediator variable is introduced in the model

Condition (2) amounts to b1 in Figure 3 becoming zero or signiregcantly less thatthan value reported for Figure 2

We used the mediated model depicted in Figure 3 to test for mediatingeffects of innovation degree innovation performance and customer loyaltyseparately on the impact of market orientation on business economicperformance We found that when either innovation performance or innovation

Figure 3Mediated model

Businesseconomic

performance

297

degree were used as mediating variable all the mediating paths weresigniregcant and that direct path from market orientation to businessperformance was not signiregcantly different from zero b1 = 004 t = 162for innovation performance b1 = 004 t = 186 for innovation degree Hencetaken separately both innovation degree and innovation performancecompletely mediate the impact of market orientation on businessperformance After regxing b1 at zero we re-estimated these two mediationalmodels The resulting parameter estimates and GFIs for these two models areshown in Table III On the other hand customer loyalty was found not to havea mediational effect between market orientation and business performanceThe parameter estimates and GFIs for this model are also given in Table III

As can be seen in Table III the mediating paths are signiregcant but thedirect path b1 is signiregcantly different from zero at a = 001 Furthermore a 99percent conregdence interval for the value for b1 reported in Table II (003 009)includes the value of b1 estimated in the mediational model using customerloyalty 004 Hence this variable does not even partially mediate on the impactof market orientation on business economic performance The standardizeddirect impact of market orientation on business performance (0408) is morethan twice the standardized impact of market orientation conveyed throughcustomer loyalty (0191)

The percentage of variance of business economic performance explained bythe model when innovation performance innovation degree or customer loyaltyare used as mediators is very similar (465 percent 453 percent and 437percent respectively)

H3The full model to be regtted corresponding to the hypothesis depicted in Figure 1is presented in Figure 3 The parameter estimates and goodness of regt testcorresponding to this model are given in Table IV (see also Figure 4) As can beseen in this Table the model regts these data very well

All the postulated relationships were found to be signiregcant at an a = 001Lagrange multiplier tests indicated that the regt of the model would notsigniregcantly improve by

adding a direct effect of market orientation to business performance nor

adding a direct effect of innovation degree on economic performance

Result (1) is in accordance with the results discussed above where we saw thatinnovation degree and innovation performance even when taken separatelycompletely mediate the impact of market orientation on business performanceResult (2) conregrms our hypothesis that innovation performance completelymediates the impact of innovation degree on business performance

Given that all effects of market orientation on business performance gothrough either innovation degree-innovation performance or through customer

IJSIM143

298

Par

amet

eres

tim

ates

Goo

dnes

sof

regt

Par

amet

erV

alue

Par

amet

erV

alue

Index

Val

ue

Innova

tion

per

form

ance

as

med

iato

rb 1

0(reg

xed

)u

117

713

[10

0](2

241

)M

FF

X2

560

(p=

035

)b 2

155

[04

0](0

42)

u2

241

4[0

84]

(71

1)S-B

X2

380

(p=

058

)b 3

263

[04

8](0

67)

u3

431

6[0

77]

(45

5)B

X2

429

(p=

051

)b 4

248

[05

9](o

81)

u4

211

3[0

65]

(45

5)df

5R

MSE

A0

03b 5

023

[05

8](0

03)

u5

180

3[0

67]

(28

0)SR

MSR

005

GF

I0

98b 6

018

[06

8](0

05)

CF

I0

99

Innova

tion

deg

ree

as

med

iato

rb

10

(regxed

)u

117

713

[10

0](2

241

)M

FF

X2

717

(p=

021

)b 2

169

[05

0](0

36)

u2

234

0[0

82]

(73

8)S-B

X2

497

(p=

043

)b 3

282

[05

1](0

82)

u3

415

6[0

74]

(45

0)B

X2

613

(p=

029

)b 4

226

[05

4](0

79)

u4

232

9[0

71]

(46

0)df

5R

MSE

A0

06b 5

012

[05

5](0

02)

u5

590

[06

9](0

78)

SR

MSR

005

GF

I0

98b 6

031

[06

7](0

09)

CF

I0

98

Cust

om

erlo

yalty

as

med

iato

rb 1

004

[04

1](0

02)

u1

177

13[1

00]

(22

41)

MF

FX

22

30(p

=0

68)

b 21

61[0

40]

(05

1)u

224

05

[08

4](7

11)

S-B

X2

155

(p=

082

)b 3

252

[04

5](0

74)

u3

447

5[0

80]

(45

5)B

X2

309

(p=

054

)b 4

265

[06

2](0

92)

u4

201

3[0

62]

(45

5)R

MSE

A0

0df

4b 5

015

[05

7](0

02)

u5

863

[06

8](1

04)

SR

MSR

002

GF

I0

99b 6

013

[03

4](0

05)

CF

I1

00

Note

s

Rob

ust

asym

pto

tic

stan

dar

der

rors

are

pro

vid

edin

par

enth

eses

st

andar

diz

edpar

amet

eres

tim

ates

are

pro

vid

edin

squar

ebra

cket

sM

FF

X2=

Min

imum

regt

funct

ion

chi-sq

uar

eSB

X2=

Sat

orra

-Ben

tler

scal

edch

i-sq

uar

eB

X2=

Bro

wnersquo

sch

i-sq

uar

eco

rrec

ted

for

non

-nor

mal

ity

RM

SE

A=

root

mea

nsq

uar

eder

ror

ofap

pro

xim

atio

n

SR

MSR

=st

andar

diz

edro

otm

ean

squar

edre

sidual

G

FI

=goo

dnes

sof

regt

index

C

FI

=co

mpar

ativ

ereg

tin

dex

Table IIIEstimation results

for the modeldepicted in Figure 3

Businesseconomic

performance

299

Par

amet

eres

tim

ates

Goo

dnes

sof

regt

R2

Par

amet

erV

alue

Par

amet

erV

alue

Index

Val

ue

Var

iable

Val

ue

b 10

12[0

55]

(22

41)

u1

177

13[1

00]

(00

2)M

FF

X2

105

9(p

=0

56)

Mar

ket

shar

e0

16b 2

013

[03

4](0

04)

u2

590

[06

9](0

78)

S-B

X2

785

(p=

080

)P

rem

ium

gro

wth

019

b 30

15[0

57]

(00

2)u

314

48

[05

4](2

59)

BX

212

67

(p=

039

)P

roreg

tabilit

y0

38b 4

078

[04

4](0

15)

u4

863

[06

8](1

04)

df

12B

usi

nes

sper

form

ance

056

RM

SE

A0

0b 5

016

[05

4](0

05)

u5

239

1[0

84]

(72

3)SR

MSR

005

Innov

atio

nper

form

ance

046

b 60

16[0

37]

(00

7)u

645

09

[08

1](1

271

)G

FI

098

Innov

atio

ndeg

ree

031

b 71

42[0

40]

(04

5)u

720

12

[06

2](4

60)

CF

I1

00C

ust

omer

loyal

ty0

32b 8

217

[04

4](0

64)

b 92

32[0

62]

(08

5)

Note

sR

obust

asym

pto

tic

stan

dar

der

rors

are

pro

vid

edin

par

enth

eses

st

andar

diz

edpar

amet

eres

tim

ates

are

pro

vid

edin

squar

ebra

cket

sM

FF

X2=

Min

imum

regtfu

nct

ion

chi-sq

uar

eSB

X2=

Sat

orra

-Ben

tler

scal

edch

i-sq

uar

eB

X2=

Bro

wnersquo

sch

i-sq

uar

eco

rrec

ted

for

non

-nor

mal

ity

RM

SE

A=

root

mea

nsq

uar

eder

ror

ofap

pro

xim

atio

n

SR

MSR

=st

andar

diz

edro

otm

ean

squar

edre

sidual

G

FI

=goo

dnes

sof

regt

index

C

FI

=co

mpar

ativ

ereg

tin

dex

R

2=

squar

edm

ult

iple

corr

elat

ions

for

endog

enou

svar

iable

s

Table IVEstimation resultsfor the modeldepicted in Figure 4

IJSIM143

300

loyalty a question arises as to the relative importance of the speciregc effectsgoing through these variables The standardized speciregc effect (computed as inBollen 1987) going through innovation degree and innovation performance is031 and 021 going through customer loyalty Hence the impact of marketorientation going through innovation is 50 percent more than that goingthrough customer loyalty

We can also see in Table IV that over 30 percent of the variance of theintermediate variables (innovation degree innovation performance andcustomer loyalty) are explained by market orientation In fact almost 50percent (464 percent to be exact) of innovation performance is explained bymarket orientation Furthermore note that the percentage of variance ofbusiness performance explained by the model is 561 percent a 52 percentincrement over what is explained by market orientation alone (see Table II) andover a 20 percent increment over what is explained by the mediational modelsconsidered previously Hence the inclusion of all three intermediate variables inthe model improves considerably our prediction of business performance

Furthermore we observe in Table IV that the direct effect of marketorientation on all three intermediate variables appear to be equal Also thedirect effects of customer loyalty and of innovation performance on economicperformance appear to be equal We re-estimated the model to test theseconstraints obtaining b1 = b2 = b3 = 0133 b5 = b6 = 0157 Satorra-BentlerX 2(15) = 8849 p = 0885 CFI = 100 GFI = 0975 SRMR = 006

ConclusionsMarket orientation can be deregned as a strategy used to reach a sustainablecompetitive advantage based on the generation and use of information within

Figure 4Full model

Businesseconomic

performance

301

organizations and on the selection of markets to be satisreged In thisframework we believe that competitive advantage results from the use ofresources and capabilities to generate differential satisfaction in proregtablemarkets Sustainability is achieved because the performance of the marketorientationrsquos behaviors requires complex organizational knowledge that cannoteasily be imitated by competitors Thus we hypothesize that the satisfaction ofproregtable markets permits the regrm to achieve a psychologically differentialposition that leads to brand loyalty and thus to higher proregts Previous studieshave

found a clear impact of market orientation on economic performance

assessed the effects of market orientation on innovation and

investigated the relationship between market orientation and relationshipmarketing variables

The present study proposes and tests a model that integrates both streams ofresearch innovation and relationship marketing More research is needed toinvestigate the role of customer satisfaction within this framework

In our necessarily partial model innovation degree innovation performanceand customer loyalty are used as intermediate variables on the effect of marketorientation on business performance Our results suggest that the addition ofthese variables help improve our predictions of business economic performance52 percent over what is explained by market orientation alone Also we foundthat innovation degree and innovation performance each taken separatelycompletely mediate the effect of market orientation on economic performanceFurthermore the impact of innovation degree on economic performance iscompletely channeled through innovation performance Customer loyalty byitself does not meditate the impact of market orientation on economicperformance but when considered along with innovation degree andinnovation performance it conveys some of the effects of market orientationon business performance This seemingly contradictory result arises from thefact that all three intermediate variables are interrelated

Our results should not be taken to imply that there are no other variablesmediate the effect of market orientation on economic performance We believethat other variables that have not been taken into account in this study such asproduct quality and customer satisfaction may also be signiregcant mediatorsHowever our results do suggest that whenever innovation degree andinnovation performance are included in the model as intermediate variablesthe effects of market orientation on business performance will mostly beconveyed through these variables Also in this study we have adopted thecurrently most widely accepted approach to measuring customer loyalty whichis based on behavioral loyalty measures (eg share of category requirementsrenew the policy probability the average customer last in the companyportfolio) It would be worth extending the present study by including not only

IJSIM143

302

behavioral measures of customer loyalty but also attitudinal measures alongthe lines of Dick and Basu (1994)

In our opinion two important contributions of the present research are outuse of objective measures of business performance and our focusing oninternational markets Despite the growing role of globalization and marketintegration and despite the increasing internationalization of corporationsmost studies on market orientation have focused on domestic markets (withnotable exceptions such as Selnes et al 1996 Webster 1994) Similarly moststudies on product innovation have also focused on domestic markets There isa lack of research yielding empirical support to the validity in an internationalsetting to research results obtained in domestic markets To regll this gap wetargeted the European Union market

Our study focused on a single industry the insurance sector Our sampleaccounted for 22 percent of the companies and 17 percent of the insurancepremiums in the targeted market An advantage of our single-industryapproach is that (with obvious reservations arising from the non-experimentalnature of our study and the fact that our sample should not be considered tohave been obtained at random) we can draw tentative predictions from ourmodel concerning the impact of market orientation on economic performance ininsurance companies operating in the European Union market An evidentdrawback of the single-industry approach adopted here is that it is not clearhow the present results extrapolate to other industries even when operating inthe same market

We have found that within the European Union insurance regrms thatconstantly monitor the evolution of current and potential customerrsquos needsmodify the attributes of the products to adapt them to the distributorsrequirements analyze competitorsrsquo marketing policy and products and knowbest of environmental trends especially technological and legal changes aremore likely to develop more new products have their new products accepted bythe market and obtain more loyal customers In turn increased customerloyalty and increased new product success will result in improved economicperformance However these conclusions must be taken with some caution aswe have used overall measures of innovation and innovation performanceFurther research is needed that takes into account the various aspects thatconstitute innovation

Previous studies have concluded that insurance companies still seethemselves as being product-focused and the industry as a whole is generallydistribution driven (Sodano 2000 Lambin 1996) This lack of closeness withcustomers has in turn contributed to the industry lack of product innovationsand product portfolios of commodity products that only compete on price Nowthe most successful regrms are redirecting their focus to the market needs andthey are beginning to exploit customer data and use market research togenerate ideas for designing new products In this context these companies

Businesseconomic

performance

303

now face choosing among too many new service options The managerialimplication of our study is that by enhancing their market orientation regrmswill know and service its customers better Thus they will generate moreinnovations by adopting a market-based product development process Alsoincreasing levels of market orientation enable regrms to discriminate more easilywhich new products have a higher success probability thus enhancing both theefregcacy and efregciency of new product development

Note

1 We considered employing the total volume of premiums for each insurance company as anadditional indicator although it seemed to us to be a better indicator of a companyrsquos sizerather than of its performance As we suspected the total volume of premiums wasuncorrelated with any of the variables considered in this study except for the companyrsquosmarket share (r = 024 p 001) Hence it is clear that volume of premiums should not beused as an indicator of insurance companiesrsquo performance

References

Armstrong JS and Overton T (1977) ordfEstimating nonresponse bias in mail surveysordm Journalof Marketing Research Vol 14 pp 396-402

Atuahene-Gima K (1996) ordfMarket orientation and innovationordm Journal of Business ResearchVol 35 pp 93-103

Baker TL Simpson PM and Sigauw JA (1999) ordfThe impact of suppliersrsquo perceptions ofreseller market orientation on key relationship constructsordm Journal of the Academy ofMarketing Science Vol 27 No 1 pp 50-7

Baron RM and Kenny DA (1986) ordfThe moderator-mediator variable distinction in socialpsychological research conceptual strategic and statistical considerationsordm Journal ofPersonality and Social Psychology Vol 51 pp 1173-82

Bentler PM (1990) ordfComparative regt indexes in structural modelsordm Psychological BulletinVol 107 pp 238-46

Bhuian SN (1997) ordfExploring market orientation in banks an empirical examination in SaudiArabiaordm The Journal of Service Marketing Vol 11 No 5 pp 317-28

Bollen KA (1987) ordfTotal direct and indirect effects in structural equation modelsordm inClogg CC (Ed) Sociological Methodology American Sociological AssociationWashington DC pp 37-69

Browne MW (1984) ordfAsymptotically distribution free methods for the analysis of covariancestructuresordm British Journal of Mathematical and Statistical Psychology Vol 37 pp 62-83

Calantone RJ di Benedetto AC and Bhoovaraghavan S (1994) ordfExamining the relationshipbetween degree of innovation and new product successordm Journal of Business ResearchVol 30 pp 143-8

Calantone RJ Schmidt JB and Song MX (1996) ordfControllable factors of new productssuccess a cross-national comparisonordm Marketing Science Vol 15 No 4 pp 341-58

Cooper RC (1994) ordfNew product the factors that drive successordm International MarketingReview Vol 11 No 1 pp 60-76

Day G (1992) ordfMarketingrsquos contribution to the strategy dialogueordm Journal of the Academy ofMarketing Science Vol 20 Fall pp 323-9

IJSIM143

304

(which has only one performance) a substantial degree of variation insubjective performance assessments In fact they report a positive relationshipbetween market orientation and judgments about the company performancewithin a single company As they point out ordfit might be that managers have amore positive view of their companyrsquos market orientation when they perceivetheir company to be performing wellordm (Van Bruggen and Smidts 1995 p 13)Hence it is important to employ objective measures of economic performanceand we shall do so in the present study

Market orientation in the services sector the European insurance industryThe insurance sector is of particular interest from a market orientationviewpoint as it works with intangible commodities in which service qualityand customer orientation are crucial elements The competitive characteristicsgenerated by the European Union provide an additional interest in studyingmarket orientation in this area The insurance sector in Europe has traditionallyoperated subject to strict regulations and strong protection from internationalcompetition However for some years now the European Commission has beenworking on the liberalization of this sector Effective implementation of this hasbrought about a major increase in competition within the sector and hasprovoked a major restructuring of insurance companies and groups Thecompetitive climate in Europe has also been inmacruenced by a downside in theeconomic cycle and changes in consumer behavior European customers nowshow greater service expectations and less loyalty As a result rivalry amongcompetitors is increasing as is the importance of competitive strategies adaptedto this sectorrsquos needs In this background the degree of orientation toward thecustomer distributors competition and the general socio-economicenvironment is becoming an increasingly important area of study not onlyfor academics but also for the business world

Lado et al (1998ab) have investigated quite extensively the marketorientation of insurance regrms within the European Union These authors havenot found signiregcant mean differences in market orientation by countryFurthermore they report substantial agreement between the factor structuresof market orientation across countries Thus it seems that the Europeaninsurance sector can be considered a homogenous population with respect tomarket orientation

DataThe population universe considered in this article is deregned as the set ofinsurance companies operating in the European Union which meet thefollowing conditions

they operate in private insurance or ordfmass insuranceordm

they have a market share of more than 005 percent and

their management is independent

IJSIM143

292

The list of European insurance companies was taken from the Financial TimesYearbook for 1996

It was assumed that senior executives were the people best qualireged to assessthe companyrsquos market orientation as well as their innovation degree innovationperformance and customersrsquo loyalty Therefore information from these variableswas gathered via a postal questionnaire submitted to the senior executive in eachof the 554 companies comprising the target population

We obtained 122 valid questionnaires giving a response rate of 22 percentThis sample accounts for over 17 percent of total insurance premiums in theEuropean Union

In order to assess response bias the questionnaires were divided intoquartiles on the basis of reception date (Armstrong and Overton 1977) Ananalysis of early and late responses did not indicate any signiregcant differencein terms of means and covariances

MeasuresBusiness economic performance is a complex construct with multiple possibleobserved indicators Here we measure this construct using three remacrectiveindicators[1]

(1) domestic market share (Jaworski and Kohli 1993 DeshpandeAcirc et al 1993Selnes et al 1996 Greenley and Foxall 1997 1998)

(2) premium growth (which is equivalent to sales growth for the insurancecompanies business (Slater and Narver 1994 Narver and Slater 1990Ruekert 1992 Greenley and Foxall 1997 1998) and

(3) proregtability per year averaged over the last three years(similar to thereturn on investment (ROI) rate (Greenley and Foxall 1997 1998))

All three indicators were expressed as percentages These data were obtainedfrom the managers responding the questionnaire Their responses were carefullycontrasted with published regnancial information (eg Reuters Insurance Brieregng)

Market orientation was measured using the Market Orientation Scale-Revised(MOS-R) This scale is a shortened version of the MOS validated by Lado et al(1998a) in the population of insurance companies of Belgium and Spain Ladoet al (1998a) shortened the original MOS scale while extending the previousvalidation study to target all insurance companies operating in the EuropeanUnion In the Appendix we provide the 30 items composing the MOS-R Eachitem is to be rated on a ten-point Likert-type scale ranging from 0 (completedisagreement) to 10 (complete agreement)

Innovation degree and innovation performance were assessed by means ofmulti-item questionnaires akin to Miller and Friesenrsquos (1982) Innovationperformance was measured by a four-item questionnaire regarding the successof a new productservice (deregned as an improved product a product extensionor a new product line) introduced by the company The questions involved

Businesseconomic

performance

293

whether the new productservice had succeeded in meeting the sales growthmarket share and proregt objectives set up by the company

Innovation degree was assessed by a three-item questionnaire that inquiredthe rate of new productsservices introduced by the company relative tocompetitors the amount of new productsservices marketed by the companyover the past three years and the nature of change of the new productsservices

Finally we used a four-item questionnaire based on existing literature (egDick and Basu 1994 Javalgi and Moberg 1997) to evaluate managersrsquoperceptions of their customersrsquo loyalty The questionnaire taps on theproportion of their customersrsquo insurance premiums taken on by the companythe average time a customer remains in the companyrsquos portfolio the probabilityof a customer renewing a premium and the overall perception of the companycustomersrsquo loyalty

Scale scores for innovation degree innovation performance and customerloyalty were obtained as an unweighted sum of the corresponding items Sincein all three cases Likert-type items on a 0-7 scale were used scale scores forthese variables range from 0-27 0-27 and 0-28 respectively For marketorientation we computed a score for each of its facets as an unweighted sum ofthe corresponding items Then a global market orientation score was obtainedas a sum of the facetsrsquo scores inversely weighted by their number of itemsHence this market orientation score assigns equal weights to each its facetsand ranges from 0-90

The scalesrsquo reliability (as assessed by coefregcient alpha) in this sample were088 (market orientation) 070 (innovation degree) 091 (innovationperformance) and 076 (customer loyalty) The means standard deviationsand correlations among all variables considered in this study are presented inTable I As can be seen in this table the three indicators of business economicperformance are signiregcantly but not largely correlated (the correlations rangefrom 020 to 029) The correlations among the hypothesized intermediatevariables (innovation degree innovation performance and customer loyalty)are not high except for innovation degree and innovation performance whichshare 36 percent of their variance The correlations of market orientation withthe intermediate variables appear signiregcantly larger (they range from 055 to058) than with the dependent variables (they range from 023 to 036) We alsoobserve in Table I that managers report on average a high degree of innovationin their businesses not so high a level of customer loyalty and a level ofinnovation performance just at the scale mean The average self-reporteddegree of market orientation is 56 on a 0-90 scale

MethodAll hypotheses were contrasted using covariance structure analysis asimplemented in LISREL 850 (see Joreskog et al 1999) Since all three indicatorsof business performance are highly positively skewed and present a high

IJSIM143

294

degree of kurtosis throughout this paper rather than attempting to transformthese variables to near-normality we shall employ an estimation approach thatis robust to non-normality of the observed variables The parameter estimateswere obtained using maximum likelihood estimation with standard errorssuitable for non-normal data (Joreskog et al 1999 Equation A24) and twotest statistics were used to assess the goodness of regt (GFI) of the model theSatorra-Bentler scaled chi-squared statistic (Satorra and Bentler 1988Equation 41) and Brownersquos (1984 Equation 220a) chi-squared statisticcorrected for non-normality To evaluate better the goodness of regt of thismodel several additional indices will also provided the root mean squarederror of approximation (RMSEA (Steiger 1990)) the standardized root meansquared residual (SRMSR (Joreskog et al 1999)) the GFI (Tanaka and Huba1985) and the Comparative Fit Index (CFI) using the independence model asbaseline (Bentler 1990 see also McDonald and Marsh 1990) Adequate to goodregt is suggested by RMSEA and SRMSR values approaching 005 For the GFIand the CFI indices values between 090 and 100 indicate adequate to excellentregt (but see Hu and Bentler 1999)

ResultsH1The model used to estimate the effects of market orientation on insurancebusinessesrsquo performance consists of a latent variable representing economicperformance with three indicators (market share premium growth andproregtability) and a single exogenous variable (market orientation) This model

MST PG PROF INNODR INNPERF LOYAL MO

MST 100PG 020 100PROF 023 029 100INNODR 030 034 035 1000INNPERF 027 032 041 062 100LOYAL 025 019 038 040 038 100MO 024 030 036 055 058 057 100x 455 788 626 1825 1422 1996 5628Std 535 749 571 356 292 520 1331

NotesN = 122 all correlations are signiregcant (a = 001) except those marked by which are onlysigniregcant at an a = 005MS = market share PG = premium growth PROF = proregtability INNODR = innovation degreeINNPERF = innovation performance LOYAL = customer loyalty MO = market orientation

Table IMeans standard

deviations andinter-correlations

Businesseconomic

performance

295

is depicted in Figure 2 The parameter estimates and GFIs for this model aregiven in Table II The model shows a good regt although note that it only hastwo degrees of freedom According to the model the best objective indicator ofbusiness economic performance is proregtability per year over 34 percent of itsvariance is accounted for by the model The standardized regressioncoefregcients reveal that proregtability per year is the best objective indicator of

Parameter estimates Goodness of regt R 2

Parameter Value Index Value Variable Value

b1 006 [061] (002) MFF X 2001 ( p = 099) Market share 015

b2 166 [039] (049) S-B X 2001 ( p = 099) Premium growth 024

b3 293 [049] (078) B X 2001 ( p = 099) Proregtability 034

b4 267 [059] (084) RMSEA 001 Business performance 037u1 17713 [100] (2241) df 2u2 2427 [085] (713) SRMSR 001u3 4248 [076] (1364) GFI 100u4 2141 [066] (515) CFI 100

NotesRobust asymptotic standard errors are provided in parentheses standardized parameterestimates are provided in square bracketsMFF X 2 = Minimum regt function chi-square SB X 2 = Satorra-Bentler scaled chi-squareB X 2 = Brownersquos chi-square corrected for non-normality RMSEA = root mean squared error ofapproximation SRMSR = standardized root mean squared residual GFI = goodness of regt indexCFI = comparative regt index R 2 = squared multiple correlations for endogenous variables

Table IIEstimation resultsfor the modeldepicted in Figure 2

Figure 2Market orientation aspredictor of economicperformance

IJSIM143

296

overall business performance Finally according to the model almost 37percent of overall business economic performance is accounted for by thedegree of market orientation

An inspection of the total effects of market orientation on the indicators ofeconomic performance suggests that unit increments of market orientation asmeasured by the MOS-R are associated with 0095 0168 and 0153 incrementsin domestic market share premium growth and proregtability per year averagedover the last three years respectively

H2A mediated model for the relationship between market orientation andbusiness performance is depicted in Figure 3 In this context a mediating effectis said to exist when

both mediating paths b5 b6 are signiregcant and

the direct effect of the exogenous variable on the outcome variablevanishes (complete mediational effect) or is signiregcantly lower (partialmediational effect) when a mediator variable is introduced in the model

Condition (2) amounts to b1 in Figure 3 becoming zero or signiregcantly less thatthan value reported for Figure 2

We used the mediated model depicted in Figure 3 to test for mediatingeffects of innovation degree innovation performance and customer loyaltyseparately on the impact of market orientation on business economicperformance We found that when either innovation performance or innovation

Figure 3Mediated model

Businesseconomic

performance

297

degree were used as mediating variable all the mediating paths weresigniregcant and that direct path from market orientation to businessperformance was not signiregcantly different from zero b1 = 004 t = 162for innovation performance b1 = 004 t = 186 for innovation degree Hencetaken separately both innovation degree and innovation performancecompletely mediate the impact of market orientation on businessperformance After regxing b1 at zero we re-estimated these two mediationalmodels The resulting parameter estimates and GFIs for these two models areshown in Table III On the other hand customer loyalty was found not to havea mediational effect between market orientation and business performanceThe parameter estimates and GFIs for this model are also given in Table III

As can be seen in Table III the mediating paths are signiregcant but thedirect path b1 is signiregcantly different from zero at a = 001 Furthermore a 99percent conregdence interval for the value for b1 reported in Table II (003 009)includes the value of b1 estimated in the mediational model using customerloyalty 004 Hence this variable does not even partially mediate on the impactof market orientation on business economic performance The standardizeddirect impact of market orientation on business performance (0408) is morethan twice the standardized impact of market orientation conveyed throughcustomer loyalty (0191)

The percentage of variance of business economic performance explained bythe model when innovation performance innovation degree or customer loyaltyare used as mediators is very similar (465 percent 453 percent and 437percent respectively)

H3The full model to be regtted corresponding to the hypothesis depicted in Figure 1is presented in Figure 3 The parameter estimates and goodness of regt testcorresponding to this model are given in Table IV (see also Figure 4) As can beseen in this Table the model regts these data very well

All the postulated relationships were found to be signiregcant at an a = 001Lagrange multiplier tests indicated that the regt of the model would notsigniregcantly improve by

adding a direct effect of market orientation to business performance nor

adding a direct effect of innovation degree on economic performance

Result (1) is in accordance with the results discussed above where we saw thatinnovation degree and innovation performance even when taken separatelycompletely mediate the impact of market orientation on business performanceResult (2) conregrms our hypothesis that innovation performance completelymediates the impact of innovation degree on business performance

Given that all effects of market orientation on business performance gothrough either innovation degree-innovation performance or through customer

IJSIM143

298

Par

amet

eres

tim

ates

Goo

dnes

sof

regt

Par

amet

erV

alue

Par

amet

erV

alue

Index

Val

ue

Innova

tion

per

form

ance

as

med

iato

rb 1

0(reg

xed

)u

117

713

[10

0](2

241

)M

FF

X2

560

(p=

035

)b 2

155

[04

0](0

42)

u2

241

4[0

84]

(71

1)S-B

X2

380

(p=

058

)b 3

263

[04

8](0

67)

u3

431

6[0

77]

(45

5)B

X2

429

(p=

051

)b 4

248

[05

9](o

81)

u4

211

3[0

65]

(45

5)df

5R

MSE

A0

03b 5

023

[05

8](0

03)

u5

180

3[0

67]

(28

0)SR

MSR

005

GF

I0

98b 6

018

[06

8](0

05)

CF

I0

99

Innova

tion

deg

ree

as

med

iato

rb

10

(regxed

)u

117

713

[10

0](2

241

)M

FF

X2

717

(p=

021

)b 2

169

[05

0](0

36)

u2

234

0[0

82]

(73

8)S-B

X2

497

(p=

043

)b 3

282

[05

1](0

82)

u3

415

6[0

74]

(45

0)B

X2

613

(p=

029

)b 4

226

[05

4](0

79)

u4

232

9[0

71]

(46

0)df

5R

MSE

A0

06b 5

012

[05

5](0

02)

u5

590

[06

9](0

78)

SR

MSR

005

GF

I0

98b 6

031

[06

7](0

09)

CF

I0

98

Cust

om

erlo

yalty

as

med

iato

rb 1

004

[04

1](0

02)

u1

177

13[1

00]

(22

41)

MF

FX

22

30(p

=0

68)

b 21

61[0

40]

(05

1)u

224

05

[08

4](7

11)

S-B

X2

155

(p=

082

)b 3

252

[04

5](0

74)

u3

447

5[0

80]

(45

5)B

X2

309

(p=

054

)b 4

265

[06

2](0

92)

u4

201

3[0

62]

(45

5)R

MSE

A0

0df

4b 5

015

[05

7](0

02)

u5

863

[06

8](1

04)

SR

MSR

002

GF

I0

99b 6

013

[03

4](0

05)

CF

I1

00

Note

s

Rob

ust

asym

pto

tic

stan

dar

der

rors

are

pro

vid

edin

par

enth

eses

st

andar

diz

edpar

amet

eres

tim

ates

are

pro

vid

edin

squar

ebra

cket

sM

FF

X2=

Min

imum

regt

funct

ion

chi-sq

uar

eSB

X2=

Sat

orra

-Ben

tler

scal

edch

i-sq

uar

eB

X2=

Bro

wnersquo

sch

i-sq

uar

eco

rrec

ted

for

non

-nor

mal

ity

RM

SE

A=

root

mea

nsq

uar

eder

ror

ofap

pro

xim

atio

n

SR

MSR

=st

andar

diz

edro

otm

ean

squar

edre

sidual

G

FI

=goo

dnes

sof

regt

index

C

FI

=co

mpar

ativ

ereg

tin

dex

Table IIIEstimation results

for the modeldepicted in Figure 3

Businesseconomic

performance

299

Par

amet

eres

tim

ates

Goo

dnes

sof

regt

R2

Par

amet

erV

alue

Par

amet

erV

alue

Index

Val

ue

Var

iable

Val

ue

b 10

12[0

55]

(22

41)

u1

177

13[1

00]

(00

2)M

FF

X2

105

9(p

=0

56)

Mar

ket

shar

e0

16b 2

013

[03

4](0

04)

u2

590

[06

9](0

78)

S-B

X2

785

(p=

080

)P

rem

ium

gro

wth

019

b 30

15[0

57]

(00

2)u

314

48

[05

4](2

59)

BX

212

67

(p=

039

)P

roreg

tabilit

y0

38b 4

078

[04

4](0

15)

u4

863

[06

8](1

04)

df

12B

usi

nes

sper

form

ance

056

RM

SE

A0

0b 5

016

[05

4](0

05)

u5

239

1[0

84]

(72

3)SR

MSR

005

Innov

atio

nper

form

ance

046

b 60

16[0

37]

(00

7)u

645

09

[08

1](1

271

)G

FI

098

Innov

atio

ndeg

ree

031

b 71

42[0

40]

(04

5)u

720

12

[06

2](4

60)

CF

I1

00C

ust

omer

loyal

ty0

32b 8

217

[04

4](0

64)

b 92

32[0

62]

(08

5)

Note

sR

obust

asym

pto

tic

stan

dar

der

rors

are

pro

vid

edin

par

enth

eses

st

andar

diz

edpar

amet

eres

tim

ates

are

pro

vid

edin

squar

ebra

cket

sM

FF

X2=

Min

imum

regtfu

nct

ion

chi-sq

uar

eSB

X2=

Sat

orra

-Ben

tler

scal

edch

i-sq

uar

eB

X2=

Bro

wnersquo

sch

i-sq

uar

eco

rrec

ted

for

non

-nor

mal

ity

RM

SE

A=

root

mea

nsq

uar

eder

ror

ofap

pro

xim

atio

n

SR

MSR

=st

andar

diz

edro

otm

ean

squar

edre

sidual

G

FI

=goo

dnes

sof

regt

index

C

FI

=co

mpar

ativ

ereg

tin

dex

R

2=

squar

edm

ult

iple

corr

elat

ions

for

endog

enou

svar

iable

s

Table IVEstimation resultsfor the modeldepicted in Figure 4

IJSIM143

300

loyalty a question arises as to the relative importance of the speciregc effectsgoing through these variables The standardized speciregc effect (computed as inBollen 1987) going through innovation degree and innovation performance is031 and 021 going through customer loyalty Hence the impact of marketorientation going through innovation is 50 percent more than that goingthrough customer loyalty

We can also see in Table IV that over 30 percent of the variance of theintermediate variables (innovation degree innovation performance andcustomer loyalty) are explained by market orientation In fact almost 50percent (464 percent to be exact) of innovation performance is explained bymarket orientation Furthermore note that the percentage of variance ofbusiness performance explained by the model is 561 percent a 52 percentincrement over what is explained by market orientation alone (see Table II) andover a 20 percent increment over what is explained by the mediational modelsconsidered previously Hence the inclusion of all three intermediate variables inthe model improves considerably our prediction of business performance

Furthermore we observe in Table IV that the direct effect of marketorientation on all three intermediate variables appear to be equal Also thedirect effects of customer loyalty and of innovation performance on economicperformance appear to be equal We re-estimated the model to test theseconstraints obtaining b1 = b2 = b3 = 0133 b5 = b6 = 0157 Satorra-BentlerX 2(15) = 8849 p = 0885 CFI = 100 GFI = 0975 SRMR = 006

ConclusionsMarket orientation can be deregned as a strategy used to reach a sustainablecompetitive advantage based on the generation and use of information within

Figure 4Full model

Businesseconomic

performance

301

organizations and on the selection of markets to be satisreged In thisframework we believe that competitive advantage results from the use ofresources and capabilities to generate differential satisfaction in proregtablemarkets Sustainability is achieved because the performance of the marketorientationrsquos behaviors requires complex organizational knowledge that cannoteasily be imitated by competitors Thus we hypothesize that the satisfaction ofproregtable markets permits the regrm to achieve a psychologically differentialposition that leads to brand loyalty and thus to higher proregts Previous studieshave

found a clear impact of market orientation on economic performance

assessed the effects of market orientation on innovation and

investigated the relationship between market orientation and relationshipmarketing variables

The present study proposes and tests a model that integrates both streams ofresearch innovation and relationship marketing More research is needed toinvestigate the role of customer satisfaction within this framework

In our necessarily partial model innovation degree innovation performanceand customer loyalty are used as intermediate variables on the effect of marketorientation on business performance Our results suggest that the addition ofthese variables help improve our predictions of business economic performance52 percent over what is explained by market orientation alone Also we foundthat innovation degree and innovation performance each taken separatelycompletely mediate the effect of market orientation on economic performanceFurthermore the impact of innovation degree on economic performance iscompletely channeled through innovation performance Customer loyalty byitself does not meditate the impact of market orientation on economicperformance but when considered along with innovation degree andinnovation performance it conveys some of the effects of market orientationon business performance This seemingly contradictory result arises from thefact that all three intermediate variables are interrelated

Our results should not be taken to imply that there are no other variablesmediate the effect of market orientation on economic performance We believethat other variables that have not been taken into account in this study such asproduct quality and customer satisfaction may also be signiregcant mediatorsHowever our results do suggest that whenever innovation degree andinnovation performance are included in the model as intermediate variablesthe effects of market orientation on business performance will mostly beconveyed through these variables Also in this study we have adopted thecurrently most widely accepted approach to measuring customer loyalty whichis based on behavioral loyalty measures (eg share of category requirementsrenew the policy probability the average customer last in the companyportfolio) It would be worth extending the present study by including not only

IJSIM143

302

behavioral measures of customer loyalty but also attitudinal measures alongthe lines of Dick and Basu (1994)

In our opinion two important contributions of the present research are outuse of objective measures of business performance and our focusing oninternational markets Despite the growing role of globalization and marketintegration and despite the increasing internationalization of corporationsmost studies on market orientation have focused on domestic markets (withnotable exceptions such as Selnes et al 1996 Webster 1994) Similarly moststudies on product innovation have also focused on domestic markets There isa lack of research yielding empirical support to the validity in an internationalsetting to research results obtained in domestic markets To regll this gap wetargeted the European Union market

Our study focused on a single industry the insurance sector Our sampleaccounted for 22 percent of the companies and 17 percent of the insurancepremiums in the targeted market An advantage of our single-industryapproach is that (with obvious reservations arising from the non-experimentalnature of our study and the fact that our sample should not be considered tohave been obtained at random) we can draw tentative predictions from ourmodel concerning the impact of market orientation on economic performance ininsurance companies operating in the European Union market An evidentdrawback of the single-industry approach adopted here is that it is not clearhow the present results extrapolate to other industries even when operating inthe same market

We have found that within the European Union insurance regrms thatconstantly monitor the evolution of current and potential customerrsquos needsmodify the attributes of the products to adapt them to the distributorsrequirements analyze competitorsrsquo marketing policy and products and knowbest of environmental trends especially technological and legal changes aremore likely to develop more new products have their new products accepted bythe market and obtain more loyal customers In turn increased customerloyalty and increased new product success will result in improved economicperformance However these conclusions must be taken with some caution aswe have used overall measures of innovation and innovation performanceFurther research is needed that takes into account the various aspects thatconstitute innovation

Previous studies have concluded that insurance companies still seethemselves as being product-focused and the industry as a whole is generallydistribution driven (Sodano 2000 Lambin 1996) This lack of closeness withcustomers has in turn contributed to the industry lack of product innovationsand product portfolios of commodity products that only compete on price Nowthe most successful regrms are redirecting their focus to the market needs andthey are beginning to exploit customer data and use market research togenerate ideas for designing new products In this context these companies

Businesseconomic

performance

303

now face choosing among too many new service options The managerialimplication of our study is that by enhancing their market orientation regrmswill know and service its customers better Thus they will generate moreinnovations by adopting a market-based product development process Alsoincreasing levels of market orientation enable regrms to discriminate more easilywhich new products have a higher success probability thus enhancing both theefregcacy and efregciency of new product development

Note

1 We considered employing the total volume of premiums for each insurance company as anadditional indicator although it seemed to us to be a better indicator of a companyrsquos sizerather than of its performance As we suspected the total volume of premiums wasuncorrelated with any of the variables considered in this study except for the companyrsquosmarket share (r = 024 p 001) Hence it is clear that volume of premiums should not beused as an indicator of insurance companiesrsquo performance

References

Armstrong JS and Overton T (1977) ordfEstimating nonresponse bias in mail surveysordm Journalof Marketing Research Vol 14 pp 396-402

Atuahene-Gima K (1996) ordfMarket orientation and innovationordm Journal of Business ResearchVol 35 pp 93-103

Baker TL Simpson PM and Sigauw JA (1999) ordfThe impact of suppliersrsquo perceptions ofreseller market orientation on key relationship constructsordm Journal of the Academy ofMarketing Science Vol 27 No 1 pp 50-7

Baron RM and Kenny DA (1986) ordfThe moderator-mediator variable distinction in socialpsychological research conceptual strategic and statistical considerationsordm Journal ofPersonality and Social Psychology Vol 51 pp 1173-82

Bentler PM (1990) ordfComparative regt indexes in structural modelsordm Psychological BulletinVol 107 pp 238-46

Bhuian SN (1997) ordfExploring market orientation in banks an empirical examination in SaudiArabiaordm The Journal of Service Marketing Vol 11 No 5 pp 317-28

Bollen KA (1987) ordfTotal direct and indirect effects in structural equation modelsordm inClogg CC (Ed) Sociological Methodology American Sociological AssociationWashington DC pp 37-69

Browne MW (1984) ordfAsymptotically distribution free methods for the analysis of covariancestructuresordm British Journal of Mathematical and Statistical Psychology Vol 37 pp 62-83

Calantone RJ di Benedetto AC and Bhoovaraghavan S (1994) ordfExamining the relationshipbetween degree of innovation and new product successordm Journal of Business ResearchVol 30 pp 143-8

Calantone RJ Schmidt JB and Song MX (1996) ordfControllable factors of new productssuccess a cross-national comparisonordm Marketing Science Vol 15 No 4 pp 341-58

Cooper RC (1994) ordfNew product the factors that drive successordm International MarketingReview Vol 11 No 1 pp 60-76

Day G (1992) ordfMarketingrsquos contribution to the strategy dialogueordm Journal of the Academy ofMarketing Science Vol 20 Fall pp 323-9

IJSIM143

304

The list of European insurance companies was taken from the Financial TimesYearbook for 1996

It was assumed that senior executives were the people best qualireged to assessthe companyrsquos market orientation as well as their innovation degree innovationperformance and customersrsquo loyalty Therefore information from these variableswas gathered via a postal questionnaire submitted to the senior executive in eachof the 554 companies comprising the target population

We obtained 122 valid questionnaires giving a response rate of 22 percentThis sample accounts for over 17 percent of total insurance premiums in theEuropean Union

In order to assess response bias the questionnaires were divided intoquartiles on the basis of reception date (Armstrong and Overton 1977) Ananalysis of early and late responses did not indicate any signiregcant differencein terms of means and covariances

MeasuresBusiness economic performance is a complex construct with multiple possibleobserved indicators Here we measure this construct using three remacrectiveindicators[1]

(1) domestic market share (Jaworski and Kohli 1993 DeshpandeAcirc et al 1993Selnes et al 1996 Greenley and Foxall 1997 1998)

(2) premium growth (which is equivalent to sales growth for the insurancecompanies business (Slater and Narver 1994 Narver and Slater 1990Ruekert 1992 Greenley and Foxall 1997 1998) and

(3) proregtability per year averaged over the last three years(similar to thereturn on investment (ROI) rate (Greenley and Foxall 1997 1998))

All three indicators were expressed as percentages These data were obtainedfrom the managers responding the questionnaire Their responses were carefullycontrasted with published regnancial information (eg Reuters Insurance Brieregng)

Market orientation was measured using the Market Orientation Scale-Revised(MOS-R) This scale is a shortened version of the MOS validated by Lado et al(1998a) in the population of insurance companies of Belgium and Spain Ladoet al (1998a) shortened the original MOS scale while extending the previousvalidation study to target all insurance companies operating in the EuropeanUnion In the Appendix we provide the 30 items composing the MOS-R Eachitem is to be rated on a ten-point Likert-type scale ranging from 0 (completedisagreement) to 10 (complete agreement)

Innovation degree and innovation performance were assessed by means ofmulti-item questionnaires akin to Miller and Friesenrsquos (1982) Innovationperformance was measured by a four-item questionnaire regarding the successof a new productservice (deregned as an improved product a product extensionor a new product line) introduced by the company The questions involved

Businesseconomic

performance

293

whether the new productservice had succeeded in meeting the sales growthmarket share and proregt objectives set up by the company

Innovation degree was assessed by a three-item questionnaire that inquiredthe rate of new productsservices introduced by the company relative tocompetitors the amount of new productsservices marketed by the companyover the past three years and the nature of change of the new productsservices

Finally we used a four-item questionnaire based on existing literature (egDick and Basu 1994 Javalgi and Moberg 1997) to evaluate managersrsquoperceptions of their customersrsquo loyalty The questionnaire taps on theproportion of their customersrsquo insurance premiums taken on by the companythe average time a customer remains in the companyrsquos portfolio the probabilityof a customer renewing a premium and the overall perception of the companycustomersrsquo loyalty

Scale scores for innovation degree innovation performance and customerloyalty were obtained as an unweighted sum of the corresponding items Sincein all three cases Likert-type items on a 0-7 scale were used scale scores forthese variables range from 0-27 0-27 and 0-28 respectively For marketorientation we computed a score for each of its facets as an unweighted sum ofthe corresponding items Then a global market orientation score was obtainedas a sum of the facetsrsquo scores inversely weighted by their number of itemsHence this market orientation score assigns equal weights to each its facetsand ranges from 0-90

The scalesrsquo reliability (as assessed by coefregcient alpha) in this sample were088 (market orientation) 070 (innovation degree) 091 (innovationperformance) and 076 (customer loyalty) The means standard deviationsand correlations among all variables considered in this study are presented inTable I As can be seen in this table the three indicators of business economicperformance are signiregcantly but not largely correlated (the correlations rangefrom 020 to 029) The correlations among the hypothesized intermediatevariables (innovation degree innovation performance and customer loyalty)are not high except for innovation degree and innovation performance whichshare 36 percent of their variance The correlations of market orientation withthe intermediate variables appear signiregcantly larger (they range from 055 to058) than with the dependent variables (they range from 023 to 036) We alsoobserve in Table I that managers report on average a high degree of innovationin their businesses not so high a level of customer loyalty and a level ofinnovation performance just at the scale mean The average self-reporteddegree of market orientation is 56 on a 0-90 scale

MethodAll hypotheses were contrasted using covariance structure analysis asimplemented in LISREL 850 (see Joreskog et al 1999) Since all three indicatorsof business performance are highly positively skewed and present a high

IJSIM143

294

degree of kurtosis throughout this paper rather than attempting to transformthese variables to near-normality we shall employ an estimation approach thatis robust to non-normality of the observed variables The parameter estimateswere obtained using maximum likelihood estimation with standard errorssuitable for non-normal data (Joreskog et al 1999 Equation A24) and twotest statistics were used to assess the goodness of regt (GFI) of the model theSatorra-Bentler scaled chi-squared statistic (Satorra and Bentler 1988Equation 41) and Brownersquos (1984 Equation 220a) chi-squared statisticcorrected for non-normality To evaluate better the goodness of regt of thismodel several additional indices will also provided the root mean squarederror of approximation (RMSEA (Steiger 1990)) the standardized root meansquared residual (SRMSR (Joreskog et al 1999)) the GFI (Tanaka and Huba1985) and the Comparative Fit Index (CFI) using the independence model asbaseline (Bentler 1990 see also McDonald and Marsh 1990) Adequate to goodregt is suggested by RMSEA and SRMSR values approaching 005 For the GFIand the CFI indices values between 090 and 100 indicate adequate to excellentregt (but see Hu and Bentler 1999)

ResultsH1The model used to estimate the effects of market orientation on insurancebusinessesrsquo performance consists of a latent variable representing economicperformance with three indicators (market share premium growth andproregtability) and a single exogenous variable (market orientation) This model

MST PG PROF INNODR INNPERF LOYAL MO

MST 100PG 020 100PROF 023 029 100INNODR 030 034 035 1000INNPERF 027 032 041 062 100LOYAL 025 019 038 040 038 100MO 024 030 036 055 058 057 100x 455 788 626 1825 1422 1996 5628Std 535 749 571 356 292 520 1331

NotesN = 122 all correlations are signiregcant (a = 001) except those marked by which are onlysigniregcant at an a = 005MS = market share PG = premium growth PROF = proregtability INNODR = innovation degreeINNPERF = innovation performance LOYAL = customer loyalty MO = market orientation

Table IMeans standard

deviations andinter-correlations

Businesseconomic

performance

295

is depicted in Figure 2 The parameter estimates and GFIs for this model aregiven in Table II The model shows a good regt although note that it only hastwo degrees of freedom According to the model the best objective indicator ofbusiness economic performance is proregtability per year over 34 percent of itsvariance is accounted for by the model The standardized regressioncoefregcients reveal that proregtability per year is the best objective indicator of

Parameter estimates Goodness of regt R 2

Parameter Value Index Value Variable Value

b1 006 [061] (002) MFF X 2001 ( p = 099) Market share 015

b2 166 [039] (049) S-B X 2001 ( p = 099) Premium growth 024

b3 293 [049] (078) B X 2001 ( p = 099) Proregtability 034

b4 267 [059] (084) RMSEA 001 Business performance 037u1 17713 [100] (2241) df 2u2 2427 [085] (713) SRMSR 001u3 4248 [076] (1364) GFI 100u4 2141 [066] (515) CFI 100

NotesRobust asymptotic standard errors are provided in parentheses standardized parameterestimates are provided in square bracketsMFF X 2 = Minimum regt function chi-square SB X 2 = Satorra-Bentler scaled chi-squareB X 2 = Brownersquos chi-square corrected for non-normality RMSEA = root mean squared error ofapproximation SRMSR = standardized root mean squared residual GFI = goodness of regt indexCFI = comparative regt index R 2 = squared multiple correlations for endogenous variables

Table IIEstimation resultsfor the modeldepicted in Figure 2

Figure 2Market orientation aspredictor of economicperformance

IJSIM143

296

overall business performance Finally according to the model almost 37percent of overall business economic performance is accounted for by thedegree of market orientation

An inspection of the total effects of market orientation on the indicators ofeconomic performance suggests that unit increments of market orientation asmeasured by the MOS-R are associated with 0095 0168 and 0153 incrementsin domestic market share premium growth and proregtability per year averagedover the last three years respectively

H2A mediated model for the relationship between market orientation andbusiness performance is depicted in Figure 3 In this context a mediating effectis said to exist when

both mediating paths b5 b6 are signiregcant and

the direct effect of the exogenous variable on the outcome variablevanishes (complete mediational effect) or is signiregcantly lower (partialmediational effect) when a mediator variable is introduced in the model

Condition (2) amounts to b1 in Figure 3 becoming zero or signiregcantly less thatthan value reported for Figure 2

We used the mediated model depicted in Figure 3 to test for mediatingeffects of innovation degree innovation performance and customer loyaltyseparately on the impact of market orientation on business economicperformance We found that when either innovation performance or innovation

Figure 3Mediated model

Businesseconomic

performance

297

degree were used as mediating variable all the mediating paths weresigniregcant and that direct path from market orientation to businessperformance was not signiregcantly different from zero b1 = 004 t = 162for innovation performance b1 = 004 t = 186 for innovation degree Hencetaken separately both innovation degree and innovation performancecompletely mediate the impact of market orientation on businessperformance After regxing b1 at zero we re-estimated these two mediationalmodels The resulting parameter estimates and GFIs for these two models areshown in Table III On the other hand customer loyalty was found not to havea mediational effect between market orientation and business performanceThe parameter estimates and GFIs for this model are also given in Table III

As can be seen in Table III the mediating paths are signiregcant but thedirect path b1 is signiregcantly different from zero at a = 001 Furthermore a 99percent conregdence interval for the value for b1 reported in Table II (003 009)includes the value of b1 estimated in the mediational model using customerloyalty 004 Hence this variable does not even partially mediate on the impactof market orientation on business economic performance The standardizeddirect impact of market orientation on business performance (0408) is morethan twice the standardized impact of market orientation conveyed throughcustomer loyalty (0191)

The percentage of variance of business economic performance explained bythe model when innovation performance innovation degree or customer loyaltyare used as mediators is very similar (465 percent 453 percent and 437percent respectively)

H3The full model to be regtted corresponding to the hypothesis depicted in Figure 1is presented in Figure 3 The parameter estimates and goodness of regt testcorresponding to this model are given in Table IV (see also Figure 4) As can beseen in this Table the model regts these data very well

All the postulated relationships were found to be signiregcant at an a = 001Lagrange multiplier tests indicated that the regt of the model would notsigniregcantly improve by

adding a direct effect of market orientation to business performance nor

adding a direct effect of innovation degree on economic performance

Result (1) is in accordance with the results discussed above where we saw thatinnovation degree and innovation performance even when taken separatelycompletely mediate the impact of market orientation on business performanceResult (2) conregrms our hypothesis that innovation performance completelymediates the impact of innovation degree on business performance

Given that all effects of market orientation on business performance gothrough either innovation degree-innovation performance or through customer

IJSIM143

298

Par

amet

eres

tim

ates

Goo

dnes

sof

regt

Par

amet

erV

alue

Par

amet

erV

alue

Index

Val

ue

Innova

tion

per

form

ance

as

med

iato

rb 1

0(reg

xed

)u

117

713

[10

0](2

241

)M

FF

X2

560

(p=

035

)b 2

155

[04

0](0

42)

u2

241

4[0

84]

(71

1)S-B

X2

380

(p=

058

)b 3

263

[04

8](0

67)

u3

431

6[0

77]

(45

5)B

X2

429

(p=

051

)b 4

248

[05

9](o

81)

u4

211

3[0

65]

(45

5)df

5R

MSE

A0

03b 5

023

[05

8](0

03)

u5

180

3[0

67]

(28

0)SR

MSR

005

GF

I0

98b 6

018

[06

8](0

05)

CF

I0

99

Innova

tion

deg

ree

as

med

iato

rb

10

(regxed

)u

117

713

[10

0](2

241

)M

FF

X2

717

(p=

021

)b 2

169

[05

0](0

36)

u2

234

0[0

82]

(73

8)S-B

X2

497

(p=

043

)b 3

282

[05

1](0

82)

u3

415

6[0

74]

(45

0)B

X2

613

(p=

029

)b 4

226

[05

4](0

79)

u4

232

9[0

71]

(46

0)df

5R

MSE

A0

06b 5

012

[05

5](0

02)

u5

590

[06

9](0

78)

SR

MSR

005

GF

I0

98b 6

031

[06

7](0

09)

CF

I0

98

Cust

om

erlo

yalty

as

med

iato

rb 1

004

[04

1](0

02)

u1

177

13[1

00]

(22

41)

MF

FX

22

30(p

=0

68)

b 21

61[0

40]

(05

1)u

224

05

[08

4](7

11)

S-B

X2

155

(p=

082

)b 3

252

[04

5](0

74)

u3

447

5[0

80]

(45

5)B

X2

309

(p=

054

)b 4

265

[06

2](0

92)

u4

201

3[0

62]

(45

5)R

MSE

A0

0df

4b 5

015

[05

7](0

02)

u5

863

[06

8](1

04)

SR

MSR

002

GF

I0

99b 6

013

[03

4](0

05)

CF

I1

00

Note

s

Rob

ust

asym

pto

tic

stan

dar

der

rors

are

pro

vid

edin

par

enth

eses

st

andar

diz

edpar

amet

eres

tim

ates

are

pro

vid

edin

squar

ebra

cket

sM

FF

X2=

Min

imum

regt

funct

ion

chi-sq

uar

eSB

X2=

Sat

orra

-Ben

tler

scal

edch

i-sq

uar

eB

X2=

Bro

wnersquo

sch

i-sq

uar

eco

rrec

ted

for

non

-nor

mal

ity

RM

SE

A=

root

mea

nsq

uar

eder

ror

ofap

pro

xim

atio

n

SR

MSR

=st

andar

diz

edro

otm

ean

squar

edre

sidual

G

FI

=goo

dnes

sof

regt

index

C

FI

=co

mpar

ativ

ereg

tin

dex

Table IIIEstimation results

for the modeldepicted in Figure 3

Businesseconomic

performance

299

Par

amet

eres

tim

ates

Goo

dnes

sof

regt

R2

Par

amet

erV

alue

Par

amet

erV

alue

Index

Val

ue

Var

iable

Val

ue

b 10

12[0

55]

(22

41)

u1

177

13[1

00]

(00

2)M

FF

X2

105

9(p

=0

56)

Mar

ket

shar

e0

16b 2

013

[03

4](0

04)

u2

590

[06

9](0

78)

S-B

X2

785

(p=

080

)P

rem

ium

gro

wth

019

b 30

15[0

57]

(00

2)u

314

48

[05

4](2

59)

BX

212

67

(p=

039

)P

roreg

tabilit

y0

38b 4

078

[04

4](0

15)

u4

863

[06

8](1

04)

df

12B

usi

nes

sper

form

ance

056

RM

SE

A0

0b 5

016

[05

4](0

05)

u5

239

1[0

84]

(72

3)SR

MSR

005

Innov

atio

nper

form

ance

046

b 60

16[0

37]

(00

7)u

645

09

[08

1](1

271

)G

FI

098

Innov

atio

ndeg

ree

031

b 71

42[0

40]

(04

5)u

720

12

[06

2](4

60)

CF

I1

00C

ust

omer

loyal

ty0

32b 8

217

[04

4](0

64)

b 92

32[0

62]

(08

5)

Note

sR

obust

asym

pto

tic

stan

dar

der

rors

are

pro

vid

edin

par

enth

eses

st

andar

diz

edpar

amet

eres

tim

ates

are

pro

vid

edin

squar

ebra

cket

sM

FF

X2=

Min

imum

regtfu

nct

ion

chi-sq

uar

eSB

X2=

Sat

orra

-Ben

tler

scal

edch

i-sq

uar

eB

X2=

Bro

wnersquo

sch

i-sq

uar

eco

rrec

ted

for

non

-nor

mal

ity

RM

SE

A=

root

mea

nsq

uar

eder

ror

ofap

pro

xim

atio

n

SR

MSR

=st

andar

diz

edro

otm

ean

squar

edre

sidual

G

FI

=goo

dnes

sof

regt

index

C

FI

=co

mpar

ativ

ereg

tin

dex

R

2=

squar

edm

ult

iple

corr

elat

ions

for

endog

enou

svar

iable

s

Table IVEstimation resultsfor the modeldepicted in Figure 4

IJSIM143

300

loyalty a question arises as to the relative importance of the speciregc effectsgoing through these variables The standardized speciregc effect (computed as inBollen 1987) going through innovation degree and innovation performance is031 and 021 going through customer loyalty Hence the impact of marketorientation going through innovation is 50 percent more than that goingthrough customer loyalty

We can also see in Table IV that over 30 percent of the variance of theintermediate variables (innovation degree innovation performance andcustomer loyalty) are explained by market orientation In fact almost 50percent (464 percent to be exact) of innovation performance is explained bymarket orientation Furthermore note that the percentage of variance ofbusiness performance explained by the model is 561 percent a 52 percentincrement over what is explained by market orientation alone (see Table II) andover a 20 percent increment over what is explained by the mediational modelsconsidered previously Hence the inclusion of all three intermediate variables inthe model improves considerably our prediction of business performance

Furthermore we observe in Table IV that the direct effect of marketorientation on all three intermediate variables appear to be equal Also thedirect effects of customer loyalty and of innovation performance on economicperformance appear to be equal We re-estimated the model to test theseconstraints obtaining b1 = b2 = b3 = 0133 b5 = b6 = 0157 Satorra-BentlerX 2(15) = 8849 p = 0885 CFI = 100 GFI = 0975 SRMR = 006

ConclusionsMarket orientation can be deregned as a strategy used to reach a sustainablecompetitive advantage based on the generation and use of information within

Figure 4Full model

Businesseconomic

performance

301

organizations and on the selection of markets to be satisreged In thisframework we believe that competitive advantage results from the use ofresources and capabilities to generate differential satisfaction in proregtablemarkets Sustainability is achieved because the performance of the marketorientationrsquos behaviors requires complex organizational knowledge that cannoteasily be imitated by competitors Thus we hypothesize that the satisfaction ofproregtable markets permits the regrm to achieve a psychologically differentialposition that leads to brand loyalty and thus to higher proregts Previous studieshave

found a clear impact of market orientation on economic performance

assessed the effects of market orientation on innovation and

investigated the relationship between market orientation and relationshipmarketing variables

The present study proposes and tests a model that integrates both streams ofresearch innovation and relationship marketing More research is needed toinvestigate the role of customer satisfaction within this framework

In our necessarily partial model innovation degree innovation performanceand customer loyalty are used as intermediate variables on the effect of marketorientation on business performance Our results suggest that the addition ofthese variables help improve our predictions of business economic performance52 percent over what is explained by market orientation alone Also we foundthat innovation degree and innovation performance each taken separatelycompletely mediate the effect of market orientation on economic performanceFurthermore the impact of innovation degree on economic performance iscompletely channeled through innovation performance Customer loyalty byitself does not meditate the impact of market orientation on economicperformance but when considered along with innovation degree andinnovation performance it conveys some of the effects of market orientationon business performance This seemingly contradictory result arises from thefact that all three intermediate variables are interrelated

Our results should not be taken to imply that there are no other variablesmediate the effect of market orientation on economic performance We believethat other variables that have not been taken into account in this study such asproduct quality and customer satisfaction may also be signiregcant mediatorsHowever our results do suggest that whenever innovation degree andinnovation performance are included in the model as intermediate variablesthe effects of market orientation on business performance will mostly beconveyed through these variables Also in this study we have adopted thecurrently most widely accepted approach to measuring customer loyalty whichis based on behavioral loyalty measures (eg share of category requirementsrenew the policy probability the average customer last in the companyportfolio) It would be worth extending the present study by including not only

IJSIM143

302

behavioral measures of customer loyalty but also attitudinal measures alongthe lines of Dick and Basu (1994)

In our opinion two important contributions of the present research are outuse of objective measures of business performance and our focusing oninternational markets Despite the growing role of globalization and marketintegration and despite the increasing internationalization of corporationsmost studies on market orientation have focused on domestic markets (withnotable exceptions such as Selnes et al 1996 Webster 1994) Similarly moststudies on product innovation have also focused on domestic markets There isa lack of research yielding empirical support to the validity in an internationalsetting to research results obtained in domestic markets To regll this gap wetargeted the European Union market

Our study focused on a single industry the insurance sector Our sampleaccounted for 22 percent of the companies and 17 percent of the insurancepremiums in the targeted market An advantage of our single-industryapproach is that (with obvious reservations arising from the non-experimentalnature of our study and the fact that our sample should not be considered tohave been obtained at random) we can draw tentative predictions from ourmodel concerning the impact of market orientation on economic performance ininsurance companies operating in the European Union market An evidentdrawback of the single-industry approach adopted here is that it is not clearhow the present results extrapolate to other industries even when operating inthe same market

We have found that within the European Union insurance regrms thatconstantly monitor the evolution of current and potential customerrsquos needsmodify the attributes of the products to adapt them to the distributorsrequirements analyze competitorsrsquo marketing policy and products and knowbest of environmental trends especially technological and legal changes aremore likely to develop more new products have their new products accepted bythe market and obtain more loyal customers In turn increased customerloyalty and increased new product success will result in improved economicperformance However these conclusions must be taken with some caution aswe have used overall measures of innovation and innovation performanceFurther research is needed that takes into account the various aspects thatconstitute innovation

Previous studies have concluded that insurance companies still seethemselves as being product-focused and the industry as a whole is generallydistribution driven (Sodano 2000 Lambin 1996) This lack of closeness withcustomers has in turn contributed to the industry lack of product innovationsand product portfolios of commodity products that only compete on price Nowthe most successful regrms are redirecting their focus to the market needs andthey are beginning to exploit customer data and use market research togenerate ideas for designing new products In this context these companies

Businesseconomic

performance

303

now face choosing among too many new service options The managerialimplication of our study is that by enhancing their market orientation regrmswill know and service its customers better Thus they will generate moreinnovations by adopting a market-based product development process Alsoincreasing levels of market orientation enable regrms to discriminate more easilywhich new products have a higher success probability thus enhancing both theefregcacy and efregciency of new product development

Note

1 We considered employing the total volume of premiums for each insurance company as anadditional indicator although it seemed to us to be a better indicator of a companyrsquos sizerather than of its performance As we suspected the total volume of premiums wasuncorrelated with any of the variables considered in this study except for the companyrsquosmarket share (r = 024 p 001) Hence it is clear that volume of premiums should not beused as an indicator of insurance companiesrsquo performance

References

Armstrong JS and Overton T (1977) ordfEstimating nonresponse bias in mail surveysordm Journalof Marketing Research Vol 14 pp 396-402

Atuahene-Gima K (1996) ordfMarket orientation and innovationordm Journal of Business ResearchVol 35 pp 93-103

Baker TL Simpson PM and Sigauw JA (1999) ordfThe impact of suppliersrsquo perceptions ofreseller market orientation on key relationship constructsordm Journal of the Academy ofMarketing Science Vol 27 No 1 pp 50-7

Baron RM and Kenny DA (1986) ordfThe moderator-mediator variable distinction in socialpsychological research conceptual strategic and statistical considerationsordm Journal ofPersonality and Social Psychology Vol 51 pp 1173-82

Bentler PM (1990) ordfComparative regt indexes in structural modelsordm Psychological BulletinVol 107 pp 238-46

Bhuian SN (1997) ordfExploring market orientation in banks an empirical examination in SaudiArabiaordm The Journal of Service Marketing Vol 11 No 5 pp 317-28

Bollen KA (1987) ordfTotal direct and indirect effects in structural equation modelsordm inClogg CC (Ed) Sociological Methodology American Sociological AssociationWashington DC pp 37-69

Browne MW (1984) ordfAsymptotically distribution free methods for the analysis of covariancestructuresordm British Journal of Mathematical and Statistical Psychology Vol 37 pp 62-83

Calantone RJ di Benedetto AC and Bhoovaraghavan S (1994) ordfExamining the relationshipbetween degree of innovation and new product successordm Journal of Business ResearchVol 30 pp 143-8

Calantone RJ Schmidt JB and Song MX (1996) ordfControllable factors of new productssuccess a cross-national comparisonordm Marketing Science Vol 15 No 4 pp 341-58

Cooper RC (1994) ordfNew product the factors that drive successordm International MarketingReview Vol 11 No 1 pp 60-76

Day G (1992) ordfMarketingrsquos contribution to the strategy dialogueordm Journal of the Academy ofMarketing Science Vol 20 Fall pp 323-9

IJSIM143

304

whether the new productservice had succeeded in meeting the sales growthmarket share and proregt objectives set up by the company

Innovation degree was assessed by a three-item questionnaire that inquiredthe rate of new productsservices introduced by the company relative tocompetitors the amount of new productsservices marketed by the companyover the past three years and the nature of change of the new productsservices

Finally we used a four-item questionnaire based on existing literature (egDick and Basu 1994 Javalgi and Moberg 1997) to evaluate managersrsquoperceptions of their customersrsquo loyalty The questionnaire taps on theproportion of their customersrsquo insurance premiums taken on by the companythe average time a customer remains in the companyrsquos portfolio the probabilityof a customer renewing a premium and the overall perception of the companycustomersrsquo loyalty

Scale scores for innovation degree innovation performance and customerloyalty were obtained as an unweighted sum of the corresponding items Sincein all three cases Likert-type items on a 0-7 scale were used scale scores forthese variables range from 0-27 0-27 and 0-28 respectively For marketorientation we computed a score for each of its facets as an unweighted sum ofthe corresponding items Then a global market orientation score was obtainedas a sum of the facetsrsquo scores inversely weighted by their number of itemsHence this market orientation score assigns equal weights to each its facetsand ranges from 0-90

The scalesrsquo reliability (as assessed by coefregcient alpha) in this sample were088 (market orientation) 070 (innovation degree) 091 (innovationperformance) and 076 (customer loyalty) The means standard deviationsand correlations among all variables considered in this study are presented inTable I As can be seen in this table the three indicators of business economicperformance are signiregcantly but not largely correlated (the correlations rangefrom 020 to 029) The correlations among the hypothesized intermediatevariables (innovation degree innovation performance and customer loyalty)are not high except for innovation degree and innovation performance whichshare 36 percent of their variance The correlations of market orientation withthe intermediate variables appear signiregcantly larger (they range from 055 to058) than with the dependent variables (they range from 023 to 036) We alsoobserve in Table I that managers report on average a high degree of innovationin their businesses not so high a level of customer loyalty and a level ofinnovation performance just at the scale mean The average self-reporteddegree of market orientation is 56 on a 0-90 scale

MethodAll hypotheses were contrasted using covariance structure analysis asimplemented in LISREL 850 (see Joreskog et al 1999) Since all three indicatorsof business performance are highly positively skewed and present a high

IJSIM143

294

degree of kurtosis throughout this paper rather than attempting to transformthese variables to near-normality we shall employ an estimation approach thatis robust to non-normality of the observed variables The parameter estimateswere obtained using maximum likelihood estimation with standard errorssuitable for non-normal data (Joreskog et al 1999 Equation A24) and twotest statistics were used to assess the goodness of regt (GFI) of the model theSatorra-Bentler scaled chi-squared statistic (Satorra and Bentler 1988Equation 41) and Brownersquos (1984 Equation 220a) chi-squared statisticcorrected for non-normality To evaluate better the goodness of regt of thismodel several additional indices will also provided the root mean squarederror of approximation (RMSEA (Steiger 1990)) the standardized root meansquared residual (SRMSR (Joreskog et al 1999)) the GFI (Tanaka and Huba1985) and the Comparative Fit Index (CFI) using the independence model asbaseline (Bentler 1990 see also McDonald and Marsh 1990) Adequate to goodregt is suggested by RMSEA and SRMSR values approaching 005 For the GFIand the CFI indices values between 090 and 100 indicate adequate to excellentregt (but see Hu and Bentler 1999)

ResultsH1The model used to estimate the effects of market orientation on insurancebusinessesrsquo performance consists of a latent variable representing economicperformance with three indicators (market share premium growth andproregtability) and a single exogenous variable (market orientation) This model

MST PG PROF INNODR INNPERF LOYAL MO

MST 100PG 020 100PROF 023 029 100INNODR 030 034 035 1000INNPERF 027 032 041 062 100LOYAL 025 019 038 040 038 100MO 024 030 036 055 058 057 100x 455 788 626 1825 1422 1996 5628Std 535 749 571 356 292 520 1331

NotesN = 122 all correlations are signiregcant (a = 001) except those marked by which are onlysigniregcant at an a = 005MS = market share PG = premium growth PROF = proregtability INNODR = innovation degreeINNPERF = innovation performance LOYAL = customer loyalty MO = market orientation

Table IMeans standard

deviations andinter-correlations

Businesseconomic

performance

295

is depicted in Figure 2 The parameter estimates and GFIs for this model aregiven in Table II The model shows a good regt although note that it only hastwo degrees of freedom According to the model the best objective indicator ofbusiness economic performance is proregtability per year over 34 percent of itsvariance is accounted for by the model The standardized regressioncoefregcients reveal that proregtability per year is the best objective indicator of

Parameter estimates Goodness of regt R 2

Parameter Value Index Value Variable Value

b1 006 [061] (002) MFF X 2001 ( p = 099) Market share 015

b2 166 [039] (049) S-B X 2001 ( p = 099) Premium growth 024

b3 293 [049] (078) B X 2001 ( p = 099) Proregtability 034

b4 267 [059] (084) RMSEA 001 Business performance 037u1 17713 [100] (2241) df 2u2 2427 [085] (713) SRMSR 001u3 4248 [076] (1364) GFI 100u4 2141 [066] (515) CFI 100

NotesRobust asymptotic standard errors are provided in parentheses standardized parameterestimates are provided in square bracketsMFF X 2 = Minimum regt function chi-square SB X 2 = Satorra-Bentler scaled chi-squareB X 2 = Brownersquos chi-square corrected for non-normality RMSEA = root mean squared error ofapproximation SRMSR = standardized root mean squared residual GFI = goodness of regt indexCFI = comparative regt index R 2 = squared multiple correlations for endogenous variables

Table IIEstimation resultsfor the modeldepicted in Figure 2

Figure 2Market orientation aspredictor of economicperformance

IJSIM143

296

overall business performance Finally according to the model almost 37percent of overall business economic performance is accounted for by thedegree of market orientation

An inspection of the total effects of market orientation on the indicators ofeconomic performance suggests that unit increments of market orientation asmeasured by the MOS-R are associated with 0095 0168 and 0153 incrementsin domestic market share premium growth and proregtability per year averagedover the last three years respectively

H2A mediated model for the relationship between market orientation andbusiness performance is depicted in Figure 3 In this context a mediating effectis said to exist when

both mediating paths b5 b6 are signiregcant and

the direct effect of the exogenous variable on the outcome variablevanishes (complete mediational effect) or is signiregcantly lower (partialmediational effect) when a mediator variable is introduced in the model

Condition (2) amounts to b1 in Figure 3 becoming zero or signiregcantly less thatthan value reported for Figure 2

We used the mediated model depicted in Figure 3 to test for mediatingeffects of innovation degree innovation performance and customer loyaltyseparately on the impact of market orientation on business economicperformance We found that when either innovation performance or innovation

Figure 3Mediated model

Businesseconomic

performance

297

degree were used as mediating variable all the mediating paths weresigniregcant and that direct path from market orientation to businessperformance was not signiregcantly different from zero b1 = 004 t = 162for innovation performance b1 = 004 t = 186 for innovation degree Hencetaken separately both innovation degree and innovation performancecompletely mediate the impact of market orientation on businessperformance After regxing b1 at zero we re-estimated these two mediationalmodels The resulting parameter estimates and GFIs for these two models areshown in Table III On the other hand customer loyalty was found not to havea mediational effect between market orientation and business performanceThe parameter estimates and GFIs for this model are also given in Table III

As can be seen in Table III the mediating paths are signiregcant but thedirect path b1 is signiregcantly different from zero at a = 001 Furthermore a 99percent conregdence interval for the value for b1 reported in Table II (003 009)includes the value of b1 estimated in the mediational model using customerloyalty 004 Hence this variable does not even partially mediate on the impactof market orientation on business economic performance The standardizeddirect impact of market orientation on business performance (0408) is morethan twice the standardized impact of market orientation conveyed throughcustomer loyalty (0191)

The percentage of variance of business economic performance explained bythe model when innovation performance innovation degree or customer loyaltyare used as mediators is very similar (465 percent 453 percent and 437percent respectively)

H3The full model to be regtted corresponding to the hypothesis depicted in Figure 1is presented in Figure 3 The parameter estimates and goodness of regt testcorresponding to this model are given in Table IV (see also Figure 4) As can beseen in this Table the model regts these data very well

All the postulated relationships were found to be signiregcant at an a = 001Lagrange multiplier tests indicated that the regt of the model would notsigniregcantly improve by

adding a direct effect of market orientation to business performance nor

adding a direct effect of innovation degree on economic performance

Result (1) is in accordance with the results discussed above where we saw thatinnovation degree and innovation performance even when taken separatelycompletely mediate the impact of market orientation on business performanceResult (2) conregrms our hypothesis that innovation performance completelymediates the impact of innovation degree on business performance

Given that all effects of market orientation on business performance gothrough either innovation degree-innovation performance or through customer

IJSIM143

298

Par

amet

eres

tim

ates

Goo

dnes

sof

regt

Par

amet

erV

alue

Par

amet

erV

alue

Index

Val

ue

Innova

tion

per

form

ance

as

med

iato

rb 1

0(reg

xed

)u

117

713

[10

0](2

241

)M

FF

X2

560

(p=

035

)b 2

155

[04

0](0

42)

u2

241

4[0

84]

(71

1)S-B

X2

380

(p=

058

)b 3

263

[04

8](0

67)

u3

431

6[0

77]

(45

5)B

X2

429

(p=

051

)b 4

248

[05

9](o

81)

u4

211

3[0

65]

(45

5)df

5R

MSE

A0

03b 5

023

[05

8](0

03)

u5

180

3[0

67]

(28

0)SR

MSR

005

GF

I0

98b 6

018

[06

8](0

05)

CF

I0

99

Innova

tion

deg

ree

as

med

iato

rb

10

(regxed

)u

117

713

[10

0](2

241

)M

FF

X2

717

(p=

021

)b 2

169

[05

0](0

36)

u2

234

0[0

82]

(73

8)S-B

X2

497

(p=

043

)b 3

282

[05

1](0

82)

u3

415

6[0

74]

(45

0)B

X2

613

(p=

029

)b 4

226

[05

4](0

79)

u4

232

9[0

71]

(46

0)df

5R

MSE

A0

06b 5

012

[05

5](0

02)

u5

590

[06

9](0

78)

SR

MSR

005

GF

I0

98b 6

031

[06

7](0

09)

CF

I0

98

Cust

om

erlo

yalty

as

med

iato

rb 1

004

[04

1](0

02)

u1

177

13[1

00]

(22

41)

MF

FX

22

30(p

=0

68)

b 21

61[0

40]

(05

1)u

224

05

[08

4](7

11)

S-B

X2

155

(p=

082

)b 3

252

[04

5](0

74)

u3

447

5[0

80]

(45

5)B

X2

309

(p=

054

)b 4

265

[06

2](0

92)

u4

201

3[0

62]

(45

5)R

MSE

A0

0df

4b 5

015

[05

7](0

02)

u5

863

[06

8](1

04)

SR

MSR

002

GF

I0

99b 6

013

[03

4](0

05)

CF

I1

00

Note

s

Rob

ust

asym

pto

tic

stan

dar

der

rors

are

pro

vid

edin

par

enth

eses

st

andar

diz

edpar

amet

eres

tim

ates

are

pro

vid

edin

squar

ebra

cket

sM

FF

X2=

Min

imum

regt

funct

ion

chi-sq

uar

eSB

X2=

Sat

orra

-Ben

tler

scal

edch

i-sq

uar

eB

X2=

Bro

wnersquo

sch

i-sq

uar

eco

rrec

ted

for

non

-nor

mal

ity

RM

SE

A=

root

mea

nsq

uar

eder

ror

ofap

pro

xim

atio

n

SR

MSR

=st

andar

diz

edro

otm

ean

squar

edre

sidual

G

FI

=goo

dnes

sof

regt

index

C

FI

=co

mpar

ativ

ereg

tin

dex

Table IIIEstimation results

for the modeldepicted in Figure 3

Businesseconomic

performance

299

Par

amet

eres

tim

ates

Goo

dnes

sof

regt

R2

Par

amet

erV

alue

Par

amet

erV

alue

Index

Val

ue

Var

iable

Val

ue

b 10

12[0

55]

(22

41)

u1

177

13[1

00]

(00

2)M

FF

X2

105

9(p

=0

56)

Mar

ket

shar

e0

16b 2

013

[03

4](0

04)

u2

590

[06

9](0

78)

S-B

X2

785

(p=

080

)P

rem

ium

gro

wth

019

b 30

15[0

57]

(00

2)u

314

48

[05

4](2

59)

BX

212

67

(p=

039

)P

roreg

tabilit

y0

38b 4

078

[04

4](0

15)

u4

863

[06

8](1

04)

df

12B

usi

nes

sper

form

ance

056

RM

SE

A0

0b 5

016

[05

4](0

05)

u5

239

1[0

84]

(72

3)SR

MSR

005

Innov

atio

nper

form

ance

046

b 60

16[0

37]

(00

7)u

645

09

[08

1](1

271

)G

FI

098

Innov

atio

ndeg

ree

031

b 71

42[0

40]

(04

5)u

720

12

[06

2](4

60)

CF

I1

00C

ust

omer

loyal

ty0

32b 8

217

[04

4](0

64)

b 92

32[0

62]

(08

5)

Note

sR

obust

asym

pto

tic

stan

dar

der

rors

are

pro

vid

edin

par

enth

eses

st

andar

diz

edpar

amet

eres

tim

ates

are

pro

vid

edin

squar

ebra

cket

sM

FF

X2=

Min

imum

regtfu

nct

ion

chi-sq

uar

eSB

X2=

Sat

orra

-Ben

tler

scal

edch

i-sq

uar

eB

X2=

Bro

wnersquo

sch

i-sq

uar

eco

rrec

ted

for

non

-nor

mal

ity

RM

SE

A=

root

mea

nsq

uar

eder

ror

ofap

pro

xim

atio

n

SR

MSR

=st

andar

diz

edro

otm

ean

squar

edre

sidual

G

FI

=goo

dnes

sof

regt

index

C

FI

=co

mpar

ativ

ereg

tin

dex

R

2=

squar

edm

ult

iple

corr

elat

ions

for

endog

enou

svar

iable

s

Table IVEstimation resultsfor the modeldepicted in Figure 4

IJSIM143

300

loyalty a question arises as to the relative importance of the speciregc effectsgoing through these variables The standardized speciregc effect (computed as inBollen 1987) going through innovation degree and innovation performance is031 and 021 going through customer loyalty Hence the impact of marketorientation going through innovation is 50 percent more than that goingthrough customer loyalty

We can also see in Table IV that over 30 percent of the variance of theintermediate variables (innovation degree innovation performance andcustomer loyalty) are explained by market orientation In fact almost 50percent (464 percent to be exact) of innovation performance is explained bymarket orientation Furthermore note that the percentage of variance ofbusiness performance explained by the model is 561 percent a 52 percentincrement over what is explained by market orientation alone (see Table II) andover a 20 percent increment over what is explained by the mediational modelsconsidered previously Hence the inclusion of all three intermediate variables inthe model improves considerably our prediction of business performance

Furthermore we observe in Table IV that the direct effect of marketorientation on all three intermediate variables appear to be equal Also thedirect effects of customer loyalty and of innovation performance on economicperformance appear to be equal We re-estimated the model to test theseconstraints obtaining b1 = b2 = b3 = 0133 b5 = b6 = 0157 Satorra-BentlerX 2(15) = 8849 p = 0885 CFI = 100 GFI = 0975 SRMR = 006

ConclusionsMarket orientation can be deregned as a strategy used to reach a sustainablecompetitive advantage based on the generation and use of information within

Figure 4Full model

Businesseconomic

performance

301

organizations and on the selection of markets to be satisreged In thisframework we believe that competitive advantage results from the use ofresources and capabilities to generate differential satisfaction in proregtablemarkets Sustainability is achieved because the performance of the marketorientationrsquos behaviors requires complex organizational knowledge that cannoteasily be imitated by competitors Thus we hypothesize that the satisfaction ofproregtable markets permits the regrm to achieve a psychologically differentialposition that leads to brand loyalty and thus to higher proregts Previous studieshave

found a clear impact of market orientation on economic performance

assessed the effects of market orientation on innovation and

investigated the relationship between market orientation and relationshipmarketing variables

The present study proposes and tests a model that integrates both streams ofresearch innovation and relationship marketing More research is needed toinvestigate the role of customer satisfaction within this framework

In our necessarily partial model innovation degree innovation performanceand customer loyalty are used as intermediate variables on the effect of marketorientation on business performance Our results suggest that the addition ofthese variables help improve our predictions of business economic performance52 percent over what is explained by market orientation alone Also we foundthat innovation degree and innovation performance each taken separatelycompletely mediate the effect of market orientation on economic performanceFurthermore the impact of innovation degree on economic performance iscompletely channeled through innovation performance Customer loyalty byitself does not meditate the impact of market orientation on economicperformance but when considered along with innovation degree andinnovation performance it conveys some of the effects of market orientationon business performance This seemingly contradictory result arises from thefact that all three intermediate variables are interrelated

Our results should not be taken to imply that there are no other variablesmediate the effect of market orientation on economic performance We believethat other variables that have not been taken into account in this study such asproduct quality and customer satisfaction may also be signiregcant mediatorsHowever our results do suggest that whenever innovation degree andinnovation performance are included in the model as intermediate variablesthe effects of market orientation on business performance will mostly beconveyed through these variables Also in this study we have adopted thecurrently most widely accepted approach to measuring customer loyalty whichis based on behavioral loyalty measures (eg share of category requirementsrenew the policy probability the average customer last in the companyportfolio) It would be worth extending the present study by including not only

IJSIM143

302

behavioral measures of customer loyalty but also attitudinal measures alongthe lines of Dick and Basu (1994)

In our opinion two important contributions of the present research are outuse of objective measures of business performance and our focusing oninternational markets Despite the growing role of globalization and marketintegration and despite the increasing internationalization of corporationsmost studies on market orientation have focused on domestic markets (withnotable exceptions such as Selnes et al 1996 Webster 1994) Similarly moststudies on product innovation have also focused on domestic markets There isa lack of research yielding empirical support to the validity in an internationalsetting to research results obtained in domestic markets To regll this gap wetargeted the European Union market

Our study focused on a single industry the insurance sector Our sampleaccounted for 22 percent of the companies and 17 percent of the insurancepremiums in the targeted market An advantage of our single-industryapproach is that (with obvious reservations arising from the non-experimentalnature of our study and the fact that our sample should not be considered tohave been obtained at random) we can draw tentative predictions from ourmodel concerning the impact of market orientation on economic performance ininsurance companies operating in the European Union market An evidentdrawback of the single-industry approach adopted here is that it is not clearhow the present results extrapolate to other industries even when operating inthe same market

We have found that within the European Union insurance regrms thatconstantly monitor the evolution of current and potential customerrsquos needsmodify the attributes of the products to adapt them to the distributorsrequirements analyze competitorsrsquo marketing policy and products and knowbest of environmental trends especially technological and legal changes aremore likely to develop more new products have their new products accepted bythe market and obtain more loyal customers In turn increased customerloyalty and increased new product success will result in improved economicperformance However these conclusions must be taken with some caution aswe have used overall measures of innovation and innovation performanceFurther research is needed that takes into account the various aspects thatconstitute innovation

Previous studies have concluded that insurance companies still seethemselves as being product-focused and the industry as a whole is generallydistribution driven (Sodano 2000 Lambin 1996) This lack of closeness withcustomers has in turn contributed to the industry lack of product innovationsand product portfolios of commodity products that only compete on price Nowthe most successful regrms are redirecting their focus to the market needs andthey are beginning to exploit customer data and use market research togenerate ideas for designing new products In this context these companies

Businesseconomic

performance

303

now face choosing among too many new service options The managerialimplication of our study is that by enhancing their market orientation regrmswill know and service its customers better Thus they will generate moreinnovations by adopting a market-based product development process Alsoincreasing levels of market orientation enable regrms to discriminate more easilywhich new products have a higher success probability thus enhancing both theefregcacy and efregciency of new product development

Note

1 We considered employing the total volume of premiums for each insurance company as anadditional indicator although it seemed to us to be a better indicator of a companyrsquos sizerather than of its performance As we suspected the total volume of premiums wasuncorrelated with any of the variables considered in this study except for the companyrsquosmarket share (r = 024 p 001) Hence it is clear that volume of premiums should not beused as an indicator of insurance companiesrsquo performance

References

Armstrong JS and Overton T (1977) ordfEstimating nonresponse bias in mail surveysordm Journalof Marketing Research Vol 14 pp 396-402

Atuahene-Gima K (1996) ordfMarket orientation and innovationordm Journal of Business ResearchVol 35 pp 93-103

Baker TL Simpson PM and Sigauw JA (1999) ordfThe impact of suppliersrsquo perceptions ofreseller market orientation on key relationship constructsordm Journal of the Academy ofMarketing Science Vol 27 No 1 pp 50-7

Baron RM and Kenny DA (1986) ordfThe moderator-mediator variable distinction in socialpsychological research conceptual strategic and statistical considerationsordm Journal ofPersonality and Social Psychology Vol 51 pp 1173-82

Bentler PM (1990) ordfComparative regt indexes in structural modelsordm Psychological BulletinVol 107 pp 238-46

Bhuian SN (1997) ordfExploring market orientation in banks an empirical examination in SaudiArabiaordm The Journal of Service Marketing Vol 11 No 5 pp 317-28

Bollen KA (1987) ordfTotal direct and indirect effects in structural equation modelsordm inClogg CC (Ed) Sociological Methodology American Sociological AssociationWashington DC pp 37-69

Browne MW (1984) ordfAsymptotically distribution free methods for the analysis of covariancestructuresordm British Journal of Mathematical and Statistical Psychology Vol 37 pp 62-83

Calantone RJ di Benedetto AC and Bhoovaraghavan S (1994) ordfExamining the relationshipbetween degree of innovation and new product successordm Journal of Business ResearchVol 30 pp 143-8

Calantone RJ Schmidt JB and Song MX (1996) ordfControllable factors of new productssuccess a cross-national comparisonordm Marketing Science Vol 15 No 4 pp 341-58

Cooper RC (1994) ordfNew product the factors that drive successordm International MarketingReview Vol 11 No 1 pp 60-76

Day G (1992) ordfMarketingrsquos contribution to the strategy dialogueordm Journal of the Academy ofMarketing Science Vol 20 Fall pp 323-9

IJSIM143

304

degree of kurtosis throughout this paper rather than attempting to transformthese variables to near-normality we shall employ an estimation approach thatis robust to non-normality of the observed variables The parameter estimateswere obtained using maximum likelihood estimation with standard errorssuitable for non-normal data (Joreskog et al 1999 Equation A24) and twotest statistics were used to assess the goodness of regt (GFI) of the model theSatorra-Bentler scaled chi-squared statistic (Satorra and Bentler 1988Equation 41) and Brownersquos (1984 Equation 220a) chi-squared statisticcorrected for non-normality To evaluate better the goodness of regt of thismodel several additional indices will also provided the root mean squarederror of approximation (RMSEA (Steiger 1990)) the standardized root meansquared residual (SRMSR (Joreskog et al 1999)) the GFI (Tanaka and Huba1985) and the Comparative Fit Index (CFI) using the independence model asbaseline (Bentler 1990 see also McDonald and Marsh 1990) Adequate to goodregt is suggested by RMSEA and SRMSR values approaching 005 For the GFIand the CFI indices values between 090 and 100 indicate adequate to excellentregt (but see Hu and Bentler 1999)

ResultsH1The model used to estimate the effects of market orientation on insurancebusinessesrsquo performance consists of a latent variable representing economicperformance with three indicators (market share premium growth andproregtability) and a single exogenous variable (market orientation) This model

MST PG PROF INNODR INNPERF LOYAL MO

MST 100PG 020 100PROF 023 029 100INNODR 030 034 035 1000INNPERF 027 032 041 062 100LOYAL 025 019 038 040 038 100MO 024 030 036 055 058 057 100x 455 788 626 1825 1422 1996 5628Std 535 749 571 356 292 520 1331

NotesN = 122 all correlations are signiregcant (a = 001) except those marked by which are onlysigniregcant at an a = 005MS = market share PG = premium growth PROF = proregtability INNODR = innovation degreeINNPERF = innovation performance LOYAL = customer loyalty MO = market orientation

Table IMeans standard

deviations andinter-correlations

Businesseconomic

performance

295

is depicted in Figure 2 The parameter estimates and GFIs for this model aregiven in Table II The model shows a good regt although note that it only hastwo degrees of freedom According to the model the best objective indicator ofbusiness economic performance is proregtability per year over 34 percent of itsvariance is accounted for by the model The standardized regressioncoefregcients reveal that proregtability per year is the best objective indicator of

Parameter estimates Goodness of regt R 2

Parameter Value Index Value Variable Value

b1 006 [061] (002) MFF X 2001 ( p = 099) Market share 015

b2 166 [039] (049) S-B X 2001 ( p = 099) Premium growth 024

b3 293 [049] (078) B X 2001 ( p = 099) Proregtability 034

b4 267 [059] (084) RMSEA 001 Business performance 037u1 17713 [100] (2241) df 2u2 2427 [085] (713) SRMSR 001u3 4248 [076] (1364) GFI 100u4 2141 [066] (515) CFI 100

NotesRobust asymptotic standard errors are provided in parentheses standardized parameterestimates are provided in square bracketsMFF X 2 = Minimum regt function chi-square SB X 2 = Satorra-Bentler scaled chi-squareB X 2 = Brownersquos chi-square corrected for non-normality RMSEA = root mean squared error ofapproximation SRMSR = standardized root mean squared residual GFI = goodness of regt indexCFI = comparative regt index R 2 = squared multiple correlations for endogenous variables

Table IIEstimation resultsfor the modeldepicted in Figure 2

Figure 2Market orientation aspredictor of economicperformance

IJSIM143

296

overall business performance Finally according to the model almost 37percent of overall business economic performance is accounted for by thedegree of market orientation

An inspection of the total effects of market orientation on the indicators ofeconomic performance suggests that unit increments of market orientation asmeasured by the MOS-R are associated with 0095 0168 and 0153 incrementsin domestic market share premium growth and proregtability per year averagedover the last three years respectively

H2A mediated model for the relationship between market orientation andbusiness performance is depicted in Figure 3 In this context a mediating effectis said to exist when

both mediating paths b5 b6 are signiregcant and

the direct effect of the exogenous variable on the outcome variablevanishes (complete mediational effect) or is signiregcantly lower (partialmediational effect) when a mediator variable is introduced in the model

Condition (2) amounts to b1 in Figure 3 becoming zero or signiregcantly less thatthan value reported for Figure 2

We used the mediated model depicted in Figure 3 to test for mediatingeffects of innovation degree innovation performance and customer loyaltyseparately on the impact of market orientation on business economicperformance We found that when either innovation performance or innovation

Figure 3Mediated model

Businesseconomic

performance

297

degree were used as mediating variable all the mediating paths weresigniregcant and that direct path from market orientation to businessperformance was not signiregcantly different from zero b1 = 004 t = 162for innovation performance b1 = 004 t = 186 for innovation degree Hencetaken separately both innovation degree and innovation performancecompletely mediate the impact of market orientation on businessperformance After regxing b1 at zero we re-estimated these two mediationalmodels The resulting parameter estimates and GFIs for these two models areshown in Table III On the other hand customer loyalty was found not to havea mediational effect between market orientation and business performanceThe parameter estimates and GFIs for this model are also given in Table III

As can be seen in Table III the mediating paths are signiregcant but thedirect path b1 is signiregcantly different from zero at a = 001 Furthermore a 99percent conregdence interval for the value for b1 reported in Table II (003 009)includes the value of b1 estimated in the mediational model using customerloyalty 004 Hence this variable does not even partially mediate on the impactof market orientation on business economic performance The standardizeddirect impact of market orientation on business performance (0408) is morethan twice the standardized impact of market orientation conveyed throughcustomer loyalty (0191)

The percentage of variance of business economic performance explained bythe model when innovation performance innovation degree or customer loyaltyare used as mediators is very similar (465 percent 453 percent and 437percent respectively)

H3The full model to be regtted corresponding to the hypothesis depicted in Figure 1is presented in Figure 3 The parameter estimates and goodness of regt testcorresponding to this model are given in Table IV (see also Figure 4) As can beseen in this Table the model regts these data very well

All the postulated relationships were found to be signiregcant at an a = 001Lagrange multiplier tests indicated that the regt of the model would notsigniregcantly improve by

adding a direct effect of market orientation to business performance nor

adding a direct effect of innovation degree on economic performance

Result (1) is in accordance with the results discussed above where we saw thatinnovation degree and innovation performance even when taken separatelycompletely mediate the impact of market orientation on business performanceResult (2) conregrms our hypothesis that innovation performance completelymediates the impact of innovation degree on business performance

Given that all effects of market orientation on business performance gothrough either innovation degree-innovation performance or through customer

IJSIM143

298

Par

amet

eres

tim

ates

Goo

dnes

sof

regt

Par

amet

erV

alue

Par

amet

erV

alue

Index

Val

ue

Innova

tion

per

form

ance

as

med

iato

rb 1

0(reg

xed

)u

117

713

[10

0](2

241

)M

FF

X2

560

(p=

035

)b 2

155

[04

0](0

42)

u2

241

4[0

84]

(71

1)S-B

X2

380

(p=

058

)b 3

263

[04

8](0

67)

u3

431

6[0

77]

(45

5)B

X2

429

(p=

051

)b 4

248

[05

9](o

81)

u4

211

3[0

65]

(45

5)df

5R

MSE

A0

03b 5

023

[05

8](0

03)

u5

180

3[0

67]

(28

0)SR

MSR

005

GF

I0

98b 6

018

[06

8](0

05)

CF

I0

99

Innova

tion

deg

ree

as

med

iato

rb

10

(regxed

)u

117

713

[10

0](2

241

)M

FF

X2

717

(p=

021

)b 2

169

[05

0](0

36)

u2

234

0[0

82]

(73

8)S-B

X2

497

(p=

043

)b 3

282

[05

1](0

82)

u3

415

6[0

74]

(45

0)B

X2

613

(p=

029

)b 4

226

[05

4](0

79)

u4

232

9[0

71]

(46

0)df

5R

MSE

A0

06b 5

012

[05

5](0

02)

u5

590

[06

9](0

78)

SR

MSR

005

GF

I0

98b 6

031

[06

7](0

09)

CF

I0

98

Cust

om

erlo

yalty

as

med

iato

rb 1

004

[04

1](0

02)

u1

177

13[1

00]

(22

41)

MF

FX

22

30(p

=0

68)

b 21

61[0

40]

(05

1)u

224

05

[08

4](7

11)

S-B

X2

155

(p=

082

)b 3

252

[04

5](0

74)

u3

447

5[0

80]

(45

5)B

X2

309

(p=

054

)b 4

265

[06

2](0

92)

u4

201

3[0

62]

(45

5)R

MSE

A0

0df

4b 5

015

[05

7](0

02)

u5

863

[06

8](1

04)

SR

MSR

002

GF

I0

99b 6

013

[03

4](0

05)

CF

I1

00

Note

s

Rob

ust

asym

pto

tic

stan

dar

der

rors

are

pro

vid

edin

par

enth

eses

st

andar

diz

edpar

amet

eres

tim

ates

are

pro

vid

edin

squar

ebra

cket

sM

FF

X2=

Min

imum

regt

funct

ion

chi-sq

uar

eSB

X2=

Sat

orra

-Ben

tler

scal

edch

i-sq

uar

eB

X2=

Bro

wnersquo

sch

i-sq

uar

eco

rrec

ted

for

non

-nor

mal

ity

RM

SE

A=

root

mea

nsq

uar

eder

ror

ofap

pro

xim

atio

n

SR

MSR

=st

andar

diz

edro

otm

ean

squar

edre

sidual

G

FI

=goo

dnes

sof

regt

index

C

FI

=co

mpar

ativ

ereg

tin

dex

Table IIIEstimation results

for the modeldepicted in Figure 3

Businesseconomic

performance

299

Par

amet

eres

tim

ates

Goo

dnes

sof

regt

R2

Par

amet

erV

alue

Par

amet

erV

alue

Index

Val

ue

Var

iable

Val

ue

b 10

12[0

55]

(22

41)

u1

177

13[1

00]

(00

2)M

FF

X2

105

9(p

=0

56)

Mar

ket

shar

e0

16b 2

013

[03

4](0

04)

u2

590

[06

9](0

78)

S-B

X2

785

(p=

080

)P

rem

ium

gro

wth

019

b 30

15[0

57]

(00

2)u

314

48

[05

4](2

59)

BX

212

67

(p=

039

)P

roreg

tabilit

y0

38b 4

078

[04

4](0

15)

u4

863

[06

8](1

04)

df

12B

usi

nes

sper

form

ance

056

RM

SE

A0

0b 5

016

[05

4](0

05)

u5

239

1[0

84]

(72

3)SR

MSR

005

Innov

atio

nper

form

ance

046

b 60

16[0

37]

(00

7)u

645

09

[08

1](1

271

)G

FI

098

Innov

atio

ndeg

ree

031

b 71

42[0

40]

(04

5)u

720

12

[06

2](4

60)

CF

I1

00C

ust

omer

loyal

ty0

32b 8

217

[04

4](0

64)

b 92

32[0

62]

(08

5)

Note

sR

obust

asym

pto

tic

stan

dar

der

rors

are

pro

vid

edin

par

enth

eses

st

andar

diz

edpar

amet

eres

tim

ates

are

pro

vid

edin

squar

ebra

cket

sM

FF

X2=

Min

imum

regtfu

nct

ion

chi-sq

uar

eSB

X2=

Sat

orra

-Ben

tler

scal

edch

i-sq

uar

eB

X2=

Bro

wnersquo

sch

i-sq

uar

eco

rrec

ted

for

non

-nor

mal

ity

RM

SE

A=

root

mea

nsq

uar

eder

ror

ofap

pro

xim

atio

n

SR

MSR

=st

andar

diz

edro

otm

ean

squar

edre

sidual

G

FI

=goo

dnes

sof

regt

index

C

FI

=co

mpar

ativ

ereg

tin

dex

R

2=

squar

edm

ult

iple

corr

elat

ions

for

endog

enou

svar

iable

s

Table IVEstimation resultsfor the modeldepicted in Figure 4

IJSIM143

300

loyalty a question arises as to the relative importance of the speciregc effectsgoing through these variables The standardized speciregc effect (computed as inBollen 1987) going through innovation degree and innovation performance is031 and 021 going through customer loyalty Hence the impact of marketorientation going through innovation is 50 percent more than that goingthrough customer loyalty

We can also see in Table IV that over 30 percent of the variance of theintermediate variables (innovation degree innovation performance andcustomer loyalty) are explained by market orientation In fact almost 50percent (464 percent to be exact) of innovation performance is explained bymarket orientation Furthermore note that the percentage of variance ofbusiness performance explained by the model is 561 percent a 52 percentincrement over what is explained by market orientation alone (see Table II) andover a 20 percent increment over what is explained by the mediational modelsconsidered previously Hence the inclusion of all three intermediate variables inthe model improves considerably our prediction of business performance

Furthermore we observe in Table IV that the direct effect of marketorientation on all three intermediate variables appear to be equal Also thedirect effects of customer loyalty and of innovation performance on economicperformance appear to be equal We re-estimated the model to test theseconstraints obtaining b1 = b2 = b3 = 0133 b5 = b6 = 0157 Satorra-BentlerX 2(15) = 8849 p = 0885 CFI = 100 GFI = 0975 SRMR = 006

ConclusionsMarket orientation can be deregned as a strategy used to reach a sustainablecompetitive advantage based on the generation and use of information within

Figure 4Full model

Businesseconomic

performance

301

organizations and on the selection of markets to be satisreged In thisframework we believe that competitive advantage results from the use ofresources and capabilities to generate differential satisfaction in proregtablemarkets Sustainability is achieved because the performance of the marketorientationrsquos behaviors requires complex organizational knowledge that cannoteasily be imitated by competitors Thus we hypothesize that the satisfaction ofproregtable markets permits the regrm to achieve a psychologically differentialposition that leads to brand loyalty and thus to higher proregts Previous studieshave

found a clear impact of market orientation on economic performance

assessed the effects of market orientation on innovation and

investigated the relationship between market orientation and relationshipmarketing variables

The present study proposes and tests a model that integrates both streams ofresearch innovation and relationship marketing More research is needed toinvestigate the role of customer satisfaction within this framework

In our necessarily partial model innovation degree innovation performanceand customer loyalty are used as intermediate variables on the effect of marketorientation on business performance Our results suggest that the addition ofthese variables help improve our predictions of business economic performance52 percent over what is explained by market orientation alone Also we foundthat innovation degree and innovation performance each taken separatelycompletely mediate the effect of market orientation on economic performanceFurthermore the impact of innovation degree on economic performance iscompletely channeled through innovation performance Customer loyalty byitself does not meditate the impact of market orientation on economicperformance but when considered along with innovation degree andinnovation performance it conveys some of the effects of market orientationon business performance This seemingly contradictory result arises from thefact that all three intermediate variables are interrelated

Our results should not be taken to imply that there are no other variablesmediate the effect of market orientation on economic performance We believethat other variables that have not been taken into account in this study such asproduct quality and customer satisfaction may also be signiregcant mediatorsHowever our results do suggest that whenever innovation degree andinnovation performance are included in the model as intermediate variablesthe effects of market orientation on business performance will mostly beconveyed through these variables Also in this study we have adopted thecurrently most widely accepted approach to measuring customer loyalty whichis based on behavioral loyalty measures (eg share of category requirementsrenew the policy probability the average customer last in the companyportfolio) It would be worth extending the present study by including not only

IJSIM143

302

behavioral measures of customer loyalty but also attitudinal measures alongthe lines of Dick and Basu (1994)

In our opinion two important contributions of the present research are outuse of objective measures of business performance and our focusing oninternational markets Despite the growing role of globalization and marketintegration and despite the increasing internationalization of corporationsmost studies on market orientation have focused on domestic markets (withnotable exceptions such as Selnes et al 1996 Webster 1994) Similarly moststudies on product innovation have also focused on domestic markets There isa lack of research yielding empirical support to the validity in an internationalsetting to research results obtained in domestic markets To regll this gap wetargeted the European Union market

Our study focused on a single industry the insurance sector Our sampleaccounted for 22 percent of the companies and 17 percent of the insurancepremiums in the targeted market An advantage of our single-industryapproach is that (with obvious reservations arising from the non-experimentalnature of our study and the fact that our sample should not be considered tohave been obtained at random) we can draw tentative predictions from ourmodel concerning the impact of market orientation on economic performance ininsurance companies operating in the European Union market An evidentdrawback of the single-industry approach adopted here is that it is not clearhow the present results extrapolate to other industries even when operating inthe same market

We have found that within the European Union insurance regrms thatconstantly monitor the evolution of current and potential customerrsquos needsmodify the attributes of the products to adapt them to the distributorsrequirements analyze competitorsrsquo marketing policy and products and knowbest of environmental trends especially technological and legal changes aremore likely to develop more new products have their new products accepted bythe market and obtain more loyal customers In turn increased customerloyalty and increased new product success will result in improved economicperformance However these conclusions must be taken with some caution aswe have used overall measures of innovation and innovation performanceFurther research is needed that takes into account the various aspects thatconstitute innovation

Previous studies have concluded that insurance companies still seethemselves as being product-focused and the industry as a whole is generallydistribution driven (Sodano 2000 Lambin 1996) This lack of closeness withcustomers has in turn contributed to the industry lack of product innovationsand product portfolios of commodity products that only compete on price Nowthe most successful regrms are redirecting their focus to the market needs andthey are beginning to exploit customer data and use market research togenerate ideas for designing new products In this context these companies

Businesseconomic

performance

303

now face choosing among too many new service options The managerialimplication of our study is that by enhancing their market orientation regrmswill know and service its customers better Thus they will generate moreinnovations by adopting a market-based product development process Alsoincreasing levels of market orientation enable regrms to discriminate more easilywhich new products have a higher success probability thus enhancing both theefregcacy and efregciency of new product development

Note

1 We considered employing the total volume of premiums for each insurance company as anadditional indicator although it seemed to us to be a better indicator of a companyrsquos sizerather than of its performance As we suspected the total volume of premiums wasuncorrelated with any of the variables considered in this study except for the companyrsquosmarket share (r = 024 p 001) Hence it is clear that volume of premiums should not beused as an indicator of insurance companiesrsquo performance

References

Armstrong JS and Overton T (1977) ordfEstimating nonresponse bias in mail surveysordm Journalof Marketing Research Vol 14 pp 396-402

Atuahene-Gima K (1996) ordfMarket orientation and innovationordm Journal of Business ResearchVol 35 pp 93-103

Baker TL Simpson PM and Sigauw JA (1999) ordfThe impact of suppliersrsquo perceptions ofreseller market orientation on key relationship constructsordm Journal of the Academy ofMarketing Science Vol 27 No 1 pp 50-7

Baron RM and Kenny DA (1986) ordfThe moderator-mediator variable distinction in socialpsychological research conceptual strategic and statistical considerationsordm Journal ofPersonality and Social Psychology Vol 51 pp 1173-82

Bentler PM (1990) ordfComparative regt indexes in structural modelsordm Psychological BulletinVol 107 pp 238-46

Bhuian SN (1997) ordfExploring market orientation in banks an empirical examination in SaudiArabiaordm The Journal of Service Marketing Vol 11 No 5 pp 317-28

Bollen KA (1987) ordfTotal direct and indirect effects in structural equation modelsordm inClogg CC (Ed) Sociological Methodology American Sociological AssociationWashington DC pp 37-69

Browne MW (1984) ordfAsymptotically distribution free methods for the analysis of covariancestructuresordm British Journal of Mathematical and Statistical Psychology Vol 37 pp 62-83

Calantone RJ di Benedetto AC and Bhoovaraghavan S (1994) ordfExamining the relationshipbetween degree of innovation and new product successordm Journal of Business ResearchVol 30 pp 143-8

Calantone RJ Schmidt JB and Song MX (1996) ordfControllable factors of new productssuccess a cross-national comparisonordm Marketing Science Vol 15 No 4 pp 341-58

Cooper RC (1994) ordfNew product the factors that drive successordm International MarketingReview Vol 11 No 1 pp 60-76

Day G (1992) ordfMarketingrsquos contribution to the strategy dialogueordm Journal of the Academy ofMarketing Science Vol 20 Fall pp 323-9

IJSIM143

304

is depicted in Figure 2 The parameter estimates and GFIs for this model aregiven in Table II The model shows a good regt although note that it only hastwo degrees of freedom According to the model the best objective indicator ofbusiness economic performance is proregtability per year over 34 percent of itsvariance is accounted for by the model The standardized regressioncoefregcients reveal that proregtability per year is the best objective indicator of

Parameter estimates Goodness of regt R 2

Parameter Value Index Value Variable Value

b1 006 [061] (002) MFF X 2001 ( p = 099) Market share 015

b2 166 [039] (049) S-B X 2001 ( p = 099) Premium growth 024

b3 293 [049] (078) B X 2001 ( p = 099) Proregtability 034

b4 267 [059] (084) RMSEA 001 Business performance 037u1 17713 [100] (2241) df 2u2 2427 [085] (713) SRMSR 001u3 4248 [076] (1364) GFI 100u4 2141 [066] (515) CFI 100

NotesRobust asymptotic standard errors are provided in parentheses standardized parameterestimates are provided in square bracketsMFF X 2 = Minimum regt function chi-square SB X 2 = Satorra-Bentler scaled chi-squareB X 2 = Brownersquos chi-square corrected for non-normality RMSEA = root mean squared error ofapproximation SRMSR = standardized root mean squared residual GFI = goodness of regt indexCFI = comparative regt index R 2 = squared multiple correlations for endogenous variables

Table IIEstimation resultsfor the modeldepicted in Figure 2

Figure 2Market orientation aspredictor of economicperformance

IJSIM143

296

overall business performance Finally according to the model almost 37percent of overall business economic performance is accounted for by thedegree of market orientation

An inspection of the total effects of market orientation on the indicators ofeconomic performance suggests that unit increments of market orientation asmeasured by the MOS-R are associated with 0095 0168 and 0153 incrementsin domestic market share premium growth and proregtability per year averagedover the last three years respectively

H2A mediated model for the relationship between market orientation andbusiness performance is depicted in Figure 3 In this context a mediating effectis said to exist when

both mediating paths b5 b6 are signiregcant and

the direct effect of the exogenous variable on the outcome variablevanishes (complete mediational effect) or is signiregcantly lower (partialmediational effect) when a mediator variable is introduced in the model

Condition (2) amounts to b1 in Figure 3 becoming zero or signiregcantly less thatthan value reported for Figure 2

We used the mediated model depicted in Figure 3 to test for mediatingeffects of innovation degree innovation performance and customer loyaltyseparately on the impact of market orientation on business economicperformance We found that when either innovation performance or innovation

Figure 3Mediated model

Businesseconomic

performance

297

degree were used as mediating variable all the mediating paths weresigniregcant and that direct path from market orientation to businessperformance was not signiregcantly different from zero b1 = 004 t = 162for innovation performance b1 = 004 t = 186 for innovation degree Hencetaken separately both innovation degree and innovation performancecompletely mediate the impact of market orientation on businessperformance After regxing b1 at zero we re-estimated these two mediationalmodels The resulting parameter estimates and GFIs for these two models areshown in Table III On the other hand customer loyalty was found not to havea mediational effect between market orientation and business performanceThe parameter estimates and GFIs for this model are also given in Table III

As can be seen in Table III the mediating paths are signiregcant but thedirect path b1 is signiregcantly different from zero at a = 001 Furthermore a 99percent conregdence interval for the value for b1 reported in Table II (003 009)includes the value of b1 estimated in the mediational model using customerloyalty 004 Hence this variable does not even partially mediate on the impactof market orientation on business economic performance The standardizeddirect impact of market orientation on business performance (0408) is morethan twice the standardized impact of market orientation conveyed throughcustomer loyalty (0191)

The percentage of variance of business economic performance explained bythe model when innovation performance innovation degree or customer loyaltyare used as mediators is very similar (465 percent 453 percent and 437percent respectively)

H3The full model to be regtted corresponding to the hypothesis depicted in Figure 1is presented in Figure 3 The parameter estimates and goodness of regt testcorresponding to this model are given in Table IV (see also Figure 4) As can beseen in this Table the model regts these data very well

All the postulated relationships were found to be signiregcant at an a = 001Lagrange multiplier tests indicated that the regt of the model would notsigniregcantly improve by

adding a direct effect of market orientation to business performance nor

adding a direct effect of innovation degree on economic performance

Result (1) is in accordance with the results discussed above where we saw thatinnovation degree and innovation performance even when taken separatelycompletely mediate the impact of market orientation on business performanceResult (2) conregrms our hypothesis that innovation performance completelymediates the impact of innovation degree on business performance

Given that all effects of market orientation on business performance gothrough either innovation degree-innovation performance or through customer

IJSIM143

298

Par

amet

eres

tim

ates

Goo

dnes

sof

regt

Par

amet

erV

alue

Par

amet

erV

alue

Index

Val

ue

Innova

tion

per

form

ance

as

med

iato

rb 1

0(reg

xed

)u

117

713

[10

0](2

241

)M

FF

X2

560

(p=

035

)b 2

155

[04

0](0

42)

u2

241

4[0

84]

(71

1)S-B

X2

380

(p=

058

)b 3

263

[04

8](0

67)

u3

431

6[0

77]

(45

5)B

X2

429

(p=

051

)b 4

248

[05

9](o

81)

u4

211

3[0

65]

(45

5)df

5R

MSE

A0

03b 5

023

[05

8](0

03)

u5

180

3[0

67]

(28

0)SR

MSR

005

GF

I0

98b 6

018

[06

8](0

05)

CF

I0

99

Innova

tion

deg

ree

as

med

iato

rb

10

(regxed

)u

117

713

[10

0](2

241

)M

FF

X2

717

(p=

021

)b 2

169

[05

0](0

36)

u2

234

0[0

82]

(73

8)S-B

X2

497

(p=

043

)b 3

282

[05

1](0

82)

u3

415

6[0

74]

(45

0)B

X2

613

(p=

029

)b 4

226

[05

4](0

79)

u4

232

9[0

71]

(46

0)df

5R

MSE

A0

06b 5

012

[05

5](0

02)

u5

590

[06

9](0

78)

SR

MSR

005

GF

I0

98b 6

031

[06

7](0

09)

CF

I0

98

Cust

om

erlo

yalty

as

med

iato

rb 1

004

[04

1](0

02)

u1

177

13[1

00]

(22

41)

MF

FX

22

30(p

=0

68)

b 21

61[0

40]

(05

1)u

224

05

[08

4](7

11)

S-B

X2

155

(p=

082

)b 3

252

[04

5](0

74)

u3

447

5[0

80]

(45

5)B

X2

309

(p=

054

)b 4

265

[06

2](0

92)

u4

201

3[0

62]

(45

5)R

MSE

A0

0df

4b 5

015

[05

7](0

02)

u5

863

[06

8](1

04)

SR

MSR

002

GF

I0

99b 6

013

[03

4](0

05)

CF

I1

00

Note

s

Rob

ust

asym

pto

tic

stan

dar

der

rors

are

pro

vid

edin

par

enth

eses

st

andar

diz

edpar

amet

eres

tim

ates

are

pro

vid

edin

squar

ebra

cket

sM

FF

X2=

Min

imum

regt

funct

ion

chi-sq

uar

eSB

X2=

Sat

orra

-Ben

tler

scal

edch

i-sq

uar

eB

X2=

Bro

wnersquo

sch

i-sq

uar

eco

rrec

ted

for

non

-nor

mal

ity

RM

SE

A=

root

mea

nsq

uar

eder

ror

ofap

pro

xim

atio

n

SR

MSR

=st

andar

diz

edro

otm

ean

squar

edre

sidual

G

FI

=goo

dnes

sof

regt

index

C

FI

=co

mpar

ativ

ereg

tin

dex

Table IIIEstimation results

for the modeldepicted in Figure 3

Businesseconomic

performance

299

Par

amet

eres

tim

ates

Goo

dnes

sof

regt

R2

Par

amet

erV

alue

Par

amet

erV

alue

Index

Val

ue

Var

iable

Val

ue

b 10

12[0

55]

(22

41)

u1

177

13[1

00]

(00

2)M

FF

X2

105

9(p

=0

56)

Mar

ket

shar

e0

16b 2

013

[03

4](0

04)

u2

590

[06

9](0

78)

S-B

X2

785

(p=

080

)P

rem

ium

gro

wth

019

b 30

15[0

57]

(00

2)u

314

48

[05

4](2

59)

BX

212

67

(p=

039

)P

roreg

tabilit

y0

38b 4

078

[04

4](0

15)

u4

863

[06

8](1

04)

df

12B

usi

nes

sper

form

ance

056

RM

SE

A0

0b 5

016

[05

4](0

05)

u5

239

1[0

84]

(72

3)SR

MSR

005

Innov

atio

nper

form

ance

046

b 60

16[0

37]

(00

7)u

645

09

[08

1](1

271

)G

FI

098

Innov

atio

ndeg

ree

031

b 71

42[0

40]

(04

5)u

720

12

[06

2](4

60)

CF

I1

00C

ust

omer

loyal

ty0

32b 8

217

[04

4](0

64)

b 92

32[0

62]

(08

5)

Note

sR

obust

asym

pto

tic

stan

dar

der

rors

are

pro

vid

edin

par

enth

eses

st

andar

diz

edpar

amet

eres

tim

ates

are

pro

vid

edin

squar

ebra

cket

sM

FF

X2=

Min

imum

regtfu

nct

ion

chi-sq

uar

eSB

X2=

Sat

orra

-Ben

tler

scal

edch

i-sq

uar

eB

X2=

Bro

wnersquo

sch

i-sq

uar

eco

rrec

ted

for

non

-nor

mal

ity

RM

SE

A=

root

mea

nsq

uar

eder

ror

ofap

pro

xim

atio

n

SR

MSR

=st

andar

diz

edro

otm

ean

squar

edre

sidual

G

FI

=goo

dnes

sof

regt

index

C

FI

=co

mpar

ativ

ereg

tin

dex

R

2=

squar

edm

ult

iple

corr

elat

ions

for

endog

enou

svar

iable

s

Table IVEstimation resultsfor the modeldepicted in Figure 4

IJSIM143

300

loyalty a question arises as to the relative importance of the speciregc effectsgoing through these variables The standardized speciregc effect (computed as inBollen 1987) going through innovation degree and innovation performance is031 and 021 going through customer loyalty Hence the impact of marketorientation going through innovation is 50 percent more than that goingthrough customer loyalty

We can also see in Table IV that over 30 percent of the variance of theintermediate variables (innovation degree innovation performance andcustomer loyalty) are explained by market orientation In fact almost 50percent (464 percent to be exact) of innovation performance is explained bymarket orientation Furthermore note that the percentage of variance ofbusiness performance explained by the model is 561 percent a 52 percentincrement over what is explained by market orientation alone (see Table II) andover a 20 percent increment over what is explained by the mediational modelsconsidered previously Hence the inclusion of all three intermediate variables inthe model improves considerably our prediction of business performance

Furthermore we observe in Table IV that the direct effect of marketorientation on all three intermediate variables appear to be equal Also thedirect effects of customer loyalty and of innovation performance on economicperformance appear to be equal We re-estimated the model to test theseconstraints obtaining b1 = b2 = b3 = 0133 b5 = b6 = 0157 Satorra-BentlerX 2(15) = 8849 p = 0885 CFI = 100 GFI = 0975 SRMR = 006

ConclusionsMarket orientation can be deregned as a strategy used to reach a sustainablecompetitive advantage based on the generation and use of information within

Figure 4Full model

Businesseconomic

performance

301

organizations and on the selection of markets to be satisreged In thisframework we believe that competitive advantage results from the use ofresources and capabilities to generate differential satisfaction in proregtablemarkets Sustainability is achieved because the performance of the marketorientationrsquos behaviors requires complex organizational knowledge that cannoteasily be imitated by competitors Thus we hypothesize that the satisfaction ofproregtable markets permits the regrm to achieve a psychologically differentialposition that leads to brand loyalty and thus to higher proregts Previous studieshave

found a clear impact of market orientation on economic performance

assessed the effects of market orientation on innovation and

investigated the relationship between market orientation and relationshipmarketing variables

The present study proposes and tests a model that integrates both streams ofresearch innovation and relationship marketing More research is needed toinvestigate the role of customer satisfaction within this framework

In our necessarily partial model innovation degree innovation performanceand customer loyalty are used as intermediate variables on the effect of marketorientation on business performance Our results suggest that the addition ofthese variables help improve our predictions of business economic performance52 percent over what is explained by market orientation alone Also we foundthat innovation degree and innovation performance each taken separatelycompletely mediate the effect of market orientation on economic performanceFurthermore the impact of innovation degree on economic performance iscompletely channeled through innovation performance Customer loyalty byitself does not meditate the impact of market orientation on economicperformance but when considered along with innovation degree andinnovation performance it conveys some of the effects of market orientationon business performance This seemingly contradictory result arises from thefact that all three intermediate variables are interrelated

Our results should not be taken to imply that there are no other variablesmediate the effect of market orientation on economic performance We believethat other variables that have not been taken into account in this study such asproduct quality and customer satisfaction may also be signiregcant mediatorsHowever our results do suggest that whenever innovation degree andinnovation performance are included in the model as intermediate variablesthe effects of market orientation on business performance will mostly beconveyed through these variables Also in this study we have adopted thecurrently most widely accepted approach to measuring customer loyalty whichis based on behavioral loyalty measures (eg share of category requirementsrenew the policy probability the average customer last in the companyportfolio) It would be worth extending the present study by including not only

IJSIM143

302

behavioral measures of customer loyalty but also attitudinal measures alongthe lines of Dick and Basu (1994)

In our opinion two important contributions of the present research are outuse of objective measures of business performance and our focusing oninternational markets Despite the growing role of globalization and marketintegration and despite the increasing internationalization of corporationsmost studies on market orientation have focused on domestic markets (withnotable exceptions such as Selnes et al 1996 Webster 1994) Similarly moststudies on product innovation have also focused on domestic markets There isa lack of research yielding empirical support to the validity in an internationalsetting to research results obtained in domestic markets To regll this gap wetargeted the European Union market

Our study focused on a single industry the insurance sector Our sampleaccounted for 22 percent of the companies and 17 percent of the insurancepremiums in the targeted market An advantage of our single-industryapproach is that (with obvious reservations arising from the non-experimentalnature of our study and the fact that our sample should not be considered tohave been obtained at random) we can draw tentative predictions from ourmodel concerning the impact of market orientation on economic performance ininsurance companies operating in the European Union market An evidentdrawback of the single-industry approach adopted here is that it is not clearhow the present results extrapolate to other industries even when operating inthe same market

We have found that within the European Union insurance regrms thatconstantly monitor the evolution of current and potential customerrsquos needsmodify the attributes of the products to adapt them to the distributorsrequirements analyze competitorsrsquo marketing policy and products and knowbest of environmental trends especially technological and legal changes aremore likely to develop more new products have their new products accepted bythe market and obtain more loyal customers In turn increased customerloyalty and increased new product success will result in improved economicperformance However these conclusions must be taken with some caution aswe have used overall measures of innovation and innovation performanceFurther research is needed that takes into account the various aspects thatconstitute innovation

Previous studies have concluded that insurance companies still seethemselves as being product-focused and the industry as a whole is generallydistribution driven (Sodano 2000 Lambin 1996) This lack of closeness withcustomers has in turn contributed to the industry lack of product innovationsand product portfolios of commodity products that only compete on price Nowthe most successful regrms are redirecting their focus to the market needs andthey are beginning to exploit customer data and use market research togenerate ideas for designing new products In this context these companies

Businesseconomic

performance

303

now face choosing among too many new service options The managerialimplication of our study is that by enhancing their market orientation regrmswill know and service its customers better Thus they will generate moreinnovations by adopting a market-based product development process Alsoincreasing levels of market orientation enable regrms to discriminate more easilywhich new products have a higher success probability thus enhancing both theefregcacy and efregciency of new product development

Note

1 We considered employing the total volume of premiums for each insurance company as anadditional indicator although it seemed to us to be a better indicator of a companyrsquos sizerather than of its performance As we suspected the total volume of premiums wasuncorrelated with any of the variables considered in this study except for the companyrsquosmarket share (r = 024 p 001) Hence it is clear that volume of premiums should not beused as an indicator of insurance companiesrsquo performance

References

Armstrong JS and Overton T (1977) ordfEstimating nonresponse bias in mail surveysordm Journalof Marketing Research Vol 14 pp 396-402

Atuahene-Gima K (1996) ordfMarket orientation and innovationordm Journal of Business ResearchVol 35 pp 93-103

Baker TL Simpson PM and Sigauw JA (1999) ordfThe impact of suppliersrsquo perceptions ofreseller market orientation on key relationship constructsordm Journal of the Academy ofMarketing Science Vol 27 No 1 pp 50-7

Baron RM and Kenny DA (1986) ordfThe moderator-mediator variable distinction in socialpsychological research conceptual strategic and statistical considerationsordm Journal ofPersonality and Social Psychology Vol 51 pp 1173-82

Bentler PM (1990) ordfComparative regt indexes in structural modelsordm Psychological BulletinVol 107 pp 238-46

Bhuian SN (1997) ordfExploring market orientation in banks an empirical examination in SaudiArabiaordm The Journal of Service Marketing Vol 11 No 5 pp 317-28

Bollen KA (1987) ordfTotal direct and indirect effects in structural equation modelsordm inClogg CC (Ed) Sociological Methodology American Sociological AssociationWashington DC pp 37-69

Browne MW (1984) ordfAsymptotically distribution free methods for the analysis of covariancestructuresordm British Journal of Mathematical and Statistical Psychology Vol 37 pp 62-83

Calantone RJ di Benedetto AC and Bhoovaraghavan S (1994) ordfExamining the relationshipbetween degree of innovation and new product successordm Journal of Business ResearchVol 30 pp 143-8

Calantone RJ Schmidt JB and Song MX (1996) ordfControllable factors of new productssuccess a cross-national comparisonordm Marketing Science Vol 15 No 4 pp 341-58

Cooper RC (1994) ordfNew product the factors that drive successordm International MarketingReview Vol 11 No 1 pp 60-76

Day G (1992) ordfMarketingrsquos contribution to the strategy dialogueordm Journal of the Academy ofMarketing Science Vol 20 Fall pp 323-9

IJSIM143

304

overall business performance Finally according to the model almost 37percent of overall business economic performance is accounted for by thedegree of market orientation

An inspection of the total effects of market orientation on the indicators ofeconomic performance suggests that unit increments of market orientation asmeasured by the MOS-R are associated with 0095 0168 and 0153 incrementsin domestic market share premium growth and proregtability per year averagedover the last three years respectively

H2A mediated model for the relationship between market orientation andbusiness performance is depicted in Figure 3 In this context a mediating effectis said to exist when

both mediating paths b5 b6 are signiregcant and

the direct effect of the exogenous variable on the outcome variablevanishes (complete mediational effect) or is signiregcantly lower (partialmediational effect) when a mediator variable is introduced in the model

Condition (2) amounts to b1 in Figure 3 becoming zero or signiregcantly less thatthan value reported for Figure 2

We used the mediated model depicted in Figure 3 to test for mediatingeffects of innovation degree innovation performance and customer loyaltyseparately on the impact of market orientation on business economicperformance We found that when either innovation performance or innovation

Figure 3Mediated model

Businesseconomic

performance

297

degree were used as mediating variable all the mediating paths weresigniregcant and that direct path from market orientation to businessperformance was not signiregcantly different from zero b1 = 004 t = 162for innovation performance b1 = 004 t = 186 for innovation degree Hencetaken separately both innovation degree and innovation performancecompletely mediate the impact of market orientation on businessperformance After regxing b1 at zero we re-estimated these two mediationalmodels The resulting parameter estimates and GFIs for these two models areshown in Table III On the other hand customer loyalty was found not to havea mediational effect between market orientation and business performanceThe parameter estimates and GFIs for this model are also given in Table III

As can be seen in Table III the mediating paths are signiregcant but thedirect path b1 is signiregcantly different from zero at a = 001 Furthermore a 99percent conregdence interval for the value for b1 reported in Table II (003 009)includes the value of b1 estimated in the mediational model using customerloyalty 004 Hence this variable does not even partially mediate on the impactof market orientation on business economic performance The standardizeddirect impact of market orientation on business performance (0408) is morethan twice the standardized impact of market orientation conveyed throughcustomer loyalty (0191)

The percentage of variance of business economic performance explained bythe model when innovation performance innovation degree or customer loyaltyare used as mediators is very similar (465 percent 453 percent and 437percent respectively)

H3The full model to be regtted corresponding to the hypothesis depicted in Figure 1is presented in Figure 3 The parameter estimates and goodness of regt testcorresponding to this model are given in Table IV (see also Figure 4) As can beseen in this Table the model regts these data very well

All the postulated relationships were found to be signiregcant at an a = 001Lagrange multiplier tests indicated that the regt of the model would notsigniregcantly improve by

adding a direct effect of market orientation to business performance nor

adding a direct effect of innovation degree on economic performance

Result (1) is in accordance with the results discussed above where we saw thatinnovation degree and innovation performance even when taken separatelycompletely mediate the impact of market orientation on business performanceResult (2) conregrms our hypothesis that innovation performance completelymediates the impact of innovation degree on business performance

Given that all effects of market orientation on business performance gothrough either innovation degree-innovation performance or through customer

IJSIM143

298

Par

amet

eres

tim

ates

Goo

dnes

sof

regt

Par

amet

erV

alue

Par

amet

erV

alue

Index

Val

ue

Innova

tion

per

form

ance

as

med

iato

rb 1

0(reg

xed

)u

117

713

[10

0](2

241

)M

FF

X2

560

(p=

035

)b 2

155

[04

0](0

42)

u2

241

4[0

84]

(71

1)S-B

X2

380

(p=

058

)b 3

263

[04

8](0

67)

u3

431

6[0

77]

(45

5)B

X2

429

(p=

051

)b 4

248

[05

9](o

81)

u4

211

3[0

65]

(45

5)df

5R

MSE

A0

03b 5

023

[05

8](0

03)

u5

180

3[0

67]

(28

0)SR

MSR

005

GF

I0

98b 6

018

[06

8](0

05)

CF

I0

99

Innova

tion

deg

ree

as

med

iato

rb

10

(regxed

)u

117

713

[10

0](2

241

)M

FF

X2

717

(p=

021

)b 2

169

[05

0](0

36)

u2

234

0[0

82]

(73

8)S-B

X2

497

(p=

043

)b 3

282

[05

1](0

82)

u3

415

6[0

74]

(45

0)B

X2

613

(p=

029

)b 4

226

[05

4](0

79)

u4

232

9[0

71]

(46

0)df

5R

MSE

A0

06b 5

012

[05

5](0

02)

u5

590

[06

9](0

78)

SR

MSR

005

GF

I0

98b 6

031

[06

7](0

09)

CF

I0

98

Cust

om

erlo

yalty

as

med

iato

rb 1

004

[04

1](0

02)

u1

177

13[1

00]

(22

41)

MF

FX

22

30(p

=0

68)

b 21

61[0

40]

(05

1)u

224

05

[08

4](7

11)

S-B

X2

155

(p=

082

)b 3

252

[04

5](0

74)

u3

447

5[0

80]

(45

5)B

X2

309

(p=

054

)b 4

265

[06

2](0

92)

u4

201

3[0

62]

(45

5)R

MSE

A0

0df

4b 5

015

[05

7](0

02)

u5

863

[06

8](1

04)

SR

MSR

002

GF

I0

99b 6

013

[03

4](0

05)

CF

I1

00

Note

s

Rob

ust

asym

pto

tic

stan

dar

der

rors

are

pro

vid

edin

par

enth

eses

st

andar

diz

edpar

amet

eres

tim

ates

are

pro

vid

edin

squar

ebra

cket

sM

FF

X2=

Min

imum

regt

funct

ion

chi-sq

uar

eSB

X2=

Sat

orra

-Ben

tler

scal

edch

i-sq

uar

eB

X2=

Bro

wnersquo

sch

i-sq

uar

eco

rrec

ted

for

non

-nor

mal

ity

RM

SE

A=

root

mea

nsq

uar

eder

ror

ofap

pro

xim

atio

n

SR

MSR

=st

andar

diz

edro

otm

ean

squar

edre

sidual

G

FI

=goo

dnes

sof

regt

index

C

FI

=co

mpar

ativ

ereg

tin

dex

Table IIIEstimation results

for the modeldepicted in Figure 3

Businesseconomic

performance

299

Par

amet

eres

tim

ates

Goo

dnes

sof

regt

R2

Par

amet

erV

alue

Par

amet

erV

alue

Index

Val

ue

Var

iable

Val

ue

b 10

12[0

55]

(22

41)

u1

177

13[1

00]

(00

2)M

FF

X2

105

9(p

=0

56)

Mar

ket

shar

e0

16b 2

013

[03

4](0

04)

u2

590

[06

9](0

78)

S-B

X2

785

(p=

080

)P

rem

ium

gro

wth

019

b 30

15[0

57]

(00

2)u

314

48

[05

4](2

59)

BX

212

67

(p=

039

)P

roreg

tabilit

y0

38b 4

078

[04

4](0

15)

u4

863

[06

8](1

04)

df

12B

usi

nes

sper

form

ance

056

RM

SE

A0

0b 5

016

[05

4](0

05)

u5

239

1[0

84]

(72

3)SR

MSR

005

Innov

atio

nper

form

ance

046

b 60

16[0

37]

(00

7)u

645

09

[08

1](1

271

)G

FI

098

Innov

atio

ndeg

ree

031

b 71

42[0

40]

(04

5)u

720

12

[06

2](4

60)

CF

I1

00C

ust

omer

loyal

ty0

32b 8

217

[04

4](0

64)

b 92

32[0

62]

(08

5)

Note

sR

obust

asym

pto

tic

stan

dar

der

rors

are

pro

vid

edin

par

enth

eses

st

andar

diz

edpar

amet

eres

tim

ates

are

pro

vid

edin

squar

ebra

cket

sM

FF

X2=

Min

imum

regtfu

nct

ion

chi-sq

uar

eSB

X2=

Sat

orra

-Ben

tler

scal

edch

i-sq

uar

eB

X2=

Bro

wnersquo

sch

i-sq

uar

eco

rrec

ted

for

non

-nor

mal

ity

RM

SE

A=

root

mea

nsq

uar

eder

ror

ofap

pro

xim

atio

n

SR

MSR

=st

andar

diz

edro

otm

ean

squar

edre

sidual

G

FI

=goo

dnes

sof

regt

index

C

FI

=co

mpar

ativ

ereg

tin

dex

R

2=

squar

edm

ult

iple

corr

elat

ions

for

endog

enou

svar

iable

s

Table IVEstimation resultsfor the modeldepicted in Figure 4

IJSIM143

300

loyalty a question arises as to the relative importance of the speciregc effectsgoing through these variables The standardized speciregc effect (computed as inBollen 1987) going through innovation degree and innovation performance is031 and 021 going through customer loyalty Hence the impact of marketorientation going through innovation is 50 percent more than that goingthrough customer loyalty

We can also see in Table IV that over 30 percent of the variance of theintermediate variables (innovation degree innovation performance andcustomer loyalty) are explained by market orientation In fact almost 50percent (464 percent to be exact) of innovation performance is explained bymarket orientation Furthermore note that the percentage of variance ofbusiness performance explained by the model is 561 percent a 52 percentincrement over what is explained by market orientation alone (see Table II) andover a 20 percent increment over what is explained by the mediational modelsconsidered previously Hence the inclusion of all three intermediate variables inthe model improves considerably our prediction of business performance

Furthermore we observe in Table IV that the direct effect of marketorientation on all three intermediate variables appear to be equal Also thedirect effects of customer loyalty and of innovation performance on economicperformance appear to be equal We re-estimated the model to test theseconstraints obtaining b1 = b2 = b3 = 0133 b5 = b6 = 0157 Satorra-BentlerX 2(15) = 8849 p = 0885 CFI = 100 GFI = 0975 SRMR = 006

ConclusionsMarket orientation can be deregned as a strategy used to reach a sustainablecompetitive advantage based on the generation and use of information within

Figure 4Full model

Businesseconomic

performance

301

organizations and on the selection of markets to be satisreged In thisframework we believe that competitive advantage results from the use ofresources and capabilities to generate differential satisfaction in proregtablemarkets Sustainability is achieved because the performance of the marketorientationrsquos behaviors requires complex organizational knowledge that cannoteasily be imitated by competitors Thus we hypothesize that the satisfaction ofproregtable markets permits the regrm to achieve a psychologically differentialposition that leads to brand loyalty and thus to higher proregts Previous studieshave

found a clear impact of market orientation on economic performance

assessed the effects of market orientation on innovation and

investigated the relationship between market orientation and relationshipmarketing variables

The present study proposes and tests a model that integrates both streams ofresearch innovation and relationship marketing More research is needed toinvestigate the role of customer satisfaction within this framework

In our necessarily partial model innovation degree innovation performanceand customer loyalty are used as intermediate variables on the effect of marketorientation on business performance Our results suggest that the addition ofthese variables help improve our predictions of business economic performance52 percent over what is explained by market orientation alone Also we foundthat innovation degree and innovation performance each taken separatelycompletely mediate the effect of market orientation on economic performanceFurthermore the impact of innovation degree on economic performance iscompletely channeled through innovation performance Customer loyalty byitself does not meditate the impact of market orientation on economicperformance but when considered along with innovation degree andinnovation performance it conveys some of the effects of market orientationon business performance This seemingly contradictory result arises from thefact that all three intermediate variables are interrelated

Our results should not be taken to imply that there are no other variablesmediate the effect of market orientation on economic performance We believethat other variables that have not been taken into account in this study such asproduct quality and customer satisfaction may also be signiregcant mediatorsHowever our results do suggest that whenever innovation degree andinnovation performance are included in the model as intermediate variablesthe effects of market orientation on business performance will mostly beconveyed through these variables Also in this study we have adopted thecurrently most widely accepted approach to measuring customer loyalty whichis based on behavioral loyalty measures (eg share of category requirementsrenew the policy probability the average customer last in the companyportfolio) It would be worth extending the present study by including not only

IJSIM143

302

behavioral measures of customer loyalty but also attitudinal measures alongthe lines of Dick and Basu (1994)

In our opinion two important contributions of the present research are outuse of objective measures of business performance and our focusing oninternational markets Despite the growing role of globalization and marketintegration and despite the increasing internationalization of corporationsmost studies on market orientation have focused on domestic markets (withnotable exceptions such as Selnes et al 1996 Webster 1994) Similarly moststudies on product innovation have also focused on domestic markets There isa lack of research yielding empirical support to the validity in an internationalsetting to research results obtained in domestic markets To regll this gap wetargeted the European Union market

Our study focused on a single industry the insurance sector Our sampleaccounted for 22 percent of the companies and 17 percent of the insurancepremiums in the targeted market An advantage of our single-industryapproach is that (with obvious reservations arising from the non-experimentalnature of our study and the fact that our sample should not be considered tohave been obtained at random) we can draw tentative predictions from ourmodel concerning the impact of market orientation on economic performance ininsurance companies operating in the European Union market An evidentdrawback of the single-industry approach adopted here is that it is not clearhow the present results extrapolate to other industries even when operating inthe same market

We have found that within the European Union insurance regrms thatconstantly monitor the evolution of current and potential customerrsquos needsmodify the attributes of the products to adapt them to the distributorsrequirements analyze competitorsrsquo marketing policy and products and knowbest of environmental trends especially technological and legal changes aremore likely to develop more new products have their new products accepted bythe market and obtain more loyal customers In turn increased customerloyalty and increased new product success will result in improved economicperformance However these conclusions must be taken with some caution aswe have used overall measures of innovation and innovation performanceFurther research is needed that takes into account the various aspects thatconstitute innovation

Previous studies have concluded that insurance companies still seethemselves as being product-focused and the industry as a whole is generallydistribution driven (Sodano 2000 Lambin 1996) This lack of closeness withcustomers has in turn contributed to the industry lack of product innovationsand product portfolios of commodity products that only compete on price Nowthe most successful regrms are redirecting their focus to the market needs andthey are beginning to exploit customer data and use market research togenerate ideas for designing new products In this context these companies

Businesseconomic

performance

303

now face choosing among too many new service options The managerialimplication of our study is that by enhancing their market orientation regrmswill know and service its customers better Thus they will generate moreinnovations by adopting a market-based product development process Alsoincreasing levels of market orientation enable regrms to discriminate more easilywhich new products have a higher success probability thus enhancing both theefregcacy and efregciency of new product development

Note

1 We considered employing the total volume of premiums for each insurance company as anadditional indicator although it seemed to us to be a better indicator of a companyrsquos sizerather than of its performance As we suspected the total volume of premiums wasuncorrelated with any of the variables considered in this study except for the companyrsquosmarket share (r = 024 p 001) Hence it is clear that volume of premiums should not beused as an indicator of insurance companiesrsquo performance

References

Armstrong JS and Overton T (1977) ordfEstimating nonresponse bias in mail surveysordm Journalof Marketing Research Vol 14 pp 396-402

Atuahene-Gima K (1996) ordfMarket orientation and innovationordm Journal of Business ResearchVol 35 pp 93-103

Baker TL Simpson PM and Sigauw JA (1999) ordfThe impact of suppliersrsquo perceptions ofreseller market orientation on key relationship constructsordm Journal of the Academy ofMarketing Science Vol 27 No 1 pp 50-7

Baron RM and Kenny DA (1986) ordfThe moderator-mediator variable distinction in socialpsychological research conceptual strategic and statistical considerationsordm Journal ofPersonality and Social Psychology Vol 51 pp 1173-82

Bentler PM (1990) ordfComparative regt indexes in structural modelsordm Psychological BulletinVol 107 pp 238-46

Bhuian SN (1997) ordfExploring market orientation in banks an empirical examination in SaudiArabiaordm The Journal of Service Marketing Vol 11 No 5 pp 317-28

Bollen KA (1987) ordfTotal direct and indirect effects in structural equation modelsordm inClogg CC (Ed) Sociological Methodology American Sociological AssociationWashington DC pp 37-69

Browne MW (1984) ordfAsymptotically distribution free methods for the analysis of covariancestructuresordm British Journal of Mathematical and Statistical Psychology Vol 37 pp 62-83

Calantone RJ di Benedetto AC and Bhoovaraghavan S (1994) ordfExamining the relationshipbetween degree of innovation and new product successordm Journal of Business ResearchVol 30 pp 143-8

Calantone RJ Schmidt JB and Song MX (1996) ordfControllable factors of new productssuccess a cross-national comparisonordm Marketing Science Vol 15 No 4 pp 341-58

Cooper RC (1994) ordfNew product the factors that drive successordm International MarketingReview Vol 11 No 1 pp 60-76

Day G (1992) ordfMarketingrsquos contribution to the strategy dialogueordm Journal of the Academy ofMarketing Science Vol 20 Fall pp 323-9

IJSIM143

304

degree were used as mediating variable all the mediating paths weresigniregcant and that direct path from market orientation to businessperformance was not signiregcantly different from zero b1 = 004 t = 162for innovation performance b1 = 004 t = 186 for innovation degree Hencetaken separately both innovation degree and innovation performancecompletely mediate the impact of market orientation on businessperformance After regxing b1 at zero we re-estimated these two mediationalmodels The resulting parameter estimates and GFIs for these two models areshown in Table III On the other hand customer loyalty was found not to havea mediational effect between market orientation and business performanceThe parameter estimates and GFIs for this model are also given in Table III

As can be seen in Table III the mediating paths are signiregcant but thedirect path b1 is signiregcantly different from zero at a = 001 Furthermore a 99percent conregdence interval for the value for b1 reported in Table II (003 009)includes the value of b1 estimated in the mediational model using customerloyalty 004 Hence this variable does not even partially mediate on the impactof market orientation on business economic performance The standardizeddirect impact of market orientation on business performance (0408) is morethan twice the standardized impact of market orientation conveyed throughcustomer loyalty (0191)

The percentage of variance of business economic performance explained bythe model when innovation performance innovation degree or customer loyaltyare used as mediators is very similar (465 percent 453 percent and 437percent respectively)

H3The full model to be regtted corresponding to the hypothesis depicted in Figure 1is presented in Figure 3 The parameter estimates and goodness of regt testcorresponding to this model are given in Table IV (see also Figure 4) As can beseen in this Table the model regts these data very well

All the postulated relationships were found to be signiregcant at an a = 001Lagrange multiplier tests indicated that the regt of the model would notsigniregcantly improve by

adding a direct effect of market orientation to business performance nor

adding a direct effect of innovation degree on economic performance

Result (1) is in accordance with the results discussed above where we saw thatinnovation degree and innovation performance even when taken separatelycompletely mediate the impact of market orientation on business performanceResult (2) conregrms our hypothesis that innovation performance completelymediates the impact of innovation degree on business performance

Given that all effects of market orientation on business performance gothrough either innovation degree-innovation performance or through customer

IJSIM143

298

Par

amet

eres

tim

ates

Goo

dnes

sof

regt

Par

amet

erV

alue

Par

amet

erV

alue

Index

Val

ue

Innova

tion

per

form

ance

as

med

iato

rb 1

0(reg

xed

)u

117

713

[10

0](2

241

)M

FF

X2

560

(p=

035

)b 2

155

[04

0](0

42)

u2

241

4[0

84]

(71

1)S-B

X2

380

(p=

058

)b 3

263

[04

8](0

67)

u3

431

6[0

77]

(45

5)B

X2

429

(p=

051

)b 4

248

[05

9](o

81)

u4

211

3[0

65]

(45

5)df

5R

MSE

A0

03b 5

023

[05

8](0

03)

u5

180

3[0

67]

(28

0)SR

MSR

005

GF

I0

98b 6

018

[06

8](0

05)

CF

I0

99

Innova

tion

deg

ree

as

med

iato

rb

10

(regxed

)u

117

713

[10

0](2

241

)M

FF

X2

717

(p=

021

)b 2

169

[05

0](0

36)

u2

234

0[0

82]

(73

8)S-B

X2

497

(p=

043

)b 3

282

[05

1](0

82)

u3

415

6[0

74]

(45

0)B

X2

613

(p=

029

)b 4

226

[05

4](0

79)

u4

232

9[0

71]

(46

0)df

5R

MSE

A0

06b 5

012

[05

5](0

02)

u5

590

[06

9](0

78)

SR

MSR

005

GF

I0

98b 6

031

[06

7](0

09)

CF

I0

98

Cust

om

erlo

yalty

as

med

iato

rb 1

004

[04

1](0

02)

u1

177

13[1

00]

(22

41)

MF

FX

22

30(p

=0

68)

b 21

61[0

40]

(05

1)u

224

05

[08

4](7

11)

S-B

X2

155

(p=

082

)b 3

252

[04

5](0

74)

u3

447

5[0

80]

(45

5)B

X2

309

(p=

054

)b 4

265

[06

2](0

92)

u4

201

3[0

62]

(45

5)R

MSE

A0

0df

4b 5

015

[05

7](0

02)

u5

863

[06

8](1

04)

SR

MSR

002

GF

I0

99b 6

013

[03

4](0

05)

CF

I1

00

Note

s

Rob

ust

asym

pto

tic

stan

dar

der

rors

are

pro

vid

edin

par

enth

eses

st

andar

diz

edpar

amet

eres

tim

ates

are

pro

vid

edin

squar

ebra

cket

sM

FF

X2=

Min

imum

regt

funct

ion

chi-sq

uar

eSB

X2=

Sat

orra

-Ben

tler

scal

edch

i-sq

uar

eB

X2=

Bro

wnersquo

sch

i-sq

uar

eco

rrec

ted

for

non

-nor

mal

ity

RM

SE

A=

root

mea

nsq

uar

eder

ror

ofap

pro

xim

atio

n

SR

MSR

=st

andar

diz

edro

otm

ean

squar

edre

sidual

G

FI

=goo

dnes

sof

regt

index

C

FI

=co

mpar

ativ

ereg

tin

dex

Table IIIEstimation results

for the modeldepicted in Figure 3

Businesseconomic

performance

299

Par

amet

eres

tim

ates

Goo

dnes

sof

regt

R2

Par

amet

erV

alue

Par

amet

erV

alue

Index

Val

ue

Var

iable

Val

ue

b 10

12[0

55]

(22

41)

u1

177

13[1

00]

(00

2)M

FF

X2

105

9(p

=0

56)

Mar

ket

shar

e0

16b 2

013

[03

4](0

04)

u2

590

[06

9](0

78)

S-B

X2

785

(p=

080

)P

rem

ium

gro

wth

019

b 30

15[0

57]

(00

2)u

314

48

[05

4](2

59)

BX

212

67

(p=

039

)P

roreg

tabilit

y0

38b 4

078

[04

4](0

15)

u4

863

[06

8](1

04)

df

12B

usi

nes

sper

form

ance

056

RM

SE

A0

0b 5

016

[05

4](0

05)

u5

239

1[0

84]

(72

3)SR

MSR

005

Innov

atio

nper

form

ance

046

b 60

16[0

37]

(00

7)u

645

09

[08

1](1

271

)G

FI

098

Innov

atio

ndeg

ree

031

b 71

42[0

40]

(04

5)u

720

12

[06

2](4

60)

CF

I1

00C

ust

omer

loyal

ty0

32b 8

217

[04

4](0

64)

b 92

32[0

62]

(08

5)

Note

sR

obust

asym

pto

tic

stan

dar

der

rors

are

pro

vid

edin

par

enth

eses

st

andar

diz

edpar

amet

eres

tim

ates

are

pro

vid

edin

squar

ebra

cket

sM

FF

X2=

Min

imum

regtfu

nct

ion

chi-sq

uar

eSB

X2=

Sat

orra

-Ben

tler

scal

edch

i-sq

uar

eB

X2=

Bro

wnersquo

sch

i-sq

uar

eco

rrec

ted

for

non

-nor

mal

ity

RM

SE

A=

root

mea

nsq

uar

eder

ror

ofap

pro

xim

atio

n

SR

MSR

=st

andar

diz

edro

otm

ean

squar

edre

sidual

G

FI

=goo

dnes

sof

regt

index

C

FI

=co

mpar

ativ

ereg

tin

dex

R

2=

squar

edm

ult

iple

corr

elat

ions

for

endog

enou

svar

iable

s

Table IVEstimation resultsfor the modeldepicted in Figure 4

IJSIM143

300

loyalty a question arises as to the relative importance of the speciregc effectsgoing through these variables The standardized speciregc effect (computed as inBollen 1987) going through innovation degree and innovation performance is031 and 021 going through customer loyalty Hence the impact of marketorientation going through innovation is 50 percent more than that goingthrough customer loyalty

We can also see in Table IV that over 30 percent of the variance of theintermediate variables (innovation degree innovation performance andcustomer loyalty) are explained by market orientation In fact almost 50percent (464 percent to be exact) of innovation performance is explained bymarket orientation Furthermore note that the percentage of variance ofbusiness performance explained by the model is 561 percent a 52 percentincrement over what is explained by market orientation alone (see Table II) andover a 20 percent increment over what is explained by the mediational modelsconsidered previously Hence the inclusion of all three intermediate variables inthe model improves considerably our prediction of business performance

Furthermore we observe in Table IV that the direct effect of marketorientation on all three intermediate variables appear to be equal Also thedirect effects of customer loyalty and of innovation performance on economicperformance appear to be equal We re-estimated the model to test theseconstraints obtaining b1 = b2 = b3 = 0133 b5 = b6 = 0157 Satorra-BentlerX 2(15) = 8849 p = 0885 CFI = 100 GFI = 0975 SRMR = 006

ConclusionsMarket orientation can be deregned as a strategy used to reach a sustainablecompetitive advantage based on the generation and use of information within

Figure 4Full model

Businesseconomic

performance

301

organizations and on the selection of markets to be satisreged In thisframework we believe that competitive advantage results from the use ofresources and capabilities to generate differential satisfaction in proregtablemarkets Sustainability is achieved because the performance of the marketorientationrsquos behaviors requires complex organizational knowledge that cannoteasily be imitated by competitors Thus we hypothesize that the satisfaction ofproregtable markets permits the regrm to achieve a psychologically differentialposition that leads to brand loyalty and thus to higher proregts Previous studieshave

found a clear impact of market orientation on economic performance

assessed the effects of market orientation on innovation and

investigated the relationship between market orientation and relationshipmarketing variables

The present study proposes and tests a model that integrates both streams ofresearch innovation and relationship marketing More research is needed toinvestigate the role of customer satisfaction within this framework

In our necessarily partial model innovation degree innovation performanceand customer loyalty are used as intermediate variables on the effect of marketorientation on business performance Our results suggest that the addition ofthese variables help improve our predictions of business economic performance52 percent over what is explained by market orientation alone Also we foundthat innovation degree and innovation performance each taken separatelycompletely mediate the effect of market orientation on economic performanceFurthermore the impact of innovation degree on economic performance iscompletely channeled through innovation performance Customer loyalty byitself does not meditate the impact of market orientation on economicperformance but when considered along with innovation degree andinnovation performance it conveys some of the effects of market orientationon business performance This seemingly contradictory result arises from thefact that all three intermediate variables are interrelated

Our results should not be taken to imply that there are no other variablesmediate the effect of market orientation on economic performance We believethat other variables that have not been taken into account in this study such asproduct quality and customer satisfaction may also be signiregcant mediatorsHowever our results do suggest that whenever innovation degree andinnovation performance are included in the model as intermediate variablesthe effects of market orientation on business performance will mostly beconveyed through these variables Also in this study we have adopted thecurrently most widely accepted approach to measuring customer loyalty whichis based on behavioral loyalty measures (eg share of category requirementsrenew the policy probability the average customer last in the companyportfolio) It would be worth extending the present study by including not only

IJSIM143

302

behavioral measures of customer loyalty but also attitudinal measures alongthe lines of Dick and Basu (1994)

In our opinion two important contributions of the present research are outuse of objective measures of business performance and our focusing oninternational markets Despite the growing role of globalization and marketintegration and despite the increasing internationalization of corporationsmost studies on market orientation have focused on domestic markets (withnotable exceptions such as Selnes et al 1996 Webster 1994) Similarly moststudies on product innovation have also focused on domestic markets There isa lack of research yielding empirical support to the validity in an internationalsetting to research results obtained in domestic markets To regll this gap wetargeted the European Union market

Our study focused on a single industry the insurance sector Our sampleaccounted for 22 percent of the companies and 17 percent of the insurancepremiums in the targeted market An advantage of our single-industryapproach is that (with obvious reservations arising from the non-experimentalnature of our study and the fact that our sample should not be considered tohave been obtained at random) we can draw tentative predictions from ourmodel concerning the impact of market orientation on economic performance ininsurance companies operating in the European Union market An evidentdrawback of the single-industry approach adopted here is that it is not clearhow the present results extrapolate to other industries even when operating inthe same market

We have found that within the European Union insurance regrms thatconstantly monitor the evolution of current and potential customerrsquos needsmodify the attributes of the products to adapt them to the distributorsrequirements analyze competitorsrsquo marketing policy and products and knowbest of environmental trends especially technological and legal changes aremore likely to develop more new products have their new products accepted bythe market and obtain more loyal customers In turn increased customerloyalty and increased new product success will result in improved economicperformance However these conclusions must be taken with some caution aswe have used overall measures of innovation and innovation performanceFurther research is needed that takes into account the various aspects thatconstitute innovation

Previous studies have concluded that insurance companies still seethemselves as being product-focused and the industry as a whole is generallydistribution driven (Sodano 2000 Lambin 1996) This lack of closeness withcustomers has in turn contributed to the industry lack of product innovationsand product portfolios of commodity products that only compete on price Nowthe most successful regrms are redirecting their focus to the market needs andthey are beginning to exploit customer data and use market research togenerate ideas for designing new products In this context these companies

Businesseconomic

performance

303

now face choosing among too many new service options The managerialimplication of our study is that by enhancing their market orientation regrmswill know and service its customers better Thus they will generate moreinnovations by adopting a market-based product development process Alsoincreasing levels of market orientation enable regrms to discriminate more easilywhich new products have a higher success probability thus enhancing both theefregcacy and efregciency of new product development

Note

1 We considered employing the total volume of premiums for each insurance company as anadditional indicator although it seemed to us to be a better indicator of a companyrsquos sizerather than of its performance As we suspected the total volume of premiums wasuncorrelated with any of the variables considered in this study except for the companyrsquosmarket share (r = 024 p 001) Hence it is clear that volume of premiums should not beused as an indicator of insurance companiesrsquo performance

References

Armstrong JS and Overton T (1977) ordfEstimating nonresponse bias in mail surveysordm Journalof Marketing Research Vol 14 pp 396-402

Atuahene-Gima K (1996) ordfMarket orientation and innovationordm Journal of Business ResearchVol 35 pp 93-103

Baker TL Simpson PM and Sigauw JA (1999) ordfThe impact of suppliersrsquo perceptions ofreseller market orientation on key relationship constructsordm Journal of the Academy ofMarketing Science Vol 27 No 1 pp 50-7

Baron RM and Kenny DA (1986) ordfThe moderator-mediator variable distinction in socialpsychological research conceptual strategic and statistical considerationsordm Journal ofPersonality and Social Psychology Vol 51 pp 1173-82

Bentler PM (1990) ordfComparative regt indexes in structural modelsordm Psychological BulletinVol 107 pp 238-46

Bhuian SN (1997) ordfExploring market orientation in banks an empirical examination in SaudiArabiaordm The Journal of Service Marketing Vol 11 No 5 pp 317-28

Bollen KA (1987) ordfTotal direct and indirect effects in structural equation modelsordm inClogg CC (Ed) Sociological Methodology American Sociological AssociationWashington DC pp 37-69

Browne MW (1984) ordfAsymptotically distribution free methods for the analysis of covariancestructuresordm British Journal of Mathematical and Statistical Psychology Vol 37 pp 62-83

Calantone RJ di Benedetto AC and Bhoovaraghavan S (1994) ordfExamining the relationshipbetween degree of innovation and new product successordm Journal of Business ResearchVol 30 pp 143-8

Calantone RJ Schmidt JB and Song MX (1996) ordfControllable factors of new productssuccess a cross-national comparisonordm Marketing Science Vol 15 No 4 pp 341-58

Cooper RC (1994) ordfNew product the factors that drive successordm International MarketingReview Vol 11 No 1 pp 60-76

Day G (1992) ordfMarketingrsquos contribution to the strategy dialogueordm Journal of the Academy ofMarketing Science Vol 20 Fall pp 323-9

IJSIM143

304

Par

amet

eres

tim

ates

Goo

dnes

sof

regt

Par

amet

erV

alue

Par

amet

erV

alue

Index

Val

ue

Innova

tion

per

form

ance

as

med

iato

rb 1

0(reg

xed

)u

117

713

[10

0](2

241

)M

FF

X2

560

(p=

035

)b 2

155

[04

0](0

42)

u2

241

4[0

84]

(71

1)S-B

X2

380

(p=

058

)b 3

263

[04

8](0

67)

u3

431

6[0

77]

(45

5)B

X2

429

(p=

051

)b 4

248

[05

9](o

81)

u4

211

3[0

65]

(45

5)df

5R

MSE

A0

03b 5

023

[05

8](0

03)

u5

180

3[0

67]

(28

0)SR

MSR

005

GF

I0

98b 6

018

[06

8](0

05)

CF

I0

99

Innova

tion

deg

ree

as

med

iato

rb

10

(regxed

)u

117

713

[10

0](2

241

)M

FF

X2

717

(p=

021

)b 2

169

[05

0](0

36)

u2

234

0[0

82]

(73

8)S-B

X2

497

(p=

043

)b 3

282

[05

1](0

82)

u3

415

6[0

74]

(45

0)B

X2

613

(p=

029

)b 4

226

[05

4](0

79)

u4

232

9[0

71]

(46

0)df

5R

MSE

A0

06b 5

012

[05

5](0

02)

u5

590

[06

9](0

78)

SR

MSR

005

GF

I0

98b 6

031

[06

7](0

09)

CF

I0

98

Cust

om

erlo

yalty

as

med

iato

rb 1

004

[04

1](0

02)

u1

177

13[1

00]

(22

41)

MF

FX

22

30(p

=0

68)

b 21

61[0

40]

(05

1)u

224

05

[08

4](7

11)

S-B

X2

155

(p=

082

)b 3

252

[04

5](0

74)

u3

447

5[0

80]

(45

5)B

X2

309

(p=

054

)b 4

265

[06

2](0

92)

u4

201

3[0

62]

(45

5)R

MSE

A0

0df

4b 5

015

[05

7](0

02)

u5

863

[06

8](1

04)

SR

MSR

002

GF

I0

99b 6

013

[03

4](0

05)

CF

I1

00

Note

s

Rob

ust

asym

pto

tic

stan

dar

der

rors

are

pro

vid

edin

par

enth

eses

st

andar

diz

edpar

amet

eres

tim

ates

are

pro

vid

edin

squar

ebra

cket

sM

FF

X2=

Min

imum

regt

funct

ion

chi-sq

uar

eSB

X2=

Sat

orra

-Ben

tler

scal

edch

i-sq

uar

eB

X2=

Bro

wnersquo

sch

i-sq

uar

eco

rrec

ted

for

non

-nor

mal

ity

RM

SE

A=

root

mea

nsq

uar

eder

ror

ofap

pro

xim

atio

n

SR

MSR

=st

andar

diz

edro

otm

ean

squar

edre

sidual

G

FI

=goo

dnes

sof

regt

index

C

FI

=co

mpar

ativ

ereg

tin

dex

Table IIIEstimation results

for the modeldepicted in Figure 3

Businesseconomic

performance

299

Par

amet

eres

tim

ates

Goo

dnes

sof

regt

R2

Par

amet

erV

alue

Par

amet

erV

alue

Index

Val

ue

Var

iable

Val

ue

b 10

12[0

55]

(22

41)

u1

177

13[1

00]

(00

2)M

FF

X2

105

9(p

=0

56)

Mar

ket

shar

e0

16b 2

013

[03

4](0

04)

u2

590

[06

9](0

78)

S-B

X2

785

(p=

080

)P

rem

ium

gro

wth

019

b 30

15[0

57]

(00

2)u

314

48

[05

4](2

59)

BX

212

67

(p=

039

)P

roreg

tabilit

y0

38b 4

078

[04

4](0

15)

u4

863

[06

8](1

04)

df

12B

usi

nes

sper

form

ance

056

RM

SE

A0

0b 5

016

[05

4](0

05)

u5

239

1[0

84]

(72

3)SR

MSR

005

Innov

atio

nper

form

ance

046

b 60

16[0

37]

(00

7)u

645

09

[08

1](1

271

)G

FI

098

Innov

atio

ndeg

ree

031

b 71

42[0

40]

(04

5)u

720

12

[06

2](4

60)

CF

I1

00C

ust

omer

loyal

ty0

32b 8

217

[04

4](0

64)

b 92

32[0

62]

(08

5)

Note

sR

obust

asym

pto

tic

stan

dar

der

rors

are

pro

vid

edin

par

enth

eses

st

andar

diz

edpar

amet

eres

tim

ates

are

pro

vid

edin

squar

ebra

cket

sM

FF

X2=

Min

imum

regtfu

nct

ion

chi-sq

uar

eSB

X2=

Sat

orra

-Ben

tler

scal

edch

i-sq

uar

eB

X2=

Bro

wnersquo

sch

i-sq

uar

eco

rrec

ted

for

non

-nor

mal

ity

RM

SE

A=

root

mea

nsq

uar

eder

ror

ofap

pro

xim

atio

n

SR

MSR

=st

andar

diz

edro

otm

ean

squar

edre

sidual

G

FI

=goo

dnes

sof

regt

index

C

FI

=co

mpar

ativ

ereg

tin

dex

R

2=

squar

edm

ult

iple

corr

elat

ions

for

endog

enou

svar

iable

s

Table IVEstimation resultsfor the modeldepicted in Figure 4

IJSIM143

300

loyalty a question arises as to the relative importance of the speciregc effectsgoing through these variables The standardized speciregc effect (computed as inBollen 1987) going through innovation degree and innovation performance is031 and 021 going through customer loyalty Hence the impact of marketorientation going through innovation is 50 percent more than that goingthrough customer loyalty

We can also see in Table IV that over 30 percent of the variance of theintermediate variables (innovation degree innovation performance andcustomer loyalty) are explained by market orientation In fact almost 50percent (464 percent to be exact) of innovation performance is explained bymarket orientation Furthermore note that the percentage of variance ofbusiness performance explained by the model is 561 percent a 52 percentincrement over what is explained by market orientation alone (see Table II) andover a 20 percent increment over what is explained by the mediational modelsconsidered previously Hence the inclusion of all three intermediate variables inthe model improves considerably our prediction of business performance

Furthermore we observe in Table IV that the direct effect of marketorientation on all three intermediate variables appear to be equal Also thedirect effects of customer loyalty and of innovation performance on economicperformance appear to be equal We re-estimated the model to test theseconstraints obtaining b1 = b2 = b3 = 0133 b5 = b6 = 0157 Satorra-BentlerX 2(15) = 8849 p = 0885 CFI = 100 GFI = 0975 SRMR = 006

ConclusionsMarket orientation can be deregned as a strategy used to reach a sustainablecompetitive advantage based on the generation and use of information within

Figure 4Full model

Businesseconomic

performance

301

organizations and on the selection of markets to be satisreged In thisframework we believe that competitive advantage results from the use ofresources and capabilities to generate differential satisfaction in proregtablemarkets Sustainability is achieved because the performance of the marketorientationrsquos behaviors requires complex organizational knowledge that cannoteasily be imitated by competitors Thus we hypothesize that the satisfaction ofproregtable markets permits the regrm to achieve a psychologically differentialposition that leads to brand loyalty and thus to higher proregts Previous studieshave

found a clear impact of market orientation on economic performance

assessed the effects of market orientation on innovation and

investigated the relationship between market orientation and relationshipmarketing variables

The present study proposes and tests a model that integrates both streams ofresearch innovation and relationship marketing More research is needed toinvestigate the role of customer satisfaction within this framework

In our necessarily partial model innovation degree innovation performanceand customer loyalty are used as intermediate variables on the effect of marketorientation on business performance Our results suggest that the addition ofthese variables help improve our predictions of business economic performance52 percent over what is explained by market orientation alone Also we foundthat innovation degree and innovation performance each taken separatelycompletely mediate the effect of market orientation on economic performanceFurthermore the impact of innovation degree on economic performance iscompletely channeled through innovation performance Customer loyalty byitself does not meditate the impact of market orientation on economicperformance but when considered along with innovation degree andinnovation performance it conveys some of the effects of market orientationon business performance This seemingly contradictory result arises from thefact that all three intermediate variables are interrelated

Our results should not be taken to imply that there are no other variablesmediate the effect of market orientation on economic performance We believethat other variables that have not been taken into account in this study such asproduct quality and customer satisfaction may also be signiregcant mediatorsHowever our results do suggest that whenever innovation degree andinnovation performance are included in the model as intermediate variablesthe effects of market orientation on business performance will mostly beconveyed through these variables Also in this study we have adopted thecurrently most widely accepted approach to measuring customer loyalty whichis based on behavioral loyalty measures (eg share of category requirementsrenew the policy probability the average customer last in the companyportfolio) It would be worth extending the present study by including not only

IJSIM143

302

behavioral measures of customer loyalty but also attitudinal measures alongthe lines of Dick and Basu (1994)

In our opinion two important contributions of the present research are outuse of objective measures of business performance and our focusing oninternational markets Despite the growing role of globalization and marketintegration and despite the increasing internationalization of corporationsmost studies on market orientation have focused on domestic markets (withnotable exceptions such as Selnes et al 1996 Webster 1994) Similarly moststudies on product innovation have also focused on domestic markets There isa lack of research yielding empirical support to the validity in an internationalsetting to research results obtained in domestic markets To regll this gap wetargeted the European Union market

Our study focused on a single industry the insurance sector Our sampleaccounted for 22 percent of the companies and 17 percent of the insurancepremiums in the targeted market An advantage of our single-industryapproach is that (with obvious reservations arising from the non-experimentalnature of our study and the fact that our sample should not be considered tohave been obtained at random) we can draw tentative predictions from ourmodel concerning the impact of market orientation on economic performance ininsurance companies operating in the European Union market An evidentdrawback of the single-industry approach adopted here is that it is not clearhow the present results extrapolate to other industries even when operating inthe same market

We have found that within the European Union insurance regrms thatconstantly monitor the evolution of current and potential customerrsquos needsmodify the attributes of the products to adapt them to the distributorsrequirements analyze competitorsrsquo marketing policy and products and knowbest of environmental trends especially technological and legal changes aremore likely to develop more new products have their new products accepted bythe market and obtain more loyal customers In turn increased customerloyalty and increased new product success will result in improved economicperformance However these conclusions must be taken with some caution aswe have used overall measures of innovation and innovation performanceFurther research is needed that takes into account the various aspects thatconstitute innovation

Previous studies have concluded that insurance companies still seethemselves as being product-focused and the industry as a whole is generallydistribution driven (Sodano 2000 Lambin 1996) This lack of closeness withcustomers has in turn contributed to the industry lack of product innovationsand product portfolios of commodity products that only compete on price Nowthe most successful regrms are redirecting their focus to the market needs andthey are beginning to exploit customer data and use market research togenerate ideas for designing new products In this context these companies

Businesseconomic

performance

303

now face choosing among too many new service options The managerialimplication of our study is that by enhancing their market orientation regrmswill know and service its customers better Thus they will generate moreinnovations by adopting a market-based product development process Alsoincreasing levels of market orientation enable regrms to discriminate more easilywhich new products have a higher success probability thus enhancing both theefregcacy and efregciency of new product development

Note

1 We considered employing the total volume of premiums for each insurance company as anadditional indicator although it seemed to us to be a better indicator of a companyrsquos sizerather than of its performance As we suspected the total volume of premiums wasuncorrelated with any of the variables considered in this study except for the companyrsquosmarket share (r = 024 p 001) Hence it is clear that volume of premiums should not beused as an indicator of insurance companiesrsquo performance

References

Armstrong JS and Overton T (1977) ordfEstimating nonresponse bias in mail surveysordm Journalof Marketing Research Vol 14 pp 396-402

Atuahene-Gima K (1996) ordfMarket orientation and innovationordm Journal of Business ResearchVol 35 pp 93-103

Baker TL Simpson PM and Sigauw JA (1999) ordfThe impact of suppliersrsquo perceptions ofreseller market orientation on key relationship constructsordm Journal of the Academy ofMarketing Science Vol 27 No 1 pp 50-7

Baron RM and Kenny DA (1986) ordfThe moderator-mediator variable distinction in socialpsychological research conceptual strategic and statistical considerationsordm Journal ofPersonality and Social Psychology Vol 51 pp 1173-82

Bentler PM (1990) ordfComparative regt indexes in structural modelsordm Psychological BulletinVol 107 pp 238-46

Bhuian SN (1997) ordfExploring market orientation in banks an empirical examination in SaudiArabiaordm The Journal of Service Marketing Vol 11 No 5 pp 317-28

Bollen KA (1987) ordfTotal direct and indirect effects in structural equation modelsordm inClogg CC (Ed) Sociological Methodology American Sociological AssociationWashington DC pp 37-69

Browne MW (1984) ordfAsymptotically distribution free methods for the analysis of covariancestructuresordm British Journal of Mathematical and Statistical Psychology Vol 37 pp 62-83

Calantone RJ di Benedetto AC and Bhoovaraghavan S (1994) ordfExamining the relationshipbetween degree of innovation and new product successordm Journal of Business ResearchVol 30 pp 143-8

Calantone RJ Schmidt JB and Song MX (1996) ordfControllable factors of new productssuccess a cross-national comparisonordm Marketing Science Vol 15 No 4 pp 341-58

Cooper RC (1994) ordfNew product the factors that drive successordm International MarketingReview Vol 11 No 1 pp 60-76

Day G (1992) ordfMarketingrsquos contribution to the strategy dialogueordm Journal of the Academy ofMarketing Science Vol 20 Fall pp 323-9

IJSIM143

304

Par

amet

eres

tim

ates

Goo

dnes

sof

regt

R2

Par

amet

erV

alue

Par

amet

erV

alue

Index

Val

ue

Var

iable

Val

ue

b 10

12[0

55]

(22

41)

u1

177

13[1

00]

(00

2)M

FF

X2

105

9(p

=0

56)

Mar

ket

shar

e0

16b 2

013

[03

4](0

04)

u2

590

[06

9](0

78)

S-B

X2

785

(p=

080

)P

rem

ium

gro

wth

019

b 30

15[0

57]

(00

2)u

314

48

[05

4](2

59)

BX

212

67

(p=

039

)P

roreg

tabilit

y0

38b 4

078

[04

4](0

15)

u4

863

[06

8](1

04)

df

12B

usi

nes

sper

form

ance

056

RM

SE

A0

0b 5

016

[05

4](0

05)

u5

239

1[0

84]

(72

3)SR

MSR

005

Innov

atio

nper

form

ance

046

b 60

16[0

37]

(00

7)u

645

09

[08

1](1

271

)G

FI

098

Innov

atio

ndeg

ree

031

b 71

42[0

40]

(04

5)u

720

12

[06

2](4

60)

CF

I1

00C

ust

omer

loyal

ty0

32b 8

217

[04

4](0

64)

b 92

32[0

62]

(08

5)

Note

sR

obust

asym

pto

tic

stan

dar

der

rors

are

pro

vid

edin

par

enth

eses

st

andar

diz

edpar

amet

eres

tim

ates

are

pro

vid

edin

squar

ebra

cket

sM

FF

X2=

Min

imum

regtfu

nct

ion

chi-sq

uar

eSB

X2=

Sat

orra

-Ben

tler

scal

edch

i-sq

uar

eB

X2=

Bro

wnersquo

sch

i-sq

uar

eco

rrec

ted

for

non

-nor

mal

ity

RM

SE

A=

root

mea

nsq

uar

eder

ror

ofap

pro

xim

atio

n

SR

MSR

=st

andar

diz

edro

otm

ean

squar

edre

sidual

G

FI

=goo

dnes

sof

regt

index

C

FI

=co

mpar

ativ

ereg

tin

dex

R

2=

squar

edm

ult

iple

corr

elat

ions

for

endog

enou

svar

iable

s

Table IVEstimation resultsfor the modeldepicted in Figure 4

IJSIM143

300

loyalty a question arises as to the relative importance of the speciregc effectsgoing through these variables The standardized speciregc effect (computed as inBollen 1987) going through innovation degree and innovation performance is031 and 021 going through customer loyalty Hence the impact of marketorientation going through innovation is 50 percent more than that goingthrough customer loyalty

We can also see in Table IV that over 30 percent of the variance of theintermediate variables (innovation degree innovation performance andcustomer loyalty) are explained by market orientation In fact almost 50percent (464 percent to be exact) of innovation performance is explained bymarket orientation Furthermore note that the percentage of variance ofbusiness performance explained by the model is 561 percent a 52 percentincrement over what is explained by market orientation alone (see Table II) andover a 20 percent increment over what is explained by the mediational modelsconsidered previously Hence the inclusion of all three intermediate variables inthe model improves considerably our prediction of business performance

Furthermore we observe in Table IV that the direct effect of marketorientation on all three intermediate variables appear to be equal Also thedirect effects of customer loyalty and of innovation performance on economicperformance appear to be equal We re-estimated the model to test theseconstraints obtaining b1 = b2 = b3 = 0133 b5 = b6 = 0157 Satorra-BentlerX 2(15) = 8849 p = 0885 CFI = 100 GFI = 0975 SRMR = 006

ConclusionsMarket orientation can be deregned as a strategy used to reach a sustainablecompetitive advantage based on the generation and use of information within

Figure 4Full model

Businesseconomic

performance

301

organizations and on the selection of markets to be satisreged In thisframework we believe that competitive advantage results from the use ofresources and capabilities to generate differential satisfaction in proregtablemarkets Sustainability is achieved because the performance of the marketorientationrsquos behaviors requires complex organizational knowledge that cannoteasily be imitated by competitors Thus we hypothesize that the satisfaction ofproregtable markets permits the regrm to achieve a psychologically differentialposition that leads to brand loyalty and thus to higher proregts Previous studieshave

found a clear impact of market orientation on economic performance

assessed the effects of market orientation on innovation and

investigated the relationship between market orientation and relationshipmarketing variables

The present study proposes and tests a model that integrates both streams ofresearch innovation and relationship marketing More research is needed toinvestigate the role of customer satisfaction within this framework

In our necessarily partial model innovation degree innovation performanceand customer loyalty are used as intermediate variables on the effect of marketorientation on business performance Our results suggest that the addition ofthese variables help improve our predictions of business economic performance52 percent over what is explained by market orientation alone Also we foundthat innovation degree and innovation performance each taken separatelycompletely mediate the effect of market orientation on economic performanceFurthermore the impact of innovation degree on economic performance iscompletely channeled through innovation performance Customer loyalty byitself does not meditate the impact of market orientation on economicperformance but when considered along with innovation degree andinnovation performance it conveys some of the effects of market orientationon business performance This seemingly contradictory result arises from thefact that all three intermediate variables are interrelated

Our results should not be taken to imply that there are no other variablesmediate the effect of market orientation on economic performance We believethat other variables that have not been taken into account in this study such asproduct quality and customer satisfaction may also be signiregcant mediatorsHowever our results do suggest that whenever innovation degree andinnovation performance are included in the model as intermediate variablesthe effects of market orientation on business performance will mostly beconveyed through these variables Also in this study we have adopted thecurrently most widely accepted approach to measuring customer loyalty whichis based on behavioral loyalty measures (eg share of category requirementsrenew the policy probability the average customer last in the companyportfolio) It would be worth extending the present study by including not only

IJSIM143

302

behavioral measures of customer loyalty but also attitudinal measures alongthe lines of Dick and Basu (1994)

In our opinion two important contributions of the present research are outuse of objective measures of business performance and our focusing oninternational markets Despite the growing role of globalization and marketintegration and despite the increasing internationalization of corporationsmost studies on market orientation have focused on domestic markets (withnotable exceptions such as Selnes et al 1996 Webster 1994) Similarly moststudies on product innovation have also focused on domestic markets There isa lack of research yielding empirical support to the validity in an internationalsetting to research results obtained in domestic markets To regll this gap wetargeted the European Union market

Our study focused on a single industry the insurance sector Our sampleaccounted for 22 percent of the companies and 17 percent of the insurancepremiums in the targeted market An advantage of our single-industryapproach is that (with obvious reservations arising from the non-experimentalnature of our study and the fact that our sample should not be considered tohave been obtained at random) we can draw tentative predictions from ourmodel concerning the impact of market orientation on economic performance ininsurance companies operating in the European Union market An evidentdrawback of the single-industry approach adopted here is that it is not clearhow the present results extrapolate to other industries even when operating inthe same market

We have found that within the European Union insurance regrms thatconstantly monitor the evolution of current and potential customerrsquos needsmodify the attributes of the products to adapt them to the distributorsrequirements analyze competitorsrsquo marketing policy and products and knowbest of environmental trends especially technological and legal changes aremore likely to develop more new products have their new products accepted bythe market and obtain more loyal customers In turn increased customerloyalty and increased new product success will result in improved economicperformance However these conclusions must be taken with some caution aswe have used overall measures of innovation and innovation performanceFurther research is needed that takes into account the various aspects thatconstitute innovation

Previous studies have concluded that insurance companies still seethemselves as being product-focused and the industry as a whole is generallydistribution driven (Sodano 2000 Lambin 1996) This lack of closeness withcustomers has in turn contributed to the industry lack of product innovationsand product portfolios of commodity products that only compete on price Nowthe most successful regrms are redirecting their focus to the market needs andthey are beginning to exploit customer data and use market research togenerate ideas for designing new products In this context these companies

Businesseconomic

performance

303

now face choosing among too many new service options The managerialimplication of our study is that by enhancing their market orientation regrmswill know and service its customers better Thus they will generate moreinnovations by adopting a market-based product development process Alsoincreasing levels of market orientation enable regrms to discriminate more easilywhich new products have a higher success probability thus enhancing both theefregcacy and efregciency of new product development

Note

1 We considered employing the total volume of premiums for each insurance company as anadditional indicator although it seemed to us to be a better indicator of a companyrsquos sizerather than of its performance As we suspected the total volume of premiums wasuncorrelated with any of the variables considered in this study except for the companyrsquosmarket share (r = 024 p 001) Hence it is clear that volume of premiums should not beused as an indicator of insurance companiesrsquo performance

References

Armstrong JS and Overton T (1977) ordfEstimating nonresponse bias in mail surveysordm Journalof Marketing Research Vol 14 pp 396-402

Atuahene-Gima K (1996) ordfMarket orientation and innovationordm Journal of Business ResearchVol 35 pp 93-103

Baker TL Simpson PM and Sigauw JA (1999) ordfThe impact of suppliersrsquo perceptions ofreseller market orientation on key relationship constructsordm Journal of the Academy ofMarketing Science Vol 27 No 1 pp 50-7

Baron RM and Kenny DA (1986) ordfThe moderator-mediator variable distinction in socialpsychological research conceptual strategic and statistical considerationsordm Journal ofPersonality and Social Psychology Vol 51 pp 1173-82

Bentler PM (1990) ordfComparative regt indexes in structural modelsordm Psychological BulletinVol 107 pp 238-46

Bhuian SN (1997) ordfExploring market orientation in banks an empirical examination in SaudiArabiaordm The Journal of Service Marketing Vol 11 No 5 pp 317-28

Bollen KA (1987) ordfTotal direct and indirect effects in structural equation modelsordm inClogg CC (Ed) Sociological Methodology American Sociological AssociationWashington DC pp 37-69

Browne MW (1984) ordfAsymptotically distribution free methods for the analysis of covariancestructuresordm British Journal of Mathematical and Statistical Psychology Vol 37 pp 62-83

Calantone RJ di Benedetto AC and Bhoovaraghavan S (1994) ordfExamining the relationshipbetween degree of innovation and new product successordm Journal of Business ResearchVol 30 pp 143-8

Calantone RJ Schmidt JB and Song MX (1996) ordfControllable factors of new productssuccess a cross-national comparisonordm Marketing Science Vol 15 No 4 pp 341-58

Cooper RC (1994) ordfNew product the factors that drive successordm International MarketingReview Vol 11 No 1 pp 60-76

Day G (1992) ordfMarketingrsquos contribution to the strategy dialogueordm Journal of the Academy ofMarketing Science Vol 20 Fall pp 323-9

IJSIM143

304

loyalty a question arises as to the relative importance of the speciregc effectsgoing through these variables The standardized speciregc effect (computed as inBollen 1987) going through innovation degree and innovation performance is031 and 021 going through customer loyalty Hence the impact of marketorientation going through innovation is 50 percent more than that goingthrough customer loyalty

We can also see in Table IV that over 30 percent of the variance of theintermediate variables (innovation degree innovation performance andcustomer loyalty) are explained by market orientation In fact almost 50percent (464 percent to be exact) of innovation performance is explained bymarket orientation Furthermore note that the percentage of variance ofbusiness performance explained by the model is 561 percent a 52 percentincrement over what is explained by market orientation alone (see Table II) andover a 20 percent increment over what is explained by the mediational modelsconsidered previously Hence the inclusion of all three intermediate variables inthe model improves considerably our prediction of business performance

Furthermore we observe in Table IV that the direct effect of marketorientation on all three intermediate variables appear to be equal Also thedirect effects of customer loyalty and of innovation performance on economicperformance appear to be equal We re-estimated the model to test theseconstraints obtaining b1 = b2 = b3 = 0133 b5 = b6 = 0157 Satorra-BentlerX 2(15) = 8849 p = 0885 CFI = 100 GFI = 0975 SRMR = 006

ConclusionsMarket orientation can be deregned as a strategy used to reach a sustainablecompetitive advantage based on the generation and use of information within

Figure 4Full model

Businesseconomic

performance

301

organizations and on the selection of markets to be satisreged In thisframework we believe that competitive advantage results from the use ofresources and capabilities to generate differential satisfaction in proregtablemarkets Sustainability is achieved because the performance of the marketorientationrsquos behaviors requires complex organizational knowledge that cannoteasily be imitated by competitors Thus we hypothesize that the satisfaction ofproregtable markets permits the regrm to achieve a psychologically differentialposition that leads to brand loyalty and thus to higher proregts Previous studieshave

found a clear impact of market orientation on economic performance

assessed the effects of market orientation on innovation and

investigated the relationship between market orientation and relationshipmarketing variables

The present study proposes and tests a model that integrates both streams ofresearch innovation and relationship marketing More research is needed toinvestigate the role of customer satisfaction within this framework

In our necessarily partial model innovation degree innovation performanceand customer loyalty are used as intermediate variables on the effect of marketorientation on business performance Our results suggest that the addition ofthese variables help improve our predictions of business economic performance52 percent over what is explained by market orientation alone Also we foundthat innovation degree and innovation performance each taken separatelycompletely mediate the effect of market orientation on economic performanceFurthermore the impact of innovation degree on economic performance iscompletely channeled through innovation performance Customer loyalty byitself does not meditate the impact of market orientation on economicperformance but when considered along with innovation degree andinnovation performance it conveys some of the effects of market orientationon business performance This seemingly contradictory result arises from thefact that all three intermediate variables are interrelated

Our results should not be taken to imply that there are no other variablesmediate the effect of market orientation on economic performance We believethat other variables that have not been taken into account in this study such asproduct quality and customer satisfaction may also be signiregcant mediatorsHowever our results do suggest that whenever innovation degree andinnovation performance are included in the model as intermediate variablesthe effects of market orientation on business performance will mostly beconveyed through these variables Also in this study we have adopted thecurrently most widely accepted approach to measuring customer loyalty whichis based on behavioral loyalty measures (eg share of category requirementsrenew the policy probability the average customer last in the companyportfolio) It would be worth extending the present study by including not only

IJSIM143

302

behavioral measures of customer loyalty but also attitudinal measures alongthe lines of Dick and Basu (1994)

In our opinion two important contributions of the present research are outuse of objective measures of business performance and our focusing oninternational markets Despite the growing role of globalization and marketintegration and despite the increasing internationalization of corporationsmost studies on market orientation have focused on domestic markets (withnotable exceptions such as Selnes et al 1996 Webster 1994) Similarly moststudies on product innovation have also focused on domestic markets There isa lack of research yielding empirical support to the validity in an internationalsetting to research results obtained in domestic markets To regll this gap wetargeted the European Union market

Our study focused on a single industry the insurance sector Our sampleaccounted for 22 percent of the companies and 17 percent of the insurancepremiums in the targeted market An advantage of our single-industryapproach is that (with obvious reservations arising from the non-experimentalnature of our study and the fact that our sample should not be considered tohave been obtained at random) we can draw tentative predictions from ourmodel concerning the impact of market orientation on economic performance ininsurance companies operating in the European Union market An evidentdrawback of the single-industry approach adopted here is that it is not clearhow the present results extrapolate to other industries even when operating inthe same market

We have found that within the European Union insurance regrms thatconstantly monitor the evolution of current and potential customerrsquos needsmodify the attributes of the products to adapt them to the distributorsrequirements analyze competitorsrsquo marketing policy and products and knowbest of environmental trends especially technological and legal changes aremore likely to develop more new products have their new products accepted bythe market and obtain more loyal customers In turn increased customerloyalty and increased new product success will result in improved economicperformance However these conclusions must be taken with some caution aswe have used overall measures of innovation and innovation performanceFurther research is needed that takes into account the various aspects thatconstitute innovation

Previous studies have concluded that insurance companies still seethemselves as being product-focused and the industry as a whole is generallydistribution driven (Sodano 2000 Lambin 1996) This lack of closeness withcustomers has in turn contributed to the industry lack of product innovationsand product portfolios of commodity products that only compete on price Nowthe most successful regrms are redirecting their focus to the market needs andthey are beginning to exploit customer data and use market research togenerate ideas for designing new products In this context these companies

Businesseconomic

performance

303

now face choosing among too many new service options The managerialimplication of our study is that by enhancing their market orientation regrmswill know and service its customers better Thus they will generate moreinnovations by adopting a market-based product development process Alsoincreasing levels of market orientation enable regrms to discriminate more easilywhich new products have a higher success probability thus enhancing both theefregcacy and efregciency of new product development

Note

1 We considered employing the total volume of premiums for each insurance company as anadditional indicator although it seemed to us to be a better indicator of a companyrsquos sizerather than of its performance As we suspected the total volume of premiums wasuncorrelated with any of the variables considered in this study except for the companyrsquosmarket share (r = 024 p 001) Hence it is clear that volume of premiums should not beused as an indicator of insurance companiesrsquo performance

References

Armstrong JS and Overton T (1977) ordfEstimating nonresponse bias in mail surveysordm Journalof Marketing Research Vol 14 pp 396-402

Atuahene-Gima K (1996) ordfMarket orientation and innovationordm Journal of Business ResearchVol 35 pp 93-103

Baker TL Simpson PM and Sigauw JA (1999) ordfThe impact of suppliersrsquo perceptions ofreseller market orientation on key relationship constructsordm Journal of the Academy ofMarketing Science Vol 27 No 1 pp 50-7

Baron RM and Kenny DA (1986) ordfThe moderator-mediator variable distinction in socialpsychological research conceptual strategic and statistical considerationsordm Journal ofPersonality and Social Psychology Vol 51 pp 1173-82

Bentler PM (1990) ordfComparative regt indexes in structural modelsordm Psychological BulletinVol 107 pp 238-46

Bhuian SN (1997) ordfExploring market orientation in banks an empirical examination in SaudiArabiaordm The Journal of Service Marketing Vol 11 No 5 pp 317-28

Bollen KA (1987) ordfTotal direct and indirect effects in structural equation modelsordm inClogg CC (Ed) Sociological Methodology American Sociological AssociationWashington DC pp 37-69

Browne MW (1984) ordfAsymptotically distribution free methods for the analysis of covariancestructuresordm British Journal of Mathematical and Statistical Psychology Vol 37 pp 62-83

Calantone RJ di Benedetto AC and Bhoovaraghavan S (1994) ordfExamining the relationshipbetween degree of innovation and new product successordm Journal of Business ResearchVol 30 pp 143-8

Calantone RJ Schmidt JB and Song MX (1996) ordfControllable factors of new productssuccess a cross-national comparisonordm Marketing Science Vol 15 No 4 pp 341-58

Cooper RC (1994) ordfNew product the factors that drive successordm International MarketingReview Vol 11 No 1 pp 60-76

Day G (1992) ordfMarketingrsquos contribution to the strategy dialogueordm Journal of the Academy ofMarketing Science Vol 20 Fall pp 323-9

IJSIM143

304

organizations and on the selection of markets to be satisreged In thisframework we believe that competitive advantage results from the use ofresources and capabilities to generate differential satisfaction in proregtablemarkets Sustainability is achieved because the performance of the marketorientationrsquos behaviors requires complex organizational knowledge that cannoteasily be imitated by competitors Thus we hypothesize that the satisfaction ofproregtable markets permits the regrm to achieve a psychologically differentialposition that leads to brand loyalty and thus to higher proregts Previous studieshave

found a clear impact of market orientation on economic performance

assessed the effects of market orientation on innovation and

investigated the relationship between market orientation and relationshipmarketing variables

The present study proposes and tests a model that integrates both streams ofresearch innovation and relationship marketing More research is needed toinvestigate the role of customer satisfaction within this framework

In our necessarily partial model innovation degree innovation performanceand customer loyalty are used as intermediate variables on the effect of marketorientation on business performance Our results suggest that the addition ofthese variables help improve our predictions of business economic performance52 percent over what is explained by market orientation alone Also we foundthat innovation degree and innovation performance each taken separatelycompletely mediate the effect of market orientation on economic performanceFurthermore the impact of innovation degree on economic performance iscompletely channeled through innovation performance Customer loyalty byitself does not meditate the impact of market orientation on economicperformance but when considered along with innovation degree andinnovation performance it conveys some of the effects of market orientationon business performance This seemingly contradictory result arises from thefact that all three intermediate variables are interrelated

Our results should not be taken to imply that there are no other variablesmediate the effect of market orientation on economic performance We believethat other variables that have not been taken into account in this study such asproduct quality and customer satisfaction may also be signiregcant mediatorsHowever our results do suggest that whenever innovation degree andinnovation performance are included in the model as intermediate variablesthe effects of market orientation on business performance will mostly beconveyed through these variables Also in this study we have adopted thecurrently most widely accepted approach to measuring customer loyalty whichis based on behavioral loyalty measures (eg share of category requirementsrenew the policy probability the average customer last in the companyportfolio) It would be worth extending the present study by including not only

IJSIM143

302

behavioral measures of customer loyalty but also attitudinal measures alongthe lines of Dick and Basu (1994)

In our opinion two important contributions of the present research are outuse of objective measures of business performance and our focusing oninternational markets Despite the growing role of globalization and marketintegration and despite the increasing internationalization of corporationsmost studies on market orientation have focused on domestic markets (withnotable exceptions such as Selnes et al 1996 Webster 1994) Similarly moststudies on product innovation have also focused on domestic markets There isa lack of research yielding empirical support to the validity in an internationalsetting to research results obtained in domestic markets To regll this gap wetargeted the European Union market

Our study focused on a single industry the insurance sector Our sampleaccounted for 22 percent of the companies and 17 percent of the insurancepremiums in the targeted market An advantage of our single-industryapproach is that (with obvious reservations arising from the non-experimentalnature of our study and the fact that our sample should not be considered tohave been obtained at random) we can draw tentative predictions from ourmodel concerning the impact of market orientation on economic performance ininsurance companies operating in the European Union market An evidentdrawback of the single-industry approach adopted here is that it is not clearhow the present results extrapolate to other industries even when operating inthe same market

We have found that within the European Union insurance regrms thatconstantly monitor the evolution of current and potential customerrsquos needsmodify the attributes of the products to adapt them to the distributorsrequirements analyze competitorsrsquo marketing policy and products and knowbest of environmental trends especially technological and legal changes aremore likely to develop more new products have their new products accepted bythe market and obtain more loyal customers In turn increased customerloyalty and increased new product success will result in improved economicperformance However these conclusions must be taken with some caution aswe have used overall measures of innovation and innovation performanceFurther research is needed that takes into account the various aspects thatconstitute innovation

Previous studies have concluded that insurance companies still seethemselves as being product-focused and the industry as a whole is generallydistribution driven (Sodano 2000 Lambin 1996) This lack of closeness withcustomers has in turn contributed to the industry lack of product innovationsand product portfolios of commodity products that only compete on price Nowthe most successful regrms are redirecting their focus to the market needs andthey are beginning to exploit customer data and use market research togenerate ideas for designing new products In this context these companies

Businesseconomic

performance

303

now face choosing among too many new service options The managerialimplication of our study is that by enhancing their market orientation regrmswill know and service its customers better Thus they will generate moreinnovations by adopting a market-based product development process Alsoincreasing levels of market orientation enable regrms to discriminate more easilywhich new products have a higher success probability thus enhancing both theefregcacy and efregciency of new product development

Note

1 We considered employing the total volume of premiums for each insurance company as anadditional indicator although it seemed to us to be a better indicator of a companyrsquos sizerather than of its performance As we suspected the total volume of premiums wasuncorrelated with any of the variables considered in this study except for the companyrsquosmarket share (r = 024 p 001) Hence it is clear that volume of premiums should not beused as an indicator of insurance companiesrsquo performance

References

Armstrong JS and Overton T (1977) ordfEstimating nonresponse bias in mail surveysordm Journalof Marketing Research Vol 14 pp 396-402

Atuahene-Gima K (1996) ordfMarket orientation and innovationordm Journal of Business ResearchVol 35 pp 93-103

Baker TL Simpson PM and Sigauw JA (1999) ordfThe impact of suppliersrsquo perceptions ofreseller market orientation on key relationship constructsordm Journal of the Academy ofMarketing Science Vol 27 No 1 pp 50-7

Baron RM and Kenny DA (1986) ordfThe moderator-mediator variable distinction in socialpsychological research conceptual strategic and statistical considerationsordm Journal ofPersonality and Social Psychology Vol 51 pp 1173-82

Bentler PM (1990) ordfComparative regt indexes in structural modelsordm Psychological BulletinVol 107 pp 238-46

Bhuian SN (1997) ordfExploring market orientation in banks an empirical examination in SaudiArabiaordm The Journal of Service Marketing Vol 11 No 5 pp 317-28

Bollen KA (1987) ordfTotal direct and indirect effects in structural equation modelsordm inClogg CC (Ed) Sociological Methodology American Sociological AssociationWashington DC pp 37-69

Browne MW (1984) ordfAsymptotically distribution free methods for the analysis of covariancestructuresordm British Journal of Mathematical and Statistical Psychology Vol 37 pp 62-83

Calantone RJ di Benedetto AC and Bhoovaraghavan S (1994) ordfExamining the relationshipbetween degree of innovation and new product successordm Journal of Business ResearchVol 30 pp 143-8

Calantone RJ Schmidt JB and Song MX (1996) ordfControllable factors of new productssuccess a cross-national comparisonordm Marketing Science Vol 15 No 4 pp 341-58

Cooper RC (1994) ordfNew product the factors that drive successordm International MarketingReview Vol 11 No 1 pp 60-76

Day G (1992) ordfMarketingrsquos contribution to the strategy dialogueordm Journal of the Academy ofMarketing Science Vol 20 Fall pp 323-9

IJSIM143

304

behavioral measures of customer loyalty but also attitudinal measures alongthe lines of Dick and Basu (1994)

In our opinion two important contributions of the present research are outuse of objective measures of business performance and our focusing oninternational markets Despite the growing role of globalization and marketintegration and despite the increasing internationalization of corporationsmost studies on market orientation have focused on domestic markets (withnotable exceptions such as Selnes et al 1996 Webster 1994) Similarly moststudies on product innovation have also focused on domestic markets There isa lack of research yielding empirical support to the validity in an internationalsetting to research results obtained in domestic markets To regll this gap wetargeted the European Union market

Our study focused on a single industry the insurance sector Our sampleaccounted for 22 percent of the companies and 17 percent of the insurancepremiums in the targeted market An advantage of our single-industryapproach is that (with obvious reservations arising from the non-experimentalnature of our study and the fact that our sample should not be considered tohave been obtained at random) we can draw tentative predictions from ourmodel concerning the impact of market orientation on economic performance ininsurance companies operating in the European Union market An evidentdrawback of the single-industry approach adopted here is that it is not clearhow the present results extrapolate to other industries even when operating inthe same market

We have found that within the European Union insurance regrms thatconstantly monitor the evolution of current and potential customerrsquos needsmodify the attributes of the products to adapt them to the distributorsrequirements analyze competitorsrsquo marketing policy and products and knowbest of environmental trends especially technological and legal changes aremore likely to develop more new products have their new products accepted bythe market and obtain more loyal customers In turn increased customerloyalty and increased new product success will result in improved economicperformance However these conclusions must be taken with some caution aswe have used overall measures of innovation and innovation performanceFurther research is needed that takes into account the various aspects thatconstitute innovation

Previous studies have concluded that insurance companies still seethemselves as being product-focused and the industry as a whole is generallydistribution driven (Sodano 2000 Lambin 1996) This lack of closeness withcustomers has in turn contributed to the industry lack of product innovationsand product portfolios of commodity products that only compete on price Nowthe most successful regrms are redirecting their focus to the market needs andthey are beginning to exploit customer data and use market research togenerate ideas for designing new products In this context these companies

Businesseconomic

performance

303

now face choosing among too many new service options The managerialimplication of our study is that by enhancing their market orientation regrmswill know and service its customers better Thus they will generate moreinnovations by adopting a market-based product development process Alsoincreasing levels of market orientation enable regrms to discriminate more easilywhich new products have a higher success probability thus enhancing both theefregcacy and efregciency of new product development

Note

1 We considered employing the total volume of premiums for each insurance company as anadditional indicator although it seemed to us to be a better indicator of a companyrsquos sizerather than of its performance As we suspected the total volume of premiums wasuncorrelated with any of the variables considered in this study except for the companyrsquosmarket share (r = 024 p 001) Hence it is clear that volume of premiums should not beused as an indicator of insurance companiesrsquo performance

References

Armstrong JS and Overton T (1977) ordfEstimating nonresponse bias in mail surveysordm Journalof Marketing Research Vol 14 pp 396-402

Atuahene-Gima K (1996) ordfMarket orientation and innovationordm Journal of Business ResearchVol 35 pp 93-103

Baker TL Simpson PM and Sigauw JA (1999) ordfThe impact of suppliersrsquo perceptions ofreseller market orientation on key relationship constructsordm Journal of the Academy ofMarketing Science Vol 27 No 1 pp 50-7

Baron RM and Kenny DA (1986) ordfThe moderator-mediator variable distinction in socialpsychological research conceptual strategic and statistical considerationsordm Journal ofPersonality and Social Psychology Vol 51 pp 1173-82

Bentler PM (1990) ordfComparative regt indexes in structural modelsordm Psychological BulletinVol 107 pp 238-46

Bhuian SN (1997) ordfExploring market orientation in banks an empirical examination in SaudiArabiaordm The Journal of Service Marketing Vol 11 No 5 pp 317-28

Bollen KA (1987) ordfTotal direct and indirect effects in structural equation modelsordm inClogg CC (Ed) Sociological Methodology American Sociological AssociationWashington DC pp 37-69

Browne MW (1984) ordfAsymptotically distribution free methods for the analysis of covariancestructuresordm British Journal of Mathematical and Statistical Psychology Vol 37 pp 62-83

Calantone RJ di Benedetto AC and Bhoovaraghavan S (1994) ordfExamining the relationshipbetween degree of innovation and new product successordm Journal of Business ResearchVol 30 pp 143-8

Calantone RJ Schmidt JB and Song MX (1996) ordfControllable factors of new productssuccess a cross-national comparisonordm Marketing Science Vol 15 No 4 pp 341-58

Cooper RC (1994) ordfNew product the factors that drive successordm International MarketingReview Vol 11 No 1 pp 60-76

Day G (1992) ordfMarketingrsquos contribution to the strategy dialogueordm Journal of the Academy ofMarketing Science Vol 20 Fall pp 323-9

IJSIM143

304

now face choosing among too many new service options The managerialimplication of our study is that by enhancing their market orientation regrmswill know and service its customers better Thus they will generate moreinnovations by adopting a market-based product development process Alsoincreasing levels of market orientation enable regrms to discriminate more easilywhich new products have a higher success probability thus enhancing both theefregcacy and efregciency of new product development

Note

1 We considered employing the total volume of premiums for each insurance company as anadditional indicator although it seemed to us to be a better indicator of a companyrsquos sizerather than of its performance As we suspected the total volume of premiums wasuncorrelated with any of the variables considered in this study except for the companyrsquosmarket share (r = 024 p 001) Hence it is clear that volume of premiums should not beused as an indicator of insurance companiesrsquo performance

References

Armstrong JS and Overton T (1977) ordfEstimating nonresponse bias in mail surveysordm Journalof Marketing Research Vol 14 pp 396-402

Atuahene-Gima K (1996) ordfMarket orientation and innovationordm Journal of Business ResearchVol 35 pp 93-103

Baker TL Simpson PM and Sigauw JA (1999) ordfThe impact of suppliersrsquo perceptions ofreseller market orientation on key relationship constructsordm Journal of the Academy ofMarketing Science Vol 27 No 1 pp 50-7

Baron RM and Kenny DA (1986) ordfThe moderator-mediator variable distinction in socialpsychological research conceptual strategic and statistical considerationsordm Journal ofPersonality and Social Psychology Vol 51 pp 1173-82

Bentler PM (1990) ordfComparative regt indexes in structural modelsordm Psychological BulletinVol 107 pp 238-46

Bhuian SN (1997) ordfExploring market orientation in banks an empirical examination in SaudiArabiaordm The Journal of Service Marketing Vol 11 No 5 pp 317-28

Bollen KA (1987) ordfTotal direct and indirect effects in structural equation modelsordm inClogg CC (Ed) Sociological Methodology American Sociological AssociationWashington DC pp 37-69

Browne MW (1984) ordfAsymptotically distribution free methods for the analysis of covariancestructuresordm British Journal of Mathematical and Statistical Psychology Vol 37 pp 62-83

Calantone RJ di Benedetto AC and Bhoovaraghavan S (1994) ordfExamining the relationshipbetween degree of innovation and new product successordm Journal of Business ResearchVol 30 pp 143-8

Calantone RJ Schmidt JB and Song MX (1996) ordfControllable factors of new productssuccess a cross-national comparisonordm Marketing Science Vol 15 No 4 pp 341-58

Cooper RC (1994) ordfNew product the factors that drive successordm International MarketingReview Vol 11 No 1 pp 60-76

Day G (1992) ordfMarketingrsquos contribution to the strategy dialogueordm Journal of the Academy ofMarketing Science Vol 20 Fall pp 323-9

IJSIM143

304

Deng S and Dart J (1994) ordfMeasuring market orientation a multi-factor multi-itemsapproachordm Journal of Marketing Management Vol 10 pp 725-42

DeshpandeAcirc R and Farley JU (1977) ordfMeasuring market orientation generalization andsynthesisordm Journal of Market Focused Management Vol 2 pp 213-32

DeshpandeAcirc R Farley JU and Webster FE Jr (1993) ordfCorporate cultures customer orientationand innovativeness in Japanese regrms a quadrad analysisordm Journal of Marketing Vol 57January pp 23-7

Dick AS and Basu K (1994) ordfCustomer loyalty toward an integrated conceptual frameworkordmJournal of the Academy of Marketing Science Vol 22 No 2 pp 99-113

Fritz W (1996) ordfMarket orientation and corporate success regndings from Germanyordm EuropeanJournal of Marketing Vol 30 No 8 pp 59-74

Gatignon H and Xuered J-M (1997) ordfStrategic orientation of the regrm and new productperformanceordm Journal of Marketing Research Vol 34 pp 77-90

Greenley GE (1995) ordfMarket orientation and company performance empirical evidence fromUKordm British Journal of Management Vol 6 pp 1-13

Greenley GE and Foxall GR (1997) ordfMultiple stakeholder orientation in UK companies andthe implications for company performanceordm Journal of Management Studies Vol 34 No 2pp 259-84

Greenley GE and Foxall GR (1998) ordfExternal moderation of association among stakeholderorientation and company performanceordm International Journal of Research in MarketingVol 15 pp 51-69

Han JK Namwoon K and Srivastava R (1998) ordfMarket orientation and organizationalperformance is innovation a missing linkordm Journal of Marketing Vol 62 No 4 pp 30-45

Harrison-Walker LJ (2001) ordfThe measurement of a market orientation and its impact onbusiness performanceordm Journal of Quality Management Vol 6 No 2 pp 139-72

Hu L and Bentler PM (1999) ordfCutoff criteria for regt indices in covariance structure analysisconventional criteria versus new alternativesordm Structural Equation Modeling Vol 6pp 1-55

Jacoby J and Chestnut RW (1978) Webster FE (Ed) Brand Loyalty Measurement andManagement Wiley New York NY

Javalgi RG and Moberg CR (1997) ordfService loyalty implications for service providersordm TheJournal of Service Marketing Vol 11 No 3 pp 165-79

Jaworski BJ and Kohli AK (1993) ordfMarket orientation antecedents and consequencesordmJournal of Marketing Vol 57 pp 53-70

Jones E Busch P and Dacin P (2002) ordfFirm market orientation and salesperson customerorientation interpersonal and intrapersonal inmacruences on customer service and retentionin business-to-business buyerplusmnseller relationshipsordm Journal of Business Research

Joreskog KG Sorbom D du Toit S and du Toit M (1999) Lisrel 8 New Statistical FeaturesSSI Chicago IL

Kamakura WA Mittal V de Rosa F and Mazzon JA (2002) ordfAssessing the service proregtchainordm Marketing Science Vol 21 No 3 pp 294-317

Kohli AK and Jaworski BJ (1990) ordfMarket orientation the construct research propositionsand managerial implicationsordm Journal of Marketing Vol 54 pp 1-18

Kotler P (1984) Marketing Management Analysis Planning and Control 5th ed Prentice-HallEnglewood Cliffs NJ

Businesseconomic

performance

305

Lado NR Maydeu-Olivares A and Martinez MA (1998a) ordfEl Nivel de la OrientacioAcircn alMercado en las Empresas Aseguradoras en EspanAumla y en el resto de Europaordm RevistaEspanAuml ola de Investigaciones en Marketing- Esic Vol 2 pp 99-113

Lado NR Maydeu-Olivares A and Rivera J (1998 b) ordfMeasuring market orientation in severalpopulations a structural equations approachordm European Journal of Marketing Vol 32No 12 pp 23-39

Lai K-H (2003) ordfMarket orientation in quality-oriented organizations and its impact on theirperformanceordm International Journal of Production Economics in press

Lambin J-J (1996) ordfThe misunderstanding about marketing today marketing is too importantto be left to sole marketing function An empirical study in the private insurance sectorordmCEMS Business Review Vol 1 No 1-2 pp 37-56

Loveman G (1998) ordfEmployee satisfaction customer loyalty and regnancial performance anempirical examination of the service proregt chain in retail bankingordm Journal of ServiceResearch Vol 1 pp 18-31

Lukas B and Ferrell OC (2000) ordfThe effect of market orientation on product innovationordmJournal of the Academy of Marketing Science Vol 28 No 2 pp 239-47

McCullough J Heng L and Khem GS (1986) ordfMeasuring the marketing orientation of retailoperations of international banksordm International Journal of Bank Marketing Vol 4 No 3pp 9-18

McDonald RP and Marsh HW (1990) ordfChoosing a multivariate model noncentrality andgoodness of regtordm Psychological Bulletin Vol 107 pp 247-55

Miller D and Friesen PH (1982) ordfInnovation in conservative and entrepreneurial regrms twomodels of strategic momentumordm Strategic Management Journal Vol 3 pp 1-25

Montoya-Weiss MM and Calantone R (1994) ordfDeterminants of new product performance areview and meta-analysisordm Journal of Product Innovation Management Vol 11pp 397-417

Narver JC and Slater SF (1990) ordfThe effect of a market orientation on business proregtabilityordmJournal of Marketing Vol 54 October pp 20-35

Nayyar PR (1990) ordfInformation asymmetries a source of competitive advantage for diversiregedservice regrmsordm Strategic Management Journal Vol 11 NovemberDecember pp 513-9

Odin Y Odin N and Valette-Florence P (2001) ordfConceptual and operational aspects of brandloyalty an empirical investigationordm Journal of Business Research Vol 53 No 2 pp 75-84

Ottum BD and Moore WL (1997) ordfThe role of market information in new productsuccessfailureordm Journal of Product Innovation Management Vol 14 pp 258-73

Pelham AM and Wilson DT (1996) ordfA longitudinal study of the impact of market structureregrm structure strategy and market orientation culture on dimensions of small-regrmperformanceordm Journal of the Academy of Marketing Science Vol 24 No 1 pp 27-43

Pitt L Caruana A and Berthon PR (1996) ordfMarket orientation and business performancesome European evidenceordm International Marketing Review Vol 13 No 1 pp 5-18

Robinson WT Fornell C and Sullivan M (1992) ordfAre market pioneers intrinsically better thanlater entrantsordm Strategic Management Journal Vol 13 No 8 pp 609-24

Ruekert RW (1992) ordfDeveloping a market orientation an organizational strategy perspectiveordmInternational Journal of Marketing Vol 9 pp 225-45

Satorra A and Bentler PM (1988) ordfScaling corrections for chi-square statistics in covariancestructure analysisordm ASA Vol 1988 pp 308-13

IJSIM143

306

Selnes F Jaworski BJ and Kohli AK (1996) ordfMarket orientation in United States andScandinavian companies a cross-cultural studyordm Scandinavian Journal of ManagementVol 12 No 2 pp 139-57

Slater SF and Narver JC (1994) ordfDoes competitive environment moderate the marketorientation-performance relationshipordm Journal of Marketing Vol 58 January pp 46-55

Sodano A (2000) ordfLeveraging CRM to build better productsordm National Underwriter Vol 104No 26 pp 23-5

Steiger JH (1990) ordfStructural model evaluation and modiregcation an interval estimationapproachordm Multivariate Behavioral Research Vol 25 pp 173-80

Steinman C Deshpande R and Farley JU (2000) ordfBeyond market orientation when customersand suppliers disagreeordm Journal of the Academy of Marketing Science Vol 28 No 1pp 109-19

Subramanian A (1997) ordfInnovativeness rederegning the conceptordm Journal of Engineering andTechnology Management Vol 13 pp 223-43

Tanaka J and Huba GJ (1985) ordfA regt index of covariance structure models under arbitrary GLSestimationordm British Journal of Mathematical and Statistical Psychology Vol 42 pp 233-9

Tufano P (1992) ordfFinancial innovation and regrst mover advantagesordm Journal of AppliedCorporate Finance Vol 1 pp 83-7

Van Bruggen G and Smidts A (1995) ordfThe measurement of market orientation a promisingtool for managementordm in Proceedings CEMS Academic Conference Vienna AustriaApril

Webb D Webster C and Krepapa A (2000) ordfAn exploration of the meaning and outcomes of acustomer-deregned market orientationordm Journal of Business Research Vol 48 pp 101-12

Webster FE Jr (1994) Market-driven Management John Wiley amp Sons New York NY

Further reading

Atuahene-Gima K (1995) ordfAn exploratory analysis of the impact of market orientation on newproduct performanceordm Journal of Product Innovation Management Vol 12 pp 275-93

Caruana A Pitt L and Berthon P (1999) ordfExcellence-market orientation link someconsequences for service regrmsordm Journal of Business Research Vol 44 pp 5-15

Diamantopoulos A and Hart S (1993) ordfLinking market orientation and company performancepreliminary evidence on Kohli and Jaworskirsquos frameworkordm Journal of Strategic MarketingVol 1 pp 93-121

Hurley R and Hult T (1998) ordfInnovation market orientation and organizational learning anintegration and empirical examinationordm Journal of Marketing Vol 62 No 3 pp 42-54

Kohli AK Jaworski BJ and Kumar A (1993) ordfMARKOR a measure of market orientationordmJournal of Marketing Research Vol 30 pp 467-77

Song MX and Parry ME (1996) ordfWhat separates Japanese new products winners from losersordmJournal of Product Innovation Management Vol 13 pp 422-39

Appendix Measurement instrument(a) Item content of the market orientation scale-revised (MOS-R)(0 = complete disagreement to 10 = complete agreement)

Analysis of the regnal customer

(1) We permanently measure our customersrsquo degree of satisfaction

(2) We constantly monitor the evolution of our current and potential customersrsquo requirements

Businesseconomic

performance

307

(3) We know the factors inmacruencing our customersrsquo purchasing habits very well

(4) We collect information necessary for detecting the appearance of new market segments(ie groups of customers with new requirements)

(5) We always have full updated information on the evolution of the image of our productsheld by our current and potential customers

Analysis of the distributor

(1) We permanently measure the degree of our distributorrsquos satisfaction

(2) We monitor the evolution of our distributorsrsquo requirements

(3) We collect information on how our products integrate into our distributorsrsquo activities

(4) We have accurate knowledge of the problems that marketing our products may cause toour distributors

(5) We always have full current information for monitoring the image of our products asheld by distributors

Analysis of the competitors

(1) We know our most dangerous competitorsrsquo aims and strategies

(2) We know our most dangerous competitorsrsquo strengths and weaknesses very well

(3) We have a system for precisely monitoring the evolution of the components of ourcompetitorsrsquo marketing policy (productsservices price communication and distribution)

Analysis of the environment

(1) We have systems enabling us to closely monitor changes in the legal social economicand technological environments

(2) We identify the sensitive and risk factors that may impact on our business

Interfunctional coordination

(1) Major market information is always spread over all the companyrsquos functional areas

(2) Marketing strategies are always drawn up in agreementwith the other business functions

(3) We have implemented actions so that each person in the company feels individuallycommitted to customer satisfaction

(4) We periodically organize interfunction meetings to analyze all important market information

(5) We encourage informal exchanges of information between the companyrsquos differentfunctions

Strategic actions on regnal customers

(1) We are quicker than the competition in responding to changes in customersrsquo requirements

(2) Our marketing plan with its necessary adaptations is very well implemented overall

(3) We give our customers complete information so they may use our products to the fulland are satisreged with them

Strategic actions on distributors

(1) We treat our distributors as though they were our actual customers

(2) We modify the attributes of our products to adapt them to our distributorsrsquo requirements

(3) We undertake actions to persuade our distributors of the beneregts they obtain fromworking with our company

IJSIM143

308

Strategic actions on competitors

(1) We quickly respond to the actions of the most dangerous competitors for our company

(2) We undertake actions to anticipate the competition

Strategic actions on the macro-environment

(1) We develop strategies to support the defense of our sectorrsquos interests through communicationand pressure groups (such as professional associations employersrsquo associations etc)

(2) We actively participate in actions whose aim is to demonstrate the social usefulness ofour sector to public opinion

(b) Item content of the innovation degree scale

(1) The rate relative to competitors of new productservice introduction by the regrm(1 = less than to 7 = greater)

(2) How many new lines of products or services has your regrm marketed in the past threeyears (1 = no new lines of product or services to 7 = very many new lines of productsor services)

(3) Change in product or lines have been (1 = been mostly of a minor nature to7 = usually been quite dramatic)

(c) Item content of the innovation performance scale(1 = not successful to 7 = very successful)

(1) To what extent has the new productservice been a success in meeting its sales growthobjectives since its launch

(2) To what extent has the new productservice been a success in meeting its market shareobjectives since its launch

(3) To what extent has the new productservice been a success in meeting its sales objectivessince its launch

(4) To what extent has the new productservice been a success in meeting its proregtobjectives since launch

(d) Item content of the customer loyalty scale

(1) Of all the types of insurance your customers have they take on with your company(1 = a minimum part to 7 = all)

(2) In comparison with your competitors the average a customer lasts in portfolio in yourcompany is (1 = much less to 7 = much more)

(3) In comparison with your competitors if a customer has taken on insurance with yourcompany the probability of hisher renewing the policy is (1 = much less to 7 = muchmore)

(4) In comparison with your competitors your customersrsquo loyalty level is (1 = much less to7 = much more)

(e) Item content of the business performance scale

(1) Domestic market share (percent)

(2) Premium growth (percent per year)

(3) Proregtability per year (percent premiums)

Businesseconomic

performance

309

Strategic actions on competitors

(1) We quickly respond to the actions of the most dangerous competitors for our company

(2) We undertake actions to anticipate the competition

Strategic actions on the macro-environment

(1) We develop strategies to support the defense of our sectorrsquos interests through communicationand pressure groups (such as professional associations employersrsquo associations etc)

(2) We actively participate in actions whose aim is to demonstrate the social usefulness ofour sector to public opinion

(b) Item content of the innovation degree scale

(1) The rate relative to competitors of new productservice introduction by the regrm(1 = less than to 7 = greater)

(2) How many new lines of products or services has your regrm marketed in the past threeyears (1 = no new lines of product or services to 7 = very many new lines of productsor services)

(3) Change in product or lines have been (1 = been mostly of a minor nature to7 = usually been quite dramatic)

(c) Item content of the innovation performance scale(1 = not successful to 7 = very successful)

(1) To what extent has the new productservice been a success in meeting its sales growthobjectives since its launch

(2) To what extent has the new productservice been a success in meeting its market shareobjectives since its launch

(3) To what extent has the new productservice been a success in meeting its sales objectivessince its launch

(4) To what extent has the new productservice been a success in meeting its proregtobjectives since launch

(d) Item content of the customer loyalty scale

(1) Of all the types of insurance your customers have they take on with your company(1 = a minimum part to 7 = all)

(2) In comparison with your competitors the average a customer lasts in portfolio in yourcompany is (1 = much less to 7 = much more)

(3) In comparison with your competitors if a customer has taken on insurance with yourcompany the probability of hisher renewing the policy is (1 = much less to 7 = muchmore)

(4) In comparison with your competitors your customersrsquo loyalty level is (1 = much less to7 = much more)

(e) Item content of the business performance scale

(1) Domestic market share (percent)

(2) Premium growth (percent per year)

(3) Proregtability per year (percent premiums)

Businesseconomic

performance

309