bestemmeling: titel, voornaam, achternaam - faculteit sociale

KATHOLIEKE UNIVERSITEIT

LEUVEN

Identity vs Performance: An Overview of Theories Explaining Trust in Government Second Report “Quality and Trust in Government”

Prof. dr. Geert Bouckaert Steven Van de Walle

Prof. dr. Bart Maddens Dr. Jarl K. Kampen

January 2002

Public Management Institute TEL: 0032 16 32 32 70 Faculty of Social Sciences FAX: 0032 16 32 32 67 Department of Political Sciences E-mail: [email protected] E. VAN EVENSTRAAT 2 A - B-3000 LEUVEN - BELGIUM www.kuleuven.ac.be/io/trust

Theories on trust in government

Bouckaert, Van de Walle, Maddens & Kampen

Quality and Trust in Government

IDENTITY VS. PERFORMANCE: AN OVERVIEW OF THEORIES

EXPLAINING TRUST IN GOVERNMENT

The research project “citizen-directed governance: quality and trust in government” has been

commissioned by the Programme for Policy Research of the Ministry of the Flemish

Community (PBO 99B/1/14). The modernisation of the public sector machinery takes quite a

central position among the general strategies to restore trust in government. Improvement

initiatives, is the idea, make the quality of public service provision more manifest and

increase it. This, in turn, leads to satisfaction among citizens, which results in increased trust

in government. This reasoning contains many steps that are neither theoretically nor

empirically ultimately supported, and are even contested. The aim of the research project is to

look into this relation between quality of public services, user satisfaction with these services

and trust in government. This report tries to give a brief overview of the theoretical

approaches to political trust, and wants to frame the so-called performance hypothesis within

the wider range of research on trust in government.

We start this report by defining the concept of trust and show where to locate it in the

framework of government-citizen interaction. We will show that ‘trust’ is in fact a very

ambiguous concept, and that its moral value and meaning is changing. In order to be able to

speak about citizens’ attitudes towards government, we first have to know more about what

‘government’ exactly is for citizens. In other words: what is the object of their trust?

Therefore we will deal with the images citizens have of government: is government seen as a

homogeneous entity or instead as a collection of institutions? Do citizens perceive a political

system or instead individual politicians? Do some parts of government figure as symbols for

the entire government?

We continue with explanations for the -perceived if not real- decline of trust in government.

We introduce these theories by making the distinction between performance- and identity-

based evaluations of government. First we will deal with the so-called performance theory:

1


2 K.U.Leuven – Public Management Institute

people have a negative attitude towards government because government does not perform

well. This evaluation depends on what citizens see (both perception and a normative

approach) as a government task, the correctness of their observations, and the importance they

give to the performance and service delivery aspect of government. The performance

hypothesis has two main parts: micro-performance (administration, functioning of politicians,

service delivery...) and macro-performance (inflation, unemployment, GNP growth, ...).

Distrust in government is however not necessarily a result of government-internal factors.

Therefore we will turn to general societal tendencies such as individualisation, modernity and

post-modernity, evolutions in interpersonal trust and social capital, etc. Furthermore we will

deal with structural factors such as information disequilibria, cultural factors such as the

nature of the democratic system (majority, proportional), and with the difference between

trust and contracts for organising society. Finally we will also look into the classical home-

team hypothesis, stating that a person distrusts government because he or she has voted for

the present opposition parties.

In the closing chapter we will discuss a number of consequences of the lack of trust in

government. First we will come back to the question whether trust is necessary, and whether

the attitude will lead to behaviour as well. Second we will discuss a number of specific

actions: exit and voice.

This is the second research report of the research project in ‘Quality and Trust in

Government’. The first report was:

1. Klantentevredenheidsmetingen bij de overheid (2001)

More information and publications can be found on www.kuleuven.ac.be/io/trust.



TABLE OF CONTENTS

Samenvatting 5

Introduction 8

CHAPTER 1 What is trust? 10

1.1 Definitions 10

1.2 What is 'trust' in 'government' 13

1.3 The changing nature of trust 15

1.4 Is trust in the institutions necessary? The constitutional vs. the political culture

approach 18

1.5 Trust opinions vs. the dominant discourse 21

1.6 Trust in government - numbers and trends 24

CHAPTER 2 The object of trust 27

2.1 Software and hardware 27

2.2 Citizen perceptions of ‘government’ 32

2.3 Generalisation vs. differentiation 34

2.4 What parts of government are more important in the formation of a (dis)trusting

attitude? 37

CHAPTER 3 Performance vs. Identity 40

3.1 Citizen roles: citizens become clients 41

3.2 Citizen roles and choice of evaluation criteria 43

CHAPTER 4 The Causes of distrust I: the performance theory 47

4.1 The tasks of government 48

4.2 The correctness of perceptions of performance 49

4.3 Process- vs. output expectations 50

4.4 Micro-performance 52

3



4.5 Macro-performance 62

4.6 Expectations and Government Overload 66

CHAPTER 5 The causes of distrust II 68

5.1 Risk, uncertainty, conflict and consensus 68

5.2 Personal characteristics and socio-demographic variables 75

5.3 Social characteristics 78

5.4 Identity 83

CHAPTER 6 The consequences of distrust 93

6.1 Causality 93

6.2 Does confidence matter? 94

6.3 Opinion and behaviour 95

6.4 Distrust, ethnocentrism and the extreme-right 99

Tables and figures 100



Samenvatting

Het concept ‘vertrouwen’ is sterk gerelateerd aan het concept ‘risico’, en is noodzakelijk voor

de omgang met onzekerheden. Wanneer er gesproken wordt over ‘vertrouwen in de overheid’,

kunnen we bezwaarlijk over ‘vertrouwen’ spreken, aangezien er in de relatie burger-bestuur

niet uitzonderlijk veel substantiële onzekerheden bestaan. Bovendien heeft het concept

vertrouwen een tijd- en cultuur- specifieke betekenis, waardoor de inhoud ervan constant

verandert. Vertrouwen en wantrouwen worden daarom gebruikt als brede termen die in feite

niet meer doen dan een algemene positieve of negatieve attitude tegenover de overheid

aanduiden.

Hoewel tegenwoordig heel veel aandacht wordt besteed aan het herstellen van het vertrouwen

in de overheid, dienen we toch eerst rekening te houden met een vrij fundamentele

vraagstelling: is vertrouwen in de overheid nodig voor het functioneren en overleven van een

politiek systeem? Afhankelijk van de gehanteerde opvatting over de aard van politiek en over

de plaats van politiek in de maatschappij bestaan hier verschillende opvattingen over. Aan de

ene kant staan zij die vertrouwen in de overheid en interpersoonlijk vertrouwen als

noodzakelijk beschouwen, en de twee dan ook met elkaar verbinden. Aan de andere kant

staan zij die wantrouwen in de overheid zien als een uiting van een gezonde democratische

attitude, aangezien de overheid en het democratisch systeem vooral een zaak is van

procedures. Het geobserveerde verschil tussen deze twee opvattingen opent tevens de

discussie over de vraag of de overheid zelf aan de basis ligt van het wantrouwen in de

overheid. Het feit dat het voor burgers gezien de sociale druk vrij moeilijk is om een

afwijkende mening in te nemen van wat gangbaar is (i.e. vertrouwen in de overheid) verklaart

waarom het zo moeilijk is strategieën te ontwikkelen om het vertrouwen in de overheid te

herstellen.

Om meer inzicht te krijgen in de oorzaken van en mogelijke oplossingen voor het

wantrouwen in de overheid dienen we meer te weten te komen over het object van het

politieke wantrouwen: is het wantrouwen gericht tegen het politieke systeem als geheel, of

alleen tegen de de politici die op dit moment aan de macht zijn? Benaderen burgers de

overheid als een amorf, homogeen geheel, of maken ze onderscheiden tussen de verschillende

instellingen en diensten? Wanneer deze laatste differentiatie voorkomt, hebben alle

5



instellingen dan een gelijkaardige impact op het beeld van de overheid als geheel, of zijn het

een beperkt aantal specifieke instellingen (parlement, openbare diensten, staatshoofd, gerecht)

die het beeld dat burgers van hun overheid hebben bepalen?

Burgers kunnen in hun beoordeling van de overheid verschillende criteria gebruiken. Deze

criteria kunnen over het algemeen worden opgedeeld in prestatie- en identiteitsgerelateerde

criteria. Daarnaast kan ook nog een normatief criterium gebruikt worden. De belangrijkste

vraag in toekomstig onderzoek dient te zijn in welke combinatie deze criteria worden gebruikt

en welke omstandigheden leiden tot het gebruik ervan. Zo wordt verondersteld dat de evolutie

naar een benadering van de burger als klant de weg heeft geopend naar een toegenomen

gebruik van prestatie-gerelateerde evaluatiecriteria.

Het project ‘burgergericht besturen’ wil nagaan of beter presterende openbare diensten zullen

leiden tot meer vertrouwen in de overheid. Opdat deze relatie geldig zou zijn, dient aan een

aantal voorwaarden te worden voldaan. Ten eerste, zoals hoger reeds vermeld, dient het

prestatiecriterium een belangrijke plaats in te nemen in de beoordeling van de overheid. Ten

tweede dient er overeenkomst te zijn over de vraag welke taken de overheid wel of niet dient

uit te voeren. Ten derde moeten percepties van het overheidsfunctioneren overeen te stemmen

met de realiteit.

De zogenaamde micro-prestatiehypothese bestaat uit twee delen: politieke en administratieve

prestaties. Er wordt verondersteld dat de prestaties van politici en administraties het politiek

vertrouwen beïnvloeden. Volgens de macro-prestatiehypothese baseren burgers hun

beoordelingen van de overheid op macro-economische indicatoren zoals werkloosheid,

economische groei, inflatie enz., uiteraard op voorwaarde dat zij de overheid als

verantwoordelijk hiervoor beschouwen. De ‘government overload’ hypothese is sterk

gerelateerd aan deze prestatiehypothesen. Deze stelt dat het voor de overheid onmogelijk is

om naar wens te presteren omdat de vragen die ten overstaan van de overheid worden gesteld

te omvangrijk zijn geworden.

Er zijn veel identiteit-gerelateerde verklaringen voor wantrouwen. De eerste die we

vermelden is van een macro-sociologische aard: de maatschappij wordt steeds complexer, en

postmateriële waardenoriëntaties nemen toe. Dit leidt tot onzekerheid en risico waarmee

burgers niet zonder meer kunnen omgaan. Zelfs wanneer contracten en procedures (naast

vertrouwen) een zekere hoeveelheid zekerheid trachten te creëren, blijft er toch een zekere

gevoel van risico bestaan.



Het blijkt moeilijk te zijn om een profiel op te maken van de vertrouwende en de wantrouwen

burger aangezien de bestaande onderzoeksresultaten heel wat contradicties aan het licht

brengen. Over de relatie tussen vertrouwen in de overheid en interpersoonlijk vertrouwen en

sociaal kapitaal bestaat reeds meer duidelijkheid, zelfs al is het nog niet duidelijk hoe deze

relatie nu precies verklaard kan worden. Een bijkomende complicatie in deze relatie is dat de

overheid zelf een stimulerende factor kan zijn in de creatie van interpersoonlijk vertrouwen.

Het verlies van traditionele identiteiten heeft een gevoel van ontworteling teweeggebracht.

Nationalisme, oorlog en sterke leiders kunnen bijdragen tot het versterken van identiteiten.

Identiteit-gerelateerde verklaringen hebben ook een politiek luik. Dit verklaart wantrouwen

als gevolg van het feit dat men het niet eens is met het beleid van de regering. Dit kan het

gevolg zijn van de inhoud van beleid in het algemeen, maar ook van bepaalde in het oog

springende beleidsdaden. Dit kan dan weer het gevolg zijn van het feit dat de

regeringspartijen niet deze zijn waarvoor iemand heeft gestemd, of van een afkeer van

politieke partijen in het algemeen. Tenslotte dienen we te verwijzen naar de discussie over het

herstel van vertrouwen door participatie te bevorderen: participatie kan ertoe bijdragen dat de

overheid deel gaat uitmaken van iemands identiteit.

We vermeldden dat er discussie bestaat over de vraag of vertrouwen in de overheid nodig is.

Onderzoek heeft ook vastgesteld dat niet alle wantrouwende attitudes resulteren in gedrag. In

het geval van wantrouwen in de overheid bestaan er voor burgers een aantal exit en voice

mogelijkheden, zoals een verandering in stemgedrag, het zich onthouden van politieke

participatie, het niet geïnteresseerd zijn in een betrekking bij de overheid enz.

7



Introduction

The research project “citizen-directed governance: quality and trust in government” has been

commissioned by the Programme for Policy Research of the Ministry of the Flemish

Community (PBO 99B/1/14). The modernisation of the public sector machinery takes quite a

central position among the general strategies to restore trust in government. Improvement

initiatives, is the idea, make the quality of public service provision more manifest and

increase it. This, in turn, leads to satisfaction among citizens, which results in increased trust

in government. This reasoning contains many steps that are neither theoretically nor

empirically ultimately supported, and are even contested. The aim of the research project is to

look into this relation between quality of public services, user satisfaction with these services

and trust in government. This report tries to give a brief overview of the theoretical

approaches to political trust, and wants to frame the so-called performance hypothesis within

the wider range of research on trust in government.

We start this report by defining the concept of trust and show where to locate it in the

framework of government-citizen interaction. We will show that ‘trust’ is in fact a very

ambiguous concept, and that its moral value and meaning is changing. In order to be able to

speak about citizens’ attitudes towards government, we first have to know more about what

‘government’ exactly is for citizens. In other words: what is the object of their trust?

Therefore we will deal with the images citizens have of government: is government seen as a

homogeneous entity or instead as a collection of institutions? Do citizens perceive a political

system or instead individual politicians? Do some parts of government figure as symbols for

the entire government?

We continue with explanations for the -perceived if not real- decline of trust in government.

We introduce these theories by making the distinction between performance- and identity-

based evaluations of government. First we will deal with the so-called performance theory:

people have a negative attitude towards government because government does not perform

well. This evaluation depends on what citizens see (both perception and a normative

approach) as a government task, the correctness of their observations, and the importance they

give to the performance and service delivery aspect of government. The performance

hypothesis has two main parts: micro-performance (administration, functioning of politicians,



service delivery...) and macro-performance (inflation, unemployment, GNP growth, ...).

Distrust in government is however not necessarily a result of government-internal factors.

Therefore we will turn to general societal tendencies such as individualisation, modernity and

post-modernity, evolutions in interpersonal trust and social capital, etc. Furthermore we will

deal with structural factors such as information disequilibria, cultural factors such as the

nature of the democratic system (majority, proportional), and with the difference between

trust and contracts for organising society. Finally we will also look into the classical home-

team hypothesis, stating that a person distrusts government because he or she has voted for

the present opposition parties.

In the closing chapter we will discuss a number of consequences of the lack of trust in

government. First we will come back to the question whether trust is necessary, and whether

the attitude will lead to behaviour as well. Second we will discuss a number of specific

actions: exit and voice.

9



CHAPTER 1 What is trust?

1.1 DEFINITIONS

Trust is a concept surrounded by conceptual vagueness since it never was a topic of

mainstream sociology (Luhmann 1998). A generally accepted working definition of trust

seems to be lacking (Lewis and Weigert 1985). A new wave of research has tried to bring

more order. The concept of trust is intimately linked to risk and expectations: trust is used as a

substitute for risk, but it also creates a risk for the truster. As Baier states it

“Trust involves the belief that others will, so far as they can, look after our interests,

that they will not take advantage or harm us. Therefore, trust involves personal

vulnerability caused by uncertainty about the future behaviour of others, we cannot

be sure, but we believe that they will be benign, or at least not malign, and act

accordingly in a way which may possible put us at risk.”(Baier 1986).

This definition implies that trust is used in situations where there is no certainty. The

explanation of an increase of distrust as a result of the evolution to a post-modern society

therefore sounds very acceptable, since people are faced with increasing uncertainty. Trust is

used to reduce uncertainty and social complexity by anticipating the future. Trusting people

behave as though the future were certain (Luhmann 1998). For Sztompka

“trust is a bet on the future contingent actions of others” (Sztompka 1999).

According to him, trust is composed of seven factors: regularity, efficiency, reliability,

representativeness, fairness, accountability and benevolence (Sztompka 1999). Gamson

defines trust as

“the probability [...] that the political system (or some part of it) will produce

preferred outcomes even if left untended” (Gamson 1968).

Risk is also an important element in Coleman’s definition: it is the expectation of gain or loss

which determines whether one will grant trust or not (Coleman 1990; Ruscio 1996). Trust is

never absolute but always conditional and contextual. If there are no expectations, choice or



uncertainty, then we cannot speak about trust. Therefore, in a relation where one of two

parties is totally inferior or subordinate, we cannot speak about trust, since there is no

freedom of choice. Clearly, in a situation of certainty, there is no need for trust. Luhmann

distinguishes between confidence and trust, where confidence refers to a more or less taken

for granted attitude that familiar things will remain stable. If you do not consider alternatives,

you are in a situation of confidence (Giddens 1996; Luhmann 1998). In interpersonal

interaction people are often ignorant about the intentions of the other, and therefore there can

be no trust nor distrust (Hardin 1998).

Most research on trust deals with interpersonal trust: trust between people. Interpersonal trust

is sometimes approached at the aggregate level, but game theory for instance approaches it

from the most basic level: interpersonal interaction between two or more people. Here

granting trust is seen as calculative behaviour: trust is a strategy to maximise one’s utility.

Granting trust is an expression of encapsulated interest: I hope the other will act in my

interest, be it for his own reasons (Hardin 1999). This rational choice approach that places

trust in a framework of self-interest is not able to explain the existence of public trust.

However, since direct and personal relations are not always possible, institutions have been

created.

According to Lewis and Weigert, trust has 3 dimensions: cognitive, emotional and

behavioural. The cognitive process does not state that additional knowledge will lead to more

trust, but instead depends on a choice: at a certain moment, there is a (socially defined)

decision not to take any further evidence or rational reasons. The emotional base is visible

when there is a betrayal of trust, since trust often requires intense emotional investments.

“Behaviourally, to trust is to act as if the uncertain future actions of others were

indeed certain in circumstances wherein the violation of these expectations results in

negative consequences for those involved” (Lewis and Weigert 1985).

11



Lewis and Weigert made a categorsiation of the types of trust according to their emotionality-

rationality composition (Lewis and Weigert 1985).

Table 1: Emotionality and rationality in trust

Levi distinguishes two main streams in research on trust: on the one hand those who argue

social bonds are at the root of trust, and on the other those who argue for an encapsulated

interest approach (estimates of the probability that others will reciprocate cooperation) (Levi

1996).

Are people’s attitudes towards government a result of previous experiences, present

observation, or future expectations? In other words, is one’s evaluation prospective or

retrospective? In Sztompka’s definition, trust is a bet on future actions: i.e. one trusts that

something will or will not happen in the future. However, this attitude is not just based on

what could happen in the future, but even more so on actual performance or experience. The

micro-performance hypothesis for instance, the research theme of the project that states that

quality of government service delivery will increase trust is clearly based on experience and

present perception. Miller and Listhaug warn against the use of prospective evaluations,

because these are projections, and therefore subject to influence from any current

dissatisfaction (Miller and Listhaug 1999).



1.2 WHAT IS 'TRUST' IN 'GOVERNMENT'

This paragraph will deal with two questions: first, is the attitude commonly described as

‘(dis)trust’ in government really ‘(dis)trust’, or is this just a catch-all term referring to a

general negative or positive attitude towards government? Second, can ‘trust in government’

be distinguished both theoretically and empirically from other attitudes such as satisfaction

with democracy, alienation, satisfaction with the way government works, political support,

legitimacy etc.?

A number of surveys, among others the Belgian General Election Study and the European

Value Studies, inquire for trust in a direct way1, by actually asking for ‘trust’. The American

National election study on the other hand uses a five-item construct of political trust-political

distrust/cynicism. It consists of the following items:

• How much of the time do you think you can trust the government in Washington to do

what is right? – just about always, most of the time, or only some of the time?

• Would you say the government is pretty much run by a few big interests looking out

for themselves or that it is run for the benefit of all the people?

• Do you think that people in the government waste a lot of money we pay in taxes,

waste some of it or don’t waste very much of it?

• Do you feel that almost all of the people running the government are smart people

who usually know what they are doing, or do you think that quite a few of them don’t

seem to know what they are doing?

• Do you think that quite a few of the people running the government are a little

crooked, not very many are, or do you think hardly any of them are crooked at all?

1 European Value Study 99: “Please look at this card and tell me, for each item listed, how much confidence you have in them, is it a great deal, quite a lot, not very much or none at all?” (Q58) Belgian general election study 95: Could you tell me, for each of these institutions, whether you trust them a lot or a little (Q130_1-Q130_13)

13



The Swedish Society, Opinion and Media Institute survey (SOM), in use since 1986 uses

questions such as:

• ‘How much trust do you have in the way the parliament is conducting their work?’

• ‘How much trust do you have in the way political parties are conducting their work?’

Here also, the trust question contains a cognitive element since it not only inquires into an

affective ‘trust’, but also into an evaluation of how work is done, and to that extent it cannot

be used to measure diffuse support in its original Eastonian meaning. The focus in the

American National Election Studies (NES) on ‘the people running the government’

encourages respondents to confuse trust in the regime with trust in political incumbents

(Mishler and Rose 1997). The single item trust question on the other hand captures the

emotional and affective content of the attitude. The Eurobarometer question ‘On the whole,

are you very satisfied, fairly satisfied, not very satisfied with the way democracy works in…’

combines a cognitive evaluation with support for the system, a possibility not distinguished

by Easton (Kuechler 1991). This question is influenced by a short-term evaluation of policy,

and is as such a bad indicator for satisfaction with democracy (Holmberg 1999). However,

according to Kuechler (Kuechler 1991) ‘mass political support’ is a general theoretical idea

that does not lend itself for measurement by a single indicator. Using a single item trust

question, by adding an affective element to the cognitive evaluations, does not give any

indication of possible reasons for (dis)trust, and perhaps a combination of affective and

cognitive measures of both specific and diffuse support is the best way to deal with the rather

mixed and intangible attitude of citizens towards the political system.

A variety of indicators are used, making comparison difficult. Even though the NES five-item

construct is summarised into a single ‘trust’ indicator, we cannot say it covers the same

content as a one-item trust question. Futhermore, focusing on ‘government’ in the question

ignores variations in trust across different political institutions (Mishler and Rose 1997). The

fact that a diverse number of indicators is used for ‘trust’ suggests that it is not just trust as

such that one wants to measure, but instead a general positive or negative attitude towards

government. Later we will also deal with the question whether the meaning of the concept of

trust remains constant over time (1.3).

To test this claim we have to reduce the conceptual vagueness surrounding the concept. On

the one hand there are claims that a lot of variation can be found between the different



indicators (e.g. satisfaction with democracy, trust, way politicians work etc…), which

indicates that they are not measuring the same (i.e. a general negative/positive attitude

towards government). Citrin for instance finds support for this in the fact that distrust is not

accompanied by the will to make changes in the system (Citrin 1974). On the other hand, this

claim is disputed vividly, by e.g. Kaase (1999) who states that satisfaction with democracy

and trust in government are indicators of the same. A minimum of conceptual clarity is

therefore required to avoid wrong conclusions.

‘Trust’ implies that there is uncertainty, because the other (here government) has a certain

amount of freedom in his/her actions. In a situation of certainty, trust is not necessary.

Nowadays, the freedom of government to act seems to be more limited as compared to

previous centuries. Laws define a framework within which the state is supposed to act. A

great deal of the attitude towards the government’s actions can thus not be explained by trust.

Only a number of marginal actions require trust (because it is the first time such a thing

happens, the law is unclear, special conditions reduce certainty). Therefore, a hypothesis

could be that the more a state is seen as a Rechtstaat, the less the attitude towards government

is one of trust. In totalitarian regimes, the entire attitude towards government is one of

(dis)trust: the chance of being imprisoned without a reason is an actual possibility, which

means personal freedom is no certainty. The best there can be is trust that one will not be

imprisoned. In our political culture, politicians still take a very modernist approach to

problems: for every problem a solution can be found. This excludes ‘risk’ from the political

discourse, making that in fact we do not measure trust in government, since real trust is absent

simply because it is not necessary.

1.3 THE CHANGING NATURE OF TRUST

The common assumption is that trust has declined. What is however often forgotten is that the

meaning and content of the term has changed as well, because points of reference have

changed. When one wants to evaluate trust, one has to find out what reference points and

what criteria are used. For satisfaction with service delivery, encounters function as the

criterion for evaluation because the services themselves are often not very concrete, making it

difficult to use them as a handy criterion (Crosby, Evans, and Cowles 1990). The trust opinion

can sometimes be seen to base itself on very tangible facts. It can be expected that in a period

15



of scandals, the moral integrity of politicians becomes important as a criterion, or that opinion

leaders become important in closed communities, etc.

Yankelovich for instance states that freedom does not worry Americans anymore, since it has

been taken for granted (Yankelovich 1991). The fear for being arrested illegally by

government is hence practically inexistent, and does therefore not lead to distrust, nor will it

lead to trust. This is because to speak about trust, there always has to be a certain amount of

risk. If the risk of loosing freedom is practically non-existent, it cannot be part of the trust

attitude towards government. According to Beck, more attention is given now to risks that are

more limited, because big risks have disappeared or have been accepted. He gives the

example of the commotion on (small-risk) nuclear plants, while traffic accidents have a bigger

impact on society. The latter however have been accepted or have been defined away. This

shows that not risks as such are important but the consciousness about them (Beck 1992). The

disappearance of traditional risks (loss of freedom, war, hunger, threats to bodily integrity…)

has opened the way for new evaluation criteria, such as service delivery quality,

responsiveness, participation etc. Here the difference between trust and confidence made by

Luhmann becomes very useful: People who trust are still aware of circumstance risks and

alternatives, whereas confidence refers to a more or less taken for granted attitude that

familiar things will remain stable (Giddens 1996; Luhmann 1998). The (perceived) total

absence of threats to personal freedom could thus be defined as confidence. Giddens describes

trust as a situation of “tacit acceptance of circumstances in which other alternatives are largely

foreclosed” (Giddens 1996).

Apart from changes in the reference points over time, there are cultural differences. Where

freedom is taken for granted by Americans, this is not necessarily the case for the

Myanmarese, while North Koreans perhaps have no possibility to compare. All this makes

research difficult. Eastern Europeans surveyed after the fall of communism will compare with

Western Europe when asked about their satisfaction with democracy, while the West itself

will use another ideal image of democracy. Italians constantly display a limited amount of

trust, but in real political life there is very little difference to be found between Italy and other

Western countries (Kuechler 1991).

Since the freedom for governments to act as they like has become very limited, trust cannot

be the attitude we have to measure to evaluate the relation between citizens and government.



A real relation of ‘trust’ can only refer to a number of marginal actions of government, those

where discretion is still possible and thus risk present. How large this freedom to act is

depends on the political system we talk about. In totalitarian regimes, governments act as they

please, and in such a case, the total relation between citizens and government can be grasped

by measuring trust. In democracies, this will not be the case, since the freedom to act is in

practice very limited, and trust may then explain little. This perhaps explains the importance

given to other measures for the relation between government and citizen (e.g. satisfaction

with service delivery, feelings of political efficacy…). In the (post-nazi) German case for

instance, asking respondents about their satisfaction with democracy is irrelevant, since the

terminology used is loaded, and nobody would dare to admit he/she is not satisfied with the

way democracy works (Fuchs 1999). Measuring this will not give any indication of the real

mood. The question remains what defines which criteria are used when. An administration

that functions well will not take an important place in the assessment of government when

police and justice are unfair. Similarly, participation in decision-making only becomes

important once there is security and enough food. However, if one of the factors that are taken

for granted disappears, the consequences on trust will be severe.

The current observation as if there is a limited amount of trust can therefore just be a result of

the lack or change of points of reference: present generations did never have the experience of

war and hunger, and therefore their criticism of government has another basis than that of

older generations. For the younger generations, thresholds and criteria for evaluating

government have changed, and perhaps we could even speak about the absence of a

foundation to guide criticism.

A possible criticism on this observation could be about the equivalence of evaluation criteria:

is a negative attitude towards government less important when it comes from a bad

functioning administration than a negative attitude as a result of a repressive government?

This brings us back to the discussion whether trust is the opposite of distrust. In the first case,

there would probably just be an evolution from trust to no trust, while in the second the

evolution would be from trust/no distrust to active distrust. Perhaps we can distinguish

between two groups of evaluation criteria: those that take away distrust, and those that create

trust. One possibility can be found below. In the first column, a number of criteria can be

found that are not used explicitly. They are taken for granted, and do as such not contribute to

trust. They are part of the basic trust in society. If however doubts arise about their taken-for-

17



grantedness, this will lead to active distrust. In the second column, a number of criteria that

can lead to trust are mentioned: their presence leads to active trust, but their absence does not

necessarily lead to distrust or dissatisfaction. This table is only given as an example and the

question in which column to place these criteria is open for speculation because of a lack of

research.

Table 2: What factors is trust constituted of?

Distrust/dissatisfaction Trust/satisfaction

Loss of freedom Lack of corruption

Threats to bodily integrity Competence

Unlawful arrest Participation

War Efficiency … …

This observation leads to two possible conclusions about ‘trust in government’-research: first,

conclusions as would trust have in- or decreased over time are difficult to prove since the

content of the term has changed, and second, what is commonly presented as trust is in fact

not trust since no uncertainty or risk is involved.

1.4 IS TRUST IN THE INSTITUTIONS NECESSARY? THE CONSTITUTIONAL VS.

THE POLITICAL CULTURE APPROACH

Democratic systems seem to be built on distrust: the existence of an extensive system of

checks and balances creates control at all levels. It therefore sounds rather illogical to expect

trust in government in a democratic system. Some authoritarian systems on the other hand are

based on trust (e.g. when a dictator gives one of his cronies full powers over a certain

province). As it is sometimes rational to trust (e.g. because you expect to benefit from it), it

can sometimes be rational not to trust (Levi 1996; Hardin s.d.).

Parry distinguishes between two approaches to political distrust: the constitutional approach

and the political culture approach (Parry 1976). The basic question is, according to him,

whether political trust is grounded on a more pervasive trust in society at large, or whether we



should seek the basis of trust in the operation of political and constitutional arrangements.

This leads him to the observation that

“The phrase ‘credibility gap’ has become a commonplace of political discussion. Yet to an older school of thought it might have seemed inappropriate to credit the politician in the first instance”.

The constitutional and political culture approaches are two very different paradigms. In the

constitutional approach, distrust is necessary for the balances in the system. One does not

speak about trusting government, but about entrusting power to government. A difference is

made between stability and consensus: consensus on the political organisation is not

necessary to have stability. The lack of trust in government therefore does not threaten the

stability of the political system. In this view, polls showing that 50 % of the citizens do not

trust government are an indicator of a healthy democratic attitude. The political culture

approach regards distrust as a kind of political alienation, and as something that threatens the

stability of the system. Communitarian styles of politics take a consensus view on politics,

and trust is a prerequisite for consensus. Nevertheless, this system, as opposed to the

constitutional approach, does also create trust because it displays a larger openness for

political experiments and deviant behaviour.

At the time when the article was written (1976), the political culture paradigm clearly

dominated and it still dominates now. If within the political culture paradigm alienation is

observed, this is thought to be due to an inadequate transmission of democratic norms through

socialisation (Borre 2000). Trusting the authorities may be a result of socialisation: it is part

of one’s civic duties (Easton 1975). Even though the underlying core of the political culture

paradigm states that distrust is bad, it remains very difficult to measure the content of a

political culture: the concept is subjective and subject to change (Dalton 2000).

19



The table below summarises the main differences between the two approaches:

Table 3: Political culture and constitutional approaches to trust

Political culture approach Constitutional approach

• Political trust emanates from trust in

society (cf. social capital and interpersonal

trust theories)

• Political trust results from procedural

measures which help to redress mistakes

and abuse of power

• Lack of trust is a problem (cynicism,

alienation…)

• Agreement on all aspects of the political

system and trust in government is not

necessary to guarantee stability of the

democratic system

• Consensus is society is necessary • Conflict can be creative

• Politicians should be benevolent, good,

honest responsible people

• A politician does not have to be good, as

long as he/she is accountable

• Creating trust can be done by creating

consensus and by eliminating conflict and

risk

• Creating trust can be done by ameliorating

the means for redress and accountability,

more openness, better information, i.e. by

improving a system’s performance

With regard to the improvement trajectories mentioned in the table, one could come to the

somewhat strange conclusion that current politics tries to improve a system’s performance

(cf. constitutional approach) in order to reach the ends of the political culture approach

(consensus).

The discussion on the position and role of politicians is related to this. When judging

government, citizens have an opinion on what government is supposed to do. The same holds

for when politicians are being judged. Citizens can have different opinions on the role of a

politician: he/she should defend the interests of his/her voters vs. he/she should work for the

general interest. Half of the American respondents thought a member of Congress should

defend what his/her district asked for, even if he/she thinks it is a bad idea (Hibbing and

Theiss-Morse 1995). Politicians are here clearly seen as delegates as opposed to the statesman

approach. Huyse found that politicians were not seen as a representative of a political party or



ideology, but as a ‘relation’, friend, influential person etc. who can ‘arrange’ something. A

politician is approached as somebody who can do personal favours (Huyse 1969). If this

politician does not cater for the individual citizen, people will not have a positive opinion

towards him/her. The liberal approach to politics does not consider it as bad when people just

delegate politics to elected politicians. Constant mass participation and involvement is in this

approach not seen as necessary. The present urge to involve people politics can therefore

conflict with a will to delegate politics (Pulkkinen 2000).

1.5 TRUST OPINIONS VS. THE DOMINANT DISCOURSE

Distrust does not necessarily have an influence on the stability of the political system or the

behaviour of the citizens and the claim one should trust government or not is disputed. One

possible explanation could therefore be that the attitudes of the citizens do not result from a

personal negative attitude vis-à-vis government, but because expressing a negative attitude

towards government could be considered as a fashion, prejudice or cultural element. Citrin

states that the “Zeitgeist” (in 1974) promotes anti-political rhetoric. He considers denigrating

speech on politicians and institutions as ritualistic negativism, and does therefore not see any

reason to be worried. He compared political distrust with the rhetoric used in a baseball game:

everyone yells at the referee and accuses him of mistakes, while this rhetoric never threatens

the game (Citrin 1974). One could even consider trust/distrust in government or in general as

a social norm, of which it is difficult to deviate.

Sztompka speaks of a ‘culture of (dis)trust’:

“When a culture of trust -or culture of distrust- appears, the people are constrained to

exhibit trust or distrust in all their dealings, independent of individual convictions,

and departures from such a cultural demand meet with a variety of sanctions”

(Sztompka 1996).

The observation that people are very critical of government and its service delivery in

everyday speech (e.g. gossip, discussions in pubs etc.), while this is not always reflected in

trust- or customer satisfaction surveys gives additional evidence for the presence of a social

21



norm (we do not give any indication of how this came into existence). Beck Jørgensen

analysed novels in which the main subject was the relation between citizens and

administration (e.g. Kafka’s novel ‘The Castle’) (Beck Jorgensen 1994). The advantage of

such an approach is that it also allows catching informal aspects of the relation, that it mainly

deals with perceptions, and that it allows tracing evolutions because of the availability of

older material. His analysis shows that these novels, when dealing with the alienation in the

relation between government/administration and citizen, never refer to concrete activities, but

to (perceived) objectives, consequences and context. Allport’s research on prejudices finds

similar processes: People always choose groups and not persons as out-groups. The

abstractness of groups allows for changing one’s attitude towards certain persons in that

group. When one then does encounter a fact that is not reconcilable with the stereotype, it is

not necessary to alter the stereotype, but one can just attribute it to a difference of the specific

person/fact (Allport 1958).

Goodsell found that citizens take a negative stand towards government as a whole, but when

‘government’ becomes more concrete in surveys, this negative attitude largely disappears

(Goodsell 1994). In fact, this symptom can be found practically everywhere: parents evaluate

the school of their own children as good, but are sceptical about the educational system.

People evaluate their own family and (working) community as good, but they still think these

institutes are disappearing (Loveless 1997). Hibbing and Theiss-Morse speak about Fenno’s

paradox: in surveys, people are positive about specific members of Congress (i.c. their own

member), but take a negative attitude towards Congress as an institute. One of the reasons

they give for this is that people use other assessment standards (Hibbing and Theiss-Morse

1995). The abstract nature of government could be another explanation, since rumours only

appear when the real facts are ambiguous or vague (Allport and Postman 1965). Instead of

rumours, we could also speak about (administrative and government-related) ‘stories’ or

urban legends.

We already mentioned the influence of the dominant discourse on the attitude towards

government. Citizens often lack experience (encounters) and knowledge to come to a well-

considered personal opinion on whether or not to trust government. The same goes for more

down to earth evaluations such as those about the quality of service delivery: people can have

an opinion even if there had been no previous contact (Stipak 1977). Therefore, it is perhaps

better to treat trust as an opinion instead of a reaction on experience. Attitudes are formed



both by personal experience and by reference groups. An aspect of the political culture

approach to political trust is that people are taught/socialised to trust government.

Socialisation is easier when the norm you are taught by family in early childhood and peer

groups corresponds to the dominant norms in society.

The spiral of silence hypothesis states that the perception of the distribution of public opinion

influences the willingness to express one’s own opinion, because one does not want to isolate

oneself by having a different opinion (Glynn, Hayes, and Shanahan 1997; Noelle-Neuman

1974). As long as the people think most people distrust government, they will express distrust

themselves, even if this perception does not correspond to reality. Minority opinions thus

become very difficult to express and are met with sanctions. The pervasiveness of mass media

contributes to this phenomenon, since these are mainly focusing on majority opinions. If

distrusting government would be a social mood or even a social norm, it becomes very

difficult to restore trust in government. Communication theory and diffusion studies could

perhaps explain the diffusion dynamics of distrust and negative experiences with government

and suggest possible strategies for reverting this trend. Diffusion of (dis)trusting opinions

could therefore be a key factor. Research on service delivery revealed that dissatisfied

customers tell on average seven people about their experience, while satisfied customers do

not. This implies that a negative opinion is diffused faster. Stories and myths about the

administration should also not be underestimated: certain stories, call them ‘urban legends’

can have an important impact on opinions. Optimal diffusion of trust or distrust is done by

trust entrepreneurs (Levi 1996).

This discourse is not only a result of the opinions of co-citizens, prejudices and rumours, it

can also be part of political culture: some political cultures regard a distrusting attitude

towards government as a virtue, others as a vice. This evaluation can change over time.

Distrust in the 60’s was considered as a progressive force, while the dominant political

discourse was conservative and right wing. Now, the dominant political discourse is

progressive and left, while many of the distrusting are conservative. This observation could

lead to the conclusion that distrust in government is not a result of government-internal

factors. Perhaps there is a sort of generalised trust/distrust, affecting all institutions of society.

Evolutions in trust in big companies, churches, trade unions and media support this

hypothesis. It could even be that a decline of social trust among citizens themselves is the

cause of declining trust in institutions. Finding support in the World Values Studies data,

23



Newton and Norris conclude however that there is a specific malaise for government and

politics, and that therefore broad and general theories of social and economic crisis are

inappropriate. The same goes for explanations focusing on the role of mass media, post-

materialism, globalisation etc. (Newton and Norris 1999). Non-political explanations of

distrust clearly go further than giving political causes to political distrust. They refer to

general tendencies in society, which also have an impact on government (social capital,

postmodernism…).

1.6 TRUST IN GOVERNMENT - NUMBERS AND TRENDS

When comparing levels of trust between European countries one often finds Belgium and

Italy at the bottom of the list. In Scandinavian countries, government always seems to enjoy a

considerable amount of support. One possible explanation could be that trust in the

institutions is just lower in Belgium and Italy. The alternative approach would be to explain

these differences by referring to cultural and structural antecedents of people’s opinion. In the

research on service satisfaction, it is said that dissatisfaction is only pronounced when quality

exceeds certain thresholds (Stipak 1979). The same could hold for trust in government.

Thresholds can differ between countries, making that the moment at which (dis)trust is

pronounced differs between countries or even individual people. The position of the

thresholds could possibly be explained by a number of cultural values, historical elements and

the dominant discourse. A number of institutional factors could also have an influence (e.g.

majority vs. consensus based systems). When one does not have information on these factors,

it could be very difficult to make cross-country comparisons.

To make meaningful statements, an approach could be only to look at relative numbers:

instead of comparing trust in local government in country A with trust in countries B, C, D,

one could look at the relative distance between trust in local and central government in all

four countries, and then assess trust in local government2. A similar approach could be to

compare trust in private to trust in public institutions.

2 Thanks to Wouter Van Dooren for mentioning this.



Politicians refer to a crisis of democracy due to constantly declining trust in government. It is

however not sure whether empirical facts support this conclusion. The scientific diagnosis on

the syndrome shows a certain evolution (Norris 1999b): in the 60’s and 70’s, there was a

‘crisis’ and the cause was government overload. The 80’s described the 70’s ‘crisis’ not as a

trend, but as a mere ‘fluctuation’; and the democratisation in a number of countries softened

fears of the end of democracy. ‘Malaise, disengagement, ennui, flight from politics and

Politikverdrossenheit’ became the key words of research in the 90’s. Nye states that there is a

lot of attention for performance theories, but that these are not able to explain the trend of

decreasing political support (Nye 1997). This shows there have been worries about distrust in

government for decades or even longer, which made Dahrendorf to conclude that

“It appears that democracy is always in crisis. Twenty-five years before the present

volume, in 1975, the Trilateral Commission published a report on its Kyoto

Conference entitled The Crisis of Democracy. […] Thirty-five years before that,

Harold Laski described the democratic crisis vividly in his Reflections on the

Revolution of Our Time. In fact, the topic is of classical parentage. Tocqueville’s

ambivalence toward what he called democracy was coupled with the hope that it

would not last. And Aristotle, of course, the father of all such theories, regarded

democracy as intrinsically unstable and therefore preferred a polity that mixes

democratic and oligarchic elements. So what is new today?” (Dahrendorf 2000)

American authors constantly refer to the results of the National Election Studies (NES), which

show that up to 70% of the respondents trusted government in the 60’s. Comparing these to

recent results, they point at the dramatic decline of trust. Hibbing and Theiss-Morse however

do not refer to the NES but to Gallup polls in the late 40’s. These showed only 15-20%

thought Congress was doing a good job (Hibbing and Theiss-Morse 1995). When one

compares present numbers to these figures trust did not really decline. So the question

remains: has trust in government really disappeared? Dalton sees a declining trend in most

countries, except for Norway and the Netherlands, be it that measurement started only at a

later date in the latter (Dalton 1999). Deschouwer analysed Eurobarometer data (satisfaction

with the way democracy functions), but did not find any trends, just fluctuations. At the same

time the differences between countries remained stable, leading him to conclude that we

should look for explanations in structural and historical factors (Deschouwer 1998). Dierickx,

25



Gijselinckx et. al. also used Eurobarometer data and found strong differences, but not trends.

They found that alienation is subject to fluctuations. The data showed the Germans to be very

distrusting in the 50’s, while they are now one of the least alienated citizens (Dierickx,

Gijselinckx, and Thijsen 1996). Newton and Norris on the other hand, working on World

Values Studies data find a decline in trust for all public institutions, with rare exceptions

(Newton and Norris 1999). Levels of satisfaction with democracy tend to fluctuate very

strongly in all of the countries, but still a similar direction can be discerned in it (Huseby

2000).

Finally, since citizens’ attitudes towards government are mainly based on subjective

evaluations, we have to take great care when dealing with the numbers. First, there seems to

be quite some fluctuation in the numbers, and the results depend on context (both

methodological and societal). Second, it are not necessarily the numbers that are important,

but the reception of the numbers. It might well be that the objective level of trust as measured

in surveys has not changed, but that there is an increased sensitivity for the issue. After all, a

problem is only a problem if defined as such. It might well be that the increased attention

given to trust in government creates awareness of reasons for distrust.



CHAPTER 2 The object of trust

Most research on trust deals with interpersonal trust. Worries about declining trust in

government imply it must be possible to trust institutions, i.c. government. This raises two

questions: is the relation towards government a relation of (dis)trust, and can we trust

institutions? The first question has been touched upon in the previous chapter. According to

Hardin, interpersonal trust is a phenomenon very different from trust in government. He

concludes that in fact we do not know whether to trust government. If one wants to trust an

institution, there are two ways to do this: one has to trust every individual in it, or the design

of roles and incentives in the organisation should be such that it works in our interest. For

government institutions this is impossible because of the scale (you can not know all persons

and their roles), and because it is impossible to have a good understanding of institutions and

structures, because the tasks of government are very complex. Hardin therefore concludes that

our views and expectations of government cannot underlie trust in government (Hardin 1998).

Furthermore, trust comes from predictability from past behaviour, while this predictability is

not reliable since it concerns human behaviour. Predictability is also difficult since the precise

intentions of government are not clear. According to Luhmann, we can only have confidence

in government, not trust (Luhmann 1979). Offe speaks of institutions as a substitute for trust.

In this way, you do not have to trust the doctor as a person, since it is sufficient that he has

graduated from a medical school. Major deviations from the normal course of events are

outruled by the built-in self-correcting mechanisms these institutions have. Of course,

institutions are of a very impersonal and abstract nature. This explains the success for populist

politics: it re-personalises institutions (Offe 1999).

2.1 SOFTWARE AND HARDWARE

Will distrust in certain politicians lead to a rejection of these politicians, or instead have an

influence on the attitude towards the entire political system? Muller and Jukam’s research

question was whether trust is a measure of (dis)approval of an incumbent administration

(those in power) or whether it is related to feelings about the legitimacy of the system of

political authority as a whole (Muller and Jukam 1977). The Miller-Citrin debate in 1974 also

27



dealt with this question: Citrin argued that trust is a measure for incumbent affect, while

Miller saw it as a measure for diffuse support for the system of government (Citrin 1974;

Miller 1974b; Miller 1974a). Citrin states that many of those who claim not to trust

government still support the present political system. Distrust is as such aimed against current

office-holders/incumbents. This would imply that person-related changes in politics would

have to have an influence on political trust, when corrected for ideological identification. It

could however also be the case that there is distrust towards political leaders in office,

notwithstanding who they are. Indeed, Mitchell and Scott assert that the ‘confidence’ gap is

one of erosion in confidence in leadership, not in the institutions (Mitchell and Scott 1987).

For Citrin, the fact that distrust of government produces neither political apathy nor political

activism for him reinforces the argument that many cynical responses merely record

opposition to incumbent officeholders or largely ritualistic expressions of fashionable clichés

(Citrin 1974). In his study of the legitimacy crisis in the USA, Yankelovich distinguished

three dimensions (Friedrichs 1980):

• Moral legitimacy crisis (integrity and quality of political leadership)

• Ideological legitimacy crisis (disillusionment with the core values of the system)

• Functional legitimacy crisis (disillusionment with the effectiveness of public

institutions)

One of the first to develop an encompassing theory on political support was David Easton. He

developed a two-dimensional framework, the first dimension containing the political

community, regime and authorities, and the other ranging from specific to diffuse support.

The idea behind it is that one can withdraw one’s support for the current government, without

withdrawing it for government. Alternatively, one can fully support the democratic values and

at the same time be very critical about the way democracy works at this moment (Easton

1965).

With regard to support for the political system, Easton distinguishes three fundamental

objects of support: the political community, the regime and the authorities. The political

community refers to a certain division of political labour, and does not necessarily have to

include a sense of community in the sociological sense. Regime refers to the operation of the

system, and the way political relationships are ordered. This division implicates that a system

may persist as a whole or instead only with respect to one of its basic objects (Easton 1965).



Norris has broadened Easton’s dimensions (Norris 1999b). She distinguishes 5 levels.

• Political community: attachment to the nation, beyond institutions

• Regime principles: values of the system

• Regime performance: how the political system functions in practice

• Regime institutions: parliament, police, civil service, …

• Political actors: e.g. evaluation of performance of particular leaders or politicians as a

class

Gamson distinguishes 4 objects of political trust (Gamson 1968):

• Incumbent authorities

• Political institutions

• Public philosophy (probably this aspect was more important in the past, since this

book was written during the Cold War with its opposition between two political

systems/ideologies)

• Political community

Support is specific when it is a direct consequence of satisfaction with system outputs

(output=transaction between system and environment). This implies that demands of the

members of the system are fulfilled. Demands, because of their often conflicting nature can

however not always be met, and sometimes there is a large time lag between input of a

demand and output. These outputs can be authoritative allocations of values, binding

decisions and actions, or transactions between a system and its environment. A sub-optimal

output can however sometimes be accepted because of the existence of diffuse support.

Diffuse support forms a reservoir of positive attitudes or goodwill. Easton himself relates

specific support to the satisfaction of members of a system with the perceived output and

performance of the political authorities, even though the knowledge of these political

authorities can be very limited. Specific support in Eastonian sense assumes that people are or

can become aware of the political authorities, even though it is enough that the members have

knowledge of the authorities as a class or undifferentiated group even if they cannot name

names or describe functions. Perceived outputs have to match articulated demands. Even

when this performance does not have an influence on trust or legitimacy, it can still generate

support for the political incumbents (Easton 1975). This means that bad performance does not

necessarily have an influence on diffuse support.

29



If there is a lack of support, generation of specific and diffuse support is necessary. To

increase specific support, only the authorities can generate specific support, by distributing

benefits (outputs) to the members of the system. Any member or group in the system however

can generate diffuse support, special attention being given to opinion leaders and mass media.

Therefore, actions taken by the authorities (i.e. outputs) only take a limited place in the repair

of diffuse support. Muller and Jukam relate the terms incumbent and political affect to

effectiveness and legitimacy. A lack of trust in the incumbent is then a result of a lack of

effectiveness, while this does not have an influence on the legitimacy of the democratic

system (Muller and Jukam 1977).

Support for the system can be generated by direct measures such as a change in the political

structures or the norms of the regime, but also by generating diffuse or specific support.

Diffuse support can be generated by a number of factors:

• Ideological principles

• Structural legitimacy (structure of the system, rules and roles for the execution of

power)

• Personal legitimacy (e.g. belief in personal qualities of the authorities)

• Belief in common interest

According to Easton, specific and diffuse support are different concepts, which makes they

will have independent determinants as well as different consequences for the functioning of a

system (Easton 1975). However, the two concepts are mixed frequently. Low output and

performance only has an effect on diffuse support in the long term, but in periods of scandals,

both specific and diffuse support often decrease. This is because diffuse support has been

under pressure for quite a long time because of continued low performance. The event or

scandal just serves as an outlet for adapting to new levels of support. Similarly, a continued

increase of specific support may eventually lead to increased diffuse support. If diffuse

support is low, it can however be that this influences the evaluations of the output of

authorities, because

“disaffection may occur not because of what each succeeding set of authorities is

supposed to have done but simply because they are perceived to be authorities - and

authorities are no longer thought worthy of trust” (Easton 1975).



It is sometimes difficult to determine where cognitive evaluations stop and affective ones

begin, as the following quote shows:

“People often forget the reason for their negative evaluations of public officials, but

the sentiments are still there, and may give rise to permanent negative evaluations of

the political system”(Huseby 2000).

Clarke, Kornberg, et. al. utter a number of criticisms on Easton’s framework. According to

them, the difference between diffuse and specific support cannot be established empirically.

This is partly because diffuse support has a symbolic content, and can as such not be

measured (Clarke, Kornberg, and Stewart 1984). According to Hibbing and Theiss-Morse,

Easton’s reasoning is tautological: he states that measures of affect (diffuse support) cannot

covariate with performance-based satisfaction. If however he finds a relation, he takes this

measure away from diffuse support and places it under specific support. Furthermore, they

argue that Easton is too output-oriented and devotes too little attention to the processes

(Hibbing and Theiss-Morse 1995). Kuechler made a categorisation of the types of support

according to Easton and concluded that a cognitive/evaluative assessment of the political

system was lacking in Easton's framework (Kuechler 1991).

Table 4: Cognitive and affective evaluations in Easton’s work

Object

Form

Incumbent

System

Cognitive/evaluative Specific ??

Emotional/affective Diffuse Diffuse

The question now remains where to place trust in the incumbent-system affect spectre. Is the

present distrust a result of a lack of specific or of diffuse support? This is important for the

evaluation of distrust -“does it matter”-, and for the conception of possible remedies.

Hetherington found that the level of political trust influences trust in the president more than

is the case the other way round, but with regard to Congress, it is trust in Congress that

influences general political trust instead of general political trust that influences trust in

Congress (Hetherington 1998). The reason he gives is that trust in the president should be

31



seen as specific support, while trust in Congress is diffuse. He considers trust as a function of

institutional support, and less as incumbent support. Muller and Jukam do not agree with this,

but this could be due to the fact that Hetherington equals trust to support, while Muller and

Jukam do not (Muller and Jukam 1977). According to Easton, legitimacy and trust should be

distinguished from each other. Both legitimacy and trust are dimensions of diffuse support.

However, it is possible for people to

“lose their trust in the ability of authorities to run the country yet not be prepared to

deny the authorities in general the moral right to rule and to expect obedience to

outputs” (Easton 1975).

2.2 CITIZEN PERCEPTIONS OF ‘GOVERNMENT’

Comparing data on trust in government in different countries requires two preconditions: both

‘trust’ and ‘government’ should have the same meaning, or the cross-national differences

should be known in order to weigh the data. It is not always clear what people see as

‘government’. Are schools part of government? Or the post-office? The railways? A survey in

the UK asked citizens which organisations they thought of as public services. 55 % mentioned

Councils, 51% public transport, 29 % police, while only 13% spontaneously mentioned

central government (Public Management Foundation 1996). This could mean central

government is thought of in political terms and not so much as service provider. Research by

the European Commission in its Continuous Tracking Survey showed a number of important

differences between European countries (Direction Générale "Information, communication,

culture, audiovisuel" 1997). Differences in the judgement of telecommunications, radio and

television, public transport etc. can be explained by a different history of privatisation,

political influence and the structure of the market. What is surprising however is that there are

-be it small- cross-country differences between judgements of police, justice etc. as public

services. 94 % of the Dutch consider the judicial system as a public service as compared to

85% of the French. Figure 1 gives some data on Belgian service providers:



Figure 1: What do citizens see as government?

Do you consider … as public services? (Belgium)

0

20

40

60

80

100

% yes 84 89 90 76 71 65

1 2 3 4 5 6

1. Health services

2. Educational system

3. Post

4. Water and electricity

distribution

5. Telecommunication services

6. Television and radio

Such a question is important to discover what kind of services are considered as part of

government, and can thus be supposed to have an influence on the opinion of trust in

government one has. If a large part of a population does not consider e.g. the post office as a

public service (as is for instance the case in the Netherlands) one could conclude that the

evaluation of the functioning of and the general attitude towards this post-office does not have

an influence on trust in government. However, such a conclusion could be false, because there

is no proof that only government-related factors constitute the opinion ’trust in government’.

In the USA, a listing of high impact agencies was developed, containing those government

agencies that are supposed to have most impact (both by frequency of contact and visibility)

on citizens. If this impact were decisive for the formation of trust in government, it would be

possible for each institution separately to calculate what influences it has on trust. The public

has difficulties to distinguish public and private sector (Dinsdale and Marson 1999), and a lot

of errors of attribution therefore occur (Swindell and Kelly 2000; Hoogland DeHoog,

Lowery, and Lyons 1990). In the next paragraph, we will argue that some institutions can

have a strong individual impact on trust, but this does not necessarily have to contradict the

existence of a generalised trust in ‘institutions’? Lipset and Schneider concluded in 1983 that

trust in the leaders of both private and public institutions showed similar tendencies (Lipset

and Schneider 1983). In addition, Norris concluded that in the USA, where data are available

from 1964 on, that trust has declined for institutions in general. That means that not only trust

in government declined, but trust in universities, medicine, major companies, press etc. as

well (Norris 1999b). Even more, during the 1980’s, the difference between confidence in

public and private institutions narrowed (Listhaug and Wiberg 1995). Inglehart saw declining

33



respect for hierarchical and authoritarian institutions in general (Inglehart 1997). Making such

distinctions requires that questions be asked about a large number of institutions in surveys.

There is an enormous amount of literature on trust in government, and the observation

mentioned above that declining trust in institutions could be broader than just government,

made that researchers do not only look at factors internal to politics, but also at broader

societal factors such as post-modern values, social capital, interpersonal trust, satisfaction

with one’s own life. This observation immediately limits the possibilities for governments to

find solutions for the credibility gap.

2.3 GENERALISATION VS. DIFFERENTIATION

In the previous paragraph, we dealt with the question whether people’s opinion on trust in

government is a direct result from government internal factors. In this and the next paragraph

we will look at the question whether trust in government refers to one amorphous unity, or

whether it is composed of trust in the composing entities of government. In most surveys,

respondents are shown a list of institutions after which they are asked how much trust they

have in each institution separately (scale from 1-4, or 1-5). It turns out that not all of these

trust opinions correlate perfectly, and that a number of clusters can be found in the list. Even

though there are differences, trust in one institution often means trust in all institutions

(Elchardus 1998). Some institutions however may have a determining impact on trust in

government in general, such as parliament, the president etc.

A number of authors have noticed a process of generalisation. There seems to be a common

factor behind the evaluations of all institutions that are related to ‘government’ (Uslaner 1999;

Dierickx et al. 1996; Mishler and Rose 1997; Stipak 1979; Norén 2000; Loveless 1997;

Pleysier, Vervaeke, and Parmentier 2000; Steen 1996; Herzog and Claunch 1997; Muller and

Jukam 1977; Kuechler 1991). This would mean that there is just one opinion of trust in

government, because people do not make distinctions conceptually. Even between some

private and public services very little difference can be found (Poister and Henry 1994). This

observation could lead to two conclusions: trust in institutions does not necessarily have

something to do with government, and, government is approached as if it was one amorphous

concept.

As for the second statement, there is no agreement among researchers. The most extreme

viewpoint is Klingemann who states there is no generalisation, and that all depends on actual



performance. If there are similarities this is due to similar performance (Klingemann 1999).

The performance hypothesis also takes this as an implicit assumption. If on the other hand

people see government as one amorphous unity, it seems improbable that specific experiences

with specific services will have a strong impact on trust in ‘government’. Another remark is

that the evaluation of government in general differs from the summation of all agencies (not

institutions) separately (Princeton Survey Research Associates and Pew Charitable Trust

2000). Blaming distrust on a bad functioning of the administration is difficult to maintain

when respondents think more often about politicians (67%) than about civil servants (16%)

when they rate government. It has to be said however that the public has difficulties to

distinguish the political from the public service (Dinsdale and Marson 1999).

Mcallister used the 1990-1991 World Value Studies data to calculate correlations between

confidence in parliament and in the civil service. He found that this correlation was in some

cases up to .80, which could indicate generalisation, while in other cases it was just .39. For

Belgium, the data indicated a .46 correlation (Mcallister 1999). Rothstein points at the

importance of output institutions to influence trust in general, since citizens have more contact

with these as compared to the input-side (voting, politicians)(Rothstein 2001). Huseby

expects a certain similarity between trust in public and private institutions due to the

disappearance of public-private ideological conflicts. In general, trust opinions seem to differ

more across institutions in the Nordic countries as compared to Southern Europe. The army

often is a case apart because of its remote and abstract character. In some research, police,

legal system, parliament and civil service seem to make up one single index (Huseby 2000).

Of course, we have to be careful with the interpretation of these differences. Citrin and Green

found that confidence in ‘the White House’ is larger than confidence in the federal executive

branch, probably because The White House has a symbolic connotation of power, while

executive branch has a rather negative connotation (bureaucracy) and is not very concrete

(Citrin and Green 1996). This means wording could explain part of the differentiation.

Concerning satisfaction with urban services, Stipak states that there is generalisation, e.g. that

specific attitudes are based on general evaluations, for instance because of lack of information

and knowledge and because political objects are in many cases rather impalpable (Stipak

1977). The Canadian Center for Management Development finds that indeed general attitudes

towards government affect perceptions of service quality (Erin Research inc. 1998).

35



Therefore, belief system differentiation (meaning that less general evaluations are used)

should occur more when respondents are better educated and politically more informed. Steen

did research in the Baltic States and found there was more trust in newer institutions (because

these were not burdened with a communist legacy, and because the population has recently

asked itself for their founding). Institutions producing a symbolic and /or diffuse output

(church, army, press…) enjoyed more confidence as compared to those with a very specific

output. This is probably because it is easier for people to have clear criteria to judge specific

outputs and because the actual outputs have deteriorated (cf. bad economic situation). Finally

he found more trust in the leaders of institutions than in the institutions themselves (Steen

1996). New Democracies Barometer data (Central and Eastern Europe) show that political

and civil institutions (with the exceptions of the church) are evaluated in a holistic way,

probably because these concepts used to be mixed under communism. Nevertheless, a

distinction could be made between 3 dimensions of trust: trust in political institutions, trust in

social and economic institutions (related to market reforms) and trust in traditional authority

(here church and army) (Mishler and Rose 1997).

Differentiation between institutions is comparable to differentiation between different levels

of government. Local or decentralised government is said to enjoy more trust than central

government, because the distance between citizen and government is smaller, and the

frequency of contact higher. This implies that citizens would differentiate between local and

central government. Uslaner finds that the degree of approval for president and Congress in

the USA influences trust in both federal and state level, thereby refuting the claim that people

trust those levels closest to them more. He concludes there is no differentiation between

different levels of government, but just a general attitude of trust in government. Therefore,

the problem of trust in government will not be solved by empowering other levels of

government (Uslaner 1999). Glaser and Denhardt find that government is government

regardless of level. Performance ratings of (US) federal and state government have a very

strong influence on perceptions and evaluations of local government (Glaser and Denhardt

1997).

As for the hypothesis of the research project, it is important to know to what extent there is

generalisation. If citizens do not make the distinction between the different institutions, it

becomes difficult to determine the independent influence of government services. The process



of generalisation, or better the opposite of it, differentiation, is part of socialisation, and

requires a learning process (Berger and Luckmann 1966). Generalisation can therefore point

at a lower level of political sophistication, or at a (deviant) form of socialisation, where

people are learnt to see government as a monolithic bloc without any differentiation, making

it thus easier to attribute it extremely good or bad characteristics.

2.4 WHAT PARTS OF GOVERNMENT ARE MORE IMPORTANT IN THE

FORMATION OF A (DIS)TRUSTING ATTITUDE?

One can say that the attitude ‘trust in government’ refers to ‘government’ in general. Because

of the fact that people do not always know what agencies are part of government and which

ones are not, or because people never have any contact with a large number of government

agencies, there are reasons to think that there are some core agencies that determine the

attitude towards government, depending on the visibility, importance or performance of these.

In states with strong leadership, the president or prime minister may take up this role.

Easton and Dennis wrote a book in 1969 on how (American) children see the political system

(Easton and Dennis 1969). Small children do not see government as one homogeneous

institute, instead they distinguish already a number of bodies, and they are able to see

differences between formal and would-be political authorities. They consider the president as

very important, but older children evaluate authority institutions such as government or the

Supreme Court as more important. The older they become, the more importance is given to

structures and impersonal institutions, and less to persons. This could be an indication of

political sophistication. They also found that even young children did not refer to the

president as the person, but as the institutionalised role. In the case of the president, he is a

strong personalisation of power, which is not the case for e.g. a Representative, who is less

able to personalise the legislative power of Congress. Personalisation requires that these

figures can be dramatised: a presidential election for the president, police actions for the

police officer etc.

This suggests that certain parts of government are more visible and judged as more important

by people, depending on, among others, political sophistication. As for the United States, the

37



President is seen as government in person, and this will have an influence on the image of

government in general. President Reagan for instance managed to increase trust in

government by emphasising ceremonial aspects of the presidency, such as symbols, the

‘grandeur’, the image of the president who stands above politics, …(Citrin and Green 1996).

Hibbing and Theiss-Morse on the other hand, in their book ‘Congress as Public Enemy’, state

that it is Congress which defines the (negative) attitude towards government in the USA

because of its visibility: all debates, compromises and opposition are too visible. A

conceptualisation of the general connections commonly made in the public mind among the

various parts of the national government showed members of Congress as part of the

(pejorative) ‘Washington system’, while the president, and certainly the members of the

Supreme Court were leaning towards the ‘constitutional system’ (Hibbing and Theiss-Morse

1995). A similar point is made in research on the Canadian situation: Feelings about

parliament and assessments of MP’s have significant effect on levels of support for the

national political community and regime, because parliament is the most salient and dramatic

symbol of the representative character of politics (Clarke et al. 1984). Certain parts of

government seem to take a more prominent place in people’s mind, but as the American

examples show, it seems difficult to bring the whole attitude/image back to one institution.

There is no reason why these images should not differ between countries and over time. In

transition countries parliament could for instance take an important place because of its

symbolisation of democracy. In a similar way, one could say that the justice system had an

important influence on trust in government in Belgium in the mid 90’s because of a number of

scandals that had a deep impact on society. Another question could be whether it are the

institutions as such which have an impact, or that these just symbolise criteria used for

judging government. In certain periods or areas, more or less importance is given to

efficiency, legitimacy, participation etc. When the pendulum swings to participation, perhaps

parliament weighs stronger, while the head of state and the administration weigh stronger

when importance is given to efficiency.

A distinction that could be made is whether people see the institutions or the persons in it. As

described above, even children seem to be able to distinguish between persons and their

function. Research in the Baltic States showed a low level of confidence in institutions,

combined with a relatively high trust in leader persons, which indicates a person-oriented

political culture. These results are matched closest by France, but differ totally from the



Norwegian ones, which indicate a more egalitarian political culture (Steen 1996). In his

research on public confidence in education, Loveless referred to research by Lipset and

Schneider who stated that dissatisfaction with government was aimed at specific persons.

Education did not really have well-known, visible persons that could be blamed, which

explains why trust in the educational system did not decline. Nowadays, more people blame

systems and not so much persons, which has made that scepticism has become broader to

include among others the educational system which had escaped till then (Loveless 1997).

Citrin suggested that because of a decrease in party loyalty, personal characteristics and the

policy of the president, the president as a person gained importance (Citrin 1974). The

research mentioned above has two methodological consequences: research should go further

than just evaluating a particular leader, and politics and political institutions should be dealt

with in a very detailed way in questionnaires.

39



CHAPTER 3 Performance vs. Identity

In theoretical approaches to trust, we find a constant tension between identity and

performance. When we defined trust (1.1), we already referred to Lewis and Weigert who saw

trust as a certain combination of rationality and emotionality (Lewis and Weigert 1985). Both

Easton (Easton 1965) and Almond and Verba (Almond and Verba 1965) have made similar

distinctions (Dalton 1999). Other grounds for evaluating government are possible:

Klingemann distinguishes three: moral, emotional/expressive and rational/instrumental

(Klingemann 1999). Attitudes towards government are thus a combination of these evaluation

criteria, and

“those intent on reshaping government to reflect the values of business are making a

mistake; citizens do not support governmental modes with a unidimensional focus on

efficiency” (Glaser and Hildreth 1999).

It remains to be investigated which if not all criteria are used on what occasion, and how or

whether citizens choose between these criteria. Furthermore, we are interested in the reasons

for choosing criteria: does it depend on structural factors, cultural factors, personal

characteristics or contingency?

The state of the society could play a role. Tönnies made the difference between Gemeinschaft

and Gesellschaft (community and association) (Toennies 1979). In Gemeinschaft, identity and

emotion-related evaluations would play a larger role (as it would perhaps be the case when

there is a charismatic (or even traditional) leadership in Weberian sense (Weber 1964)). In

Gesellschaft on the other hand, modernisation and differentiation processes would lead to the

introduction of instrumental evaluations. Harisalo and Miettinen made a similar difference

between the ‘communitarian system’ and the ‘business village’, where in the former trust is

often blind faith, while in the latter it is a result of participation and experiences. They added

that government has evolved from a communitarian system to a business village model

(Harisalo and Miettinen 1997). Hibbing and Theiss-Morse stress the importance of feelings

and emotions in political attitudes (Hibbing and Theiss-Morse 1995), and asked their

respondents how angry/afraid/disgusted/uneasy/proud they were vis-à-vis a number of

political actors. The decline of traditional society and of (national, not regional) nationalism,



and the disappearance of dominant ideologies and clear enemies (e.g. capitalism vs.

communism), has opened the way for performance evaluations of government. At the same

time, most of the traditional identity-related factors mentioned above have made way for

factors related to the loss of identity.

Because of the focus of the research project “citizen directed governance”, we will now direct

our attention to the reasons why citizens would use performance-related criteria when judging

government. The micro performance theory implies that citizens have an interest in efficiency

and quality service delivery in general, and that this is their prime criterion for judging

government. Citizens whose prime identification is that of a user of government services,

demand performance and will judge government on its services. Others however will perhaps

use other criteria, depending on whether citizens expect performance from government. The

increased attention for performance in government institutions is rather recent, and perhaps

citizens do not see government as something that has to perform in the first place.

3.1 CITIZEN ROLES: CITIZENS BECOME CLIENTS

The New Public Management literature started to approach users of public services as clients

(Pollitt 1993), and public sector reform initiatives favour business-like government, including

the treatment of users as clients (Moe 1994; Osborne and Gaebler 1992). Epstein, Wray et al.

distinguish 6 roles citizens can take (Epstein, Wray, Marshall, and Grifel 2000):

• customers

• owners/shareholders: is the job getting done (value for money, problem solved, in an

ethical way)

• issue framers: not only concerned about services and performance, but also about

values and aspirations

• co-producers

• service quality evaluators: e.g. as a participant in a client survey. It concerns a much

more active and result-oriented role as compared to the customer-role

41



• independent outcome trackers: e.g. grassroots movements concerned about quality of

life, sustainable development, …

Because of the fact that service delivery probably takes a more important place in the relations

between government and citizen than for instance voting, it can be expected

“…that a good service relationship -in other words, a good experience of services as

they are planned, designed, delivered and monitored- is a foundation upon which

new ‘citizen’ roles can be built” (Corrigan 1996).

Despite all the attention for participation, stakeholdership, business-like government etc.,

“regardless of class or ideology, most [British] people do not like to be called

‘customer’ or, even less, ‘stakeholder’. Both are words invented by politicians with

special meaning to support campaigns or reform and change; neither seem to have

resonance with the public. Instead, they prefer terms as ‘citizen’ or ‘member of the

public’” (Public Management Foundation 1996).

Box describes the constant tension between individualism and collectivism in American

citizenship, and observes a movement away from the ‘citizen as customer’ terminology:

“The idea is for citizens to become part of their community’s governance, taking

responsibility for their local governments rather then treating them as something

separate, either customers to be “served” or antagonists to be opposed” (Box 1998).

The new public service approach has extensive criticism on the reduction of citizens to clients

(Denhardt and Denhardt 2000; Chapin and Denhardt 1995). According to Gaster and Deakin,

one can, roughly speaking, consider oneself as a citizen or as a customer in one’s relation

towards public services. Customers are then described as active users (willing or not), past

users, and those wishing or needing to use but who are excluded. Citizens on the other hand

have an interest in service quality because they are member of a broader community. This

citizenship is inclusive, since citizens also have an opinion on the services they do not use

themselves. Most quality management programmes take a top-down approach and only speak

about customers (technical elements of quality), while neglecting democratic elements



(subjective factors, involvement) (Gaster and Deakin 1998). Deakin himself is even more

critical about the client-supplier relationship:

“The weakness of consumer-based models of action, like the (so-called) Citizen’s

Charter, is that they exclude in another way, by admitting only half the concerns of

citizens as legitimate (those affecting them in their capacity as individual users of

services) but do not recognise the accountability of service providers to the

community at large” (Deakin 1994).

Using the term client obscures the fact that governments have multiple customers with

competing perspectives (Pegnato 1997). Citizens have substantial rights, including the right to

speak about service quality in general. Consumers have procedural rights, and are entitled the

right to be heard by the administration, to receive redress etc., be it only in specific, concrete

cases.

3.2 CITIZEN ROLES AND CHOICE OF EVALUATION CRITERIA

The previous paragraph has shown that citizens can take up different roles. In this paragraph,

we will now address the question whether the roles that citizens take (client, citizen, …) have

an influence on the criteria they use for judging government or government performance3.

Ringeling distinguishes 4 criteria people can use for judging government (Ringeling 1993):

• Instrumental: effectiveness, efficiency

• Bureaucratic: legality, justice, possibilities for discretion in policy

• Contingency: representativity, receptiveness

• Symbols/values: political order, distribution of values

This indicates there are criteria other than performance related ones that can be used for

judging governments. These criteria often contradict each other: more efficiency can mean

less representativity and vice versa. The criteria used by citizens to judge government can also

change over time: the attention for performance is rather recent, while perhaps an anti-

3 The distinction between citizens and clients, and between the criteria they use, could perhaps also be useful for researching the NIMBY-syndrome: people favour better public transport, but do not want a railway in their backyard.

43



communist attitude or constituency service and case work were indicators for success in the

past. European research partly illustrates this point: when asked what are the principal

characteristics of public services, we can notice at least some differentiation among European

countries (Direction Générale "Information, communication, culture, audiovisuel" 1997)4. It is

difficult to find objective criteria for assessing performance, since different dimensions and

values are involved in government performance. We therefore can choose between a

multitude of assessment criteria, such as efficiency, effectiveness, equity, responsiveness,…

The assessment of performance is always value-laden, which for example explains the

importance given to efficiency since the start of the reinvention movement in the USA (Glaser

and Denhardt 1997). Efficiency cannot be the only criterion, since a government that is too

efficient can become too powerful (Pegnato 1997). For Klingemann, the evaluation of

government performance is an instrumental attitude that should be distinguished from moral

and expressive attitudes that also determine the relation with government (Klingemann 1999).

If the performance criterion is judged as an important one by citizens, they may support the

current structure of the administrative state, but cynically withhold trust as long as

(government) goods and services reach them in a sufficient way (Mitchell and Scott 1987). A

different opinion is that trust is largely independent from the advantages people obtain from

particular policy outputs, but that it stems from a favourable perception of the conduct of

political affairs in general (Gabriel 1998; Easton 1965). This means that citizens make their

evaluations as citizens and not as customer (customers of goods and services, but also

customers of policy and politics).

The existence of cross-national differences in the criteria used for judging government actions

is demonstrated by Bouckaert’s comparison of public service and citizen charters: In the

British Charter, the relation between citizen and state is similar to that between client and

service deliverer, and is thus entirely rooted in market ideology. The French and Belgian

Charters on the other hand give more attention to values as neutrality and legitimacy, and treat

the common good as a basic condition in order to start speaking about client-supplier

relations. Basic rights are a central concept in the French Charter, while the Belgian one also

speaks about basic duties (Bouckaert 1995a).

4 The characteristics among which the respondents could choose were: accessible for all, provided by the state, works for the general interest, not intended to make profit.



If governments want to restore trust, they can improve performance, but this implies the

performance is also perceived as a success by the citizens. An alternative strategy could

therefore be to try to manipulate the criteria for success (Citrin and Green 1996). The way

government sees its citizens acts as a mirror for these citizens, and thereby gives them a

certain role (Beck Jorgensen 1994). The conception of service charters can for instance be an

attempt to promote performance as a ground for judgement, possible to the detriment of other

criteria. Deakin and Wright state that,

“in its preoccupation with the economy and efficiency of public services, the Right

has altogether evaded the further and fundamental question of the effectiveness of

such services.” (Deakin and Wright 1990)

A number of authors did research on what criteria citizens used to judge government, and

whether this has an influence on the perception and evaluation of performance. Rothstein

asked respondents about their support for a number of welfare services. Two institutions

seemed to enjoy less confidence: those providing housing allowances and social assistance.

The reason he gives is that these are the two most selective programmes, since decisions are

made on a case basis, with a considerable degree of bureaucratic discretion, and a focus on

individual clients. The other welfare services were less selective and were seen as working for

the community as a whole. These findings somewhat undermine the hypothesis that modern

citizens see themselves as mere clients (Rothstein 2001).

In his research on local governance, Rose categorised Norwegian respondents into three

categories: citizen/voters, taxpayers and consumers, and found that, depending on the

category they were placed in, these people used other criteria to judge: citizen/voters

emphasise democracy to a greater degree, while taxpayers for instance emphasised efficiency

and (local) autonomy. Citizens with a taxpayer orientation seemed to have less confidence in

politicians (Rose 1999; Rose and Pettersen 2000). This could lead to the conclusion that

efficiency considerations only influence confidence when efficiency is seen as an important

feature of government.

The fact whether or not bad performance has an influence on trust therefore depends on

whether performance is used as a main criterion for judgement. The same goes for other

criteria: people who feel alienated, powerless and indifferent to politics, can still be trusting,

since the relation between alienation and distrust depends on whether political efficacy and

45



participation is important to the person concerned, and whether he/she actually blames

political objects for this alienation. Glaser and Hildreth found for instance that citizens did not

notice productivity improvements in public services, but instead asked for more

responsiveness (Glaser and Hildreth 1999).

We observed that citizen roles and the related criteria used for judgement do play an

important role. Therefore, it could be important to know what determines why a certain

criterion is used, and by whom. First, the time factor is important: Recently, both efficiency

and participation became more important. Second, cultural and national differences determine

the choice of criteria: If citizens see good performance as taken for granted or as totally

lacking, perhaps performance does no longer serve as a criterion. In this case, only

experiences totally opposed to expectations will influence the attitude towards government.

Third, the context in which citizens are asked to give an evaluation may influence the choice

of criteria. When e.g. asked about trust in government in the context of a negatively evaluated

service encounter, performance-based evaluations of government may become more

important. Fourth, socio-demographic factors might have an influence. According to Rose,

consumers are young, highly educated, female, have a high income, vote left or centre and

work in the public sector. Taxpayers are older, less educated, vote right and work in private

sector. Citizen/voters do not have clearly distinguishable characteristics, but tend to live in

small villages and vote right (Rose 1999).



CHAPTER 4 The Causes of distrust I: the performance theory

Distrust is often related to the bad performance of government (Bok 2001). This partially

explains the attention given to quality and reform initiatives in government5. The relation

between performance and trust has however not been tested as extensively compared to other

hypotheses such as the influence of party preference or the media. The performance approach

to trust in government has two main parts: Macro-performance theory explains variations in

trust across countries and over time as due to variations in unemployment rates, economic

growth, inflation, the stability of governments etc. (Brown and Coulter 1983; Listhaug and

Wiberg 1995; Newton and Norris 1999; Kuechler 1991; Miller and Listhaug 1999; Anderson

1995; Kornberg and Clarke 1994; Zussman 1997). Micro-performance theory relates

variations in trust to changes in (the quality of, or the perception of) government service

delivery (Norén 1997; Norén 2000; Rose and Pettersen 2000; Hoogland DeHoog et al. 1990;

Glaser and Hildreth 1999; Kobi 1998).

The key to investigating satisfaction with performance (that in turn leads to trust) is the

disparity between expectations and performance, or the perception of it. Anderson describes

four different theories dealing with this disparity (Anderson 1973).

• Cognitive dissonance (assimilation): discrepancies between expectations and

performance will be minimised (assimilated) by adjusting perceptions or expectations.

• Contrast theory: This is the converse of assimilation: if performance does not match

expectations, the evaluation will be less favourable than if there were no expectations

at all.

• Generalised negativity: Any discrepancy between reality and expectations, even if

small, will lead to a generalised negative state.

• Assimilation-contrast: depending on the extent of the discrepancy, this discrepancy

will either be magnified or be minimised. One does no longer accept a disparity

between expectations and performance/perception above a certain threshold. This

threshold depends on the commitment to a product, or, in our case, the affiliation with

the mission of an agency or the reservoir of diffuse support.

47



In this chapter, we will deal with the so-called micro- and macro-performance hypotheses. In

order to be able to determine the relation between performance and trust, it is necessary to

know what citizens consider as government tasks before we can speak in a meaningful way

about these issues. Secondly, it is important that performance perceptions somehow

correspond to reality. If not, it becomes very difficult to stick to the performance hypothesis.

Even though the performance hypothesis mainly deals with outputs and outcomes, processes

often influence perceptions and evaluations. The micro-performance hypothesis can be

approached from a political and administrative side (service delivery). The macro-

performance hypothesis presupposes a relation between macro-indicators (mainly macro-

economic ones) and political trust.

4.1 THE TASKS OF GOVERNMENT

Citizens will take a negative attitude towards government if government does not perform its

tasks in a satisfactory way. But what if government performs a task perfectly, while citizens

do not agree this is a task for government? Moreover, will not performing a certain task have

the same influence on trust as performing it badly? A related claim is not that government

performs badly, but that government has become too big. Here the rise of distrust should be

related to the growth of government. But despite complaints about big government, very few

citizens mention concrete areas where government could withdraw (Privy Council Office and

CCMD 1996). There is a relation between quality of government and economic performance.

The best performing governments are larger and collect higher taxes, since creating a high

quality government costs money, and therefore big government should not be equated to bad

government (Rothstein 2001). Whether government is big or small, less and less citizens

believe in the problem solving capacity of government (Kobi 1998).

The new politics and post materialism hypothesis explains distrust by referring to unfulfilled

new demands that are made towards government. Post materialists do not want less

government, but want government to devote attention to other priorities (Dahrendorf 1994;

Borre 2000). It is not a matter of doing things well, but of doing the good things. The

evaluation of government services seems to be related to the fact whether one thinks a certain

5 E.g. Copernicus in Belgium, reinventing government in the USA, la Relève in Canada, …



service is a task of government (Princeton Survey Research Associates and Pew Charitable

Trust 2000). Uslaner finds a relation between the state of the economy and trust in

government, but only when respondents thought government could actually exert influence on

the state of the economy (Uslaner 1999). If citizens do not hold government responsible for

something, it is unlikely this will influence evaluations. However, we do see that citizens hold

government responsible for quality of life and numerous economic and social concerns

(Glaser and Denhardt 1997). Lipset and Schneider found a relation between budget deficit and

confidence, and found that citizens relate budget deficit to political performance, and do not

consider it as an effect of the general economic conditions (Lipset and Schneider 1983).

Establishing the relation between performance and trust thus becomes difficult because

citizens have other opinions on whether government should do a certain task, and whether

they think government can have an impact on certain things (e.g. the economy). Furthermore,

there are also real differences: in some countries health care is a government task and in

others not, certain tasks are more critical than others, …(Miller and Listhaug 1999). All this

makes comparison very difficult.

4.2 THE CORRECTNESS OF PERCEPTIONS OF PERFORMANCE

Despite a trend of increasing government performance and quality initiatives in government

services, support is declining (Loveless 1997). There are however also critics who have

doubts about this performance (Bok 1997). This shows that performance as such is

insufficient to explain low levels of trust. To determine whether there is a relation between

performance and trust, it is important to know how citizens perceive performance. If

perception does not correspond to the actual situation, an improvement of performance will

not influence citizens’ attitudes. The same holds when government is not aware of the criteria

citizens use for evaluating performance.

If performance is, according to a number of objective criteria, good or increasing, and

citizens’ evaluations remain bad, one often attributes this to the lack of information.

Performance should not only be good, it should also be seen to be good. The best way to

make prejudices disappear is to confront them with reality. Confronting people with the

inaccuracy of their views creates an internal conflict, and could lead to the disappearance of

the prejudice (Allport 1958). However, more information will not necessarily help:

perceptions are always selective. The primordial attitude towards government could influence

49



perceptions: sceptics will fail to see positive aspects of government, even if these abound, and

instead focus on the few negative points. In this case, additional information could just be

irrelevant for people’s attitudes:

“The public does not realize how many staffers members [of Congress] have, how

much money members make, or how long members have served, yet it still wants to

cut staffs, salaries, and careers.” (Hibbing and Theiss-Morse 1995).

A place where information can have a beneficial influence is in the reduction of errors of

attribution. In a number of occasions, the public is not aware whether government provides a

service or not. In the worst case, they see good performing agencies as private and bad

performing ones as public. Research has shown that on sites where many errors of attribution

occur, satisfaction with services is high (Hoogland DeHoog et al. 1990). This shows that

when measuring satisfaction, and consequently trust, it is necessary to know what is

considered as government.

Making the public judge government performance in a positive way is difficult: it often only

becomes aware of the existence of a certain service when it fails (Berman 1997). The reaction

of citizens on bad performance is not necessarily negative: first, it depends on expectations.

Second, it depends on whether the citizen attributes the cause of the problem to the service

provider: If however no explanation is given for bad performance, the user will attribute the

error to the service provider and not to an external cause. When a solution for the bad

performance is proposed, this reinforces the feeling of the user that the service provider is

responsible. Finally, there will be less dissatisfaction when the user judges the mistake as a

coincidence. The bad reputation of government therefore leads to dissatisfaction in case of

mistakes since users seem to consider the occurrence of a mistake as normal (Bitner 1990).

4.3 PROCESS- VS. OUTPUT EXPECTATIONS

Performance can or cannot be a criterion to judge government. Within performance however,

one can still choose between a number of criteria. The main choice is between output and

process. Is one satisfied with the outcome, or does the process followed to come to that

outcome also matter? Are the expectations one has about output (i.e. I got what I wanted at

the town hall), or about process (friendliness, responsiveness…)? Easton clearly focuses on



outputs, while the conversion (process) is not dealt with in a detailed way. However, Easton

also states that trust does not necessarily result from clear outcomes, but that it can also be a

result of symbolic satisfaction with the processes that generate these outcomes (Easton 1965;

Easton 1975).

Rothstein also focuses on output institutions, but then as opposed to input institutions (votes,

politicians), since citizens have a more frequent contact with output institutions (Rothstein

2001). Hibbing and Theiss-Morse find that dissatisfaction toward government is due to

policy, but toward Congress, the dissatisfaction is due to process. Some disliked processes

(e.g. bargaining, compromising) are however inhere to politics (Hibbing and Theiss-Morse

1995). While Weberian bureaucracy gave a lot of attention to processes and procedures, New

Public Management reintroduced the importance of outputs. A British survey report stated,

“…public management must focus on outcome. The public does not give a damn

about management per se” (Public Management Foundation 1996).

Glaser and Denhardt come to a different conclusion, and find the process-variable

responsiveness to be the most important one:

“There is, for example, an assumption that making a government that works better

and costs less (that is a government that is more efficient) will increase trust in

government […] issues of equity and responsiveness are far more critical in affecting

the level of trust in government” (Glaser and Denhardt 1997).

Recent developments again acknowledge the importance of processes, which corresponds to

the re-integration of customer and citizen roles. Miller and Listhaug give the example of

injustice: citizens can evaluate government performance in this field by judging the absolute

output of injustice (e.g. weighing what the average citizen receives against ethical principles),

or by judging the criteria used for coming to decisions, this is the perception of procedural

injustice. They suppose the latter has a more profound impact on trust. Even if the output is

clearly sub-optimal, it does not necessarily influence opinion as long as government follows

its own rules and procedures (Miller and Listhaug 1999).

Due to the nature of government performance, output and process are often intertwined.

Services are produced and consumed at the same time, making a distinction between output

51



and process aspects impossible in practice. Therefore, we will use the term performance as

encompassing both process and output.

4.4 MICRO-PERFORMANCE

In this chapter, we will deal with the micro performance hypothesis. We distinguish two parts:

performance of politicians and performance of government agencies. The first part deals with

how people judge politicians, whether they keep promises and are judged as reliable and

honest. The second part with how government agencies, service providers, government

administration etc. work.

Political micro-performance

In this paragraph, we will deal with political micro-performance: how do citizens judge the

performance of politicians and the political decision-making system? There are similarities

between the aspects dealt with here and other paragraphs, so we will limit ourselves to a brief

description of the main issues.

Because of the importance given to the aspect of certainty in creating trust, keeping election

promises is an important aspect in the evaluation of politicians. Election promises create

expectations, and may as such contribute to government overload.

Since citizens entrust politicians with a task, failure to perform these task could be regarded

as a breach of trust. Therefore it can be expected that a premature end of government

coalitions and parliaments legislatures will have an influence on trust. Listhaug and Wiberg

related confidence in government and parliament to the number of governments in the

previous decade, the duration of a single government and the number of parties in government

as indicators of government stability. The only conclusion they could make was that these

performance indicators might have a negative impact on confidence, and that they could not

give any definitive answers (Listhaug and Wiberg 1995). At first sight, there seems to be a

relation between satisfaction with democracy and the stability of governments in Europe. If

however Italy, which had frequent government changes, is left out of the analysis, this

relation largely disappears (Huseby 2000).



An important process evaluation is the inclination to take decisions in dark backrooms with a

limited number of actors. Evaluations of political performance can also be based on personal

characteristics of politicians. In 5.4 we will explain that similarity leads to trust. The

relationship between sympathy for political leaders and perceptions of politicians in general

seems to be weak but significant (Huseby 2000).

A decline of trust very often gets associated with political scandals, such as corruption.

Political scandals certainly influence levels of trust in the short term, but there is no evidence

that they also have influence in the long term. If however there has been continued discontent

with perceived performance, thus influencing Eastonian specific support, a scandal may lead

to it that this decreased specific support gets transferred to diffuse support, thus also

influencing system loyalties (Easton 1975). Using Eurobarometer data, Huseby for instance

found that ‘satisfaction with democracy’ is highly unstable, suggesting that these opinions are

based on short time standards of evaluation (Huseby 2000). Therefore, it is difficult to

distinguish between fluctuations caused by scandals, and other fluctuations. Dierickx,

Gijselinckx et. al. wonder whether we should look at scandals as the cause of political

alienation. Perhaps, they claim, scandals and mediocrity of politicians are just symptoms of

underlying evolutions, or perhaps the alienated public has become more eager for scandals

(Dierickx et al. 1996).

Perceptions and evaluations of political micro-performance are not necessarily objective,

since they are very often influenced by ideological factors. There seems to be unwillingness

to rate political performance favourably when the ideological orientation of the party in power

and the respondent differs. We will deal with this aspect in detail in 0.

Bureaucratic and administrative micro-performance

The impact of bureaucratic performance on trust in government depends on a number of

factors: actual performance should be the most important factor according to the micro-

performance theory. Other factors could be the size of the bureaucracy: if the administration is

small, it will perhaps have less impact on citizens’ views of government. Another factor are

the stereotypes surrounding bureaucracy, further enforced by bureaucratic bashing: the

existence of stereotypes does not promote genuine perceptions and evaluations of

performance.

53



The assessment of government performance does not only depend on mere performance

variables. This has implications for the availability of possibilities for improvements. On

some factors, government has an influence, on others not. Controllable variables include

information flow, media coverage, co production, government contacting,… Non-

controllables include the state of the economy (this is especially true for local government,

less for central government), quality of life, social issues, etc. Both types seem to have similar

abilities for prediction (Glaser and Denhardt 1997).

Bureaucratic satisfaction is related to government satisfaction in the abstract. Citizens’

opinion is much worse when asked about bureaucracy as compared to asking about specific

performance (Serra 1995). Respondents who have not contacted government over the last

three months, those without a computer, fax or modem, with a lower income and less

education, and whose family situation is a household are the least satisfied with government

performance in a number of areas. Satisfaction was highest in the areas of agriculture, safe

transportation system and health care, the lowest with regard to taxation and unemployment

and related areas. The most important shortcomings were dealing with the amount of red tape,

promptness and openness (Privy Council Office and CCMD 1996). Note that other research

finds that people who have less frequent contact with government assess performance better

than those with frequent contacts (Glaser and Denhardt 1997). When surveying for

satisfaction with service delivery, we have to keep in mind that not all citizens have equal

inclination to contact bureaucracy. This depends on the need for services, the awareness (of

their existence), political efficacy, political interest, social involvement and political

participation (Serra 1995).

Bureaucratic bashing

Politicians to garner votes have used criticism about the performance of the government

administration. This in turn has helped citizens to confirm their (negative) opinion (Haque

1998). But when citizens are asked what forms their opinion on confidence, most think about

politicians, and not about civil servants, be it that in general citizens have problems to

distinguish public service and politics (Dinsdale and Marson 1999). Misleading information

on administrative functioning is seen as necessary to get support for reform initiatives (Moe

1994). Complaining about the inefficiency of government can be a strategy: there is

continuity between citizen assessment of government performance and willingness to pay for



performance. Some authors claim that the main reason why people demand better

performance is to allow for lower taxes (Glaser and Hildreth 1999).

Public-private stereotypes

Evaluation of the public service and government agencies is not only based on performance.

Stereotypical images of civil servants perhaps play an even more important role, as does the

artificial opposition between public and private sector. According to Deakin and Wright, all

what is considered as public sector, is considered bad

”Instead of a drive for better public services, the Right has launched a drive for fewer

public services. In justification, it has simply relied on an Orwellian incantation of

“public bad, private good”. Instead of seeking ways to extend choice, improve

quality, and strengthen accountability within public services for all, it has preferred

to find ways to give more choice to some and even (as with the opt-out provisions

for schools) more opportunities for some to exit altogether” (Deakin and Wright

1990).

The tendency to privatise gains and socialise loss has perhaps contributed to this. It is

therefore crucial to know what the citizen considers as a public service. Dinsdale and Marson

advise to make within public services a distinction between agencies providing a service for

individual clients and those benefiting citizens in general (e.g. policy, building of dikes, food

safety…) (Dinsdale and Marson 1999). When citizens are asked to compare a number of

public and private bodies on their service delivery, answers were more constant in public

sector, while there were lots of ups and downs in the evaluations of private sector service

delivery (Privy Council Office and CCMD 1996).

Research by CCMD on service delivery found that almost three quarters of Canadian civil

servants thinks the population stereotypes civil servants as lazy and uncaring (Privy Council

Office and CCMD 1996). This is especially the case for front-line staff, civil servants with a

lower education and morale, and those working in a department using a traditional service

mechanism. Focus groups however revealed that even though it is thought the public holds

these stereotypes in general, individual clients were not seen as having stereotypes. Very few

civil servants believed bureaucracy is actually diminishing.

55



Indicators for Micro-performance

For judging performance, one uses a number of criteria. In the case of administrative micro-

performance, this criterion could be quality, but other criteria are available as well, such as the

existence of a particular service, the level of service etc. It is however important to know

whether citizens regard quality services as normal. If citizens expect quality, only the lack of

it will lead to a negative evaluation, and the presence of it will not necessarily lead to a

positive evaluation. Users and providers use different criteria for judging quality: while users

are concerned with freedom, health, security etc, service providers are taking a rational and

economic reasoning, which leads to a disconnection in values (Privy Council Office and

CCMD 1996).

In this paragraph on indicators for micro-performance, we will base ourselves on research

about client satisfaction measurement. Quality and levels of performance is not sufficient to

explain satisfaction. Instead, a large number of variables are involved, and we will try to bring

these into a model on quality-satisfaction and trust (Bouckaert, Kampen, Maddens, and Van

de Walle 2001).

Exit-voice-loyalty

A theory often used in satisfaction research is the exit-voice-loyalty theory (Hirschman 1970).

A dissatisfied client has two options: complaining (voice), or refusing to use the service or

product again (exit). When dealing with government services, the exit-option is often not

available, what implies that loyalty in the public sector has another meaning than in private

sector. In the private sector, exit is the best way for measuring client satisfaction: users simply

go to the competitors. In the public sector, this is impossible in most cases, be it that there

exists a limited possibility to express one’s displeasure (e.g. by moving to another

municipality when the present one does not perform sufficiently – voting with your feet

(Tiebout 1956), or by starting a company in a tax paradise). Exit is thus a hypothetical

concept. Therefore, a loyal user does not necessarily have trust in the service or government

concerned. Voice is an important option with regard to public services. A number of

possibilities exist to voice one’s (dis)satisfaction with service delivery: complaints, press,

ombudsman, recommending services, answers in satisfaction and other surveys, etc. This

voice should then be used as input for quality improvement initiatives. Whether one chooses

for exit or voice depends on the available possibilities, and on the loyalty one has vis-à-vis the



agency. The more loyal, the smaller the chance to use exit. It is not only important whether

one uses voice. The perceived effectiveness of voice is even more important. Feedback on

complaints can in its turn induce a (dis)satisfaction.

Perceptions, expectations and output

Satisfaction can be determined by process and output evaluations. Because of their

intangibility, it can be expected that processes will influence satisfaction to a larger extent

than outputs. Expectations on output depend on the knowledge one has about the service. If

the user does not know exactly what he or she wants, this can have both positive and negative

consequences. A person who is aware of the tasks and functioning of a service will perhaps be

more understanding when something goes wrong. Frequency of use considerations and their

impact on satisfaction are very similar to this: frequent experiences with an agency will

influence expectations. If dissatisfaction however correlates with a limited knowledge about

the service, communication could be the problem.

It may be clear already that satisfaction does not only depend on objective service quality, but

also on how this quality is being perceived by users. The role of expectations should not be

underestimated. A user who expected he would have to wait for 30 minutes, but who can

already leave after 15 minutes will be satisfied, while another user who did not expect he

would have to wait will be displeased with a 15 minutes waiting time. Psychological

marketing research states that dissatisfaction is caused by a discrepancy between expectations

and perception/experience. Summarised (Maister 1985): Satisfaction= perception –

expectation

It may be clear by now that quality management as such will not necessarily increase

satisfaction. Apart from quality management, there is a need for perception and expectation

management (Bouckaert 1995b). Dissatisfaction, is the common opinion, results from a

malfunctioning of the administration, and therefore it is the administration that should

improve its service delivery. Research has shown this is not always the perception of citizens

(Council for Excellence in Government 1999). Asking users about their perception of

responsibility for the functioning of an agency is important, because it gives us information

on images citizens have, and because this can give us information on how to design

improvement strategies and perception management. Expectation can be unrealistic and a

57



‘tyranny of the user’ should be avoided, especially in a government context. When in a

satisfaction survey users are asked to compare the agency with a hypothetical ideal service

provider, this creates expectations as if a perfect service delivery would exist and as if there

are no conflicting quality aspects in this perfect service (is more participation and friendliness

possible without affecting efficiency?). This means that even though expectations are

important, they should not be the single most important factor in satisfaction surveys. Apart

from the client’s expectation-perception gap, a number of other gaps are as important

(Zeithaml, Parasuraman, and Berry 1990).

• Gap 1: Expected service vs. management perceptions of customer expectations

• Gap 2: Service quality specifications vs. management perceptions of customer

expectations

• Gap 3: Service delivery vs. service quality specification

• Gap 4: External communications to customers vs. service delivery

• Gap 5: Expected service vs. perceived service

Process variables

A number of items can practically always be found in satisfaction surveys: reliability,

responsiveness, competence, courtesy, tangibles, credibility, security, access, communication,

understanding the customer (Zeithaml et al. 1990). These items give information on

shortcomings in service and give a better insight into factors determining satisfaction. There

are two ways for measuring satisfaction: direct, by asking “how satisfied are you in general

about…” or indirect, by using some or all of the indicators mentioned above and to combine

these to form a measure of satisfaction in general. Using both measures at the same time is the

best option for a number of reasons. First, because we only suppose the separate items

combined are equal to a general satisfaction question. This clearly reflects a producer view on

satisfaction and quality, since it excludes other factors that could influence satisfaction from

the survey. By comparing the combined separate items with the general satisfaction question,

we are able to determine whether there are other –unknown- influencing factors. If the list

would be exhaustive, these items would explain the total variance of the general trust item.

Secondly, the way satisfaction is measured influences satisfaction. A survey directly related to

the use of the service generally generates a higher satisfaction as compared to a general

survey (Goodsell 1994). This discrepancy is often used by administrations to discredit the



citizens’ ability to give a correct opinion. Some of the difference is explained by the lack of

harmonisation of the questionnaires, and by the fact that the population differs (citizens in

general vs. customers of the specific service).

Mission and nature of services

Identification of the user and characteristics of the agency and customer-agency relations can

influence satisfaction. We will deal here with the extent to which the user identifies with the

mission of the agency involved, directness and frequency of use, visibility, homo- or

heterogeneity. Satisfaction depends on whether one has sympathy for what the agency does

(mission), and whether one thinks what the agency does is good for society. Research by the

Pew Research Center found a relation between satisfaction and whether one could identify

with the mission of the agencies concerned (Princeton Survey Research Associates and Pew

Charitable Trust 2000). Such an item is very similar to the questions on the importance of

service for users (Schmidt and Strickland 1998). On the one hand we would thus have those

who have a firm belief in what the agency does and who have no criticism on the service

delivery, and on the other hand those who think the agency is redundant and useless, who do

not agree with this agency’s policy, and who are critical about every aspect of service

delivery.

Depending on service characteristics, user opinions differ. Services can be categorised

according to frequency of use (frequent/infrequent), home- or heterogeneity, and directness

(direct/indirect) (Das, Das, and Mckenzie 1995; Roth, Bozinoff, and MacIntosh 1990).

Heterogeneous user demands and frequency of use seems to lead to less satisfaction.

Satisfaction surveys on services with which most citizens only have an indirect contact lead to

more convergence (i.e. less variance) in satisfaction ratings across users. A meta-analysis of

customer surveys in the USA for instance showed that fire services always get a better grade

than road repair services (Miller and Miller 1991). This means that a fire services which,

according to objective performance measures, performs bad will still receive a higher

subjective rating than a well performing road repair service.

Directness and frequency of contact allow distinguishing between high and low experience

services (Das et al. 1995), a concept similar to the importance a service has for the user. A

number of factors determine satisfaction: Users who use a service only once, and for who this

fact is only of minor importance will often not bother to give a negative judgement, even if

59



performance is objectively spoken bad (Schmidt and Strickland 1998). If however the user

knows he or she will have to use the service again, he or she will judge this service as more

important, and come to a more outspoken attitude (mostly negative). High experience services

are called high impact agencies in the USA (Weather service, Food Safety Inspection,

Veteran’s Health Care, etc.), since they are supposed to have an important and visible impact

on the lives of citizens. Importance of a service should however not turn into absolute

dependence, something that is important in the discourse on –often social- services. Absolute

dependence also has an impact on the attitude of the service providing professionals towards

users (Soss 1999). External factors such as the social position of the user in society probably

define where to draw a line between importance and dependence (Fitzgerald and Durant

1980). Furthermore, this process is being influenced by the amount of identification and

attachment of the user with the agency. There are however also citizens with limited or absent

contacts. It turns out that non-users often have a more negative opinion, which could be a

consequence of a lack of knowledge of and attachment to the agency and the image the

agency has.

Users of services provided by local authorities will be more satisfied with service delivery

when they display a strong attachment to their local community, and when they value local

autonomy over efficiency. This could of course have to do with the higher visibility of local

service delivery. Visibility however should not necessarily lead to a higher satisfaction.

Services one does hardly know, and whose function is not clear to most citizens, let alone that

they know which level of government is responsible for it, enjoy more favourable views than

visible services (Hoogland DeHoog et al. 1990). What is more, increased visibility implies

bad quality gets noticed easier as well. If and when this plays a role in judging public services

is not entirely clear, because some authors state that government is approached as a

homogeneous entity, making distinctions between services, and their different qualities,

useless (Stipak 1977; Glaser and Denhardt 1997). In this case, the a-priori attitude towards

government comes to take an important place in the formation of opinions of satisfaction.

Trust and satisfaction

The main question in the micro-performance theory is whether service delivery satisfaction

should be seen as an honest evaluation of the service provision, or whether satisfaction is

heavily influenced by one's attitude towards policy and government in general. An example



will make this more clear: is a user of a government agency providing scholarships satisfied

because of a quality service provision process, because he/she got the scholarship that was

asked for, because he/she agrees with the government’s scholarship policy, because he/she

agrees that government gives scholarships at all, or because of the existence of a favourable

stereotypical image of government in general.

Figure 2: Quality perception and improvement

All the factors mentioned in this paragraph can be summarised in a model (Bouckaert et al.

2001). The model is a hypothetical flow-chart where government (right) tries to improve

services based on internal performance data and users’ and citizens’ feedback (centre). We

start in the centre of the model, where a user receives a service. The most important factors at

the client level are perception, the client’s opinion on the service quality, the choice between

exit, voice and loyalty, and the a posteriori trust in this service and government in general. A

number of a priori factors are also present in the model: a priori trust in government, resulting

from -amongst others- the affinity with government policy, and affinity with the mission of

the agency. Perception is furthermore influenced by service characteristics, knowledge about

the agency and process and output expectations. The agency itself receives feedback on its

61



performance by own internal performance measurements and client feedback. This feedback

is then used as input for quality initiatives.

4.5 MACRO-PERFORMANCE

Until recently, research on trust devoted little attention to performance, since it mainly

focused on political culture explanations for (dis)trust. The difference between the two

approaches is that economic performance evaluations are done based on collective instead of

individual judgements. These judgements are retro- instead of prospective. This means

aggregate data are used (Mcallister 1999). The use of aggregate data helps to cope with the

fact that even while citizens are critical towards e.g. social policy, they do not complain about

their own entitlements (Huseby 2000). The macro performance approach states that citizens

have very clear expectations vis-à-vis government: it has to keep inflation and unemployment

down, take care of economic growth and the environment and provide a sufficient amount of

social services. This approach helps explaining short term evolutions in trust, but it is only

when performance remains bad for a longer period, that evaluations will influence trust in a

lasting way. The macro-performance approach does neither allow explaining cross-national

differences. One has to make a difference between the objective performance of government

and the way that this performance is (subjectively) evaluated. In reality, objective and

subjective evaluation match very well.

Better economic conditions lead to support for the incumbent, but this does not lead to more

support for third parties (i.e. those who are not in government) (Hetherington 1999). Others

find that a bad economic situation does not only influence support for the incumbent, but for

politicians in general (Kornberg and Clarke 1992). Within the relation performance-trust,

government performance is also sometimes seen as a dimension of regime legitimacy

(Weatherford 1992). There seems to be a relation between how government manages the

economy and the perception of the functioning of democracy (Anderson and Guillory 1997).

Kornberg and Clarke find that evaluations of the national economy have more influence on

support for the political community than on support for the incumbents, while other research

finds the opposite (Kornberg and Clarke 1992; Clarke, Dutt, and Kornberg 1993; Huseby



2000). Although performance influences attitudes towards regime processes and political

authorities, this does not make people change their mind towards the democratic system:

“Negative evaluations of government performance might lead to a disapproval of the

incumbent government and votes for different parties, however, people do not

necessarily lose confidence in their systems ability to solve these problems.

Moreover, even if citizens lose faith in their political authorities or institutions, they

do not immediately assume an alternative political system will be more efficient or

better capable of solving these problems” (Huseby 2000).

Kornberg and Clarke however find that retrospective judgements about how government

manages the national economy have no significant relation with trust (Kornberg and Clarke

1994). Economic conditions have no impact on US congressional elections, but they do

influence presidential elections, be it that an indicator such as national unemployment appears

to have little effect on the latter (Himmelberg and Wawro 2001).

Importance of issues within the social-economic context

In previous paragraphs, we have said that there is considerable disagreement on whether

government should perform certain tasks. This factor could influence the extent to which

performance determines trust in government. According to Huseby, the performance

hypothesis is only valid when applied to issues on which there is a consensus that government

should perform them, and people should see them as an important and not just as a secondary

task of government (Huseby 2000). As to issues on which there is no consensus, the policy

distance hypothesis fits better to explain trust (policy distance is the perceived distance

between one’s own opinion/ideology, and government policy). Policy distance has not so

much to do with evaluations of outcomes or success of a certain policy, but with the choice of

policy itself.

This of course explains nothing about the fact that people can attach less importance to certain

aspects. For some caring for the environment is more important than a low inflation, for

others it is the other way round. Citrin and Green found that there is a relation between the

rate of employment, macro-economic performance and distrust. However, under the Reagan

63



government, even those whose opinion it was the economic situation had deteriorated started

to have more trust. This was because Reagan combated inflation instead of unemployment

etc., which influenced people’s opinion (Citrin and Green 1996). We have to keep in mind

that evaluations of the macro-economic situation can be influenced by personal characteristics

such as life satisfaction, personal economic situation etc., since these are influenced by the

macro-economic situation. Education could influence the evaluation criteria used: people with

a lower education will be affected more easily by unemployment, and probably this criterion

will outweigh others. People who would like government to spend more on law enforcement,

environment and welfare display less trust. This can be explained by using the policy distance

hypothesis, but also by using performance theory (Borre 1995).

As mentioned in the introductory paragraphs of the performance chapter, it is important to

know what criteria are used for judging and how important certain issues are for citizens.

Ideological orientations could be an important factor here: government performance in basic

social policy will be closer related to assessments of the political system in socialist welfare

regimes, than in liberal or conservative welfare regimes (Huseby 2000). As for micro-

performance, an important aspect is whether the perception of the economic situations

corresponds to reality. In general, subjective evaluations match macro-economic indicators,

even though perception is influenced by a number of factors such as knowledge and

information, the relation to the incumbent and policy distance (Huseby 2000).

There seems to be agreement on the fact that performance evaluations only have a short-term

impact on support and trust.

“It is the short-term assessments of the political system that are most likely to be

influenced by evaluations of economic performance. The more profound support

there is for the principles of the regime and the national community, the less inclined

is that support likely to be influenced by these evaluations. In these regards, the

confidence in the political institutions may constitute a border line for the impact of

government performance evaluation” (Huseby 2000).



Indicators for Macro-performance

As mentioned above, different people can use different criteria for evaluating performance.

However, since research on the relation between macro-performance and trust mainly works

with aggregate data, this hardly creates problems. On a number of indicators, there is some

sort of consensus: unemployment rates, inflation, economic growth etc. Others are open for

debate: health care, environmental policy, elderly care etc. This choice depends on whether

people recognise these as something that is done by government and that government can

influence and whether people accept that it is done by government. Furthermore, citizens need

to have enough information to make a judgement, which is not necessarily the case for

complicated issues such as environmental policy. When doing international comparative

research, a number of additional variables become helpful: the level of civil and political

rights, years of democracy in the country, the relative size of government spending, quality if

life, human development index etc. (Lane and Ersson 1994).The helpfulness of these

indicators (as is in fact also the case for those mentioned earlier) does to a large extent depend

on the possibilities for the population to compare their situation to that in other countries.

As mentioned when dealing with micro-performance, there is a kind of grievance

asymmetry/negative bias: inflation, unemployment etc. do have a negative impact on trust,

but when government scores well on these issues, people do not notice this, and it therefore

does not affect the level of support (Huseby 2000). Visibility is sometimes very important.

The increasing number of elderly people has made government’s role in providing elderly

care more visible. Therefore, elderly care perhaps takes a more important position now than it

did before. We have to make sure not only to look at absolute output data. In the field of

social policy, not how much is being redistributed, but how this is done is important (Huseby

2000). Indeed, Uslaner, using the Gini index and World Value Studies data found that

economic inequality is a strong determinant of trust (Uslaner 2000). Kuechler found a relation

between the state of the economy (unemployment rates and inflation) and satisfaction with

democracy, but he adds that this relation is far from perfect, and that other -political- factors

also have an influence (Kuechler 1991). Holmberg could not give any conclusive evidence

whether the economic situation influenced trust. At the aggregate level, trust declined fastest

in a period before a major economic downturn started. At the individual level, it are people

with a bleak economic outlook who have less confidence (Holmberg 1999). Miller and

Listhaug argue against the use of GDP per capita as an indicator for macro-performance,

65



because they think budget deficit has a more direct link with government performance. In any

case, they also find a strong relation between performance and confidence (Miller and

Listhaug 1999). Himmelberg and Wawro think it is wrong to use national indicators, since

people only know these from newspapers. Instead, they argue, we should look at local

economic conditions, since these are more tangible for people (Himmelberg and Wawro

2001). However, the assessment of the general economic situation shows a higher relation to

satisfaction with democracy as compared to the assessment of the personal economic situation

(Anderson and Guillory 1997).

4.6 EXPECTATIONS AND GOVERNMENT OVERLOAD

A current explanation for decreasing trust is that expectations are rising, making it impossible

for governments to honour these: there is government overload (Crozier, Huntington, and

Watanuki 1975). Eastern Europeans are dissatisfied because their expectations were set too

high in the euphoric period directly following the fall of the Berlin wall (Sztompka 1999).

Belgians thought the legal system would change immediately after the White March, etc…

The conflicting nature of certain popular demands also makes it difficult to make policy that

pleases all. One possible solution could thus be to lower expectations (Miller and Listhaug

1999). A survey in Canada in 1996 revealed that up to 60% of the respondents agreed with the

statement that “We simply can’t afford to maintain our current level of government programs

and services” (Privy Council Office and CCMD 1996). This is closely related to the opinion

on the scope of government: what are the tasks of government? Overload can mean that

governments are not able anymore to fulfil their tasks, but also that people think that

government has become too big, or at least has taken up the wrong tasks.

On the one hand, democratic systems cause government to grow, because of the multitude of

often-conflicting demands. On the other hand, however, democracy allows that people can

live under a condition of stable expectations, which is not the case in less predictable political

systems. What’s more, democratic systems can, without a lot of resistance or overt

propaganda, change expectations (Levi 1996). This is how marketing works in private sector:

producers themselves create most expectations, which makes they can operate within a

framework of very clear expectations.



Citizens want to see their objectives realised, but they often disregard that choosing one

option excludes other options. This implies it is impossible to satisfy all demands. An

alternative for choosing is compromising, but the incapacity to deal with compromises creates

an image of a government that does not dare to choose. Recent government modernisation

initiatives create great expectations. If however these are only rhetoric or born out of the will

for political profiling, the outcomes of modernisation could even be worse than the present

situation. The creation of new expectations has to be accompanied with performance

matching these expectations. The rise of anti-system parties has however made it necessary to

concentrate on a large number of issues, making good performance in all of them impossible.

It could however also be that the problem is to be found with a lack of communication about

performance.

A key institution on perceptions and expectations is the media. Since the gap between politics

and citizens has become larger (or is perceived as such), very few opinions on government

performance can be a result of direct personal experience. One possibility is that media give a

good picture of the real world, the other is that what the media shows has little to do with real

personal experiences (Kaase 1999). In the latter case, the media creates unrealistic

expectations, and by focusing on negative aspects, it distorts perception.

67



CHAPTER 5 The causes of distrust II

So far, this report only dealt with failing government performance as a cause of distrust. Apart

from performance, identity-related aspects have an influence on the level of trust. This means

we have to broaden our scope and devote more attention to societal factors.

5.1 RISK, UNCERTAINTY, CONFLICT AND CONSENSUS

New value orientations and postmaterial attitudes

Inglehart is seen as the leading researcher linking personal value orientations to trust. This

link implies that distrust can only partially be blamed on government. There is an evolution in

society towards postmaterial values: individualism, needs for belonging, self-esteem, self-

realisation... These new value orientations create new/other societal demands towards

government (new politics hypothesis) (Inglehart 1990). Responding to a rising individualism

is difficult for governments, but it is possible for governments to focus on the new demands

made by citizens (environmental concerns, migration, north-south relations...).

Postmaterialists in general feel politically more competent, have more interest in politics and

participate more actively in society. While their feelings of political efficacy are stronger, they

still have to function in a dominant materialist environment. The political efficacy hypothesis

thus would lead to the conclusion that postmaterialists should trust government more, while

the performance hypothesis and the fact that there is incongruence between demands towards

government and the supply lead to the conclusion that postmaterialists trust less (Gabriel

1998).

In research on value orientations and political trust and efficacy, Gabriel found

postmaterialism to be an antecedent of civic competence and internal efficacy, while it is

unrelated to trust in government. The study of other value orientations such as religiosity and

libertarianism lead him to conclude that value orientations do not have a uniform impact on

beliefs in government. According to him, materialism and postmaterialism are irrelevant in

the explanation of trust in political leaders (Gabriel 1998).

Postmaterialist orientations however result in another approach towards politics: traditional

tangible outputs of government become less important, unless these are able to create a



feeling of community (e.g. the impact of big public works on the attitude towards

government). Instead, symbolic and ceremonial aspects of politics become important. ‘What

is being done’ is replaced by ‘how it is done’. Ceremony does not only attract citizens with

postmaterialist attitudes. It helps to promote identification with the state for other citizens as

well. Citrin and Green found that one of the reasons why trust in government increased under

president Reagan’s reign in the USA was that Reagan emphasised ceremonial and dignified

aspects of the presidency by stressing the imperial aura of the presidency and by presenting it

as above politics and as presenting the general will (Citrin and Green 1996).

A problem of terminology occurs here: re-symbolisation of politics is referred to as

postmodern. However, this re-symbolisation also gives politics an aura of inassailableness,

while postmodernism often refers to a constant questioning and genuine participation at all

levels. What is meant with re-symbolisation is that certain aspects of parliamentary decision-

making should not be brought into the open (compromises, power-based negotiations), and

that big theoretical debates and final solutions should be kept for the public.

The importance given to performance has reduced politics to just another sector in society.

This implies that citizens are supposed to be able to look at government outputs without

relating their evaluations to ideological orientations and identities. It is not sure whether

citizens are able to do this and to what extent they need these symbols. Clearly, the concept of

re-symbolisation goes against the movement for performance, transparency and

accountability. One of the reasons why we could argue for it will be explained in the

following paragraph: there is an increased awareness that politics is a matter of choice: for

every problem, rational arguments can lead to more than one possible solution, which makes

it impossible for government to make a rational choice between proposed solutions. If

however the appearance of rationality is upheld, it is made easier to challenge decisions by

rational arguments, which immediately makes the ‘arbitrariness’ of the decision visible.

Dealing with a complex society

Giddens states that it are not postmodern, postmaterialist and other new value orientations that

lead to distrust, but instead that modernisation itself is to blame. Modernisation leads people

to believe that all problems can be solved and all risks known and controlled. Now this

modernisation has come to its extreme, leading to an incapability of people to deal with risk

(Giddens 1996). He speaks of a high modernity, were the consequences of modernity are

69



becoming radicalised and universalised. In this approach the problem is that there are not

enough, or even no, postmodern attitudes. A genuine postmodern view of the world implies

an ability to cope with uncertainty and difference and a willingness to accept this constantly

changing environment and hierarchy of norms and values (Pulkkinen 2000).

People cannot deal with complexity and uncertainty. Society in general and government more

specific has become more complicated. In a traditional, i.e. not complex, society, there is no

need for trust since there is knowledge. Increasing complexity creates a demand for trust

(Luhmann 1979), serving as artificial knowledge and created certainty. In this sense trust is a

social construction, which answers to our need for security (Lewis and Weigert 1985). Trust

is tolerance of uncertainty. Complexity makes us hand over tasks to specialists. Whenever we

are no longer able to solve problems, we want to delegate them, e.g. to government, who can

then solve problems in our place (Kobi 1998; Inglehart 1999). Reducing complexity is done

by creating systems and delegating the authority to search for a solution to these systems, of

which government is one of the most important (and NGO’s, religions etc. as well).

In traditional 19th and early 20th century bureaucracy, complexity was successfully reduced by

the concept of government infallibility. This infallibility could not be challenged by admitting

mistakes because this would endanger certainty. If corrections were necessary, the hierarchy

would solve this. Recent attention for responsiveness in government requires discretion from

administrations, and the replacement of hierarchy by other steering mechanisms has taken

away some certainties (Beck Jorgensen 1994). Flexibility and change have replaced

standardised rules and procedures in government, and this could have caused the loss of trust

in government (Sztompka 1996). Efforts to restore trust by only focusing on service delivery

are in fact an effort to reduce complexity: citizens are seen as mere clients. The annihilation of

a citizen role in this perspective shows that the complexity reduction is insufficient, since it

has become clear that this solution is only able to solve a limited number of problems. This

could also explain the impact of media use on trust: the media give an image of society’s

problems as if there are simple solutions, while people observe a government that is not able

to solve these problems. The belief in simple solutions for problems helps to explain the rise

of populist parties.



Trust vs. contracts. Trust and created certainty

Our present society increasingly uses contracts to regulate relations. Interpersonal trust is

being replaced by formal procedures. In this way, making implicit assumptions explicit

creates certainty. Trust, and interpersonal and person-institution attitudes in general, are thus

being reduced to procedural matters. This allows for the public interest to become the sum of

individual interests, and for organisational and procedural interests to become more important

than the content of decisions. This procedural constructed trust makes that citizens and

governments can interact as if there was certainty. However, while the formalisation (and the

resulting equal treatment and certainty) is seen as a solution for the lack of trust, it is also a

symptom of the lack of trust. Contracts, charters etc. make expectations more visible, while

this would not be necessary in real trust-relations. Furthermore, the ever increasing number of

laws, codes of conduct and commission may simply increase the number of possibilities to act

untrustworthy (Newton 1999).

Situations in which all relations are guided by law, as for instance is the case in a so-called

Rechtstaat create certainty, but such a system is very sensitive to situations in which the world

is becoming more and more complex. The need for certainty and stability in order to create

trust seems to conflict with quality measures which instead call for administrative discretion

and flexible responses. In fact, the use of charters and similar instruments to lay down

government-citizen relations could be seen as a symptom of the lack of trust. Charters are a

kind of contracts that are used to show one’s trustworthiness by making the norms and limits

explicit. The fact these things have to be made public means that these things were not so

obvious in the first place. In a real relation of trust, expectations are much more clear. A

continuous pattern of the new form of interactions will slowly shape expectations allowing

the contractual relation to be replaced by one of trust.

As mentioned before, there is a danger that contracts do not longer reflect the real world:

contracts stand for certainty while the real world is being faced with change and complexity.

Where there is a mismatch between the real situation (complexity and ambiguity) and the

created situation (contract, simplicity, certainty), distrust can emerge. Intra-organisational a

similar process has been described with regard to the discrepancy between highly

standardised and formalised control and quality procedures on the one hand and complex and

ambiguous tasks on the other. This shows that formal control mechanisms may foster distrust,

while in fact they were instated as a remedy for the lack of trust. However, most research on

71



trust so far has mainly focused on the advantages of formalisation for creating trust (mainly

game theory, principal-agent theories, transaction cost theory). On the other hand, research on

e-government and privacy stresses the detrimental effects of control on mutual trust.

Apart from pure contracting, another solution which is discussed is the (re)proceduralisation

of politics (De Schutter, Lebessis, and Paterson 2001). Substantive politics is in this way

being replaced by a legally established process of consultation and participation. This could

be seen as an attempt to divert attention from the content of decision to the method of coming

to it. In this way, citizens are no longer faced with the uncertainties and complexities of

policy, which is never absolute, but only with a time- and place-specific reality. Instead,

citizens can rely upon concrete procedures that do not change. Clearly, the debate whether

politics should be about procedure or about content goes back to two very different political-

philosophical traditions, as we have mentioned earlier (1.4).

Game theory and principal-agent approaches

Game theory has dealt with the concept of trust in an extensive way. It uses the concept of

negotiation based on rational interest. It should therefore be seen as a rational choice approach

to trust. Even though game theory is especially able to explain interpersonal trust (encounters

between two persons), trust in government can be analysed as a series of repeated encounters

and past experiences. The game-theoretical approach to trust starts from a two-player basic

trust game, in which there is asymmetric information. A trustee should be seen as trustworthy

by the truster in order to be trusted. This perceived trustworthiness depends on the possible

pay-offs for the trustee, which are not in their totality known to the truster (Bacharach and

Gambetta 1997).

Trust gives actors freedom to act, since it creates stable expectations and reduces uncertainty,

allowing people to calculate expected loss or gain when engaging in an encounter. This

created certainty helps to reduce transaction costs in encounters (Coulson 1998). By using

game theory, the micro-performance theory sounds very credible: quality improvements and

repeated positive encounters will lead to stable expectations, based on experience. The

repeated experience of cooperation creates trust (Crosby et al. 1990).



Both strength and weakness of the approach is that it approaches individuals as rational

actors. It mainly deals with trust as rational behaviour and to a large extent disregard trust as a

normative concept. Game theory seems to exclude norms, moral discourse and emotions from

the discussion. Rational choice theorists are not yet able to bring motives for acting other than

private interest into their models. Game theory denies the emotional nature of trust, and only

deals with cognitive aspects (Lewis and Weigert 1985). As Ruscio states it

“the common understanding of trust suggests a suspension of calculation, rather than

an anticipation of benefits” (Ruscio 1996).

However, Coleman regards the inclination not to break one’s promise as a mere strategic

constraint, and not as a normative one, since breaking promises might endanger future

encounters (Coleman 1990). Anfang and Urban however find a way out of this problem by

stating that once a relation of trust has been established between two actors, we can no longer

speak of a game of simple cooperation, but instead of a normatively defined situation (Anfang

and Urban 1994). When we approach trust from a game-theoretical view, the problem of

generalised interpersonal trust and trust in government, becomes one of collective action

(Rothstein 2001). Axelrod found that in his ‘tit-for-tat’ strategy, everyone started the game by

trusting, and nobody took advantage of the other player unless the other player had done so

first (Axelrod 1984; Coulson 1998). Individuals can easily perform actions destroying trust,

while actions creating trust require cooperation and involvement of free riders. Stopping with

giving bribes becomes very difficult unless everybody does. These collective action problems

become even more severe when we take into account the fact that most government actions do

not benefit citizens as individuals, but instead the public interest. Here social capital, ideas

and identities become important.

Game theory implies that actors in an encounter have freedom to act. However, in real-life

situations there often is a power- and information-disequilibrum. Relation asymmetries

become especially visible in layperson-expert relationships. A patient visiting a doctor clearly

has less information, and therefore depends on the doctor. In such a situation, trust is required.

In a situation of absolute knowledge, trust is not necessary anymore. The existence of expert

systems, the fact that man is not a rational being, and the complexity of politics implies that

trust is somehow necessary. Both knowledge and possibilities for control are different for

principal and agent.

73



A great deal of the research on trust deals with certainty and information. The principal-agent

theory is built around the structural consequences of an information asymmetry in principal-

agent relations (Pratt and Zeckhauser 1991). Whether citizens are satisfied with public service

delivery depends on a complex principal-agent problem: administrations should be responsive

to both politicians (those who give orders) and citizens (clients), while politicians are

responsive to citizens. This can create incompatibilities. The existence of such relation

asymmetry or, in this case, the advantage bureaucrats have because of their expertise, might

hinder the development of collaborative administration-citizen relations.

Risk- and consensus-orientations

Not all people react in a similar way to complexity. To trust one needs knowledge and a

shared understanding that one is relying on the other. Increasing knowledge creates security,

while giving trust, relying on others and creating connections creates harmony. According to

Braithwaite, people as well as institutions lean towards one of both value-orientations

(Braithwaite 1998). Those with an inclination to control outcomes, calculate risk and collect

knowledge have a security orientation. Those who value openness, concern, value

congruence, identification etc. have a harmony orientation. Trust, and therefore co-operation,

can be exchange based (security values) and identity/communal based (harmony values).

Citizen’s trust in an institution is highest when there is congruence in the norms and values

underlying the institution and one’s value orientation. Institutions such as those in the social

sector are based on harmony values, others, such as the police should be seen as security

institutions. Though this approach is different from most research on trust, some problems

remain. Most institutions, and indeed citizens as well, do have a combination of values. The

judicial system for instance could be categorised as a security institution because it is

responsible for law and order and because it punishes. On the other hand, the ultimate

function of a court is creating harmony by trying to re-integrate norm-transgressors in society.

We can see some similarities between this research and the categorisation of trust Lewis and

Weigert made, where two dimensions, rationality and emotionality, are to be found (Lewis

and Weigert 1985)(1.1). Perhaps the personal value orientation has an influence on the

definition of citizen roles (see 3.1). Such an approach sheds doubt on the ability of

governments to remedy distrust, since it could be embedded into people’s value orientation



and distrust could be due to insurmountable value incongruence between (certain) citizens and

built-in characteristics of institutions.

Majoritarian vs. Proportional/consensus systems

According to Rotter, trust among people is low in a competitive environment (Rotter 1967).

There seems to be no reason why we could not generalise this to political trust. Therefore,

institutional structures have the capacity to influence attitudes towards the regime (Norris

1999a). Consensus and majority systems structure democratic institutions in a different way,

and therefore winning and losing have different consequences in both systems. Anderson and

Guillory based themselves on Lijphart’s distinction between consensual and majoritarian

systems. Losers (e.g. those who were not elected, whose issue did not become policy…) in a

consensus system are more satisfied with the way democracy works than those in a

majoritarian system (Anderson and Guillory 1997). Using another dataset however, Norris

concluded the opposite (Norris 1999a). Norris’s conclusions are perhaps supported by

research by Hibbing and Theiss-Morse who state that compromising is not valued by citizens,

since this reminds them that there are no absolute answers and certainties (Hibbing and

Theiss-Morse 1995).

5.2 PERSONAL CHARACTERISTICS AND SOCIO-DEMOGRAPHIC VARIABLES

Most surveys include a number of items dealing with socio-demographic characteristics. With

these items, researchers try to create profiles of (dis)trusting people. In this paragraph, we will

look at a number of these and point at a number of research results.

The older generation, born before or during the war, had other experiences than younger

generations. From the 70’s on, generation X, which seems to have a more negative attitude,

came to take an important position in society, since older people are dying and declining birth

rates limit the number of young people (Brehm and Rahn 1997). Age indeed seems one of the

only demographic variables to correlate with trust in a systematic way.

Socio-economic status seems to have a rather weak and changing influence on trust (Citrin

and Green 1996), and except for income and educational level, most socio-demographic

variables have little explanatory power. The impact of education and income could perhaps be

explained by the fact that better education helps to replace cynicism by political realism.

75



However, there is no reason why political realism would lead to positive evaluations, and not

to negative ones (Rose and Pettersen 2000). Steen found that the higher educated have less

trust in institutions, but that there is no effect of education on trust in the leaders of these

institutions (Steen 1996). This is surprising, since it are those with a lower education that trust

less in most cases. According to Elchardus and Smits, this is one of the challenges of future

research on trust in government: most research provided answers for the low amount of trust

among the lower educated (e.g. less political sophistication, different media use…), but these

theories are not able to provide answers for the increasing distrust among the higher educated

(Elchardus and Smits 2001). In Turkey, those with a higher education seem to have more

trust, while in Lithuania they have less trust (Newton 1999). The relation between education

and trust therefore is not direct and one-dimensional.

Satisfaction with one’s own life is sometimes closely related to SES. If we can find a clear

relation between political distrust and dissatisfaction with one’s own life, income, economic

well-being, personal competence and political efficacy, this might shed new light on the

research. It would mean that the explanation for political distrust has to be looked for beyond

politics. As Brehm and Rahn state it

“Americans transfer their unhappiness about their own lives onto confidence about

federal institutions, after controlling for ideology, presidential popularity, economic

conditions” (Brehm and Rahn 1997).

Newton found that social trust is highest with those having a central position in society, which

makes him to conclude that trust is a prerogative of winners. He however also found that

political trust is randomly distributed in society, and that it mainly correlates with political

variables (Newton 1999). This leads to two observations: first, it is important to distinguish

social and political trust in research, and second, it suggests that social trust might be related

to wealth and economic conditions. This however seems not to be supported by the facts: The

World Values Survey data show a positive correlation between subjective well-being and

economic development, however only up to a certain extent. First, the economic development

has to reach a certain level to have an influence, and second, once GNP/capita rises above a

certain level, the relation with subjective well-being seems to disappear (Inglehart 1999).

Uslaner observed that the relation should not be looked for in absolute numbers of economic

development and personal prosperity, but instead in existing inequalities. The lower the



economic inequality in a country, the higher the level of interpersonal trust. This also seems

to hold for egalitarianism in society in general (Uslaner 2000).

Huyse has drawn attention to environmental factors determining political participation. He

found differences in participation depending on the place of residence of respondents. Every

region seemed to have specific cultural and structural factors determining participation:

Participation could depend on the presence of political personalities in villages, local power

relations, specific political historical factors, etc. Structurally, it means that when a lot of

people are participating in a certain area, the threshold becomes lower for others to participate

as well, while in villages with low participation, it becomes more difficult to do the first step

(Huyse 1969).

Other observations show a strong relation between the strength of religious

ideological/philosophical orientations and trust: deeply religious or secularist/free thinkers

people trust more6 (Elchardus and Smits 2001).

As we can see, research results on the relation between socio-demographic characteristics and

trust are often contradictory. Since distrust seems to originate in all segments of the

population, the recovery of trust will have to happen in exactly the same way (Citrin and

Green 1996). This difficult relation could lead to the conclusion that we should look at other

factors than socio-demographic ones to explain distrust. One solution could be to suppose that

only political factors determine political trust (Norén 2000). Another is to look at cultural or

group characteristics. In any case, hoping to find clear relations between a certain socio-

demographic variable and trust would be too obvious to be true, and therefore one should

always look at these variables in combination with others, e.g. party affiliation, media use

etc… Borre for instance finds that those with distrust are to be found in middle and higher

social classes, but this is explained by policy distance: the measurement was done whit a

social-democratic government in power, and only very few of their voters are to be found in

these social strata (Borre 2000). Newton and Norris even suggest that trust could be a

personality trait on its own. Trust then is an affective orientation, which makes there are

trusters and cynics (Newton and Norris 1999).

6 We refer to the dutch words “levensbeschouwelijke oriëntatie” and “vrijzinnig”, used in their own specific context.

77



5.3 SOCIAL CHARACTERISTICS

Interpersonal trust

A lot of research, especially that on social capital, regards interpersonal trust –trust between

people- as a prerequisite for trust in government. Interpersonal trust is also seen as prior to the

formation of secondary associations. Therefore, we should deal with this concept before

treating other theoretical approaches to trust in government. We refer to the paragraph on

game theory for one possible explanation for the coming into existence of interpersonal trust

(0).

According to Hardin, there is a decrease of interpersonal trust because in the past we knew

less people in the past, which allowed getting to know them better and therefore trust them.

Nowadays, we know many people, all of whom are trusted a bit. At the same time, the

complexity of society made government less able to protect citizens against each other,

leaving less sanctions against poor behaviour (Hardin 1999). Putnam actually found that those

born before 1945 tend to trust their fellow citizens more than those born after 1945 (Putnam

1995). Inglehart however refers to research stating that distrust characterises traditional

societies, and not trust. The reason given for this is that traditional societies can survive if one

only trusts those who one knows personally, while a modern society can only function if one

does not constantly assume that all strangers are enemies (Inglehart 1990).

The literature does not agree on whether there is a relation between interpersonal trust and

political trust. Brehm and Rahn find a relation (Brehm and Rahn 1997), but this relation is

very small (Uslaner 2000; Orren 1997), be it consistently positive (Kaase 1999; Newton and

Norris 1999). Others however deny there is a causal relation (Newton 2000). Newton and

Norris conclude that the relation between social/interpersonal trust and institutional

confidence operates at the societal and not at the individual level (i.e. the relation can only be

found using aggregate data), and that the relation is not mediated by social capital (Newton

and Norris 1999).

Social capital

The paragraph on interpersonal trust is related to this one on social capital since groups

facilitate trusting other people (for instance because of group norms and shared values), and

since experiences in groups and associations learn how to cooperate with and trust other



people. The two approaches are very difficult to distinguish, and this is why a number of

aspects dealt with here apply as well to the paragraph on interpersonal trust.

The social capital theory states that by being involved in all kinds of associations, people

learn to co-operate and to trust each other, because people from different backgrounds are

brought together. This trust is generalised to people outside the association (interpersonal

trust), and, subsequently, trust in government. Trust in government therefore depends on the

level of involvement of people in associations. Berman makes a similar observation. He found

that in small cities, citizens were less cynical, since there was more personal involvement of

citizens. However, large cities with rich inhabitants and a high economic growth also scored

well on cynicism, but this he explained by the possibility of these cities to provide good

services (Berman 1997).

When civic engagement declines, trust in government will decline as well. Brehm and Rahn

explain this phenomenon by referring to the problem of collective action (Brehm and Rahn

1997). Frequent participation and engagement leads to a closer web of social interaction and

trust, making free-riding less probable. If a citizen expects others to cooperate, he/she is more

likely to engage in cooperative endeavours.

Since social capital is more than the sum of individuals, this creates some problems for the

analysis. Bourdieu’s notion of social capital is clearly different from the one used here

(Bourdieu 1979; Decoster 2000). His notion refers to individual social capital, while social

capital in this context here is a characteristic of social systems, not of individuals. Using this

theory, the analysis of trust is best done at the aggregate level. It will therefore be possible to

find a relation between trust in institutions and social trust at the aggregate level, while it will

not always be possible to find this relation at the individual level. The main drawback in the

social capital approach is that it is a normative approach: trust = good; trust is a virtue. The

social capital approach therefore states citizens have to trust government, as opposed to

traditional liberalism which states citizen should distrust government (Hardin 1999).

Putnam worries about the decline of social capital (Putnam 2000), while Zussman states it just

seems as if social capital is declining, since the current participation is different from

participation we knew in the past. This means that new kinds of organisations and

participation, often more informal, have emerged (Zussman 1997). Blokland states that while

the sheer number of group memberships is increasing, these at the same time become more

79



meaningless (Blokland 2001). A number of explanations are given for the decline of social

capital (Brehm and Rahn 1997):

• Decrease of the level of schooling: this does not correspond to reality

• Change in economic resources (since engagement demands a certain investment):

these fluctuate and can thus not create a decrease

• Increased TV consumption and decreased psychological involvement

Newton found that those with social trust are not the same as those with political trust, and

that there are different forms of trust which do not form a single syndrome, e.g. trust in

government, interpersonal trust, trust in family, etc. This is because political socialisation

happens in an indirect way (mainly via the media), while one can learn social trust directly

(own experience). Political trust is an evaluation of the performance of political life, while

social trust should be seen as a personality trait. Therefore, political and social trust is

expressed by different groups for different reasons. He finds further support for his thesis in

the fact that confidence in parliament is in the first place related to political variables: interest

in politics, national pride, incumbency support etc.7 He finds that, compared to other groups

(family, peer groups), people spend relatively little time in organisations, which makes their

influence on trust must be limited, especially when we take into account the fact that people

who spend a lot of time in organisations are not more trusting than those who spend only little

time (Newton 2000). Other authors found as well that social and political trust were not

related (Orren 1997). Almond and Verba found that while social trust is transferred to

political trust in the US and UK, this is not the case in Germany and Italy (Almond and Verba

1965; Parry 1976). In most cases, social and political capital go together, but they do not have

to. According to Billiet, Cambré et. al. it are only political organisations that transfer an

attitude of trust in government to their members. Social organisations in general might a well

socialise their members to become distrustful of government. Involvement in organisations as

such is therefore not enough. It depends on the kind of organisation (Billiet, Cambré, and

Swyngedouw 1997).

7 Though we have to remark here that Newton does not make clear distinctions between civic mindedness,

participation, citizenship, involvement, political interest, concern with public/common good, tolerance, ability to

compromise, confidence in political institutions, etc. He practically treats them as synonyms. Finding a strong

relation between them is therefore no surprise.



Government as facilitator of social trust

In general, the social capital hypothesis states that social capital and interpersonal trust leads

to trust in government. However, as has been mentioned in the previous paragraph, there are

some doubts about this relation, and the monodirectional causality is still no more than a

hypothesis. An alternative approach is not to see social capital/interpersonal trust as leading to

trust in government, but to see a trusted government as a facilitator for social

capital/interpersonal trust. With this we mean that government creates the conditions under

which co-operation with limited risks becomes feasible. Tocqueville did not agree with this

point: the state did not and could not create trust (De Tocqueville 1946). Nowadays, the state

is larger and more powerful, which allows it to set the conditions of modern life, including the

social and political institutions within which trust might be sustained. The absence of so

called ‘thick trust’, based on family and organisational life does no longer take the place it

used to take and has been replaced by a reduction of risk by secondary institutions (Newton

1999).

“Institutions, if appropriately designed, can enable us to trust persons whom we have

never had contact with and with whom we share no relevant communal allegiance”

(Deakin 1994).

Most obvious are the functions of police and justice. In this logic, the decline of interpersonal

trust should not worry us. What is more important is the decline of trust in the so-called order

institutions. Rothstein notices a decline of trust in parliament and government in Sweden, but

does not find this matched by a decline in generalised trust. With regard to trust in the

institutions of law and order, that is, the courts and the police, he does find the strongest

correlations with generalised trust (Rothstein 2001). Similarly, in her research on trust, Norén

treats the police and the judicial system as the core functions of the state, because it is the

belief in the just procedures these provide that explain why people obey the law (Norén

2000). Listhaug and Wiberg also notice a decline in support for ‘order institutions’, but relate

this to a weakening of traditional belief systems (Listhaug and Wiberg 1995). This does not

necessarily mean that countries with efficient judicial systems become more trusting. In any

case, the legal system should be seen as the mechanism that translates trust in government

into interpersonal trust (Uslaner 2000). The state can produce trust in four ways (Levi 1996):

81



• Coercion of those who are not compliant

• Universalistic policies

• Establishment of credible courts & impartial institutions

• Participation

Offe makes the following categorisation of possibilities to create trust, be it by government or

not (Deakin 1994):

Table 5: Possibilities to create trust

Truth Justice

Passive Truth-telling Fairness

Active Promise-keeping Solidarity

Since trusting someone implies a risk on behalf of the truster, the state can help to reduce that

risk, making it easier or safer for people to trust each other. The core issue here is ‘justice’.

This means that police and justice should not only perform well, but should also be seen to do

so. Trust in government does not disappear because of poor performance, but because of a

lack of fairness and impartiality. It is not enough that states treat citizens with concern and

respect. They should treat them with equal concern and respect (Rothstein 2001). Perhaps,

according to Hardin, government today, because of the changed environment, is less able to

protect citizens from each other, which explains declining trust (Hardin 1999).

Does this mean that the absence of (trust in) government in general and its order institutions

more specific would result in a loss of trust among people? Hobbes would speak of a war of

all against all (Hobbes and MacPherson 1978). Rothstein however takes a supply and demand

approach to this question: If the supply of trust diminishes while demand remains constant,

other actors will fill this vacuum. If justice fails to deliver, the demand for a trustworthy

protective institution will not disappear (Rothstein 2001). This has also been shown by



Gambetta in describing the rise and role of the Mafia8 (Gambetta 1998), and the increasing

number of self-help groups and of lawyers who sell ‘artificial trust’ (Putnam 2001).

This discussion can be understood better when seen within the framework of collective action

and social dilemmas. For individual citizens it is quasi impossible to change practices such as

corruption. The optimal solution would be if nobody paid bribes, but as an individual, you do

not have any stimulus to be the first to do so. The same goes for ‘justice’. An individual’s

short-term advantage is very limited. Bringing people together to destroy collective

institutions on the other hand is easier (Rothstein 2001).

5.4 IDENTITY

Identity as a framework of reference

Distrust can be a result of a loss of identity. People lose their traditional frameworks of

reference, which confronts them with complexity. Traditional identity gave people certainty

about their place in society, and this (hierarchical) ordering could already ‘solve’ a number of

problems. There seems to be an increase in alienation. Seeman distinguished 5 dimensions in

alienation: powerlessness, meaninglessness, normlessness, isolation and self-estrangement.

Isolation may lead to assigning low value to goals or beliefs that are typically highly valued in

a society. This can be a source of innovation, but more frequently is regarded as an act of

rebellion. It is becoming increasingly unclear for the individual what he or she ought to

believe (Seeman 1959).

Similarity leads to trust. If people realise that their interests are similar, this makes it safer for

them to trust, as we have explained in the paragraph on game theory. Taylor already noticed

that trust only emerges when the parties involved realise that their fundamental interests are

not necessarily antagonistic (Taylor 1967). Marketing research showed that similarity among

individuals in a relational context influences relationship satisfaction (Crosby et al. 1990). In

research on trust in the institutions, students did not really distinguish between police, courts,

and politics, but students in criminology had more trust in the police, law students more in

courts (Pleysier et al. 2000). Since perceived similarity leads to trust, the absence of

8 Be it that the trust produced by the Mafia is not a kind of trust that increases by use, but instead is exclusive. Trusting someone else outside the Mafia means less trust in the Mafia, as opposed to the social capital

83



similarities may lead to distrust. This perhaps helps to explain the relation between

ethnocentrism and trust: the appearance of immigrants somehow challenges the traditional

identity.

Elchardus and Smits found that religious people were more trusting (Elchardus and Smits

2001). Religiosity is said to strengthen trust in government and civicness. Declining trust in

government has followed the pattern of secularisation in Western Europe (Gabriel 1998). This

can be explained by a feeling of security, and by the integration in a larger collectivity

because of religious feelings. The declining role of religion in society is joined by the

observation of a decline in utopian thinking (Multatuli Expert Seminar 2001), both leading to

a feeling of aimlessness and disconnectedness.

Nationalism, rallying around the flag etc. promote cohesion in society, and thereby strengthen

citizens’ identity. A very clear and simple identity that is not challenged by any dissident

views creates stability and certainty. Although it is not necessary that such an identity is

formed by contrasting it to other identities, some speak about a need to have enemies and

allies (Volkan 1994). The disappearance of communism as the ‘enemy’ in the West could

therefore have contributed to the rise of distrust.

Nationalism and patriotism are often used as indicators for political community support

(Norris 1999b). Patriots, because of their strong affective integration in the political

community, display more support for this political community. This explains the lower

amount of trust in government in cases where nation and state do not coincide (Elchardus and

Smits 2001). Patriotism is not always a good indicator: The Oklahoma bombing in the USA

showed that fervent patriots could have a deep distrust of government. In Germany, rejection

of nationalist feelings was till lately an indicator for system support (Kuechler 1991).

Another aspect we should mention here has to do with big projects initiated by government:

preparing for war, big infrastructure projects (e.g. Tennessee Valley Authority), and other

great events (landing on the moon, coronation...) help to create a feeling of connectedness and

identity, and creates trust. This helps to explain why strong leadership seems to be able to

restore trust. Strong leadership may be a means to create an identity. The leader symbolises

the nation and the population identifies itself with the leader. Here, trust in government is

indirect: there is trust in the leader symbolising the state, while there probably is no direct

relation between citizens’ attitudes and trust in government.

hypothesis that states that trust breeds trust.



The loss of collective identity coincides with an increasing individualisation. Putnam explains

the relation between individualisation and distrust in government as follows: The dominating

framework in modern Western societies is the individual. Government, as a collective

undertaking, therefore does not fit into this framework, making that it is impossible to take a

positive attitude towards government (Putnam 2001).

Somehow, it is easy to see a relation between the aspects referred to here and social capital.

Social capital offers individuals a possibility to have a feeling of belonging. Within the

discussion on identity and trust in government, another factor is important as well: the

influence of political identity, which we will deal with in the following paragraph.

Political ideology

Political ideology is an important factor in explaining trust. In this paragraph, we will not only

deal with mere party affiliation or even liberal/socialist/conservative etc. orientations. We will

also focus on the general attitude towards the core of a state. In the chapter on performance

theory, we mentioned attitudes towards big or small government, opinions on which tasks

government should and should not do etc. Most theorising on trust focuses on incumbents and

regimes, while less is done on the existence of distrust as a result of displeasure with the

political system as a whole.

Kornberg and Clarke did research in Canada and asked citizens about their conceptions of

democracy, and measured the degree of satisfaction with the political system (Kornberg and

Clarke 1994). The views on the most important characteristics of democracy in Canada can be

reduced to four categories:

1. Collective security vs. individual rights (securing lives and property vs. protecting

freedom of expression)

2. Political and economic opportunities for all

3. Elections, capitalism

4. Equality of group influence

They then compared these characteristics with the ‘real’ political organisation in Canada. The

congruence hypothesis then states that there will be more satisfaction with the system if

people’s beliefs are more consistent with reality. This approach can be criticized. First, other

aspects of democracy were emphasised depending on background variables: the lower

85



educated stressed equal opportunities, those with a higher education equality of influence.

Older respondents saw democracy as a system to guarantee security. Second, a difficult phase

in this research design is to determine what the ‘real’ situation actually is, since this always

involves a certain amount of discretion. The closest we can get for Belgium is referring to the

so-called elite consensus (Dewachter 1995).

Fuchs distinguished between a liberal democracy which is procedural and a socialist

democracy which is substantial (Fuchs 1999). Incongruence between one’s own opinion and

the nature of the political system would then lead to dissatisfaction with the political system.

Fuchs tested this in Germany, and found that the two approaches were in fact more or less

mixed in the respondents’ mind. We however have to bear in mind that in West Germany

democracy as a system and democratic performance used to be separated because of historical

factors. Support for democracy as such therefore becomes irrelevant. West Germans seemed

to adhere more to liberal democracy (freedom), while East Germans adhered to a social

democratic system (equality). This means that when gauging for the level of support one

should measure both support for outputs and support for the process. It also means that the

democratic system bears some inhere contradictions. Especially the relation between

democracy and a capitalist system is dubious (Friedrichs 1980). More than dissatisfaction

with government performance and distrust in incumbents, inconsistencies between personal

ideologies and the official underlying principles of the political system are potential threats to

the stability of the political system.

Policy distance hypothesis

An alternative to explanations defining distrust as an attitude undermining the existence of the

state, the political system and democracy itself, is the hypothesis that distrust is a result of

disagreement with government policy. This means that people distrust government because of

ideological reasons. In this paragraph, we will deal with three aspects of this ‘policy distance’

hypothesis. First, there can be policy distance because the party in government is different

from the respondent’s preferred party. Second, the reason could be less ideological and be a

result of disagreement on a number of critical issues. Third, distrust can be a result not of a

perceived distance between one’s own opinion and the parties in power, but of a rejection of

the party system in general.



A first explanation is that distrust emanates from the fact that one’s own party is not part of

government, resulting in a distance between the actual and the preferred policy. This means

that a liberal will distrust a socialist government (and vice versa), while he/she will have trust

in government with a liberal government in power. In the Miller/Citrin debate, it was a central

theme that we tend to trust those who agree with us, i.e. governments whose policy we can

agree with (Miller 1974b; Citrin 1974; Miller 1974a). For instance, those who have voted for

the party of the major in municipalities have more confidence in the municipal council (Rose

and Pettersen 2000).

The policy distance cannot be observed by just looking at the parties in power. One also has

to look at the policy output. Parties in power are often forced to take centre positions and to

compromise, which alienates the extreme elements in the party. This ‘centre policy’ also

explains why the extreme left and right is less trusting. Therefore it is not distrust which leads

one to vote extreme, but the extreme vote which alienates one from government because of

the large policy distance (Miller and Listhaug 1999). This means distrust can also be a

consequence of the feeling one does not have a choice: all parties are perceived as too similar

and as undistinguishable, which results in an increased number of politically homeless voters.

If one is dissatisfied with the policy of the government or one’s own party, and there are no

real alternatives, then there will be cynicism about the whole system, and not just about the

party (Miller and Listhaug 1990).

Such a hypothesis is difficult to test in countries with a proportional system because

governments are composed of different parties, and the multitude of parties in the political

game combined with a high number of unloyal voters makes it difficult to make clear

ideological distinctions between the parties. This explains why the policy distance hypothesis

has mainly been brought forward and tested in the USA, where the existence of only two

parties makes research easier. Another methodological problem is that when one tries to

measure trust and policy distance that respondents project their own preferences on the

politicians they have elected, making it difficult to conclude whether one is agreeing or not

with government policy, since this evaluation is a combination of policy approval,

performance, emotional factors etc. (Citrin and Green 1996). The aspect of policy distance is

important when measuring trust in politicians: asking for trust in politicians in general

disregards this policy distance hypothesis, and therefore this question should distinguish

87



between politicians of the party one is ideologically related to and other politicians (Rose and

Pettersen 2000).

Despite the attractiveness of the policy distance hypothesis, some research has questioned its

relevance. Holmberg for instance notices that trust was declining equally among partisans for

all parties in Sweden, whether in government or not, be it that less people now identify

themselves with a political party (Holmberg 1999). Nevertheless, absolute levels of trust

differ between parties and it could therefore be that “political distrusters are looking for

certain parties” (Norén 2000). Nevertheless, there is a relation between trust and the

ideological distance to the parties in power in periods of elections, since the increased contact

in this period between voters and parties stimulates party identification. The existence of this

relation does not have to contradict Kuechlers’ observation that satisfaction with democracy

does not increase in electoral periods (cycle of attention hypothesis) (Kuechler 1991). The

observation that trust was declining equally for all parties indicates another drawback of the

policy distance hypothesis: Even though it identifies a possible cause for distrust, it is not able

to give explanations for the long term evolutions of trust, since this hypothesis should show

mere fluctuations in trust and in the ideological profile of those who distrust.

An alternative to the home team hypothesis has been brought forward by Holmberg: distrust

can also be a result of discontent with one’s own party’s policy, since it is this party on which

one has conferred trust (Holmberg 1999). This brings us to a related hypothesis. Policy

distance is not so much a result of disagreement with general policy or ideology, but of

disagreement with certain critical issues. Perceived policy distance is not always an

evaluation of the policy as a whole, or of the ideological leanings of the parties in power.

Instead, this evaluation can be a factor of a limited number of critical policy issues that

determine the attitude towards government. Holmberg refers to the Swedish membership of

the EU and immigrant policies as critical issues. With regard to a number of critical issues,

this hypothesis comes close to the performance hypothesis and the importance given to a

number of government tasks (4.1). Whether critical issues alone are a good way to measure

policy distance is open for debate, but it is in this way policy distance is often measured:

respondents are asked to place themselves and a number of governmental actors on a scale

with on both sides policy statements which stand in opposition. The problem with this method

of measurement is that it is not always clear to find policy options that clearly stand in



opposition, and the differing context and urgency of the issues makes comparison difficult.

Borre observed a distance between a cosmopolitan political elite and a

nationalist/isolationalist mass9 in Denmark on a number of critical issues such as EU, NATO,

immigration, asylum seekers etc. This means that on issues having to do with these values,

there will be a considerable amount of disagreement. Even though alienation is lower when

there is more party competition, thus when voters feel they have an influence on policy, this is

secondary to these issue positions. This explains why people can feel alienated within their

own party (Borre 2000).

While the policy distance hypothesis states that people do not trust government because they

do not agree with the government’s policy, or because their preferred political party is not part

of the government, an alternative hypothesis states that distrust is not a consequence of the

content of party policies, but of the mere existence of political parties. Distrusting citizens

consider political parties as redundant and reject the party-system. This could be considered

as a specific case of the policy distance based on critical issues as described above.

It is necessary to distinguish in questionnaires between trust in politicians and trust in political

parties, because these are different concepts (Rose and Pettersen 2000). Most surveys

however do not allow for making this distinction10. Deschouwer concludes that there are no

signs indicating a general distrust in politics, but he finds a crisis of most political parties

instead, especially the traditional ones. This however does not mean the concept of political

party is in crisis, but that there is a crisis of most of the parties separately (Deschouwer 1998).

It has to be said however that people have a rather ambiguous attitude towards political

parties: Rose and Pettersen find that in small municipalities respondents think political parties

are redundant, but think the existing ones function well. In large municipalities on the other

hand, respondents state political parties don’t function well, but they think they are necessary

(Rose and Pettersen 2000). Stokes found political trust to be unrelated to party identification,

but related with the rejection of the party system (Stokes 1962). His indicator of the rejection

of the party system could however be flawed, since it only measured the ability to identify

oneself along the partisan spectrum. This indicator does perhaps do no more than indicating

9 Other research has also shown this distinction between a cosmopolitan elite and a nationalist mass (Elchardus and Smits 1998) 10 In Belgium for instance, there are only questions on ‘trust in political parties’ in the general election survey and the APS survey.

89



value incongruence. Hetherington found that in the American presidential elections the

distrustful voters support candidates from the non-incumbent major party (e.g. if the president

is a Democrat, they will vote Republican). If however there is a third viable candidate, his/her

party gets the votes of the distrustful (Hetherington 1999).

It is a fact that people nowadays feel less attached to (traditional) political parties. This makes

it easier for them to switch votes to other (anti-system) parties. Aberbach gives the example of

the candidacy of Goldwater in the 1964 American elections (Aberbach 1969). Goldwater took

an anti-system attitude, and the votes he received indicated that many voters set aside their

traditional party loyalties for their basic attitude towards the political system. The anti-

(party)system attitude took the upper hand over party-affiliation. This shows that the presence

of an anti-system party gives a possibility for the expression of political distrust.

Miller and Listhaug compared trust in government in Sweden, the USA and Norway (Miller

and Listhaug 1990). They concluded that discontent with political parties is depending on the

fact whether voters feel they have a real choice. That explained the difference between

Sweden and the USA on the one hand and Norway on the other by the fact that the Norwegian

party system allowed for parties to canalise the discontent, and to deal with problems within

the existing decision-making structures. In the USA on the other hand, alternatives were

difficult because of the strong position of the two main parties, and in Sweden, those with low

trust showed a detachment to all parties, because there were too many political quarrels

among the parties. In the United States this was illustrated with the example of Black voters

in the 60’s and 70’s who did not fully agree with the policy of the Democratic party, but for

whom voting Republican was not an option (Miller and Listhaug 1990).

Interest in politics and participation

Trust is a means for reducing risk and uncertainty. One could therefore hypothesise that the

amount of trust in government one has will be influenced by knowledge of politics. In a

situation of absolute knowledge however, trust is not necessary. Yankelovich makes a

difference between public opinion and public judgement. Public opinion is in this opinion

regarded as inferior, since it is not backed by political knowledge. Public judgement on the

other hand means that citizens have decided where they stand cognitively, emotionally and

morally, and that they have reconciled inconsistencies in their thinking, taken all relevant



facts into account and grasped the consequences of the choices they have made (Yankelovich

1991). Quality opinions therefore do not mean that people are informed about everything, but

instead that the opinion has inner coherence and reflects an abstract vision of politics. To

decide on whether opinions are consistent and firm, he uses a ‘mushiness index’, which

depends on

• Personal involvement in the issue

• How knowledgeable one feels in the issue

• Whether or not authority figures have taken a firm stand on the issue

• The firmness of opinions of friends, neighbours, co-workers (e.g. discussed with

others)

• Whether the issue fits in an already established belief system

• Whether or not people have worked through their own ambivalence about the issue

The observation that knowledge helps to reduce risk and uncertainty may help to explain why

participation in itself helps to increase trust: participation involves people in situations they

would otherwise know nothing about. Identification can be stimulated by participation and

involvement. Political trust seems to be related to feelings of efficacy (i.e. feelings whether

one can influence politics). Despite discontent with political processes, there is a strong desire

to participate in politics (Epstein et al. 2000). A possible explanation for the relation between

participation and trust could be that the decision to participate is a personal decision, which

makes people integrate the object of participation into their identity. Therefore, it becomes

very difficult to maintain a negative attitude towards the object of participation. The same

process could explain the fact that people assess their children’s school better than the

schooling system, the Member of Parliament they have elected better than parliament etc. The

first object is always linked very closely to one’s own choice, and a process of cognitive

dissonance prevents them from evaluating it in a negative way. The policy distance

hypothesis would as well fit into this framework. The observation that those with a higher

attachment to local government give better evaluations of local government services also does

this (Rose 1999). In client satisfaction measurement, we find that monopolist companies get

lower ratings than non-monopolists because the use of the latter companies depends on a free

choice (Fornell, Anderson, Cha, and Bryant 1996). Therefore, the monopolistic situation of

many government services alone could explain part of the dissatisfaction.

91



Holmberg found that trust in politicians increases in election periods, possibly because

politicians become more visible and citizens more knowledgeable (Holmberg 1999). Political

participation is often regarded as the opposite of political passivity. Political passivity

however is in many cases a process of cognitive dissonance: people are interested in politics,

but see it as unreachable, which makes them reduce their interest, which influences political

knowledge, making politics de facto unreachable (Hooghe 1999). Estimates of potential

participation are perhaps more important than participation itself. This subjective political

competence is called political efficacy.

Just as trust however, knowledge about and interest in politics is seen as a positive and

necessary value in society. However, we see that some are informed very well about politics

while others are not. Hibbing and Theiss-Morse regard political knowledge as just any other

form of knowledge:

“Some people are highly knowledgeable about politics, knowing facts and having

opinions on a wide variety of political issues, just as some people know everything about

cars, cooking, backpacking or baseball” (Hibbing and Theiss-Morse 1995).



CHAPTER 6 The consequences of distrust

6.1 CAUSALITY

The implicit hypothesis in recent quality initiatives in Western government administrations is

that quality leads to satisfaction and satisfaction to trust. We have shown already that the first

relation (quality-satisfaction) is not so obvious, because of the multitude of other factors

involved in the creation of a (dis)satisfied attitude, and because of the importance of

perceptions and expectations (Bouckaert et al. 2001). The second relation is even more

complicated. Even if we suppose that trust is determined only by performance, it might well

be that a trusting attitude leads to a better perception of service delivery. In most instances,

there are theoretical reasons for indicating the direction of the causality, but in research on

trust, there are often theoretical justifications for both directions (e.g. satisfaction→trust or

trust→satisfaction; social capital→trust or trust→social capital).

Hetherington finds that it is the level of political trust in general that influences trust in the

president, and not so much trust in the president that influences political trust in general,

while the opposite is true for the relation between trust in Congress and political trust in

general. Research on stereotypes has found that feelings about the sum affect feelings about

parts more than feelings about the parts affect the sum (Hetherington 1998). Huseby states

that

“the survey data on the relationship between evaluations of government performance

and political support is incapable of establishing the direction of causality. It is

uncertain whether citizens give negative responses to questions on government

performance because they do not trust the government, or if they loose faith in

government because they evaluate the economic performance as poor” (Huseby

2000).

When doing research, one should thus clearly state whether trust will be dealt with as a

dependent or independent variable, since trust can be both cause and effect (Ruscio 1996).

93



6.2 DOES CONFIDENCE MATTER?

The main question in all research is whether distrust in government does actually matter. This

depends on the different approaches to trust in government. Some say that having too much

trust is dangerous. After all, the system of checks and balances in a democratic system shows

that a democracy is built on distrust. Others (political community approach) state trust is

necessary for the functioning of government and society. Deceasing trust in political

authorities does not necessarily mean support for democracy is decreasing as well. This brings

us back to the discussion on the difference between incumbent and system support. There

however is quite some disagreement on this matter. Some say support for democracy as an

abstract system of governing is declining (Crozier et al. 1975), while others do not find this,

and only see evolutions in support for political authorities and some aspects of the regime

(Klingemann and Fuchs 1995). Dissatisfaction does not necessarily endanger democracy,

since it can be used as a force for reform or improvement of the democratic system.

Approaching the lack of trust as a problem per se implies a conservative and static attitude

towards society since it considers the present situation as the best possible one and neglects

the creative possibilities of distrust. One could therefore say that trust in government is not

really necessary, as long as there is no active distrust (Hardin 1998). Furthermore, when

citizens are asked what they consider as important in their life, politics ranks very low (Kaase

1999). Distrust only becomes important when the distrustful plan to go into politics or to exert

overt behaviour in any other way (Easton 1965).

Instead of looking at the risks of distrust, we could also look at the benefits of trust in

government. Trust enables governments to act without having to resort to coercion or the use

of force for every single decision (Gamson 1968). Moesen found a positive relation between

the level of interpersonal trust and the economic performance of certain regions (Moesen, Van

Puyenbroeck, and Cherchye 1999; Moesen 2000).

We will look into the possible consequences of a lack –or alternatively, presence- of trust in

government. First, we want to know whether distrusting attitudes will also lead to distrusting

behaviour, second we will look at the potential attitudes or behavioural expressions. We will

finish by touching upon the relation between the lack of trust in government and the rise of



extreme-right parties, processes which should, according to politicians and opinion-makers,

be seen as cause and effect.

6.3 OPINION AND BEHAVIOUR

Is distrust symbolic or instrumental?

Both surveys and research have indicated that trust is declining. “So what?”, one could say.

Perhaps distrust is just a state of mind without further consequences. For Easton, political

support can take two forms: overt (actions) and covert (attitudes) (Easton 1965). He adds that

overt actions of support are not always meant like that, since it is difficult to predict all

consequences of a certain intention. To assess the potential impact of attitudes on the stability

of the political system, we have to take into account the variance and intensity of the attitude,

the location of the attitudes in the total population and the capacity of this part of the

population to transform attitudes into actions. This means that a quantitatively small group

(elite) is able to invest a sufficient amount of supportive energy to allow for the continued

existence of the system. This is why Huyse divided his respondents into participants and

spectators: spectators can have a critical attitude, but they will not resort to action (Huyse

1969). The system only comes under stress, when support sinks under a certain level.

The fact that attitudes do not always lead to action makes it difficult to evaluate these

attitudes. Behaviour does in most cases not follow attitudes: Parents complain about the

public school system, while private schools do not increase their market share and success

rates in universities increase (Loveless 1997). Students constantly complain about their

university restaurant but keep going there. If distrust is only symbolic, one could say there is

no reason to worry. It is however difficult to establish whether an attitude is merely symbolic.

Structural hindrances may hinder the expression of an attitude, and these expressions are often

faced with social punishments or benefits. Muller and Jukam found that the relation between

political support and aggressive behaviour is weak for those with a centrist ideological

attitude, while this relation is strong for the ideological left and right. A further difference

between left and right is that left-bending persons will strike and participate in

demonstrations, while right-bending ones will resort to other means for expressing their

distrust (Muller and Jukam 1977). Changes in behaviour could take two forms: conventional

95



and unconventional acts. Conventional acts include changing vote choice etc., unconventional

ones strikes, political violence etc. The lower trust is, the more chance that political

participation will be of non-institutionalised, i.e. unconventional nature (Kaase 1999). We

will speak about exit, voice and loyalty (Hirschman 1970).

Exit/voice/loyalty

Exit

There is only a limited possibility to “exit” in cases of distrust. Leaving the state is no option

in most but the extreme cases. Tiebout however has spoken about the possibility to vote with

one’s feet. His hypothesis was that if people do not agree with the policies, and more

specifically, the tax-level in their municipality, they would leave this municipality and

relocate (Tiebout 1956). One can however also leave the state in a symbolic way by refraining

from participation in elections, since election are considered to be the core of a democratic

state. Other exit-options are to stop paying taxes, to refrain from political participation or to

refuse jobs in government.

Election turnout: Whether the low turnout in elections is a result of distrust is a matter of

discussion. Low turn-out can also indicate that people trust government and think government

is doing a good job, and that they therefore do not bother about going to the ballot box, since

there are no burning issues which can mobilise voters (Hardin 1999). However, those who are

not satisfied with policy and government seem to abstain from voting, be it that this relation is

weak, making that we can not speak of a crisis (Rose and Pettersen 2000). Citrin found that

people who trust government do not vote more, nor participate more in political campaigns

nor keep track of political current affairs more intensively (Citrin 1974). Empirical data do

not suggest the occurrence of a massive exit from politics, c.q. voting. There is a trend of

declining voter turnout, but it remains very limited, while absolute percentages are quite high.

Only in Switzerland, the USA and with regard to the elections for the European parliament

declining voter turnout can be observed (Deschouwer 1998). However, in countries with

compulsory voting (e.g. the Netherlands), voter turnout dropped dramatically when the

principle was abandoned (Maddens 1995).



Political participation: Distrust can influence people’s willingness to participate in politics.

Indeed, Deschouwer remarks that the crisis of confidence is mainly one of the loss of

confidence in the existing political parties (Deschouwer 1998). Willingness to be a candidate,

membership in and donating to political parties is declining. This in turn decreases the

political staff recruitment potential, which can have repercussions on the quality of policy

making and in its turn influence trust in government.

Working for government: The willingness to work for government is related to the point

mentioned above. A number of questionnaires have asked respondents whether they would

want their children to work for government (National Performance Review (NPR), Kennedy,

and The Henry J. Kaiser Family Foundation 2000; The Pew Research Center for the People

and the Press 1998). A negative government image is believed to have an impact on the

recruitment capabilities of government and consequently on the quality of public employees

(Directie Personeelsmanagement Rijksdienst 1998).

Paying taxes: Distrust is supposed to lead to a loss of legitimacy of government, which makes

it difficult for government to make people obey the law and pay taxes. A decrease in the

willingness to pay taxes may be due a general distrust in government, but can also, and we

refer to the performance hypothesis, be due to the observation that government agencies spoil

money by not being efficient. Glaser and Hildreth use the term tax-performance continuity to

indicate a (dis)continuity between citizen assessment of government performance and

willingness to pay for this performance. They conclude that if the only reason why people are

asking for more and better performance is to be able to cut taxes, there will always be

discontinuity between the willingness to pay taxes and perceptions of performance (Glaser

and Hildreth 1999). Not paying taxes should however not only be seen as an egoistic effort to

save money, since it has been observed that even while people stop paying taxes, they

increase their support for charity (Sztompka 1996). Kobi found that the amount of people

complaining about paying more taxes than they get in return from government (in

Switzerland) has not increased but instead remained constant. If there were a connection with

trust in government, the number of people complaining about taxes would have to increase

jointly with distrust. Complaints on general tax pressures however seem to have increased

(Kobi 1998).

97



Voice

Voting has always been the traditional way of airing voice in a democratic system. In recent

years people became less loyal to parties, so we can observe quite some volatility in party

choice between elections (Deschouwer 1998). Some voice, i.c. changing party preference can

be considered as an alternative way of exit: voting for anti-system parties can be interpreted as

exit-behaviour. In two-party elections, distrusting people tend to vote for the challenging

party or candidate, or, if there is a possibility, they vote for third parties which often question

existing political norms (Hetherington 1999). The existence of anti-system parties offers

political distrust a means of expression (Aberbach 1969). The absence of anti-system parties

could make political distrust irrelevant, but this depends on the opinion the citizen has about

the actual influence of parties and elections on politics. Political distrust leads to a decline in

traditional political participation, while unconventional political activity is increasing, that is,

if political distrust does not express itself in total passivity. We mentioned the new politics

hypothesis earlier where new political groups emerged around new political issues

(environment, women’s rights...). Similarly, a number of new pressure groups could pop up

dealing with issues not successfully dealt with within the existing political system. Action

groups focusing on local issues could be both an expression of political distrust and an

expression of political interest and connectedness. Other possible indicators for political

distrust could be increases in demonstrations, terrorism, strikes etc.

Loyalty

We have mentioned before that distrust may lead to a decrease in the willingness to pay taxes

and obey the law (Tyler 1990). The amount of loyalty determines whether and which

responses citizens will develop: exit and voice.

“To the extent people can trust government, they are more likely to comply

willingly, to give their behavioral consent” (Levi 1996).

Distrust is therefore likely to increase transaction costs in society, since there will be more

occasions in which trustworthiness has to be checked and controlled (Anfang and Urban

1994). Furthermore, because of the decrease in legitimacy, political elites’ manoeuvring space

becomes smaller, making innovation more difficult (Kobi 1998).



When distrust develops to become the basic attitude towards the state, this affects perceptions.

All aspects related to government will be perceived in a negative way. This can be a partial

explanation for the causality problem we have dealt with before: it could be that distrusting

people are simply unable to see good government performance. Distrust can even lead to false

observations. Schema theory explains this as follows:

“information processing is therefore conceptualised as theory-driven rather than

data-driven, that is, it relies on people’s prior expectations, preconceptions and

knowledge about the social world in order to make sense of new situations and

encounters” (Augoustinos and Walker 1996).

6.4 DISTRUST, ETHNOCENTRISM AND THE EXTREME-RIGHT

In a number of countries, one of the reasons why governments started worrying about trust in

government is the rise of extreme right parties and ethnocentrism. Distrust and alienation are

supposed to increase the potential for extreme-right and populist parties. Even though

extreme-right rhetoric is mainly blaming Western governments for what they do -or don’t-

(choice of policy, functioning of administrations and police...), a great deal of the explanation

should be looked for in identity/emotion-related explanations of distrust. Rising numbers of

immigrants and other people who are ‘different’ in one way or another and the increased

(created) awareness of the existence of these groups have lead to a crumbling of the image of

the own population as homogeneous and stable.

A common feature of most extremist and populist parties is the attention they give to

identifying common problems and proposing solutions to these problems. This clearly fits

into Giddens’ observation about high modernity, where people demand a solution for every

possible problem, and cannot cope with unsolved problems, risk and difference (Giddens

1996). The fact that these parties seem to offer actual solutions makes them so attractive.

Again, a question of causality arises. In the efforts to restore trust, it is supposed that because

people distrust government they vote extreme right. Another approach however could be that

people who are extreme right become distrustful, among other things because governments do

not perform according to their wishes and because of an anti-government socialisation (Miller

and Listhaug 1999).

99



Tables and figures

LIST OF TABLES

Table 1: Emotionality and rationality in trust 12

Table 2: What factors is trust constituted of? 18

Table 3: Political culture and constitutional approaches to trust 20

Table 4: Cognitive and affective evaluations in Easton’s work 31

Table 5: Possibilities to create trust 82

LIST OF FIGURES

Figure 1: What do citizens see as government? 33



BIBLIOGRAPHY

Aberbach, Joel D. 1969. "Alienation and Political Behavior." American Political Science Review 63(1):86-99.

Allport, Gordon W. 1958. The Nature of Prejudice. New York: Doubleday.

Allport, Gordon W. and Leo Postman. 1965. The Psychology of Rumor. New York: Russell & Russell.

Almond, Gabriel A. and Sidney Verba. 1965. The Civic Culture: Political Attitudes and Democracy in Five

Nations. Princeton, NJ: Princeton University Press.

Anderson, Christopher. 1995. Blaming the Government: Citizens and the Economy in Five European

Democracies. Armonk : ME Sharpe.

Anderson, Christopher J. and Christine A. Guillory. 1997. "Political Institutions and Satisfaction With

Democracy: a Cross-National Analysis of Consensus and Majoritarian Systems." American Political

Science Review 91(1):66-81.

Anderson, Rolph E. 1973. "Consumer Dissatisfaction: the Effect of Disconfirmed Expectancy on Perceived

Product Performance." Journal of Marketing Research 10(1):38-44.

Anfang, Peter and Pieter Urban. 1994. ""Vertrauen"-Soziologisch Betrachtet. Ein Beitrag Zur Analyse Binärer

Interaktionssysteme." Stuttgart: Institut für Sozialwissenschaften. Abteilung für Soziologie I.

Augoustinos, Martha and Iain Walker. 1996. Social Cognition: an Integrated Introduction. London: Sage.

Axelrod, Robert M. 1984. The Evolution of Cooperation. New York: Basic books.

Bacharach, Michael and Diego Gambetta. 1997. "Trust in Signs." .

Baier, Annette. 1986. "Trust and Anti-Trust." Ethics 96:231-60.

Beck Jorgensen, Torben. 1994. "The Giaconda Smile of Authorities: An Essay of Fictional Pictures of Public

Administration and Citizens." Pp. 267-303 in Good Novels, Better Management. Reading

Organizational Realities, eds. Barbara Czarniawska-Jourges and Pierre Guillet de Monthoux. Chur:

Harwood academic publishers.

Beck, U. 1992. Risk Society: Towards a New Modernity. London: Sage.

Berger, Peter L. and thomas Luckmann. 1966. The Social Construction of Reality: a Treatise in the Sociology of

Knowledge. Garden City New York: Doubleday.

Berman, Evan M. 1997. "Dealing With Cynical Citizens." Public Administration Review 57(2):105-12.

101



Billiet, Jaak, Bart Cambré, and Marc Swyngedouw. 1997. "De Kennis Van De Federale Logica En Van De

Bevoegdheden En Beleid Van De Vlaamse Overheid: Een Verklaringsmodel." Res Publica( 4):609-27.

Bitner, Mary J. 1990. "Evaluating Service Encounters: the Effects of Physical Surroundings and Employee

Responses." Journal of Marketing 54(2):69-82.

Blokland, Hans. 2001. Een Rehabilitatie Van De Politiek 1: De Modernisering En Haar Politieke Gevolgen:

Weber, Mannheim En Schumpeter. Amsterdam: Boom.

Bok, Derek. 1997. "Measuring the Performance of Government." Pp. 55-76 in Why People Don't Trust

Government, eds. Joseph S. Jr. Nye, Philip D. Zelikow, and David C. King. Cambridge: Harvard

University Press.

Bok, Derek. 2001. The Trouble With Government. Cambridge, MA: Harvard University Press.

Borre, Ole. 1995. "Scope-of-Government Beliefs and Political Support." Pp. 343-66 in The Scope of

Government, eds. Ole Borre and Elinor Scarbrough. Oxford: Oxford University Press.

———. 2000. "Critical Issues and Political Alienation in Denmark." Scandinavian Political Studies 23 (4):285-

309.

Bouckaert, Geert. 1995a. "Charters As Frameworks for Awarding Quality: The Belgian, British and French

Experience." Trends in Public Service Renewal: Recent Developments and Concepts of Awarding

Excellence, eds. Herman Hill and Helmut Klages. Frankfurt a. M.: Lang.

———. 1995b. "Remodeling Quality and Quantity in a Management Context." Public Productivity Through

Quality and Strategic Management, eds. A. Halachmi and Geert Bouckaert. IOS Press.

Bouckaert, Geert, Jarl K. Kampen, Bart Maddens, and Steven Van de Walle. 2001.

Klantentevredenheidsmetingen Bij De Overheid: Eerste Rapport Burgergericht Besturen: Kwaliteit En

Vertrouwen in De Overheid. Leuven: Instituut voor de Overheid.

Bourdieu, Pierre. 1979. La Distinction: Critique Sociale Du Jugement. Paris: Ed. de Minuit.

Box, Richard C. 1998. Citizen Governance: Leading American Communities into the 21st Century. London:

Sage.

Braithwaite, Valerie. 1998. "Communal and Exchange Trust Norms: Their Value Base and Relevance to

Institutional Trust." Pp. 46-74 in Trust and Governance, eds. Valerie Braithwaite and Margaret Levi.

New York: Russell Sage Foundation.

Brehm, John and Wendy Rahn. 1997. "Individual-Level Evidence for the Causes and Consequences of Social

Capital." American Journal of Political Science 41(3):999-1023.



Brown, Karin and Philip B. Coulter. 1983. "Subjective and Objective Measures of Police Service Delivery."

Public Administration Review 43(1):50-58.

Chapin, Linda W. and Robert B. Denhardt. 1995. "Putting "Citizens First!" in Orange County, Florida." National

Civic Review.

Citrin, Jack. 1974. "Comment: The Political Relevance of Trust in Government." American Political Science

Review 68(3):973-88.

Citrin, Jack and Donald P. Green. 1996. "Presidential Leadership and the Resurgence of Trust in Government."

British Journal of Political Science 16(4):431-53.

Clarke, Harold D., Nitish Dutt, and Allan Kornberg. 1993. "The Political Economy of Attitudes Toward Polity

and Society in Western European Democracies." The Journal of Politics 55(4):998-1021.

Clarke, Harold D., A. Kornberg, and M. C. Stewart. 1984. "Parliament and Political Support in Canada."

American Political Science Review 78(2):452-69.

Coleman, James S. 1990. Foundations of Social Theory. Cambridge, Mass.: Harvard University Press.

Corrigan, P. 1996. "Recreating the Public: a Responsibility for Local Government." .

Coulson, Andrew. 1998. " Trust and Contract in Public Sector Management." Pp. 9-34 in Trust and Contracts:

Relationships in Local Government, Health and Public Services, ed. Andrew Coulson. Bristol: The

Policy Press.

Council for Excellence in Government. 1999. "America Unplugged: Citizens and Their Government." .

Crosby, Lawrence A., Kenneth R. Evans, and Deborah Cowles. 1990. "Relationship Quality in Services Selling:

an Interpersonal Influence Perspective." Journal of Marketing 54(3):68-81.

Crozier, M., S. Huntington, and J. Watanuki. 1975. The Crisis of Democracy. New York: University Press.

Dahrendorf, R. 1994. "The Changing Quality of Citizenship." Pp. 10-19 in The Condition of Citizenship, ed. B.

Van Steenbergen. London: Sage.

Dahrendorf, Ralf. 2000. "Afterword." Pp. 311-14 in Disaffected Democracies: What's Troubling the Trilateral

Countries, eds. Susan J. Pharr and Robert D. Putnam. Princeton: Princeton University Press.

Dalton, Russell J. 1999. "Political Support in Advanced Industrial Democracies." Pp. 57-77 in Critical Citizens:

Global Support for Democratic Governance, ed. Pippa Norris. Oxford: Oxford University Press.

———. 2000. "Citizen Attitudes and Political Behavior." Comparative Political Studies 33(6/7):912-40.

103



Das, Hari, Mallika Das, and Francis Mckenzie. 1995. "Assessing the "Will of the People": an Investigation into

Town Delivery Satisfaction." Canadian Public Administration 38(1):77-93.

De Schutter, Olivier, Notis Lebessis, and John Paterson, eds. 2001. Governance in the European UnionEuropean

Commission.

De Tocqueville, Alexis. 1946. Democracy in America. London: Oxford Univeristy Press.

Deakin, Nicholas. 1994. "Accountability: the Superstructure." Policy Studies 15(3):1-15.

Deakin, Nicholas and Anthony Wright. 1990. "Introduction." Pp. 1-16 in Consuming Public Services, eds.

Nicholas Deakin and Anthony Wright. London: Routledge.

Decoster, Kristof. 2000. Sociaal Kapitaal Voor Politologen: Een Contradictio in Terminis? Antwerpen: UIA.

Departement politieke en sociale wetenschappen.

Denhardt, Robert B. and Janet V. Denhardt. 2000. "The New Public Service: Serving Rather Than Steering."

Public Administration Review.

Deschouwer, Kris. 1998. "Op Verkenning in De Kloof." Pp. 77-100 in Wantrouwen En Onbehagen, ed.

Elchardus Mark. Brussel: VUBPress.

Dewachter, Wilfried. 1995. Besluitvorming in Politiek België. Leuven: ACCO.

Dierickx, Guido, Caroline Gijselinckx, and Peter Thijsen. 1996. "Culturele Deprivatie En Politieke Aliënatie:

Een Tussentijds Rapport." Res Publica 38(3-4):631-56.

Dinsdale, Geoff and Brian D. Marson. 1999. "Citizen/Client Surveys: Dispelling Myths and Redrawing Maps." :

Canadian Centre for Management Development.

Directie Personeelsmanagement Rijksdienst. 1998. Arbeidsmarktmonitor Rijksdienst 1998. Ministerie van

Binnenlandse Zaken en Koninkrijksrelaties.

Direction Générale "Information, communication, culture, audiovisuel". 1997. "L'Opinion Des Européens Sur

Les Services Publics ." Brussel.

Easton, David. 1965. A Systems Analyis of Political Life. New York.

———. 1975. "A Re-Assessment of the Concept of Political Support." British Journal of Political Science

5(4):434-57.

Easton, David and Jack Dennis. 1969. Children in the Political System: Origins of Political Legitimacy. New

York: McGraw-Hill Book Company.



Elchardus, Mark. 1998. " Was u Vandaag Al Slachtoffer of Werd u Al Beschuldigd? Gebruiksaanwijzing Voor

Een Wantrouwige Samenleving." Pp. 37-76 in Wantrouwen En Onbehagen, ed. Elchardus Mark.

Brussel: VUBPress.

Elchardus, Mark and Wendy Smits. 1998. "Het Vertrouwen Van De Vlamingen in De Instellingen." De

Gemeente(6-7):38-43.

Elchardus, Mark and Wendy Smits. 2001. "Een Wantrouwig Landje: Maatschappelijk Vertrouwen in

Vlaanderen." Vlaanderen Gepeild: De Vlaamse Overheid En Burgeronderzoek 2001, Brussel:

Ministerie van de Vlaamse Gemeenschap.

Epstein, Paul, Lyle Wray, Martha Marshall, and Stuart Grifel. 2000. "Engaging Citizens in Achieving Results

That Matter: A Model for Effective 21st Century Governance." .

Erin Research inc. 1998. "Citizens First." : Canadian Centre for Management Development.

Fitzgerald, Michael R. and Robert F. Durant. 1980. "Citizen Evaluations and Urban Management: Service

Delivery in an Era of Protest." Public Administration Review 40(6):585-94.

Fornell, Claes, Eugene W. Anderson, J. Cha, and B. E. Bryant. 1996. "The American Customer Satisfaction

Index: Nature, Purpose and Findings." Journal of Marketing 60:7-18.

Friedrichs, David O. 1980. "The Legitimacy Crisis in the United States: A Conceptual Analysis." Social

Problems 27(5):540-555.

Fuchs, Dieter. 1999. "The Democratic Culture of Unified Germany." Pp. 123-45 in Critical Citizens: Global

Support for Democratic Governance, ed. Pippa Norris. Oxford: Oxford University Press.

Gabriel, Oscar W. 1998. "Political Efficacy and Trust." Pp. 357-89 in The Impact of Values, Jan W. Van deth

and Elinor Scarbrough. New York: Oxford University Press.

Gambetta, Diego. 1998. "Mafia: the Price of Distrust." Pp. 158-75 in Trust: Making and Breaking Cooperative

Relations, ed. Diego Gambetta. Oxford: Basil Blackwell.

Gamson, William A. 1968. Power and Discontent. Homewood, Illinois: The Dorsey Press.

Gaster, Lucy and Nicholas Deakin. 1998. "Quality and Citizens." Pp. 143-56 in Trust and Contracts:

Relationships in Local Governement, Health and Public Services, ed. Andrew Coulson. Bristol: The

Policy Press.

Giddens, Anthony. 1996. The Consequences of Modernity. Cambridge: Cambridge university press.

Glaser, Mark A. and Robert B. Denhardt. 1997. "Local Government Performance Through the Eyes of Citizens."

105



Neward: University of Delaware.

Glaser, Mark A. and Bartley W. Hildreth. 1999. "Service Delivery Satisfaction and Willingness to Pay Taxes."

Public Productivity and Management Review 23(1):48-67.

Glynn, Carroll J., Andrew F. Hayes, and James Shanahan. 1997. "Perceived Support for One's Opinion and

Willingness to Speak Out: A Meta-Analysis of Survey Studies on the "Spiral of Silence"." Public

Opinion Quarterly 61:452-63.

Goodsell, Charles T. 1994. The Case for Bureaucracy: a Public Administration Polemic. Chatham: Chatham

House Publishers Inc.

Haque, M. S. 1998. "Legitimation Crisis: a Challenge for Public Service in the Next Century." International

Review of Administrative Sciences 64(1):13-26.

Hardin, Russell. 1998. " Trust in Government." Trust and Governance, eds. Valerie Braithwaite and Margaret

Levi. New York: Russell Sage Foundation.

Hardin, Russell. 1999. "Do We Want Trust in Government?" Pp. 22-41 in Democracy and Trust, ed. Mark E.

Warren. Cambridge: Cambridge University Press.

———. s.d. "Distrust." .

Harisalo, Risto and Ensio Miettinen. 1997. Trust Capital: The Third Force of Entrepreneurship. Tampere:

University of Tampere, Department of administrative science.

Herzog, Richard J. and Ronald G. Claunch. 1997. "Stories Citizens Tell and How Administrators Use Types of

Knowledge." Public Administration Review 57(5):374-79.

Hetherington, Marc J. 1998. "The Political Relevance of Political Trust." American Political Science Review

92(4):791-808.

Hetherington, Marc J. 1999. "The Effect of Political Trust on the Presidential Vote, 1968-96." American

Political Science Review 93(2):311-26.

Hibbing, John R. and Elizabeth Theiss-Morse. 1995. Congress As Public Enemy: Public Attitudes Toward

American Political Institutions. Cambridge: Cambridge University Press.

Himmelberg, Charles P. and Gregory Wawro. 2001. "Is All Politics Local? National Elections and Local

Economic Conditions." .

Hirschman, Albert O. 1970. Exit, Voice, and Loyalty: Response to Decline in Firms, Organizations and States.

Cambridge, MA: Harvard University Press.



Hobbes, Thomas and C. B. MacPherson. 1978. Leviathan. Harmondsworth: Penguin books.

Holmberg, Soren. 1999. " Down and Down We Go: Political Trust in Sweden." Pp. 103-22 in Critical Citizens:

Global Support for Democratic Governance, ed. Pippa Norris. Oxford: Oxford University Press.

Hooghe, Marc. 1999. ""Politiek, Da's Niet Voor Mij". Een Explorerend Kwalitatief Onderzoek Naar Adaptieve

Preferentievorming Als Causaal Mechanisme Bij Politike Passiviteit." Tijdschrift Voor Sociologie

20(1):51-72.

Hoogland DeHoog, Ruth, David Lowery, and William E. Lyons. 1990. "Citizen Satisfaction With Local

Government: A Test of Individual, Jurisdictional, and City-Specific Explanations." Journal of Politics

52(3):807-37.

Huseby, Beate M. 2000. Government Performance and Political Support: a Study of How Evaluations of

Economic Performance, Social Policy and Environmental Protection Influence the Popular

Assessments of the Political System. Trondheim: Institutt for sosiologi og statsvitenskap.

Huyse, Luc. 1969. De Niet-Aanwezige Staatsburger: De Politieke Apathie Sociologisch in Kaart Gebracht.

Antwerpen: Standaard.

Inglehart, Ronald. 1990. Culture Shift in Advanced Industrial Society. Princeton: Princeton University Press.

———. 1997. "The Erosion of Institutional Authority and Post-Materialist Values." Why People Don't Trust

Government, eds. Joseph S. Nye, Philip D. Zelikow, and David C. King. Cambridge Mass.: Harvard

University press.

———. 1999. "Postmodernization Erodes Respect for Authority, but Increases Support for Democracy." Pp.

236-56 in Critical Citizens: Global Support for Democratic Governance, ed. Pippa Norris. Oxford:

Oxford University Press.

Kaase, Max. 1999. "Interpersonal Trust, Political Trust and Non-Institutionalised Political Participation in

Western Europe." West European Politics 22(3):1-21.

Klingemann, Hans-Dieter. 1999. "Mapping Political Support in the 1990s: A Global Analysis." Pp. 31-56 in

Critical Citizens: Global Support for Democratic Governance, ed. Pippa Norris. Oxford: Oxford

University Press.

Klingemann, Hans-Dieter and Dieter Fuchs, eds. 1995. Citizens and the State. Oxford: Oxford University Press.

Kobi, Silvia. 1998. "Y a-t-Il Un Lien Entre Le Déclin De La Confiance Citoyenne Et L'Évaluation Que Font Les

Gouvernés Du Rapport "Couts-Bénéfices" De L'Action Étatique?" .

Kornberg, Allan and Harold D. Clarke. 1992. Citizens and Community: Political Support in a Representative

107



DemocracyCambridge University Press.

———. 1994. "Beliefs About Democracy and Satisfaction With Democratic Government: The Canadian Case."

Political Research Quarterly(3):537-64.

Kuechler, Manfred. 1991. "The Dynamics of Mass Political Support in Western Europe: Methodological

Problems and Preliminary Findings." Pp. 275-93 in Eurobarometer: the Dynamics of European Public

Opinion: Essays in Honour of Jacques-René Rabier, Karlheinz Reif and Ronald Ingelhart. Houndmills:

Macmillan.

Lane, Jan-Erik and Svante Ersson. 1994. Comparative Politics: an Introduction and New Approach. Cambridge:

Polity Press.

Levi, Margaret. 1996. A State of Trust. Florence: European University Institute/Robert Schumann Centre.

Lewis, David J. and Andrew Weigert. 1985. "Trust As a Social Reality." Social Forces 63(4):1985.

Lipset, Seymour M. and William C. Schneider. 1983. The Confidence Gap: Business, Labor, and Government in

the Public Mind. New York: Free Press.

Listhaug, Ola and Matti Wiberg. 1995. "Confidence in Political and Private Institutions." Pp. 298-322 in Citizens

and the State, eds. Klingemann Hans-Dieter and Dieter Fuchs. Oxford: Oxford University Press.

Loveless, Tom. 1997. "The Structure of Public Confidence in Education." American Journal of Education

105(2):127-59.

Luhmann, Niklas. 1979. Trust and Power. Chicester: John Wiley & Sons.

Luhmann, Niklas. 1998. "Familiarity, Confidence, Trust: Problems and Alternatives." Pp. 94-107 in Trust:

Making and Breaking Cooperative Relations, ed. Diego Gambetta. Oxford: Basil Blackwell.

Maddens, Bart. 1995. Drie Nieuwe Parlementen. Een Praktische Gids Voor Kiezers En Kandidaten. Leuven:

Davidsfonds.

Maister, D. 1985. "The Psychology of Waiting Lines." The Service Encounter: Managing Employee-Customer

Interaction in Service Business, eds. J. Czepiel, M. Solomon, and C. Surprenant. Lexington: Lexington

Books.

Mcallister, Ian. 1999. "The Economic Performance of Governments." Pp. 188-203 in Critical Citizens: Global


Miller, Arthur H. 1974a. "Rejoinder to "Comment" by Jack Citrin: Political Discontent or Ritualism." American

Political Science Review 68(3):989-1001.



Miller, Arthur H. 1974b. "Political Issues and Trust in Government: 1964-1970." American Political Science

Review 68(3):951-72.

Miller, Arthur H. and Ola Listhaug. 1990. "Political Parties and Confidence in Government: a Comparison of

Norway, Sweden and the United States." British Journal of Political Science 20(3):357-86.

———. 1999. "Political Performance and Trust." Pp. 204-16 in Critical Citizens, (ed.) Pippa Norris. Oxford:

Oxford university press.

Miller, Thomas I. and Michelle A. Miller. 1991. "Standards of Excellence: US Residents' Evaluations of Local

Government Services." Public Administration Review 51(6):503-13.

Mishler, William and Richard Rose. 1997. "Trust, Distrust and Skepticism: Popular Evaluations of Civil and

Political Institutions in Post Communist Societies." The Journal of Politics 59(2):418-51.

Mitchell, Terence R. and William G. Scott. 1987. "Leadership Failures, the Distrusting Public, and Prospects of

the Administrative State." Public Administration Review 47(6):445-52.

Moe, Ronald C. 1994. "The "Reinventing Governement" Exercise: Misinterpreting the Problen, Misjudging the

Consequences." Public Administration Review 54(2):111-22.

Moesen, Wim. 2000. "Maatschappelijk Kapitaal En Economische Performantie." Leuven.

Moesen, Wim, Tom Van Puyenbroeck, and Laurens Cherchye. 1999. "Trust As Societal Capital: Economic

Growth in European Regions." Leuven: Centre for Economic Studies.

Muller, Edward M. and Thomas O. Jukam. 1977. "On the Meaning of Political Support." American Political

Science Review 71:1561-77.

Multatuli Expert Seminar. 2001. "Angst En Onzekerheid in De Moderne Samenweving." .

National Performance Review (NPR), John F. Kennedy, and The Henry J. Kaiser Family Foundation. 2000.

"Attitudes Towards Government." .

Newton, Kenneth. 1999. " Social and Political Trust in Established Democracies." Pp. 169-87 in Critical

Citizens: Global Support for Democratic Governance, ed. Pippa Norris. Oxford: Oxford University

Press.

Newton, Kenneth. 2000. "Social Trust and Social Capital: Political Trust and Political Capital." .

Newton, Kenneth and Pippa Norris. 1999. "Confidence in Public Institutions: Faith, Culture or Performance." .

Noelle-Neuman, Elizabeth. 1974. "The Spiral of Silence: A Theory of Public Opinion." Journal of

109



Communication 24:43-51.

Norris, Pippa. 1999a. "Institutional Explanations for Political Support." Pp. 217-35 in Critical Citizens: Global


Norris, Pippa. 1999b. "Introduction: the Growth of Critical Citizens." Pp. 1-30 in Critical Citizens: Global


Norén, Ylva. 1997. "Hur Sköter Kommunstyrelsen Sin Uppgift?" Mangfald Bilder Av En Storstadsregion, ed.

Lennart Nilsson .

———. 2000. "Explaining Variations in Political Trust in Sweden: Lessons Learned by Theories and Data." 30

p.

Nye, Joseph S., ed. 1997. Why People Don't Trust Government. Cambridge (Mass.): Harvard university press.

Offe, Claus. 1999. "How Can We Trust Our Fellow Citizens?" Pp. 42-87 in Democracy and Trust, ed. Mark E.

Warren. Cambridge: Cambridge University Press.

Orren, Gary. 1997. "Fall From Grace: the Public's Loss of Faith in Government." Pp. 77-108 in Why People

Don't Trust Government, eds. Joseph S. Nye, Philip D. Zelikow, and David C. King. Cambridge:

Harvard University Press.

Osborne, David and Ted Gaebler. 1992. Reinventing Government. How the Entrepreneurial Spirit Is

Transforming the Public SectorWilliam Patrick Books.

Parry, Geraint. 1976. "Trust, Distrust and Consensus." British Journal of Political Science 6(2):129-43.

Pegnato, Joseph A. 1997. "Is a Citizen a Customer?" Public Productivity and Management Review 20(4):397-

404.

Pleysier, S., G. Vervaeke, and S. Parmentier. 2000. "'Het' Vertrouwen in 'De' Instelling: Enkele

Beschouwingen." Panopticon 21(1):63-79.

Poister, Theodore H. and Gary T. Henry. 1994. "Citizen Ratings of Public and Private Service Quality: a

Comparative Perspective." Public Administration Review 54(2):155-60.

Pollitt, Christopher. 1993. Managerialism and the Public Services. Second edition ed. Oxford: Blackwell.

Pratt, J. W. and R. J. Zeckhauser. 1991. Principals and Agents: the Structure of Business. Boston: Harvard.

Princeton Survey Research Associates and Pew Charitable Trust. 2000. "Customers Judge the Performance of

Federal Government Agencies: Report on the Findings." .



Privy Council Office and CCMD. 1996. "Perceptions of Government Service Delivery. Report of the Task Force

on Service Delivery." Ottawa: Privy Council Office//CCMD.

Public Management Foundation. 1996. The Glue That Binds: the Public Value of Public Services. London:

Public Management Foundation//MORI.

Pulkkinen, Tuija. 2000. The Postmodern and Political Agency. Jyväskylä: SoPhi.

Putnam, Robert D. 1995. "Tuning in, Tuning Out: the Strange Disappearance of Social Capital in America." PS:

Political Science and Politics:664-83.

———. 2000. Bowling Alone: the Collapse and Revival of American Community. New York: Simon and

Schuster.

———. 2001. "Multatuli Lecture." .

Ringeling, Arthur. 1993. Het Imago Van De Overheid: De Beoordeling Van Prestaties Van De Publieke Sector.

's-Gravenhage: VUGA.

Rose, Lawrence. 1999. "Citizen (Re)Orientations in the Welfare State: From Private to Public Citizens?" Pp.

131-48 in Citizenship and Welfare State Reform in Europe, Jet Bussemaker. London: Routledge.

Rose, Lawrence E. and Per A. Pettersen. 2000. "The Legitimacy of Local Government: What Makes a

Difference? Evidence From Norway ." Pp. 25-66 in Citizen Responsive Government, eds. Keith

Hoggart and Terry Nichols Clark. Amsterdam: Elsevier Science.

Roth, Victor J., Lorne Bozinoff, and Peter MacIntosh. 1990. "Public Opinion and the Measurement of Consumer

Satisfaction With Government Services." Canadian Public Administration 33(4):571-83.

Rothstein, Bo. 2001. "Creating Trust From Above: Social Capital and Institutional Legitimacy." .

Rotter, Julian B. 1967. "A New Scale for the Measurement of Interpersonal Trust." Journal of Personality

35(4):651-65.

Ruscio, Kenneth P. 1996. "Trust, Democracy, and Public Management: a Theoretical Argument." Journal of

Public Administration Research and Theory 6(3):461-77.

Schmidt, Faye and Teresa Strickland. 1998. "Client Satisfaction Surveying: Common Measurements Tool." :

Canadian Centre for Management Development.

Seeman, Melvin. 1959. "On the Meaning of Alienation." American Sociological Review 24(6):783-91.

Serra, George. 1995. "Citizen-Initiated Contact and Satisfaction With Bureaucracy: a Multivariate Analysis."

111



Journal of Public Administration Research and Theory 5(2):175-88.

Soss, Joe. 1999. "Welfare Application Encounters: Subordination, Satisfaction, and the Puzzle of Client

Evaluations." Administration and Society 31( 1):50-94.

Steen, Anton. 1996. "Confidence in Institutions in Post-Communist Societies: the Case of the Baltic States."

Scandinavian Political Studies 19(3):205-55.

Stipak, Brian. 1977. "Attitudes and Belief Systems Concerning Urban Services." Public Opinion Quarterly

41(1):41-56.

Stipak, Brian. 1979. "Citizen Satisfaction With Urban Services: Potential Misuse As a Performance Indicator."

Public Administration Review 39(1):46-52.

Stokes, Donald E. 1962. "Popular Evaluations of Government: an Empirical Assessment." Ethics and Bigness:

Scientific, Academic, Religious, Political and Military, eds. Harlan Cleveland and Harold D. Lasswell.

New York: Harper and Brothers.

Swindell, David and Janet M. Kelly. 2000. "Linking Citizen Satisfaction to Performance Measures: a

Preliminary Evaluation." Public Performance and Management Review 24(1):30-52.

Sztompka, Piotr. 1996. " Trust and Emerging Democracy: Lessons From Poland." International Sociology

11(1):37-62.

———. 1999. Trust: A Sociological Theory. Cambridge: Cambridge University Press.

Taylor, Frederic W. 1967. The Principles of Scientific Management. New York: Norton.

The Pew Research Center for the People and the Press. 1998. "Decosntructing Distrust: How Americans View

Government." .

Tiebout, C. M. 1956. "A Pure Theory of Local Expenditures." Journal of Political Economy 64(5):416-24.

Toennies, Ferdinand. 1979. Gemeinschaft Und Gesellschaft: Grundbegriffe Der Reinen Soziologie. Darmstadt:

Wissenschaftliche Buchgesellschaft.

Tyler, Tom. 1990. Why People Obey the Law. New Haven: Yale university press.

Uslaner, Eric M. 1999. " Is Washington Really the Problem." .

———. 2000. "Trust, Democracy, and Governance." .

Volkan, Vamik D. 1994. The Need to Have Enemies and Allies: From Clinical Practice to International

Relationships. Northvale: Jason Aronson Inc.


Bouckaert, Van de Walle, Maddens & Kampen 113

Weatherford, M. S. 1992. "Measuring Political Legitimacy." American Political Science Review 86(1).

Weber, Max. 1964. Wirtschaft Und Gesellschaft I: Die Wirtschaft Und Die Geselltschaftlichen Ordnungen Und

Mächte. Köln: Kiepenheuer & Witsch.

Yankelovich, D. 1991. Coming to Public Judgement: Making a Democracy Work in a Complex World.

Syracuse, NY: Syracuse University Press.

Zeithaml, Valarie A., A. Parasuraman, and Leonard L. Berry. 1990. Delivering Quality Service: Balancing

Customer Perceptions and Expectations . London: The Free Press.

Zussman, David. 1997. "Do Citizens Trust Their Government." Canadian Public Administration 40(2):234-54.

bestemmeling: titel, voornaam, achternaam - faculteit sociale

Documents