exchange rate volatility and firm … fluctuation in exchange rates became an ... growth has...

Proceedings of the Australia-Middle East Conference on Business and Social Sciences 2016, Dubai

(in partnership with The Journal of Developing Areas, Tennessee State University, USA)

ISBN 978-0-9925622-3-6

653

EXCHANGE RATE VOLATILITY AND FIRM PERFORMANCE

IN NIGERIA: A DYNAMIC PANEL REGRESSION APPROACH

Ikechukwu Kelilume Lagos Business School, Pan-Atlantic University, Lagos, Nigeria

ABSTRACT

This paper investigates the effects of volatility clustering in exchange rate on firm’s performance in Nigeria, examining cross sectional data for the most active 20 companies listed on the Nigerian Stock Exchange. The empirical investigation develops three dynamic panel models that account for heterogeneities among the companies and it extends recent research by allowing international investors and corporations to base their investment decisions on the exchange rate volatilities between the Nigerian Naira and their home country currencies. The results show that exchange rate volatility has significant negative impacts on the rate of return on assets, asset turn ratio and the portfolio activity & resilience, thus, showing the significant negative impact of exchange rate volatility on firm performance in Nigeria between 2004 and 2013. Overall, this study suggests that the higher the exchange rate volatility, the less is the firm efficiency and performance. JEL Classifications: C34, D21, D51, D92, F31 Keywords: Exchange Rate Volatility, Firm Performance, Panel Data Regression, Rate of Return on Assets, Asset Turn Ratio, Portfolio Activity & Resilience Corresponding Author’s Email Address: [email protected]

INTRODUCTION Firrm performance has played a central role in management research. A series of important studies has allowed us to have a robust technical knowledge on key issues such as on the main determinants of firm performance (see, for example, Green, Maggioni, & Murinde, 2000; Dellas & Hess, 2002; Yeh, Lee & Pen, 2002; Min, 2002; Jun, Marathe & Shawky, 2003; Tse, Wu & Young, 2003; Du & Wei, 2004; Bae, Chan & Ng, 2004; Lesmond, 2005). There are two levels of determinants of firms’ performance: the first relates to external factors beyond the control of the firms and are economy-wide, the second relates to internal factors within the control of the firms (Babatunde & Olaniran, 2009). Most studies have taken a micro approach by examining internal factors within the control of the firms. However, the current context of globalization has raised competitiveness concerns and necessitates an in-depth examination of the macro variables.

The principal objective of this paper is to gain knowledge on the effects of exchange rate volatility on firm’s performance in Nigeria, using two key performance variables, cost of goods sold and gross profit before tax. More specifically, this research investigates the effects of volatility clustering in exchange rate on firm’s performance in Nigeria, examining cross sectional data for the most active 20 companies listed on the Nigerian Stock Exchange and using the panel data regression approach.

Firm performance in Nigeria has not particularly received much attention from macroeconomic point of view. The few existing studies on the subject have leaned more on individual firm performance in relation to micro variables. Yet, in the empirical literature, several scholars’ contend that firms can take advantage of changes in macroeconomic aggregates to influence business performance (Greer, Ireland & Wingender 2001; Navarro, Bromiley & Sottile, 2010). One of such macroeconomic aggregates is exchange rate. However, real and nominal exchange rates have fluctuated widely since the collapse of the Bretton Woods system of fixed exchange rates (Baum, Caglayan, & Barkoulas, 2001).

This fluctuation in exchange rates became an issue of great concern to corporate establishments and policy makers in Nigeria in the wake of the 2007-2008 global financial crises. At the WDAS1, exchange rate during the period (January 2, 2008), opened at N115.00/US$ and closed at N130.32/US$ in the period (December 31, 2008) representing a depreciation of 13.32 percent. In November 2011, the naira exchanged at N153.5/US$ representing a depreciation of over 33.5 percent over the period 2008-2011. By February 2015, volatility levels have risen to the highest levels in a decade (Fig. 1). The question is: what is the impact of these shifts in volatility on firm performance in Nigeria? Can the relationship between exchange rate volatility and firm performance be captured?



ISBN 978-0-9925622-3-6

654

FIG. 1: DOLLAR TO NAIRA EXCHANGE RATE (2004-2015)

Data Source: Central Bank of Nigeria (2015). Authors’ calculations using Microsoft Excel (2010) and graphical

analysis results.

Notes: This chart shows the volatile movements of the exchange rate in Nigeria. It as well depicts the drastic fall

in the value of the naira in the past decade.

As a result of the volatile nature of the movement of exchange rate in Nigeria, over the past decades, real output growth has averaged 6.5 percent with the manufacturing sector growth averaging less than 3.5 percent over the same period. This presupposes a declining firms’ performance in the face of changing macroeconomic environment. The volatile nature of the movement of exchange rate in Nigeria and the paucity of research in the area of exchange rate volatility and firm performance provides us a good opportunity for studying the effects of exchange rate volatility on firm performance in Nigeria.

A step further is necessary in the study of the effects of exchange rate volatility on firm’s performance. Instability in the foreign exchange market can pose significant risk to companies and enterprises conducting businesses in foreign markets. Through massive depreciation of the domestic currency of a country, firms operating in the domestic market can be subjected to either translation or transactions risks translating into significant shrinkage in company’s earning, market share, cash flow and firm’s balance sheet. Therefore, understanding the firm’s sensitivity in terms of exchange rate volatility for an African market, such as Nigeria, is therefore a key concern for corporate future and policy makers’ agendas, since the influx of investors from Europe, the United States, and a swelling number of emerging destinations is high. To the best of the authors’ knowledge, this type of research has not been yet undertaken for the case of Nigeria.

This paper proceeds as follows: Section 2 reviews the major theoretical and empirical literature on exchange rate volatility and firm performance. Section 3 discusses the cross-sectional data and the dynamic panel methodology employed in the study. Section 4 summarizes the empirical results. Section 6 covers caveats and possible future research, and Section 7 concludes. LITERATURE REVIEW Firm Transactions, Performance and Exchange Depreciation Recent international financial crises have underscored the import and significance of the international monetary mechanism to corporations. As a consequence, external variables such as exchange rate fluctuations have become of great weight in determining the character of firm performance. From Harris’s (2001) viewpoint, exchange rate depreciation is a necessary factor, influencing the gap in firms’ productivities. In agreement, Auer & Chaney (2007) suggest that the market power of a given firm depends not only on the prices and qualities of its close competitors and on the prices of other closely related firms, but also on the exchange rate movement which is hugely influenced by export transactions of low quality goods, which are in high demand internationally, since producers of such good are in steep competition and face the same exchange rate shocks (Fung, 2004).

Different firms react differently to exchange rate volatility. This heterogeneity has serious distorting influence on the levels of exchange rate movements in the economy. Berman, Martin & Mayer (2008) depict

100

110

120

130

140

150

160

170

180

190

200

200

4

200

5

200

6

200

7

200

8

200

9

201

0

201

1

201

2

201

3

201

4

201

5

Oil Price Fall

Global Financial Crisis



ISBN 978-0-9925622-3-6

655

that high and low productivity firms react differently to exchange rate depreciation, in the presence of distribution costs in the export market. That is, firms with high performance optimally raise their markup rather than the volume they export; while low productivity firms choose the opposite strategy, thus distorting the levels of exchange rate movements in the economy.

Clear insights from the concept of fixed cost to export show that exporting requires a high productivity, an attribute which in turn gives an incentive to firms to react to exchange rate depreciation by increasing their export price rather than their sales (Berman, et al. 2008). Berman et al. (2008) found that high performance firms react to depreciation in exchange rate by increasing their export price rather than their export volume but low performance-exporting firms do otherwise.

The literature has revealed that more firms enter the export market following depreciation: these firms are small, their productivity is low and their extensive margin response to exchange rate movement is small when compared with other firms at the aggregate level. This literature shows that prices are rigid in the currency of the export market and the levels of uncertainties as a result of the nature of the levels of imperfection in the competitive market and the influence of distribution cost (i.e. Campa, Goldberg & Gonzalez-Minguez, 2005; Corsetti & Dedola, 2005; Gopinath, Oleg & Rigobon, 2010). In this wise, local distribution costs basically lower the pass-through to consumer prices and generate uncertainties in productive capacity to remedy further reductions (Corsetti & Dedola, 2005; Berman, et al. 2008). Empirical Literature The study of firm performance has yielded a vast body of literature, showing that firm performance is affected by myriads of factors such as inventory (Thille, 2006), quality of the banking system (Dellas & Hess, 2002), liquidity risk (Min, 2002; Jun, Marathe & Shawky, 2003; Lesmond, 2005), number of regulations and their imbedded costs (Green, Maggioni, & Murinde, 2000), number of informed agents (Du & Wei, 2004), information asymmetry (Tse, Wu & Young, 2003), segmentation (Yeh, Lee & Pen, 2002), the impact of investibility (Bae, Chan & Ng, 2004). Much of the literature have proposed that the main factors that are important in order for firms to be competitive include the ability to export, effective policy regulation, management style, ownership structure, technology and human capital (Bartelsman & Doms 2000; Girma, Greenaway, & Kneller 2002). This literature clearly has under-emphasized the impacts of macro variables such as exchange rates.

Shapiro’s (1975) theoretical study on the profitability effect of exchange rate uncertainty on the value of multinational firms is often cited as the earliest study, implicitly or explicitly, on the relationship between exchange rate volatility and firm performance. Using a partial equilibrium model to determine the variables affecting a multinational firm’s exchange rate risk and the change in the value of the corporation, Shapiro found that exporting firms profit from a domestic currency depreciation.

After the pioneering study of Shapiro (1975), much empirical research have tried to explain the relationship between exchange rate volatility and firm performance on one hand, and stock market returns on the other, using different regression models and empirical methodologies. The new century has already started with two influential studies, Chatterjee, Carneiro & Vichyanond (2010) and Baggs, Beaulieu, Fung & Lapham (2011), which mainly focused on exchange rate movements and firm’s performance using advanced econometric modeling techniques. Chatterjee et al.’s (2010) study on the effect of exchange rate shocks on pricing decision of multi-product firms and its impact on firm’s performance constructed a model using the quantity of the scope of the product, to analyze firms’ price adjustment in circumstances of exchange rate depreciation. The study revealed that, in the event of exchange depreciation, most firms increase the prices of products closer to their core competency. Chatterjee et al. claims that this kind of adjustments enhances firms’ performance. As well, Baggs, Beaulieu, Fung & Lapham’s (2011) study on the impact of exchange rate on retail firms and its impact on different industry groups showed that the effect of exchange rate movement was most adverse on profitability and the obtained effect diminished over time in accordance to the location of the firm, i.e. nearness to the market. The real exchange rate distance interaction term maintain the same strong positive sign for all the firms observed. Bagg et al. observed a negative exchange rate effect due to a net effect on the prices of input as a result of an increase in the domestic rate of exchange, which caused the prices of inputs to fall and as a consequence, the retail price of the good to be reduced. This indicates that a real appreciation of a currency reduces the level of sales, increase the labor supply and firm profit.

Studies such as Baggs et al. (2011) has shown that exchange rate volatility influence the levels of sales: sales decrease as the rate of exchange appreciates while sales increase as the rate of exchange depreciates. Subsequently, it can be inferred that the mixed effect of exchange rate movement have significant effects on firm’s survival. That is, the real exchange rate effect has substantial implications for the sales, employment and profits of firms.

As well, exchange rate has a significant positive impact on export volumes, which varies across firms; although it is significantly reduced for low performing firms (Berman, Martin & Mayer. 2008). This suggests that high and low productive firms have distinct strategies for various circumstances of exchange rate changes. Moreover, in a model with heterogeneous consumers, exchange rate pass through can be incomplete and



ISBN 978-0-9925622-3-6

656

heterogeneous across goods of different quality. This distinction is as well shown in Auer and Chaney (2008) who suggest that exporters sell goods of different qualities to consumers who have heterogeneous preferences for quality. For this reason, the level of productivity is subject to decreasing return to scale. The level of supply and intense competition react to cost changes due to exchange rate fluctuations.

The literature has also examined the responses of firms to maintaining stable profit margins and sales as the rate of exchange fluctuates. For example, Klitgaard (1999) found that the short-run response of profit margins to exchange rate movements appears to be more prominent in firms associated with high valued commodities, perhaps due to the fact that the prices of these commodities are majorly invoiced in foreign currencies such that when they are valued in terms of the local currency, they respond automatically to exchange rate swings. Klitgaard also found that the direction of the movement of the local currency have no effect on firms behavior in analyzing sale-profit relationship, this is because, these firms are used to adjusting their profit margins to stabilize prices in foreign markets, only raising their profit margins when the local currency depreciates and cutting margins when the currency appreciates.

It must be noted that there is no conclusive evidence on the impact of exchange rate volatility on firms’ performance. While a strand of the literature claim that exchange rate volatility provides little or no explanation for stock performance (Jorion, 1990 and 1991; Bartov & Bodnar, 1994; Bernard & Galati, 2000), another strand contend that stock performance is significantly affected by exchange rate volatility (Dumas & Solnik, 1995; Choi et al., 1998; Doukas, Hall & Lang, 1999; Patro, Wald Wu, 2002). Studies such as Dumas and Solnik (1995), Aquino (2005, 2006), and Yau & Nieh (2006) claim that exchange rate volatility account for much of the volatility of equity markets. This evidence suggests that the volatility of the local stock markets increase with exchange rate volatility.

As well, like other empirical literature, different researchers have used different methodologies and different models to study firm performance, exchange rate volatility and the relationship between them. More advanced econometric techniques have been applied to firm performance studies in the 1990s and 2000s compared to the studies between the 1960s and 1980s, contributing to improvements in the understanding of corporate performance. Some of the most advanced techniques have been applied to improve, among others, the regression estimates. For example, studies such as Park, Yang, Shi & Jiang (2006) constructed firm-specific exchange rate shocks based on the pre-crisis destinations of firms’ exports to evaluate the impact of exportation on firm productivity and performance. Their results show that although export growth increase firm productivity and performance, exchange rate depreciation slow growth in export. The introduction of firm-specific exchange rate shocks Park, at al. (2006) is made possible by the availability of information on firm-specific export country destination for foreign-invested firms in China’s industrial. This was used to represent the timing and pattern of devaluations which were unforeseen due to the crisis. This identification strategy controlled for sector fixed effect and thus rules out bias from unobserved regional or sectoral changes. Also, it made it possible to determine how instrumented changes in exports affect measures of firm performance. Three key measures of performance considered by this study are the value of total factor productivity, total sales and return to asset. They were known to increase as the levels of export increased. Firms with greater currency depreciation have slower growth in exports and export growth was noticed to increase firm productivity and performance.

It can be seen that Park, at al.’s (2006) findings are hugely different from Jorion (1990) and (1991), Bartov & Bodnar (1994), and Bernard & Galati (2000) who demonstrate that exchange rate volatility provides little or no explanation for stock performance. Thus, in the literature, different methodologies used have led to different conclusions. In the next section, the data and the specification of the dynamic panel model that we use and the econometric methodology are described in specific details. METHODOLOGY AND DATA Data Set A balanced panel of 10 annual observations from 20 companies over the period of 2004-2013 was used in this study. The sample of companies represents all major companies in Nigeria. The company data comprises cross sectional yearly observations of company performance indicators for twenty most active companies listed on the Nigeria Stock Exchange. The twenty companies selected for the study are Fortes, Ashaka Cement, Cadbury Nig. Plc, Conoil, Flour Mills, Guinness Nig. Plc, John Holt, Julius Berger, Mobil Oil Nig. Plc, Nestle Foods, Nigerian Breweries, Oando, PZ, Texaco, Total Nig Plc, UAC, Unilever, Lafarge Cement, Transnational Corporation and Dangote Sugar.

The variables used in the study to capture firm performance are the rate of return on assets (RRA), the asset turn ratio (ATR) and portfolio activity & resilience (PAR). RRA is usually obtained by simply dividing the firm’s profits by the total assets of the business while ATR is obtained by dividing the firm’s sales revenue by the assets employed in the business. The third measure, PAR, is obtained by dividing the percentage change in sales by the percentage change in GDP. These three measures produce excellent metrics of assessing the firms’ performance over a number of years and of comparing several companies.



ISBN 978-0-9925622-3-6

657

Any empirical analysis of the impact of exchange rate volatility on corporate performance has to control for the influence of other economic variables that are correlated with exchange rates. The macroeconomic variables that move in tandem with exchange rates are Crude Oil Price, Prime Lending Rate, Imports, Federal Reserves and Total Government Expenditure. Accordingly, they are all included as explanatory variables alongside the exchange rate volatility variable in our three panel models for this study. As well, the exchange rate volatility variable used is the square of the mean adjusted relative change in the official exchange rate. The data set was obtained from Nigerian Stock Exchange Fact Book (2014), companies’ annual report and statements of accounts and the National Bureau of Statistics Nigeria. Panel Unit Root Test According to Pesaran and Smith (1995), Pesaran (1997), and Pesaran & Shin (1999), the ARDL approach is valid irrespective of whether the regressor are endogenous or exogenous, and regardless of whether the variables are I (0) or I (1). In order to guarantee appropriate specification, this study execute panel unit root tests on the dependent and independent variables, following the approach of Im, Pesaran, and Shin (IPS) (1995) who established a panel unit root test for the joint null hypothesis that every time series in the panel model is non-stationary. Moreover, the Im, Pesaran and Shin (2003) (IPS) test is adopted because the companies are heterogeneous1. The IPS test is based on this model:

ititjtiij

p

jtiiit XYYY i ,11, (1) For i=1,…,N and t=1,…,T. Equation (8) is an error correction model. If |∂i|<1, the series is trend stationary. The Dynamic Panel Model The purpose of this section is to construct models of the relationship between exchange rate volatility and firm performance in Nigeria, as described by equations (2-4).

itititititititit uTEXPRESVOILPIMPTPLREXCRVRRA 7654321 (2)

itititititititit uTEXPRESVOILPIMPTPLREXCRVATR 7654321 (3)

itititititititit uTEXPRESVOILPIMPTPLREXCRVPAR 7654321 (4)

Where the subscript i denotes the ith company (i = 1,...,20) and the subscript t denotes the tth year (t = 1,...,10). RRAit is the Rate of Return on Assets for company i at time t. ATRit is the Asset Turn Ratio for company i at time t. PARit is the Portfolio Activity & Resilience for company i at time t. EXCRVt is the Exchange Rate Volatility at time t. PLRt is the Prime Lending Rate at time t. OILPt is the log of Crude Oil Price at time t. IMPTt is the Import as a percentage of GDP at time t. RESVt is the log of Federal Reserves at time t. TEXPt is the log of Total Government Expenditure at time t. RRA, ATR, PAR, EXCRV and PLR are not used in log forms because they are either percentages, ratios or rates.

In order to capture the dynamic processes between exchange rate volatility and firm performance in Nigeria, a dynamic panel data analysis method was used. This method has a lot of advantages: a dynamic panel data analysis factors in the dynamic processes between the dependent and independent variables (Baltagi, 1995); panel data estimation controls for both missing and unobserved variables/relationships; it allows for identification of company-specific effects (Arellano-Bond, 1991; Matyas & Sevestre, 1996); and dynamic effects and feedback from current or past shocks are captured in the model.

According to Baltagi (2005, pp. 135), Many economic relationships are dynamic in nature and one of the advantages of panel data is that they allow the researcher to better understand the dynamics of adjustment. Let yit be the dependent variable in company i, and xit be the vector of company-specific regressors. Then, a simple dynamic panel data model in levels can be represented as (Hsiao, 2003, pp. 75):

itittiit xyy 1, i = 1,. . . N; t = 1,. . . T. (5)

∂ is a scalar, μi denotes the ith individual’s effect. The uit follows a one-way error component model such that, uit = ηi + vit. (6)

1 Panel unit root tests are divided into two, based on the assumption of homogeneity and heterogeneity. Examples of studies based on the assumption of a homogeneous model are Breitung (2000), Hadri (2000) and Levin, Lin and Chu (2002). Studies based on the assumption of a heterogeneous model are Im, Pesaran and Shin (2003), Maddala & Wu (1999), Choi (2001).



ISBN 978-0-9925622-3-6

658

where ηi ∼ IID(0,σµ2) and νit ∼ IID(0,σν

2) independent of each other and among themselves (Baltagi, 2005). µi is a vector of unobserved common factors. Further, it is assumed that E(ηi) = 0, E(vit) = 0, E(vitηi) = 0 for all i = 1, . . . , N and t = 2, . . . , T . (7) E(vitvis) = 0 for all i = 1, . . . , N and t ≠ s (8) E(yi1vit) = 0 for all i = 1, . . . , N and t = 2, . . . , T . (9)

An assumption of no correlation between the regressors and the composite error term has been made. On the other hand, the inclusion of the lagged dependent variable yt-1 in the models breaks down the condition of zero correlation between explanatory variables and the error term. This is better explained by Baltagi (2005, pp. 135) that:

The dynamic panel data regression is characterized by two sources of persistence over time. Autocorrelation due to the presence of a lagged dependent variable among the regressors and individual effects characterizing the heterogeneity among the individuals.

In order to ensure a convincing and robust estimation of the equation (11), Arellano & Bond (1991) proposed the generalized method of moments (GMM) estimator. The benefit of the GMM estimator lies in the ability to sweep the across-time individual-specific effect by taking first differences:

itititit vxyy 1 (10)

Where ∆yit = yit − yit−1 for i = 1. . . N and t = 2. . . T.

Nonetheless, this estimation method is further extended by Arellano & Bover (1995) and Blundel & Bond (1998). Blundell & Bond (1998) applies a set of moment conditions using the first differenced equation, and another set of moment conditions in levels. The first differences of two or more period lagged-dependent variables are valid instruments in levels; as well, two or more period lagged-dependent variables in levels are relevant instruments in first differences (Blundell & Bond, 1998; Blundell et al., 2000). Moreover, in order to generate more instrumental variables for estimation, some or all the explanatory variables (xmit) are either predetermined or exogenous.

In this study, we acknowledge that firm performance is very likely to be correlated with the firm-specific effects and the shocks to the firm performance in the previous periods. Therefore, we use the following moment conditions to identify the valid instruments in first differences: E(yit−s∆uit) = 0 for t = 3, . . . , T and 2 ≤ s ≤ t − 1 (11) E(xit− s∆uit) = 0 for t = 3, . . . , T and 1 ≤ s ≤ t − 1 (12) In addition, to identify the instruments in levels, we use the moment conditions: E(uit∆yit−1) = 0 for t = 3, . . . , T (13) and E(uit∆xit−1) = 0 for t = 3, . . . , T (14) This is the idea behind Arrelano-Bover and the GMM estimators are consistent for large N and finite T, and therefore more efficient than the Arellano & Bond estimator.

Since an important assumption of the validity of GMM estimation is that the instruments are exogenous, we confirm validity of the instruments using the Sargan test. Further, an important assumption of the consistency of the GMM estimator is that the idiosyncratic errors are serially non-correlated. Therefore, we use the Arellano & Bond (1991) test to check for second-order autocorrelation. As well, we apply small-sample corrections to the covariance matrix estimates, and the standard errors, which are robust to heteroskedasticity. Therefore, our panel specification allows for a significant degree of cross-company heterogeneity, due to the fact that the effect of exchange rate volatility on corporate performance could vary across companies, depending on company-specific factors such as efficiency, management and assets. Since our major goal in this study is to determine the impact of exchange rate volatility on corporate performance in Nigeria, our methods do not dwell on the specific dynamics that might be germane to a specific company. EMPIRICAL INVESTIGATION Panel Unit Root Test An important issue before making the appropriate specifications, often ignored by previous studies, is to determine if the variables are stationary or not. We carried out IPS panel unit root tests on the dependent and independent variables; the obtained results are as shown in Table 3. The results show that we can reject the null hypothesis of a unit root in favour of stationarity at the 5% level of significance level. Hence, we can safely begin the panel data estimation.



ISBN 978-0-9925622-3-6

659

TABLE 1: PANEL UNIT ROOT TESTS

Source: Authors’ calculation using STATA 11

Notes: By Schwarz criterion, the lag length was 1. (**) and (*) indicate stationarity at significance

levels 1% and 5% respectively.

The panel unit root tests has established that the variables are I(0) or I(1). The dynamic approach is valid

regardless of whether the regressors are endogenous or exogenous, and regardless of whether the variables are I

(0) or I (1) (Pesaran & Smith, 1995; Pesaran, 1997; Pesaran & Shin, 1999).

Dynamic Panel Estimation

Each of the three performance measures of the firms is regressed on Exchange Rate Volatility, Crude Oil Price,

Prime Lending Rate, Imports as a % of GDP, Reserves and Total Government Expenditure in order to examine

the contemporaneous effect of Exchange Rate volatility on firm performance The Least Squares estimates

obtained are reported for two cases2:

(a) Arellano-Bond dynamic panel-data and,

(b) Arrelano-Bover/Bundell-Bond system dynamic panel-data.

Panel estimates of the effects of Exchange Rate Volatility on the Rate of Return on Assets using both Arrelano-

Bond & Arellano-Bover GMM estimation methods are shown in Table 2.

2 Individual company estimates are available on request, but note that they are likely to be individually unreliable considering the fact that the time dimension of the panel is relatively small.

Variables IPS Statistics Prob. Values

RRA -2.915* * 0.002

OILP -2.152* 0.016

PAR -4.089** 0.000

ATR -2.933** 0.002

EXCRV -3.636** 0.000

PLR -2.058* 0.020

IMPT -3.142** 0.001

RESV -13.046** 0.000

TEXP -4.209** 0.000



ISBN 978-0-9925622-3-6

660

TABLE 2: PANEL ESTIMATES OF THE EFFECTS OF EXCHANGE RATE VOLATILITY ON THE

RATE OF RETURN ON ASSETS, 2004-2013

Arelano-Bond Dynamic Panel Arrelano-Bover/Blundell-Bond System dynamic panel

Lagged RRA 0.024

(0.645)

0.007

(0.908)

EXCRV -0.134**

(0.002)

-0.145*

(0.024)

PLR -0.105

(0.549)

-0.093

(0.628)

IMPT 0.122*

(0.040)

0.134*

(0.047)

OILP -0.068*

(0.025)

-0.072*

(0.043)

RESV -0.032*

(0.034)

-0.035**

(0.004)

TEXP 3.047*

(0.011)

4.142*

(0.030)

N 160 180

Wald χ2 1693.68* 1384.42*

Sargan test 97.307 77.241

AB test -0.324 -0.293

Source: Authors’ calculation using STATA 11.

Notes: The (**) signifies variable significant at 1%, (*) significance at 5%. Values in brackets are probabilities.

The values in parentheses are probabilities. AB test is Arellano and Bond test for AR(2). The Sargan test reports

that under the null the over-identified restrictions are valid. The estimations were conducted with two-step

efficient GMM and small sample corrections to the covariance matrix estimate.

The results across both Arrelano-Bond & Arrelano-Bover/Blundell-Bond System GMM specifications

are not materially different. The estimates suggest an inverse relationship between Exchange Rate volatility and

Rate of Return on Assets. The coefficients of Exchange Rate volatility are negative and always statistically

significant, with their values ranging from -0.134 (the Arrelano-Bond model) to -0.145 (Arrelano-

Bover/Blundell-Bond model).

From the estimation results of both models, the one period lagged Rate of Return on Assets has a

positive but insignificant effect on the current Rate of Return on Assets, suggesting a weak adjustment dynamics

in the effect and behavior of previous Rate of Return on Assets.

No significant effect is observed for Premium Lending Rate, suggesting no impact of Premium

Lending Rate on the Rate of Return on Assets. Premium Lending Rate seems to be less important. Imports has a

particularly positive and significant impact on the Rate of Return on Assets. Based on our Arrelano-

Bover/Blundell-Bond model, increases in Imports lead to an increase (coefficient of 0.134) in the Rate of Return

on Assets. As well, Federal Reserves and Total Government Expenditure are important factors when explaining

the Rate of Return on Assets. What is interesting, though, is that the Federal Reserves has a negative significant

impact on the Rate of Return on Assets. That is, the higher the Federal Reserves, the lower the Rate of Return

on Assets.

For both models, the Sargan test rejects the null of misspecification. As well, the Arellano–Bond (AB)

second order autocorrelation test rejects the null of serial correlation in the idiosyncratic error.

Panel estimates of the effects of Exchange Rate Volatility on the Asset Turn Ratio using both Arrelano-

Bond & Arellano-Bover GMM estimation methods are shown in Table 3.



ISBN 978-0-9925622-3-6

661


ASSET TURN RATIO, 2004-2013

Arelano-Bond Dynamic Panel Arrelano-Bover/Blundell-Bond System

dynamic panel

Lagged ATR 0.723*

(0.000)

0.113*

(0.000)

EXCRV -0.137*

(0.002)

-0.113**

(0.048)

PLR -0.009*

(0.862)

-0.058*

(0.551)

IMPT 0.010

(0.814)

0.012*

(0.900)

OILP -0.051

(0.180)

-0.045

(0.359)

RESV -0.037*

(0.012)

-0.028**

(0.238)

TEXP 2.972*

(0.016)

1.359*

(0.251)

N 160 180

Wald χ2 278.15* 773.98*


AB test -0.262 -0.389




that under the null the overidentified restrictions are valid. The estimations were conducted with two-step


As shown in Table 3, the results across both Arrelano-Bond & Arrelano-Bover/Blundell-Bond System

GMM specifications are not too different. The estimates suggest an inverse relationship between Exchange Rate

volatility and Asset Turn Ratio. The coefficients of Exchange Rate volatility are negative and always

statistically significant, with their values ranging from -0.137 (the Arrelano-Bond model) to -0.113 (Arrelano-

Bover/Blundell-Bond model). In other words, the higher the exchange rate volatility, the lower the Asset Turn

Ratio.

From the estimation results of both models, the one period lagged Asset Turn Ratio has a positive and

significant effect on the current Asset Turn Ratio, suggesting a strong adjustment dynamics in the effect and

behavior of previous Asset Turn Ratio. This suggests that companies with less than Asset Turn Ratio would not

experience persistent decline. The coefficients of the lagged Asset Turn Ratio are between zero and one,

implying partial catch-up.

As well, Premium Lending Rate, Oil Price and Reserves have a significant negative impact on Asset

Turn Ratio. That is, the higher the Premium Lending Rate, Oil Price and Reserves, the less the Asset Turn Ratio.

Of particular interest is the fact that Reserves has a negative significant impact on the Rate of Return on Assets.

That is, the higher the Federal Reserves, the lower the Rate of Return on Assets. No significant effect is

observed for Imports, suggesting no impact of Imports on the Asset Turn Ratio. That is, Imports seems to be

less important. Total Government Expenditure has a particularly positive and significant impact on the Asset

Turn Ratio. For both models, the Sargan test rejects the null of misspecification. As well, the Arellano–Bond

(AB) second order autocorrelation test rejects the null of serial correlation in the idiosyncratic error.

Panel estimates of the effects of Exchange Rate Volatility on the Portfolio Activity & Resilience using

both Arrelano-Bond and Arellano-Bover GMM estimation methods are shown in Table 4.



ISBN 978-0-9925622-3-6

662


PORTFOLIO ACTIVITY & RESILIENCE, 2004-2013

Arelano-Bond Dynamic Panel Arrelano-Bover/Blundell-Bond System

dynamic panel

Lagged ATR 0.171*

(0.037)

0.117**

(0.000)

EXCRV -0.701*

(0.042)

-0.532**

(0.000)

PLR -2.417**

(0.000)

-1.113*

(0.047)

IMPT 0.272

(0.494)

0.110

(0.402)

OILP -0.401*

(0.035)

-0.251**

(0.006)

RESV -0.213**

(0.001)

-0.164**

(0.000)

TEXP 2.035*

(0.024)

1.130**

(0.003)

N 160 180

Wald χ2 415.21** 975.73*


AB test -0.125 -0.17




that under the null the overidentified restrictions are valid. The estimations were conducted with two-step


Table 4 shows the estimates for the impact of Exchange Rate Volatility on the Portfolio Activity &

Resilience. The results across both Arrelano-Bond and Arrelano-Bover/Blundell-Bond System GMM

specifications are similar. The estimates suggest an inverse relationship between Exchange Rate volatility and

Portfolio Activity & Resilience. The coefficients of Exchange Rate volatility are negative and always

statistically significant, with their values ranging from -0.701 (the Arrelano-Bond model) to -0.117 (Arrelano-

Bover/Blundell-Bond model). In other words, the higher the exchange rate volatility, the lower the Portfolio

Activity & Resilience.

From the estimation results of both models, the one period lagged Portfolio Activity & Resilience has a

positive and significant effect on the current Portfolio Activity & Resilience, suggesting a strong adjustment

dynamics in the effect and behavior of previous Portfolio Activity & Resilience. This suggests that companies

with less than adequate Portfolio Activity & Resilience would not experience persistent decline. The coefficients

of the lagged Portfolio Activity & Resilience are between zero and one, implying partial catch-up.

As well, Premium Lending Rate, Oil Price and Reserves have a significant negative impact on Portfolio Activity

& Resilience. That is, the higher the Premium Lending Rate, Oil Price and Reserves, the less the Portfolio

Activity & Resilience. No significant effect is observed for Imports, suggesting no impact of Imports on

Portfolio Activity & Resilience. That is, Imports seems to be less important. Total Government Expenditure has

a particularly positive and significant impact on Portfolio Activity & Resilience. For both models, the Sargan

test rejects the null of misspecification. As well, the Arellano–Bond (AB) second order autocorrelation tests

rejects the null of serial correlation in the idiosyncratic error.

Overall, this study suggests that exchange rate volatility and other macroeconomic parameters (i.e. premium

lending rate, oil price and reserves) adversely affect rate of return on assets, asset turn ratio, portfolio activity &

resilience and, consequently, firm performance, consistent with the empirical literature.



ISBN 978-0-9925622-3-6

663

CONCLUSION, LIMITATIONS AND FUTURE RESEARCH

With three dynamic panel models, we investigated the impact of Exchange Rate volatility on firm performance

in Nigeria. The models include firm efficiency dependent variables such as Rate of Return on Assets, Asset

Turn Ratio, and Portfolio Activity & Resilience calculated from data for 20 most active companies listed on the

floor of Nigerian Stock Exchange. This study is also a first for Nigeria since previous models use the profit after

tax and cost of goods sold as a proxy for firm performance. One statistically significant economic relation

characterizes the Arelano-Bond Dynamic Panel and Arrelano-Bover/Blundell-Bond System Dynamic Panel

models adopted in this study. This is: exchange rate volatility has significant negative impacts on the rate of

return on assets, asset turn ratio and the portfolio activity & resilience, thus, showing the significant negative

impact of exchange Rate volatility on firm performance in Nigeria between 2004 and 2013. Overall, this study

suggests that the higher the exchange rate volatility, the less firm efficiency and performance.

A number of policy implications can be drawn from this analysis for investors and financial market

participants. Because all firms are not uniformly susceptible to exchange rate volatility, risk diversification

possibilities across industries are recommended. Information on firm vulnerability, relative immunity or strength

in the face of exchange rate volatility can be used to inform portfolio strategies on exchange rate risk. When

exchange rate shocks are imminent or the foreign exchange environment changes, investors and market

participants can alter or rebalance their portfolios with stocks of dissimilar firms by looking at the response of

the firms to exchange rate volatility.

Despite the robust findings of this study, this study faced a number of issues such as strong

homogeneity assumptions across the companies, and not factoring in the dynamics of the feedback effects from

firm performance to exchange rate volatility, and the error cross-sectional dependencies that exist across the

firms, due to unobserved common factors that tend to increase at times of currency crises. Due to the intrinsic

cross-company heterogeneities, it can be said that the effect of exchange rate volatility on firm performance may

be firm-specific and estimation of a panel model that pooled together all observations across the companies may

not be helpful to investors intent on a specific company and their usage could even be misleading. Actually, it is

possible in a number of ways to relax the homogeneity assumption, but there is one obvious challenge when it

comes to estimation of specific firms: the non-linearity of the relationships, identification and estimation for a

particular firm may require much more time series data than are presently obtainable.

The findings of the study offer several avenues for future research. For future research, we will investigate

whether and how exchange rate volatility affects firm performance in Sub-Saharan Africa. To explore this, we

will rebuild the three dynamic panel models used in this study into a customised dynamic panel model for Sub-

Saharan Africa, where all the efficiency variables are determined from individual country stock exchange data

for the biggest companies in Sub-Saharan Africa.

REFERENCES

Aquino, RQ 2006, ‘A Variance Equality Test of the ICAPM on Philippine Stocks: Post-Asian Financial Crisis Period’, Applied Economics, vol. 38, no. 3, pp. 353-362.

Arellano, M & Bond, S 1991, ‘Some tests of specification for panel data: Monte carlo evidence and an application to employment equations’, Review of Economic Studies’, vol. 58, no. 2, pp. 277–297

Auer, R & Chaney, T 2007, ‘How Do the Prices of Different Goods Respond to Exchange Rate Shocks? A Model of Quality Pricing-to-Market’, University of Chicago and NBER Working Paper, December. Auer, R & Chaney, T 2009, ‘Exchange Rate Pass-Through in a Competitive Model of Pricing-to-Market’,

Journal of Money, Credit and Banking, vol. 41, no. 1, February, pp. 151-175. Babatunde, MA & Olaniran, O 2009, ‘The effects of internal and external mechanism on governance and

performance of corporate firms in Nigeria’, Corporate ownership & control, vol. 7, no. 12, pp. 330-343.

Bae, K, Chan, K & Ng, A 2004, ‘Investibility and return volatility’, Journal of Financial Economics, vol. 71, no. 2, pp. 239-263.

Baggs, J, Beaulieu, E, Fung, L, & Lapham, B 2011, ‘Exchange Rate Movements and Firm Dynamic in Canadian Retail Industries’, Econometrics society, North America, Winter Meeting.

Baltagi, BH 1995, Econometric analysis of panel data, New York, John Wiley and Sons. Bartelsman, EJ & Doms, M 2000, ‘Understanding productivity: lessons from longitudinal micro databases’,

Journal of Economic Literature, vol. 38, no. 3, pp. 569-594 Bartov, E, & Bodnar, GM 1994, ‘Firm Valuation, Earnings Expectations, and the Exchange-Rate Exposure

Effect’, Journal of Finance, vol. 49, no. 5, pp. 1755-1785. Baum, CF, Caglayan, M & Barkoulas, JT 2001, ‘Exchange rate uncertainty and firm profitability. Journal of

Macroeconomics’, vol. 23, no. 4, pp. 565-576.



ISBN 978-0-9925622-3-6

664

Berman, N, Martin, P, & Mayer, T 2008, ‘How do different firms react to exchange rate changes? Prices, quantities, entry and exit’, Paris School of Economics, University of Paris-1 Pantheon Sorbonne and CEPR Working Paper, December.

Bernard, HJ & Galati, GE 2000, ‘The Co-movement of US Stock Markets and the Dollar’, BIS Quarterly Review, pp. 31-34.

Blundell, R & Bond, S 1999, ‘GMM estimator with persistent panel data: an application to production functions’, IFS Working Paper No. W99/4.

Blundell, R, & Bond, S 2000, ‘GMM estimation with persistent panel data: an application to production functions’, Econometric reviews, vol. 19, no. 3, pp. 321-340.

Breitung, J 2000, ‘The local power of some unit root tests for panel data, In B. Baltagi (ed.), Nonstationary Panels, Panel Cointegration, and Dynamic Panels’, Advances in Econometrics, vol. 15, pp. 161-178. .

Campa, JM, Goldberg, LS & Gonzalez-Minguez, JM 2005, ‘Exchange rate pass-though to import prices in the Euro area’, Staff Report 219, Federal Reserve Bank of New York.

Chatterjee, A, Carneiro, RD & Vichyanond, J 2010, ‘Multi-Product Firms and Exchange Rate Fluctuations’, Princeton University Working Paper, August.

Choi, I 2001, ‘Unit roots tests for panel data’, Journal of International Money and Finance, vol. 20, no. 02, pp. 229–272.

Choi, JJ, Hiraki, T & Takezawa, N 1998, ‘Is Foreign Exchange Risk Priced In The Japanese Stock Market?’, Journal of Financial and Quantitative Analysis, vol. 33, no. 03, pp. 361-382.

Clerides, SK, Lach, S & Tybout, J 1998, ‘Is learning by exporting important? Micro-dynamic evidence from Colombia, Mexico and Morocco’, Quarterly Journal of Economics, vol. 113, no.3, pp. 903-947.

Corsetti, G & Dedola, L 2005, ‘A Macroeconomic Model of International Price Discrimination’, Journal of international Economics, vol. 67, no.1, pp. 129-155.

Dellas, H & Hess, M 2002, ‘Financial Development and Stock Returns: A Cross Country Analysis, CEPR Discussion Paper, no. 3681 (January).

Doukas, J, Hall, P. & Lang, L. 1999, ‘The Pricing of Currency Risk in Japan’ Journal of Banking and Finance, vol. 23, no. 1, pp. 1-20.

Du, J & Wei, S 2004, ‘Does insider trading raise market volatility?’, The Economic Journal, vol. 114, no. 498, pp. 916-942.

Dumas, B & Solnik, B 1995 ‘The World Price of Foreign Exchange Risk’, Journal of Finance, vol. 50, no. 2, pp. 445-477.

Farinas, JC & Martin-Marcos, A 2003, ‘Exporting and Economic Performance: Firm-level evidence for Spanish Manufacturing’, Centre for Research on Globalisation and Economic Policy, Working Paper, University of Nottingham.

Fung, L 2004, ‘Large real exchange rate movements, firm dynamics, and productivity growth, mimeo’, University of Alberta, Canada.

Girma, S, Greenaway D & Kneller, R 2002 ‘Does exporting lead to better performance? A micro-econometric analysis of matched firms’, GEP Research Paper 02/09, University of Nottingham.

Gopinath, G, Oleg, I & Rigobon, R 2010, ‘Currency Choice and Exchange Rate Pass-through’, American Economic Review, vol. 100, no. 1, March, pp. 304-336.

Green, CJ, Maggioni, P & Murinde, V 2000, ‘Regulatory lessons for emerging stock markets from a century of evidence on transactions costs and share price volatility in the London Stock Exchange’, Journal of Banking & Finance, vol. 24, no. 4, pp. 577-601.

Greer, CR, Ireland, TC & Wingender, JR 2001, ‘Contrarian human resource investments and financial performance after economic downturns’, Journal of Business Research, vol. 52, no. 3, pp. 249-261.

Hadri, K 2000, ‘Testing for stationarity in heterogeneous panels’, Econometrics Journal, vol. 3, no. 2, pp. 148-161

Harris, G. (2001) Is there a Case for Exchange Rate Induced Productivity Changes? Canadian Institute for Advanced Research Working Paper 164.

Hsiao, C 2003, Analysis of panel data, 2nd Edition, Cambridge: Im, K., Pesaran H & Shin, Y 2003, ‘Testing for unit roots in heterogeneous panels’, Journal of Econometrics,

vol. 115, no. 1, pp.53–74. Im, K., Pesaran, H, & Shin, Y 1995, ‘Testing for Unit Roots in Heterogeneous Panels’, DAE Working Papers

Amalgamated Series, no. 9526, University of Cambridge. Jorion, P 1990 ‘The Exchange Rate Exposure of U.S. Multinationals’, Journal of Business, vol. 63, pp. 331-45. Jorion, P 1991, ‘The Pricing of Exchange Rate Risk in the Stock Market’, Journal of Financial and Quantitative

Analysis, vol. 26, no. 03, pp. 363-376. Jun, SG, Marathe, A & Shawky, H 2003, ‘Liquidity and stock returns in emerging equity Markets’, Emerging

Markets Review vol. 4, no. 1, pp. 1-24. Klitgaard, T 1999, ‘Exchange Rates and Profit Margins: The Case of Japanese Exporters’, FRBNY Economic

Policy Review, Federal Reserve Bank of New York, April. Lesmond, D 2005, ‘Liquidity of emerging markets’, Journal of Financial Economics vol. 77, no. 2, pp. 411-452



ISBN 978-0-9925622-3-6

665

Levin, A, Lin C & Chu, J 2002, ‘Unit roots tests in panel data: asymptotic and finite sample properties’, Journal of Econometrics, vol. 108, no. 1, pp. 1-24

Maddala, GS & Wu, S 1999, ‘A comparative study of unit root tests with panel data and a new simple test’, Oxford Bulletin of Economics and Statistics, vol. 61, pp. 631-652.

Matyas, L & Sevestre, P 1996, The econometrics of panel data, 2nd Edition. Min, J 2002, ‘Program trading and intraday volatility in the stock index futures market and spot market: The

case of Korea’, Journal of Asia-Pacific Business, vol. 4, no. 3, pp. 53-68. Navarro, P, Bromiley, P, & Sottile, P 2010, ‘Business cycle management and firm performance: Tying the

empirical knot’, Journal of Strategy and Management, vol. 3, no. pp. 1, 50-71. Park, A, Yang, D, Xinzheng, S & Jiang, Y 2006 ‘Exporting and firm performance: Chinese Exporters and the

Asian Financial Crisis’, Research Seminar in International Economics, University of Michigan, Discussion Paper No. 549.

Patro, DK, Wald, JK & Wu, Y 2002, ‘Explaining exchange rate risk in world stock markets: A panel approach’, Journal of banking & finance, vol. 26, no. 10, pp. 1951-1972.

Shapiro AC 1975, ‘Exchange Rate Changes, Inflation, and the Value of the Multinational Corporation’, J. Financ., vol. 30, no. 2, pp. 485-502.

Thille, H 2006, ‘Inventories, market structure, and price volatility’, Journal of Economic Dynamics and Control, vol. 30, no. 7, pp. 1081-1104.

Tse, Y, Wu, C, & Young, A 2003, ‘Asymmetric information transmission between a transition economy and the U.S. market: evidence from the Warsaw Stock Exchange’, Global Finance Journal, vol. 14, no. 3, pp. 319-332.

Yau, HY & Nieh, C 2006, ‘Interrelationships among Stock Prices of Taiwan and Japan and NTD/Yen Exchange Rate’, Journal of Asian Economics, vol. 17, no. 3, pp. 535-552.

Yeh,Y, Lee,T, Pen, J 2002, ‘Stock Returns and Volatility under Market Segmentation: The Case of Chinese A and B Shares’, Review of Quantitative Finance and Accounting, vol. 18, no. 3, pp. 239-257.

exchange rate volatility and firm … fluctuation in exchange rates became an ... growth has...

Documents