Bribery analysis through firm size, firm age and gender of the top manager

(1)

Bribery analysis through firm size, firm age and

gender of the top manager

MSc Thesis International Business & Management

University of Groningen

International Business and Management Faculty of Economics and Business

Nettelbosje 2 9747 AE Groningen The Netherlands June 20, 2014 By Jelmer Oenema S2231476 j.oenema.1@student.rug.nl J.C. Kapteynlaan 33A 9714 CM Groningen +31 651520820

(2)

Abstract

Nowadays, corruption is a common problem in doing business. This research extends existing literature on determinants of corruption by investigating the effect of the factors firm size, firm age and gender of the top manager on bribery. This might give new insights to firms and managers which can help them in the future while doing business. The relationship between the three factors and bribery is measured by using firm-level data from 27 different countries in Eastern Europe and Central Asia. A Logistic regression is performed over a total of 16243 observations, in order to empirically test the hypotheses about the relationship between the three factors and bribery. Institutional theory, anomie theory and social role theory are used to draw the analysis. The findings of this study show a positive and significant relationship between firm size and bribery. Furthermore, the findings show that male top managers are significantly more likely to bribe than female top managers. However, no evidence is found for a relationship between bribery and firm age.

(3)

1. Introduction

The above words are from Huguette Labelle, chair of the board of directors of ‘Transparency International’, an organization that monitors corruption. That organization publishes a list of worldwide corruption every year, called the Corruption Perceptions Index (CPI). According to Transparency International (2014), many companies hide corrupt acts behind secret partnerships and subsidiaries, while others exploit tax laws, construct cartels or abuse legal loopholes. Even in the modern world there are no exceptions. The Bribe Payers Index (2011) of Transparency International shows us that corruption is everywhere. Bribing occurs in small countries, large countries, developing economies and developed economies. None of the countries on the Bribe Payers Index (2011) is bribing free. That makes corruption a worldwide problem. However, there are differences between and within countries in the level of corruption. A research on corruption between countries is called country level research, whereas a research on corruption within countries is better known as firm level research. Moreover, if we measure the influence of a top manager on corruption we speak about an individual level research. This study contains a research on the latter two: firm level and individual level.

(6)

6 corruption delays private investment and so affects growth of the economy. Furthermore, Fisman and Svensson (2007) prove that bribing firms have worse financial outcomes. The reasons why some empirical findings show that corruption enhances growth and why some corrupt countries have high foreign direct investment rates are still obscure.

(7)

7 focus on the demand side, our research provides additional insights into the influence of firm characteristics on bribery using new, exploratory data. As mentioned before, the influence of firm characteristics on bribery has mainly been researched from the demand side and thus creates support for further research. Furthermore, existing research on the influence of individual characteristics on bribery includes the factors education, work experience and income of the manager. Other individual factors are underexposed in existing literature and would therefore be worthy to research. In consequence of this, this study is setup to examine the relationship and link between other determinants and bribery. The other, in current literature underexposed determinants, are firm size, firm age (both firm-level) and gender of the top manager (individual-level). This study has chosen to test firm size and firm age due to previous literature. Many authors (e.g. Glancey, 1998; Davidsson et al., 2002; Almus & Nerlinger, 2000; Wijewardena & Tibbits, 1999) find a negative relationship between firm growth and both firm size and firm age. Due to the fact that corruption limits growth it will be interesting to study the direct impact of firm size and firm age on corruption. Furthermore, individual characteristics as determinants of corruption are not much researched yet. Therefore this study chose to add gender of top manager as third independent variable. Moreover, research on linking firm-level data with individual-level data is rare and could therefore lead to interesting and unveiling outcomes. Our study tends to further this new field of research by addressing the research gap of the relation between firm and individual characteristics and bribery on the supply side.

(8)

8 of gender on corruption at the demand side. From that point of view it is argued that males are more corrupt than females (Dollar et al., 2001; Swamy et al., 2001; Torgler & Valey, 2006). However, in our research we will look at the supply side of bribery, hence it will offer a better understanding of the relationship between corruption and gender. The focus of this study will be on how these three determinants might influence corruption and how all the factors are interrelated. This study will use data from 27 Eastern European and Central Asian countries (list of countries can be found in Appendix 1), because countries from that areas are usually ranked among the most corrupt countries. According to Tonoyan et al. (2010), the reason for this is that corruption is prevalent in transition economies. All the 27 countries in this study are officially defined as a transition economy. Therefore Eastern Europe and Central Asia are appropriate areas for this study, which will afford a better insight in why some firms do bribe more than others. Consequently, the main research question of this study is:

Why are some firms more corrupt than others?

In order to answer the main research question, this study will first analyse the existing literature about corruption. Subsequently, the hypotheses will be formulated. Thereafter, the methodology section will be introduced and afterwards the empirical results and conclusions will be drawn. In furtherance of answering the main research question this study has sub research questions, which are stated as follows:

What is corruption and how is it measurable? How can firm characteristics influence bribery? How can individual characteristics influence bribery?

(9)

(10)

10

2. Literature review

2.1 Corruption

(11)

11 have a bigger probability to be corrupt than developed countries. From the demand side of view, Schleifer and Vishny (1993) argue that weak governments with unstable political institutions have a tough time in avoiding their agents from demanding bribes. Furthermore, Yui and Lau (2008) state that, for an entrepreneur, it is extremely important to have political connections in less-developed economies. Bribery is an investment that entrepreneurs need to make in order to be successful in an institutionally weak transition economy (Peng & Heath, 1996). Svensson (2005) argues that corruption turns up in reaction to good-natured rules when individuals pay bribes to avoid penalties for injurious behaviour or when the monitoring of rules is deficient.

(12)

12 The amount and the size of the bribes depend among other on the characteristics of the firm and the individual manager, because even if firms face the same institutional environment, the paid amount of bribes may differ (Svensson, 2003). This is why country-level research cannot explain corruption differences within countries. Moreover, Meagher and Thomas (2004) find that within some governments corruption is systematic, while within others it is individualized. According to Aidis et al. (2012), corruption can be transformed into social norms and behaviour whereby it can be seen as an informal institution. An individual-level research is necessary to identify the foundation of bribery. For instance, Granovetter (2005) states that the socially superior person initiates the bribe. Hence, corruption can be researched at country-level, firm-level, and individual-level. Therefore, bribery can considerably vary, due to the various institutional conditions and pressures that a firm faces (Martin et al., 2007).

More and more countries are trying to eliminate corruption within their economy, because it gives firms an unfair unjustified advantage in securing their business (D’Souza, 2012). Therefore, the OECD started its own anti-bribery convention, existing of forty countries. According to Getz and Volkema (2001), (non)governmental and international organizations have undertaken comprehensive exertions to fight corruption and set up specific guidelines to prevent bribery and other unethical routines in international business.

2.2 Measuring corruption

(13)

13 A measurement of corruption on the individual level is necessary in order to clarify why there are differences in bribing by managers that operate in the same environment. Moreover, de Jong et al. (2012) state that micro-level research is required if you want to show bribery heterogeneity in a country. It is hard to collect useful data at the individual level, because top managers mostly do not want to talk about their own corrupt behavior. Reinikka and Svenson (2006) call it impossible to collect reliable data virtually. However, when some effort is spend on building a trustful relationship between the interviewer and manager, it becomes possible to integrate corruption questions. According to Reinikka and Svensson (2006), the questions about corruption need to be asked in different sections of a survey, in order to get trustworthy answers. Furthermore, the questions need to be formulated in an indirect approach, to avoid that the manager gets the feeling that the interviewer thinks he is corrupt. The Business Environment and Enterprise Surveys (BEEPS) succeeded to integrate corruption questions in their firm surveys, which lead to reliable answers. In the first survey sections of the BEEPS trust will be created, after which in one of the last sections of the survey the manager is asked about corrupt behavior and bribery. Jensen et al. (2010) argue that corruption is a sensitive subject for managers, which may cause governmental vengeance and therefore enhances the probability of false response of the managers. This hold up problem also accounts for the BEEPS research data.

The survey from BEEPS researches administrative corruption, bribery and to until what extent both aspects affect the formation of policies. Administrative corruption means the colluding of taxpayers and collectors to reduce remissions (Flatters & MacLeod, 1995). The measurement of BEEPS is able to investigate intra country corruption on an individual-level, which makes it an appropriate measurement.

2.3 Bribery at the firm and individual-level

(14)

14 differs per firm, even if the firms are active in the same market or environment. Chen et al. (2008) find that firm size, market competition and the volume of exports have influence on the probability of paying bribes. However, according to Martin et al. (2007), beside firm characteristics also the institutional context, business environment and individual characteristics play a role in the decision whether or not to bribe.

(15)

15 brings both sides together. According to them, in the public sector are the ones who demand the bribes and in the private sector are the ones who supply the bribes.

(16)

16

3. Development of hypotheses

3.1 Firm characteristics and bribery

This section includes the hypotheses about the relationship between bribery and the firm characteristics size and age.

3.1.1. The effect of firm size on bribery

(17)

17 the higher the pursue of interested actions by managers will be. These self-interested actions include aspects of management and administrative practice, which can be explained by institutional theory. According to Eisenhardt (1988), those practices become embedded in the organizational culture and therefore will be legitimized within the organization and society. Moreover, Rodriguez et al. (2005) state that bribery influences organizational legitimacy through its consequences on informal and formal businesses. Bribery becomes ingrained by this process of legitimization. Hence, the behavior of managers that pursue self-interests within their organization will lead to a higher ability to pay bribes.

Young et al. (2011) argue that small firms have less bargaining power than large firms and are therefore forced to pursue bribery methods, due to the fact that less bargaining power gives them limited options in managing rent-seeking processes. Furthermore, Peng and Qi Zhou (2011) state that small firms are likely to be affected harder by bribery than large firms, due to limited resources. Moreover, Jovanovic (1982) argues that large firms have better access to resources and are therefore able to have large economies of scale, which makes them more efficient than small firms. In terms of efficiency, by the use of firm-level data McArthur and Teal (2002) proved a relationship between low efficiency and corruption. Small firms are more likely to be inefficient than large firms and are therefore more likely to be corrupt. However, Jovanovic (1982) also reveals an opposite possibility. When a firm increases in size it may face coordination problems, which will cause a downgrade of efficiency and therefore a higher possibility to become exposed to corruption.

We can see that there are various meanings and findings of authors about the effects of firm size on bribery. Even the explanation of institutional theory leads to a doubt about the influence. However, based on determinants such as bargaining power, resources, efficiency and legal systems, it could be argued that smaller firms are showing a higher level of corruption than large firms. Athanasouli et al. (2012) researched the relationship between firm size and corruption at Greek firms and also found that firm size and growth are negatively associated with corruption. Hence, corruption will decrease when firm size increases. Therefore the following hypothesis will be:

(18)

18 3.1.2 The effect of firm age on bribery

In existing literature firm age is amply used as control variable, but not often as an independent variable. In our research firm age is an independent variable and in this section we will use anomie theory to explain firm age differences regarding bribery. According to Merton (1968), this theory explains how social and cultural systems of an organization force the individuals within the organization to engage in deviating forms of behavior. Merton (1968) argues that individuals might pursue social and cultural norms instead of institutional norms, especially when they face pressure. The deviating forms of behavior which the individuals might adopt do also include unethical forms of behavior. Unethical behavior does not mean that it is not functional and therefore organizational goals might be accomplished in an unethical, amoral, and anomie way (Merton, 1968). According to Martin et al. (2007), when pressure in organizations becomes higher, individuals within the organization may choose any possible way to achieve the goals, irrespective of legality. Bribery can be such a way. Regarding bribery, anomie theory explains the difference between young and old firms in the following manner. Because of the fact that in comparison younger firms lack in resources and skills, it becomes more difficult for such firms to gain competitive advantage (Thornhill & Amit, 2003). Furthermore, Thornhill and Amit (2003) state that older firms face less difficulties in obtaining returns than younger firms. Younger firms have a disadvantage of novelty; therefore it might occur that they engage in unethical behavior to achieve their main goals. This disadvantage of novelty entails a smaller relational network. Younger firms lack in external relations and must develop that to become a redoubtable player on the market. This external pressure creates engagement in deviant behavior. For that reason it is stated that younger firms are more likely to be corrupt than older firms. However, some authors argue that bribes rise when firms become older. For instance, according to Lee et al. (2010), younger firms are higher motivated to stay away from corruption and have a leaning to pay less bribes than older firms. Thus, in existing literature there is no universal consensus about the effects of firm age on corruption. Nevertheless, based on the anomie theory, there is a negative relationship and therefore as firms become older corruption will decrease. This lead to the following hypothesis:

(19)

19

3.2 Managerial traits and bribery: The role of sex

(20)

20 characteristics and males cultivate a more assertive and choice making character, it can be assumed that females will be less active in bribery than males. According to Eagly (1987), more equality of status between women and men would lead to a decrease in the effect of gender differences on corruption over time. However, Torgler and Valev (2006) have a contradicted view to social role theory. According to them, males are less willing to comply than females and are more likely to agree that corruption can be justified. They argue that there is no chance of a decline in gender differences over time, even if status between men and women becomes more equal. Nevertheless, since males are more assertive and better able to fulfil a leadership role than women, we would expect that males have a higher likelihood to bribe than women. Therefore, the third formulated hypothesis in this study is:

Hypothesis 3: Firms with male managers are more likely to pay bribes than firms with female managers.

3.3 Conceptual model and summary of hypotheses

This conceptual model shows the aim of this study on the basis of the hypotheses. The aim of this study is to get a better understanding of how firm size, firm age, and gender affect bribery at firms in various different economies. In our study we will do research on the effect of these factors within a firm on corruption. The factors in the model are called variables and contain firm size, firm age and gender of the top manager. We expect that the size of the firm, expressed in the number of employees, will have a negative effect on corruption. This means that if the firm size will become

Firm swwicountr y with Corruption H1 ─ H2 ─ H3

+

Gender of top manager Increase of firm age Increase of firm size

(21)

(22)

22

4. Data and method

4.1 Data Source

Data from www.enterprisesurveys.org will be used to test the hypotheses in this study. This website provides BEEPS data, which is an abbreviation for Business Environment and Enterprise Performance Surveys. The BEEPS is created by the European Bank for Reconstruction and Development in cooperation with the World Bank and comprises a database of firm information, gathered with the aid of surveys. Among other things BEEPS contains data on corruption, trade, taxation, sales, performance and technology. The answers on the survey questions are given by top managers and firm owners of registered firms. State-owned enterprises are not part of the survey participants. The data of BEEPS has an advantage over other databases because it contains data from emerging and developing countries. Since corruption is a serious problem in such countries the database of BEEPS is useful for this study. Only firms with five employees or more were asked by the World Bank to participate. Furthermore, the size of the economy has been determinative for the number of conducted surveys. In large-sized economies more surveys have been conducted than in small and medium-sized economies. BEEPS have split up two sections of surveys, namely service and manufacturing firms. Nevertheless, most of the survey questions are similar. Some of the questions about corruption are highly sensitive. Therefore, the questions are not asked by government workers to ensure the reliability of the answers.

(23)

23 sample per country varies per industry from 150 until 1320 and is separated by geographic region, industry and firm size. However, the used methodology differs across the countries, which is a limitation of the BEEPS.

4.2 Countries

BEEPS in total offers data on 135 countries from all over the world. This research will focus on 27 countries in Eastern Europe and Central Asia. Since corruption is prevalent in transition economies (Tonoyan et al., 2010), the geographical area of Eastern Europe and Central Asia is appropriate for conducting this research. In previous research many authors found evidence for the hypothesis that states that firms in transition economies are more likely to bribe than firms in developed economies. The geographical area of Eastern Europe and Central Asia can be seen as a transition area. Transition economies are changing from centrally planned economies to free market economies, which lead to economic liberalization and therefore the elimination of trade barriers.

According to the Corruption Perception Index (CPI) 20131, only four from the twenty-seven Eastern European and Central Asian countries in this study have a sufficient score on corruption (55 or higher). The total scores on the list range from 1 - 100, where 1 is very corrupt and 100 is not corrupt. The four countries which score 55 points or higher are Estonia, Poland, Lithuania and Slovenia. The most corrupt country of the 27 is Uzbekistan, with a score of 17. In total, the 27 countries are ranked from 28 to 168 on a total of 177 countries, with 1 being less corrupt and 177 being very corrupt. Table 1 shows the table with all the 27 countries and their CPI scores.

(24)

24 Table 1: Countries and their CPI ranking and scores

Ranked (1 = not corrupt, 177 = very corrupt) Country Score (1 = very corrupt, 100 = not corrupt) 28 Estonia 68 38 Poland 60 43 Lithuania 57 43 Slovenia 57 47 Hungary 54 49 Latvia 53 55 Georgia 49 57 Croatia 48 57 Czech Republic 48 61 Slovakia 47 67 Macedonia 44 67 Montenegro 44 69 Romania 43

72 Bosnia and Herzegovina 42

72 Serbia 42 94 Armenia 36 102 Moldova 35 111 Kosovo 33 116 Albania 31 123 Belarus 29 127 Azerbaijan 28 127 Russia 28 140 Kazakhstan 26 144 Ukraine 25 150 Kyrgyzstan 24 154 Tajikistan 22 168 Uzbekistan 17

In total 16,243 top managers and firm owners are surveyed by the BEEPS.

4.3 Dependent variable: Bribery

(25)

25 “It is said that establishments are sometimes required to make gifts or informal payments to public officials to “get things done” with regard to customs, taxes, licenses, regulations, services etc. On average, what percentage of total annual sales, or estimated total annual value, do establishments like this one pay in informal payments or gifts to public officials for this purpose?”

Since some questions about corruption are very sensitive, participants have the option of refusing the answer. Besides, there exists a probability that some answers to above question are not accurate. Therefore, outliers and unfilled answers in the survey are removed from the estimation. The data of above question is converted to a binary scale, which points out whether a firm bribes or not.

4.4 Key independent variables

The independent variables in this study are firm size, firm age and gender of the top manager. These variables may affect bribery. In order to scope the influence of the independent variables on bribery by firms, it is essential to operationalize the variables.

4.4.1 Firm size (H1)

In order to measure firm size, the following question will be used: ‘At the end of fiscal year 2009, how many permanent, full-time employees did this establishment employ?’ It is a continuous variable that is measured in the amount of employees per firm.

4.4.2 Firm age (H2)

The question: ‘In what year did this establishment begin operations?’ is used to measure the age of the firm, which is a continuous variable. The amount of years will be calculated through the use of next formula: firm age = conducted survey year – year of establishment.

4.4.3 Gender (H3)

(26)

26

4.5 Control variables

(27)

27 Table 2: Variables

4.6 Method assumptions

(28)

28 to Hill et al. (2009), heteroscedasticity leads to biased estimates of the standard errors, which in turn lead to incorrect hypotheses and anomalous results. For these reasons, this study will not test for assumptions of linearity, normality and homoscedasticity. However, this study will focus on testing for endogeneity and multicollinearity. 4.6.1 Endogeneity

Endogeneity occurs when the error term is correlated with one or more of the independent variables. If this is the case, the influence of the independent variables on the dependent variable will be constantly overestimated. According to Hill et al. (2009), there will never be a correlation between the independent variables and the residuals of the sample, since the residuals should be randomly taken. Hence, it will not be appropriate to use those independent variables for testing the assumption of endogeneity and therefore a theoretical analysis is needed to satisfy the assumption (Hill et al, 2009). Since all the data regarding to the variables in this study can be observed and judged by firms and management it is not expected that the variables in this study correlate with unobserved variables in the error term. However, if a problem of endogeneity should occur, there is a high chance that this is due to the judgment or a measurement error made by the respondents. In that case it is likely to assume that the error term is correlated with indicators that are based on the judgment by respondents. Unobserved variables could affect such indicators. Nevertheless, in the sample are no indicators present as independent variables. If this would be the case, there would be two options to master the endogeneity problem. The first one is two-stage least squares (2SLS) and the second one is generalized least squares (GLS). When it is tough to directly observe the indicators, instrumental variables can be used for the measurement (Hill et al., 2009). However, also this kind of variables is not present in the sample. It can be assumed that the independent variable is exogenous, since generic endogeneity tests depend on such instrumental variables that are not present in the sample of this study. Therefore, it has to be concluded that an endogeneity problem would not occur.

4.6.2 Multicollinearity

(29)

29 which in turn makes is difficult to see the relationship. Due to this problem the standard errors will inflate and the estimates of the regression model emerge as doubtful. An additional problem is that removing or adding one independent variable to the regression model will cause substantial changes to the model. In the worst case scenario, when there is substantial multicollinearity, the program may not find a solution to the analysis (Hill et al., 2009). There are two main approaches to measure multicollinearity. The first one is tolerance. Tolerance is the percentage of variance in the independent variable that is not accounted for by the other independent variables. The second approach to measure multicollinearity is the Variance Inflation Factor (VIF). The VIF gives an indication to the degree to which the standard errors will be inflated due to the levels of collinearity. According to Neter et al. (1985), if the values of the variables in the VIF exceed a value of 10, multicollinearity will be problematic. Table 3 shows the results of the tolerance and VIF tests for multicollinearity in this study.

Table 3: Tolerance and Variance Inflation Factor results

Model (dependent variable: percent of total

annual sales paid in informal payments)

Collinearity statistics

Tolerance (1/VIF) VIF Firm age Firm size Gender Export Business sector Ownership structure Lobbying Obstacle Country .976 .997 .965 .978 .967 .982 .971 .996 .883 1.025 1.003 1.037 1.023 1.035 1.019 1.030 1.004 1.151

From the table we can read that none of the variables exceed the cut off value of 10, which mean that multicollinearity will not be an issue.

4.6.3 Goodness of fit

(30)

30 model in this study should not encounter problems of endogeneity and multicollinearity. In order to demonstrate how well the regression model fits the data, a Chi-squared goodness of fit-test is conducted. The outcomes of the test, which can be found in Appendix 3, show us a significance level lower than 0.0001. Furthermore, the outcome value of Chi-Square is 580.065. Therefore, we can conclude that the model fits the data well and that the independent variables are significantly related to the dependent variable. In the next section the empirical results will be provided.

4.7 Descriptive statistics

Table 4 provides the descriptive statistics of this study. All the variables are included. In this section the mean, minimum, maximum and standard deviation of the variables will be described and discussed. The total sample (N) in this study is 16,243 firms.

Table 4: Descriptive statistics

Variable N Mean Minimum Maximum Std. Deviation

(31)

31 Dependent variable

From a total of 16,243 surveyed firms, 6589 firms admit that they pay bribes, which is more than 40 percent. However, 1244 firms of them pay bribes lower than 1.00% of their total annual sales. The other 5345 firms thus pay bribes which account for 1.00% or more of the total annual sales of the firm. 60 percent of the surveyed firms, which are 9654 firms, state that they do not bribe.

Independent variables

In the sample the firm size varies between a minimum of 1 employee and a maximum of 18,208 employees. The average size of the surveyed firms consists of a number of 112 full-time employees. In terms of the age of the firms, the youngest firm in the sample is 1 year old, whereas the oldest firm is 207 years old. The average firm age is 15 years. Among the top managers, 17% are women and 83% are men. Two percent indicate that they do not know.

Control variables

(32)

32 Appendix 5 gives us an overview of the correlations between the variables. Our independent variables firm size and firm age are significantly correlated, whereas our independent variable gender of the top manager is not significantly correlated with both firm size and firm age. Firm size shows significant correlations with all control variables, except lobbying. Firm age correlates with export, ownership structure and country. Gender of the top manager only correlates with the control variables business sector and country. However, country is significantly correlated with all used variables. The significant correlation between country and all other variables is not a surprise and therefore we will use separate models in the Logistic regression analysis. Some of the models include the country variables and in some of the models the variables are excluded.

(33)

33

5. Empirical results

5.1 Baseline results

This section provides the results of the logistic regression and will discuss three models. Model 1 explains the effect of the independent variables on bribery, in model 2 only the control variables are explained in order to estimate the influence of these variables on the likelihood of bribery and in model 3 we have taken out the country variables to avoid data redundancy.

(34)

34 Table 5: Effect of all used variables on bribery

Standard error in parentheses p<0.05

Model 1 (a) Model 1 (b)

Variable β Exp(β) Sig. Variable β Exp(β) Sig.

INDEPENDENT Firm size Firm age Gender of top manager CONTROL Export Sector Lobbying Ownership Obstacle COUNTRY Uzbekistan (reference) Belarus Georgia Tajikistan Ukraine Albania Russia Poland Romania 0.007 (0.011) -0.059 (0.094) 0.279 (0.259) -0.008 (0.002) 0.417 (0.070) 0.004 (0.003) 0.031 (0.050) -0.068 (0.037) -0.731 (0.584) -1.666 (0.585) -1.604 (0.588) -0.548 (0.585) -0.899 (0.580) -0.970 (0.586) -0.441 (0.585) -2.224 (0.580) 1.007 1.061 1.322 0.992 1.512 1.003 1.032 0.934 0.487 0.181 0.205 0.576 0.395 0.380 0.663 0.137 0.037 0.521 0.011 0.000 0.000 0.232 0.001 0.063 0.000 0.209 0.003 0.007 0.343 0.115 0.099 0.439 0.005 Serbia Kazakhstan Moldova Bosnia Herz. Azerbaijan Macedonia Armenia Kyrgyz Estonia Czech Rep. Hungary Latvia Lithuania Slovakia Slovenia Bulgaria Croatia Montenegro -1.407 (0.582) -1.339 (0.586) -0.988 (0.583) -1.255 (0.587) -1.903 (0.590) -1.403 (0.585) -1.421 (0.590) -2.071 (0.584) -0.326 (0.585) -2.305 (0.593) -1.965 (0.591) -1.694 (0.587) -1.598 (0.579) -1.366 (0.584) -1.267 (0.579) -2.917 (0.585) -2.287 (0.584) -2.911 (0.585) 0.255 0.267 0.378 0.281 0.143 0.244 0.238 0.128 0.823 0.096 0.144 0.188 0.198 0.267 0.293 0.056 0.104 0.059 0.019 0.028 0.098 0.034 0.000 0.013 0.018 0.002 0.589 0.001 0.003 0.003 0.009 0.046 0.021 0.001 0.001 0.000 Observations Pseudo R² -2 Log likelihood Chi² 16243 0.166 2882.385 543.779

(35)

35 caused a decrease in value of Pseudo R² and Chi². This indicates a good fit of the model. The first control variable discussed is export. Model 2 shows a negative export coefficient, meaning that the more a firm exports, the less the likelihood will be that it bribes. Furthermore, the relationship is significant (β= -0.012; p<0.05). The second variable controls for the service or manufacturing business sector in which the firm is operating. The variable is positive and significant, meaning that a difference between both sectors exist and that bribery is more prevalent in the service sector (β= 0.475; p<0.05). This result is in line with the expectation, which assumed a stronger correlation between bribery and the service sector than bribery and the manufacturing sector, caused by the fact that the service sector is older. The third control variable contains the time a manager spend with public officials in order to get things done. In our research we call it lobbying. The results show that lobbying is positively related to the probability of bribery. The positive relationship means that if a firm spends more of its senior management’s time (%) on dealing with government regulations, the likelihood that the firm bribes will increase. However, the result is not significant (β= 0.005; p>0.05). Therefore we cannot assume that this proposition is correct. Table 6 indicates a positive and significant relationship between bribery and the dummy variable ‘type of ownership of the firm’ (β= 0.155; p<0.05). This means that a difference exist in the ownership structures and their influence on the probability of bribery. The results show that single owned firms are less sensitive to bribery than firms with a partnership structure or other forms of multiple ownership structures. This result is not in line with the assumption that single owned firms would be more likely to bribe than firms with other ownership structures. The fifth control variable represents the probability of bribing at firms that specify corruption as the biggest obstacle while doing business. The coefficient is negative, which means that such firms have a lower probability of bribery. However, the variable is insignificant (β= -0.094; p>0.05). The sixth and last control variable is the country variable. Each country of this study is represented in table 6. Uzbekistan is used as the reference country since it is the most corrupt one of all countries according to the CPI 20132. The table indicates that all countries are negatively related to bribery in reference to Uzbekistan, which is used as reference point (most corrupt country). Furthermore, in

(36)

36 20 out of 27 countries the relation is significant. Only in Belarus, Ukraine, Albania, Russia, Poland, Moldova and Estonia the results show an insignificance level.

Table 6: Effect of control variables on bribery Standard error in

parentheses p<0.05

Model 2 (a) Model 2 (b)

Variable β Exp(β) Sig. Variable β Exp(β) Sig.

(37)

37 5.1.3 The effect of all used variables (except countries) on the likelihood that a firm bribes

In order to avoid redundancy of the data, model 3 in table 7 below excludes the country variables. With this removal we see a change in coefficients. The effect on bribery of the independent variable firm age becomes positive. However, it remains insignificant (β= 0.057; p>0.05). Firms size still remains positive and significant (β= 0.013; p<0.05). The third independent variable, the gender of the top manager, also still remains positive and significant (β= 0.284; p<0.05). In terms of the control variables, removing the country variables causes some changes. First, the effect of export becomes positive, but also turns into an insignificant level (β= 0.015; p>0.05). Secondly, the effect of lobbying becomes negative and remains insignificant (β= -0.016; p>0.05). The variable ownership also becomes negatively related to bribery. Besides, this variable becomes significant. This means that removing the country variables results in new evidence which assumes that single owned firms have a higher probability of bribery than firms with partnerships or other multiple ownership structures. This is contradictory to the results of the Logistic regression included country variables. The corruption as biggest obstacle variable still remains negatively and insignificant correlated to bribery. If we take a look at the Pseudo R² and Chi² in model 3, we see a decrease in value compared to Pseudo R² and Chi² in model 1 and 2. This is caused by the use of fewer variables in the third model. Therefore we can conclude that the model is a good fit.

(38)

38 Table 7: Effect of independent and control variables on bribery, except countries

Standard error in parentheses p<0.05

Model 3

Variable β Exp(β) Sig.

INDEPENDENT Firm size Firm age

Gender of top manager CONTROL Export Sector Lobbying Ownership Obstacle 0.013 (0.025) 0.057 (0.091) 0.284 (0.259) 0.015 (0.046) 0.632 (0.721) -0.016 (0.020) -0.180 (0.341) -0.051 (0.036) 1.018 1.059 1.328 1.015 1.882 0.985 0.836 0.898 0.024 0.662 0.048 0.755 0.035 0.441 0.009 0.168 Observations Pseudo R² -2 Log likelihood Chi² 16243 0.075 3267.422 107.376 5.2 Robustness test

(39)

39 Table 8: Probit regression outcomes

Model 4 (a) Model 4 (b)

β Exp(β) Sig. β Exp(β) Sig.

(40)

40

5.3 Interaction effect

In this section we will analyze whether there exists an interaction effect between firm size and firm age. An interaction effect occurs if the effect of one of the independent variables on the dependent variable differs, depending on the level of a moderator variable. This moderator variable is another independent variable. In this study, we use firm size as moderator variable. In our research firm size and firm age are both tested on a negative relationship with bribery. However, the findings for firm age did not show significant evidence to support the proposition in contrast to the findings for firm size. The results for firm size show a significant and positive relationship between firm size and bribery. Therefore, this study will test whether the effect of firm age on bribery will differ when it depends on the value of firm size. For that reason this study will test for an interaction between firm size and firm age.

Several authors have studied the role of firm size on firm growth and found a significant relationship (Hall, 1987; Mata, 1994). Dunne et al. (1989) added the role of firm age to the test and concluded that both firm size and age are negatively related to growth. Furthermore, also Evans (1987) studied the influence of firm size and age on growth by using firm data from the manufacturing sector in the United States. The findings of that study show that small sized and younger firms are more likely to grow. Liu et al. (1999) find the same negative relationship in Taiwan. Since it is commonly known that corruption affects growth it might be assumed that firm age and size together are positively related to bribery. According to existing literature, increasing firm size and age does not lead to higher growth. Furthermore, growth is affected by corruption and according to our study increasing firm size is significantly related to corruption. Therefore we expect the interaction effect will show a positive correlation between firm size and firm age regarding to bribery. The smaller and younger the firms are, the higher the growth. The bigger and older the firms are, the lower the growth and the higher the probability of bribery.

(41)

41 Pseudo R² and Chi² increases in model 5b compared to model 5a. The interaction between firm size and age is narrowly positive and insignificant, meaning that there is not enough evidence to support the statement that firm age, when interacting with firm size, has a positive relationship with the probability of bribery.

Table 9: Interaction Effect

Model 5 (a) Model 5 (b)

Variable: β Exp(β) Sig. β Exp(β) Sig.

Firm size Firm age Gender

Firm size * Firm age

0.005 (0.007) 0.118 (0.112) 0.289 (0.214) 1.005 1.125 1.334 0.023 0.294 0.039 0.009 (0.016) 0.055 (0.125) 0.287 (0.215) 0.003 (0.004) 0.991 1.057 1.332 1.003 0.031 0.501 0.044 0.467 Observations Pseudo R² -2 Log Likelihood Chi² 16243 0.073 3256.808 104.774 16243 0.075 3259.002 105.024

(42)

42

6. Discussion and conclusion

Existing literature have discussed many negative and positive effects of corruption. Corruption is a huge problem across many countries in the world, especially in transition economies. According to the BEEPS (2013), one out of four of all by the World Bank surveyed firms in the world are expected to engage in bribery. Furthermore, it is proven that there is a high probability of an increase in business costs when bribery occurs at firms. Therefore, bribery is a severe problem in the world.

(43)

(44)

44 likelihood to engage in bribery than females. As can be derived from the regression analyses, gender seems positively related to bribery, which means that males have a higher probability of paying bribes than females. Furthermore, a significant support is found for the results and therefore we have enough evidence to accept the third hypothesis. To summarize, hypothesis one is rejected since firm size seems to be positively related to bribery. Hypothesis two cannot be rejected or accepted since we did not found enough significant evidence for it. Finally, hypothesis three is accepted since firms with male top managers seem to be more likely to engage in bribery than firms with female top managers. On the basis of these findings we can answer our main research question: Why are some firms more corrupt than others? This study found support for the statement that firm size and gender of the top manager are determinants of corruption at firms. That makes that firms with many employees (big size) or/and a male top manager (gender) are more corrupt than other firms.

6.1 Added value

(45)

45 characteristics size and age of the firm. We were wondering whether a correlation should exist between those characteristics and the probability that a firm pays a bribe. This study reveals several strategic implications for policymakers, firms and their managements. Firstly, it indicates which kind of firm has a high probability of engaging in bribery to get things done. Managers can use this information when doing business. For example, when a firm wants to make an investment, they now know what to look out for in terms of corruption. Secondly, male managers seem to be more corrupt than female managers. Therefore, firms should more intently monitor male managers than female managers on corrupt behavior. Furthermore, when a firm forms a new management, it should consider putting a female in charge as top manager, in order to lower the probability of corrupt behavior. For governments and policy makers, there are three implications. Firstly, based on firm size and gender of the top manager of that firm, governments have an overview of the kind of firms that bribe. This can be useful for governments in order to reduce corruption in their country. Secondly, the findings of this study indicate which types of firms have a higher probability of bribing. Therefore, policy makers can intensively monitor that kind of firms in order to combat corruption. Thirdly, policy makers could introduce new policies that give women privileges for better positions in the managements of firms. Since males are more likely to pay bribes than females such policies could reduce the corruption problem.

6.2 Limitations and further research

(46)

(47)

47

Appendix 1 - List of countries

1 Albania 2 Armenia 3 Azerbaijan 4 Belarus

(48)

48

Appendix 2 - Questions of the BEEPS

Bribery:

It is said that establishments are sometimes required to make gifts or informal payments to public officials to “get things done” with regard to customs, taxes, licenses, regulations, services etc. On average, what percentage of total annual sales do establishments like this one pay in informal payments or gifts to public officials for this purpose? …….% of bribes paid

Dummy: Does the firm pay bribes?

1 = Yes 0 = No

Size:

At the end of fiscal year (2002, 2005, 2009), how many permanent, full-time employees did this establishment employ?

……….(no. of employees)

Age:

In what year did this establishment begin operations?

………. (year)

Gender:

Is the Top Manager female?

1 = Yes 2 = No -9 = Do not know

Export:

What percentage of this establishment’s sales was exported directly (without the use of a third party: …………..(%)

Sector:

1 = Manufacturing 3 = Other services 2 = Service (retail)

Ownership structure:

1 = Shareholder ownership with shares traded in the stock market

6 = Other 2 = Shareholder ownership with shares traded

privately 7 = Single member private ltd. Co.

3 = Sole proprietorship 8 = Business permit

4 = Partnership 9 = Unregistered

(49)

49 Dummy:

1 = Single ownership 0 = Other

Lobbying:

What percentage of total senior management’s time was spent on dealing with requirements imposed by government regulations?

………. % of senior management’s time spent on dealing with regulations 0 = No time was spent -9 = Do not know

Corruption as the biggest obstacle:

Which of the following elements of the business environment, if any, currently represents the biggest obstacle faced by this establishment?

1 = access to finance 10 = labor regulations

2 = access to land 11 = political instability

3 = business licensing and permits 12 = practices of competitors in the informal sector

4 = corruption 13 = tax administration

5 = courts 14 = tax rates

6 = crime, theft and disorder 15 = transport

7 = customs and trade regulations 16 = telecommunications

8 = electricity 17 = economic and regulatory policy

uncertainty

9 = inadequate educated workforce 18 = macro-economic instability Dummy:

(50)

50

Appendix 3 - Goodness of fit

In order to test if the regression model fits the data, a Chi-squared goodness of fit-test is conducted.

Chi-Square Test

Bribery Observed N Expected N Residual

YES 6589 8124.0 -1535.0 NO 9654 8124.0 1535 Total 16243 Test Statistics E Chi-Square 580,065 a Df 1 Asymp. Sig. 0.000

a. 0 cells (0.0%) have expected frequencies less than 5. The minimum expected cell frequency is 8124.0

Pseudo R-Square

-2 Log Likelihood 96.886 a

Cox and Snell 0.028

Nagelkerke 0.287

(51)

51

Appendix 4 – Country frequency table

(52)

52

Appendix 5 - Correlation table

Pearson correlation

• = Correlation is significant at the 0.05 level (2-tailed). ** = Correlation is significant at the 0.01 level (2-tailed).

PEARSON Correlation Bribery Yes/No Firm size Firm age

Gender % Export Business Sector

(53)

53

References

Aidis, R., Estrin, S. & Mickiewicz, T. 2012. Size matters: entrepreneurial entry and government. Small Business Economics, 39(1): 119–139.

Alesina, A., Devleeschaauwer, A., Easterly, W., Kurlat, S., & Wacziarg, R. 2003. Fractionalization. Journal of Economic Growth, 8(2): 155-194.

Alesina, A. & Zhuravskaya, E. 2011. Segregation and the quality of government in a cross section of countries. American Economic Review, 101(5): 1872-1911. Ali, A. M. & Isse, H. S. 2003. Determinants of economic corruption: A cross-country

comparison. Cato Journal, 22: 449–466.

Almus, M. & Nerlinger, E. 2000. Testing Gibrat’s law for young firms: Empirical results from West Germany. Small Business Economics, 15: 1-12.

Andreasson, S. 2011. Africa’s prospects and South Africa’s leadership potential in the emerging markets century. Third World Quarterly, 32(6): 1165-1181.

Arvis, J. F. & Berenbeim, R. 2003. Fighting Corruption in East Asia: Solutions From the

Private Sector. Washington DC: published by the World Bank.

Athanasouli, D., Goujard, A. & Sklias, P. 2012. Corruption and firm performance: Evidence from Greek firms. International Journal of Economic Sciences and Applied

Research, 5(2): 43-67.

Baumann, H. D. & Kaen, F. R. 2003. Firm Size, Employees and Profitability in U.S. Manufacturing Industries. Working paper, Social Science Research Network, University of New Hampshire.

Berman, B. J. 1998. Ethnicity, patronage and the African state: the politics of uncivil nationalism. African Affairs, 97: 305-341.

Blumberg, B., Schindler, P. & Cooper D. 2011. Business Research Methods. New York: McGraw-Hill Higher Education.

Clarke, G. & Xu, L. C. 2004. Privatization, Competition and Corruption: How Characteristics of Bribe Takers and Payers Affect Bribe Payments to Utilities. Journal of Public

Economics, 88(9–10): 2067–2097.

Cuervo-Cazurra, A. 2006. Who cares about corruption? Journal of International Business

Studies, 37: 807–822.

Dacin, M. T., Goodstein, J. & Scott, W. R. 2002. Institutional Theory and Institutional Change: Introduction to the Special Research Forum. Academy of Management

Journal, 45(1): 45–57.

Davidson, P. B., Kirchhoff, A., Hatemi, J. & Gustavsson, H. 2002. Empirical analysis of business growth factors using Swedish data. Journal of Small Business

(54)

54 De Jong, G., Ahn Tu, P. & van Ees, H. 2012. Which Entrepreneurs bribe and what do they get from it? Exploratory Evidence from Vietnam. Entrepreneurship: Theory & Practice, 36(2) 323–345.

De Vaal, A. & Ebben, W. 2011. Institutions and the Relation between Corruption and Economic Growth. Review of Development Economics, 15: 108–23.

Dollar, D., Fisman, R. & Gattia, R. 2001. Are women really the “fairer” sex? Corruption and women in government. Journal of Economic Behaviour and Organization, 46(4): 423–429.

Dong, B., Dulleck, U. & Torgler, B. 2012. Conditional corruption. Journal of Economic

Psychology, 33: 609-627.

Dowden, R. 2008. Africa, Altered States, Ordinary Miracles. London (UK): Portobello Books.

Dunne, T., Roberts, M. J. & Samuelson, L. 1989. The growth and failure of U.S. Manufacturing plants. Quarterly Journal of Economics, 104: 671-698.

Eagly, A. H. 1987. Sex differences in social behavior: A social role interpretation. Hillsdale, NJ: Erlbaum.

Eagly, A. H., & Wood, W. 1991. Explaining sex differences in social behavior: A meta- analytic perspective. Personality and Social Psychology Bulletin, 17: 306-315. Eagly, A. H., Wood, W. & Diekman, A. 2000. Social role theory of sex differences and similarities: A current appraisal. In Eckes, T. & Trautner, H. The developmental

social psychology of gender. Mahwah, NJ: Erlbaum.

Easterly, W. & Levine, R. 1997. Africa’s growth tragedy: policies and ethnic divisions.

Quarterly Journal of Economics, 112(4): 1203-1250.

Eisenhardt, K. M. 1988. Agency and Institutional Theory Explanations: The Case of Retail Sales Compensation. Academy of Management Journal, 31(3): 488.

Evans, D. S. 1987. The relationship between firm growth, size and age: estimates for 100 Manufacturing industries. Journal of Industrial Economics, 35: 567-581. Feingold, A. 1994. Gender differences in personality: A meta-analysis. Psychological

Bulletin, 116: 429-456.

Fisman, R. & Svensson, J. 2007. Are Corruption and Taxation Really Harmful to Growth? Firm Level Evidence. Journal of Development Economics, 83(1): 63-75.

Flatters, F. & MacLeod, W. B. 1995. Administrative corruption and taxation. International

Tax and Public Finance, 2: 397-417.

(55)

55 Getz, K. & Volkema, J. 2001. Culture, perceived corruption, and economics. Business &

Society, 40(1): 7–30.

Glaeser, E. L. & Saks, R. E. 2006. Corruption in America. Journal of Public Economics, 90: 1053-1072.

Glancey, K. 1998. Determinants of growth and profitability in small entrepreneurial firms.

International Journal of Entrepreneurial Behavior and Research, 4: 18-27.

Granovetter, M. 2005. The Social Construction of Corruption. Working Paper, Stanford University.

Gray, C. W. & Kaufmann, D. 1998. Corruption and development. Finance and Development, 35(1): 7-10.

Guerrero, M. & Rodríguez-Oreggia, E. 2008. On the individual decisions to commit corruption: A methodological complement. Journal of Economic Behavior and

Organization, 65: 357-372.

Gyimah-Brempong, K. 2002. Corruption, economic growth and income in- equality in Africa.

Economics of Governance, 3: 183-209.

Hall, B. 1987. The relationship between firm size and firm growth in the US manufacturing sector. Journal of Industrial Economics, 35: 583-606.

Herrera, A. M. & Rodriguez, P. 2003. Bribery and the Nature of Corruption. Working Paper, Department of Economics, Michigan State University, MI.

Heywood, P. 1996. Continuity and change: analysing political corruption in modern Spain. From: Little, W. & Posada-Carbó, E. Political Corruption in Europe and Latin

America, New York: St Martin’s Press.

Hill, R. C., Griffiths, W. E. & Lim, G. C. 2009. Principles of Econometrics 4th_edition.

Hoboken (NJ): John Wiley & Sons Ltd.

Hodge, A., .Shankar, S., Prasada Rao, D. S. & Duhs, A. 2011. Exploring the Links between Corruption and Growth. Review of Development Economics, 15: 474–90.

Hunt, J. & Laszlo, S. 2012. Is Bribery Really Regressive? Bribery’s Costs, Benefits, and Mechanisms, World Development, 40(2): 355–372.

Huntington, S. P. 1968. Political Order in Changing Societies. New Haven, CT: Yale University Press.

Jain, A. 2001. Corruption: a review. Journal of economic surveys, 15(1): 71 – 121

Jensen, N., Li, Q. & Rahman, A. 2010. Understanding corruption and firm responses in cross- national firm-level surveys. Journal of International Business Studies, 41: 1481– 1504.

(56)

56 Johnson, S., Kaufmann, D., McMillan, J. & Woodruff, C. 2000. Why do firms hide? Bribes

and unofficial activity after communism. Journal of Public Economics, 76: 495– 520.

Jónsson, B. 2007. Does the size matter? The relationship between size and profitability of Icelandic firms. Bifröst Journal of Social Science, 1: 43–55.

Jovanovic, B. 1982. Selection and Evolution of Industry. Econometrica, 50(2): 649-670. Kahan, D. M. 1998. Social Meaning and the Economic Analysis of Crime. The Journal of

Legal Studies, 27: 609-622.

Keefer, P. & Knack, S. 1997. Why don.t poor countries catch up? A cross-national test of an institutional explanation. Economic Inquiry, 35: 590-602.

LaPorta, R., Lopez-de-Silanes, F., Shleifer, A. & Vishny, R. 1999. The quality of government. Journal of Law, Economics, and Organization, 15: 222– 279. Lee, S. H., Oh, K. & Eden, L. 2010. Why do firms bribe? Insights from residual control

theory into firms’ exposure and vulnerability to corruption. Management

International Review, 50: 775–796.

Leff, N. 1964. Economic Development Through Bureaucratic Corruption. American

Behavioral Scientist, 8: 8.

Li, H., Xu, L. C. & Zou, H. 2000. Corruption, income distribution and growth. Economics

and Politics, 12: 155-182.

Liu, J. T., Tsou, M. W. & Hammit, J. K. 1999. Do small plants grow faster? Evidence from Taiwan Electronics Industry. Economics Letters, 65: 121-129.

Lui, F. T. 1985. An Equilibrium Queuing Model of Bribery. Journal of Political Economy, 93(4): 760-81.

Martin, K., Cullen, J., Johnson, J. & Parboteeh, K. P. 2007. Deciding to bribe: a cross-level analysis of firm and home country influences on bribery activity. Academy of

Management Journal, 50(6): 1401–1422.

Marvel, M. & Lumpkin, G. 2007. Technology entrepreneurs’ human capital and its effects on innovation radicalness. Entrepreneurship Theory and Practice, 6: 807–828.

Mata, J. 1994. Firm growth during infancy. Small Business Economics, 6: 27-39.

Mauro, P. 1995. Corruption and Growth. Quarterly Journal of Economics, 110: 681–712. McArthur, J. & Teal, F. 2002. Corruption and Firm Performance in Africa. Working

Paper Series, Centre for the Study of African Economies, University of Oxford.

Meagher, P. & Thomas, M. A. 2004. A Corruption Primer: An Overview of Concepts in the Corruption Literature. Working Paper, the IRIS Center at the University of

Maryland, College Park, MD.

Meon, P. G. & Sekkat, K. 2005. Does Corruption Grease or Sand the Wheels of Growth?

(57)

57 Meon, P. G. & Weill, L. 2010. Does Better Governance Foster Efficiency? An Aggregate

Frontier Analysis. Economics of Governance, 6: 75–90.

Merton, R. 1968. Social Theory and Social Structure. New York: Free Press

Messner, S., & Rosenfeld, R. 1997. Political restraint of the market and levels of criminal homicide: A cross-national application of institutional-anomie theory. Social Forces, 75: 1393–1416

Mo, P. H. 2001. Corruption and economic growth. Journal of Comparative Economics, 29: 66-97.

Mocan, N. 2008. What determines corruption? International evidence from micro data.

Economic Inquiry, 46(4): 493-510.

Myrdal, G. 1968. Asian drama; an inquiry into the poverty of nations. New York: Pantheon Books.

North, D. C. 1994. Economic Performance through Time. American Economic Review, 84(3): 359-368.

Olken, B. A. 2006. Corruption and the costs of redistribution: Micro evidence from Indonesia.

Journal of Public Economics, 90: 853–870.

Pellegrini, L. & Gerlagh, R. 2008. Causes of corruption: a survey of cross-country analyses and extended results. Economics of Governance, 9(3): 245-263.

Peng, M. W. & Heath, P. S. 1996. The Growth of the Firm in Planned Economies in Transition: Institutions, Organizations, and Strategic Choice. The Academy of

Management Review, 21(2): 492-528.

Peng, M. W. & Luo, Y. 2000. Managerial ties and firm performance in a transition economy: The nature of a micro-macro link. Academy of Management Journal, 43: 486–501. Peng, M. W. & Qi Zhou, J. 2011. Does bribery help or hurt firm growth around the world.

Working paper, Jindal School of Management, University of Texas.

Peters, S., Kinsey, P., & Malloy, T. 2004. Gender and leadership perceptions among African Americans. Basic and Applied Social Psychology, 26: 93-101.

Reinikka, R. & Svensson, J. 2006. Using Micro-Surveys to Measure and Explain Corruption.

World Development, 34(2): 359–370.

Rodriguez, P., Uhlenbruck, K. & Eden, L. 2005. Government Corruption and the Entry Strategies of Multinationals. Academy of Management Review, 30(2): 383–96. Sachs, J. D. & Warner, A. M. 1997. Sources of slow growth in African economies. Journal

of African Economics, 6: 335-376.

Schleifer, A. & Vishny, R. W. 1993. Coruption. The Quarterly Journal of Economics, 108: 599–617.

(58)

58 D’Souza, A. 2012. The OECD Anti-Bribery Convention: Changing the currents of trade.

Journal of Development Economics, 97: 73–87.

Spencer, J. & Gomez, C. 2011. MNEs and corruption: the impact of national institutions and subsidiary strategy. Strategic Management Journal, 32: 280-300

Svensson, J. 2003. Who must pay bribes and how much? Evidence from a cross section of firms. The quarterly Journal of Economics, 118(1): 207–230.

Svensson, J. 2005. Eight questions about corruption. Journal of Economic Perspectives, 19: 19–42.

Swamy, A., Knack, S., Lee, Y. & Azfar, O. 2001. Gender and corruption. Journal of

Development Economics, 64(1): 25–55.

Tanzi, V. & Davoodi, H. 1997. Corruption, public investment, and growth. IMF Working

Paper, no. 97/139, International Monetary Fund, Washington, DC.

Thomas, A. & Bendixen, M. 2000. The management implications of ethnicity in South Africa. Journal of International Business Studies, 31(3): 507-519.

Thornhill, S. & Amit, R. 2003. Learning About Failure: Bankruptcy, Firm Age, and the Resource-Based View. Organization Science, 14(5): 497–509.

Tonoyan, V., Strohmeyer, R., Habib, M. & Perlitz, M. 2010. Corruption and

Entrepreneurship: How formal and informal institutions shape small firm behavior in transition and mature market economies. Entrepreneurship Theory and Practice, 34(5) 803–831.

Torgler, B. & Valev, N. T. 2006. Women and illegal activities: gender differences and women’s willingness to comply over time. Working Paper No. 2006-15, Center for Research in Economics, Management and the Arts (CREMA).

Torgler, B. & Valev, N. T. 2010. Gender and public attitude toward corruption and tax evasion. Contemporary Economic Policy, 28: 554-568.

Treisman, D. 2000. The causes of corruption: a cross-national study. Journal of Public

Economics, 76(3): 399–457.

Uhlenbruck, K., Rodriguez, P., Doh, J., & Eden, L. 2006. The impact of corruption on entry strategy: Evidence from telecommunication projects in emerging economies.

Organization Science, 17: 402–414.

Weber, J. & Getz, K. 2004. Buy bribes or bye-bye bribes: the future status of bribery in international commerce. Business Ethics Quarterly, 14(4): 695-711

Wei, S. J. 2000. How Taxing is Corruption on International Investors? Review of Economics

and Statistics, 82(1): 1-11.

Wijewardena, H. & Tibbits, G. E. 1999. Factors contributing to the growth of small

Bribery analysis through firm size, firm age and gender of the top manager