The effects of culture on capital structure : a cross-country analysis : is culture a missing factor affecting the level of leverage a firm takes on?

(1)

The effects of culture on capital structure A cross-country analysis

Is culture a missing factor affecting the level of leverage a firm takes on?

Author: C. Zhou

Student number: 10532129

Faculty: Economics & Business Institution: University of Amsterdam Supervisor: Dr. N. Martynova

Date: February 20th, 2016

Abstract

This study studies the effects of culture on capital structure. Individualism, uncertainty avoidance and long-term orientation are used to carry out a cross-country analysis on leverage of firms in 50 countries. We find evidence that capital structure is affected by uncertainty avoidance in most periods, especially in the most recent periods. According to literature, uncertainty avoidance shows the tolerance for uncertainty of a society. Uncertainty avoidance leads to greater emphasis on financial stability, which leads to a decreased use of debt finance. Debt usage is viewed to be too risky, since it puts more stress on the financial stability. A robustness check also shows the affection of uncertainty avoidance on capital structure on the most recent period. However, the sign on the coefficient is negative, which goes into literature and economic intuition. Further investigation is needed.

(2)

2 Introduction

Capital structure is an important aspect of a firm. One has to make a decision between equity or debt financing. Thus there are a lot of possible combinations. And since there are a lot of possibilities, there must an optimal one, where a balance between costs and benefits of debt is found. This lead to extensive research on capital structure, which lead to three different main views: the classic trade-off theory between financial distress costs and interest tax shield benefits of debt, the agency theory which adds agency costs and benefits, and the pecking order hypothesis in which firms prioritize their financial sources. Through this extensive research, it has become clear that firm-specific, as well as country-specific variables play a role in determining capital structure (Öztekin, 2015; Franke et al., 1991). However, culture is not included in studies done on country-specific effects, when culture does differ across countries, while there are studies done on behavioral biases. These studies provide evidence that these biases have effects on managerial decisions. This leads us to our study, where we will incorporate individualism, uncertainty avoidance, and long-term orientation as cultural values to examine the effects of culture on capital structure. Malmendier et al. (2011) argue that manager’s belief significantly affect financial policies beyond traditional determinants of capital structure. Parsons and Titman (2008) also provide evidence that managerial preferences influence capital structure. These behavior biases will be used to link culture to capital structure. We use data on firms in 50 different countries to do a cross-country analysis on firm level, to look whether culture has a significant influence on capital structure or not. We use country-level analysis for robustness checks. I believe that the outcome of this research can give further insights to the problem of distortion of the optimal level of debt. The next sections discusses prior literature on capital structure. In section 3, we discuss the link between culture and capital structure. In section 4, we describe the dataset we use, followed by the results of our data in section 5. We conclude in section 6, and discuss our findings.

(3)

3

Based on theories on capital structures surveyed by Harris and Raviv (1991), there are many testable determinants of capital structure with an economic intuitive relationship between the determinant and leverage. They however argue that only a small number of those potential factors can be labeled as general principles. Other studies also show differences between determinants. Table 1 in the appendix shows some studies done on determinants of leverage. It shows us that there are indeed a lot of potential determinants of capital structure, but profitability is the only factor that is agreed on to have a negative effect on leverage. This is according to Harris and Raviv (1991), only a small number of potential determinants of capital structure can be labeled as general principles.

The signs of the determinants can be explained by the main theoretical views on capital structure. Profitability is the only factor which has a consistent negative sign. According to the trade-off theory, benefits of the interest tax shield are weighted against the costs of financial distress. The more the firm makes profit, the more the firm can make use of the interest tax shield benefits. Financial distress is also less likely, which reduces the costs. Thus, profitability is negatively correlated to leverage. According to pecking-order theory, firms prefer to use retained earnings first, then debt, and finally debt. A profitable firm has more retained earnings, and would thus need less debt. According to agency theory, agency costs and benefits are added to the trade-off between debt and equity. The negative sign can be explained by the fact that profitable firms face more agency costs, as empire building and management entrenchment, as it is more attractive to hold on to profitable firms. The difference in signs of the other determinants in table 1 can be explained by these theories as well, but further elaboration is not the focus point of this study. We seek reliable determinants for our study. Recent research done by Özetekin (2015) showed the determinants of capital structure that are internationally reliable. The sample size of his study consisted of 37 countries. He concluded that the reliable firm-specific determinants for leverage are tangibility, profitability, firm size, liquidity, and industry, inflation and the quality of the countries’ institutions.

Because of inflation and countries’ institutions, it is clear that country-specific factors are important as well. A survey on managerial views on the determinants of capital structure across European countries done by Bancel and Mittoo (2004), suggests that factors determining financing decisions are similar between European and US firms. But they suggest that there are indeed country-specific factors. Differences arise across countries, especially between Scandinavian and non-Scandinavian countries. Their

(5)

5 concern is that biases and measurement problems arise because of survey data. They have found that Scandinavian managers have a different view on capital structure than others. Bancel and Mittel believe that moral and ethical norms among others may be one of the factors that could have caused the differences in views on capital structure. This is not taken into accounting in their study.

Another study done on international capital structure by Rajan and Zingales (1995) also suggest that there are differences across countries. They have done multiple regressions on the G7 countries, with leverage as the dependent variable to determine the strength of tangibility, market-to-book, size, and profitability. They have found that most of these factors affecting leverage in the United States appear to be similarly correlating with leverage in other countries as well. However, while these factors are fairly similar across the G-7 countries, there are still differences to be found. This suggests that institutional environment is important to the determination of capital structure.

A study done La Porta et al. (1997) shows that civil law has weaker investor protections and less developed capital markets than common law countries.

This takes us to this study. In this study, cultural factors will be included as country-specific effect variables affecting capital structure. This can give us a clearer picture of the differences in capital structure worldwide. The next section discusses literature on culture and capital structure.

Culture and capital structure

To analyze the effect of culture on capital structure, we first need to define culture and establish the link between culture and capital structure. This link is explained by behavioral biases. To capture cultural effects, we use individualism, uncertainty avoidance and long-term orientation.

Behavioral bias

Hofstede (1991) defined culture as the collective programming of the mind distinguishing the members of one group or category of people from another. And according to Adler (1997), culture influences behavior. Traditional economic framework assumes rationality in decisions of economic agents. If people behave differently because of culture however, rationality may be distorted. This is confirmed by research on psychology. Malmendier et al. (2011) argue that manager’s belief and early-life experiences significantly affect financial policies beyond traditional determinants of capital structure. Parsons and Titman (2008) also argue that managerial preferences may affect

(6)

6 capital structure (Antonczyk and Salzmann, 2013). People also tend to be optimistic. According to Larwood and Whittaker (1997), behavioral biases are especially strong for executives, which influence their decisions. These differences in behavior stem from culture. It seems that it is indeed reasonable to assume that culture is a significant factor affecting capital structure.

Individualism

Individualism and collectivism are heavily discussed topics in the psychology literature. Individualism describes the individual attitude towards one another and society. It is the degree to which individuals are integrated into groups (Hofstede). In societies where individualism prevail, individuals seem not to care more about themselves than others. The expectation exists to only look after oneself and the immediate family. On the opposite, collectivism, the group is more important. In societies where collectivism prevail, individuals seem to be more integrated into strong groups and extended families such as uncles and aunts. According to Heine and Lehman (1995), the individualistic culture has an independent view, versus the collectivistic one, which has an interdependent self-view. People with an independent self-view, are more optimistic than people with an interdependent self-view. Thus, individualism is linked to optimism. In another study done by Heine and Lehman (1999), it is shown that individualistic individuals, tend to even inflate their confidence, which leads to overly positive beliefs. In contrast to collectivistic individuals, who have more self-discipline. Hackbarth (2008) and Heaton (2002) both show that managerial traits, such as optimism, can cause variation in capital structures. Optimism induces risk perception bias, which causes debt to be viewed as undervalued, and equity be viewed as overvalued by the market. This implies that the more optimistic, and thus individualistic, the higher their leverage ratios.

Hypothesis 1: Leverage is positively related to individualism

Uncertainty avoidance

Uncertainty avoidance shows the tolerance for uncertainty of a society. It indicates the extent people feel either comfortable or uncomfortable in unstructured situations, situations which are unknown, different than usual. Cultures that avoid uncertainty, would like to prevent those unstructured situations by laws and rules, and safety and security measures. In cultures that accept uncertainty however, there is a higher tolerance for unknown and different opinions (Hofstede). Chui et al. (2002) finds that uncertainty avoidance leads to greater emphasis on financial stability, which leads to a decreased use of debt finance. They further imply that uncertainty avoiding cultures view debt usage too

(7)

7 risky, since it puts more stress on the financial stability. Baum et al. (2009) investigates the relationship between leverage and macroeconomic uncertainty. They concluded that firms use less debt when uncertainty increases. They also imply that this effect is different for higher and lower leveraged firms. This implies that the more you avoid uncertainty, the less debt you want to use.

Hypothesis 2: Leverage is negatively related to uncertainty avoidance.

Long-term orientation

According to Hofstede, long-term orientation tells us that long-term oriented cultures, are more sighted towards future rewards. They possess values as patience and persistence. Short-term oriented cultures hold more onto the present, such as respect for tradition. Antonczyk et al. (2014) say that when investors favor long-term profitability, managers can pursuit long-term projects more at ease. They are not continuously forced to yield short-term positive returns. A term investment environment focuses on long-term strategies and opportunities. This will lead to better relationships with banks, which makes external bank financing easier. They indeed find that bank finance is positively correlated with the degree of long-term orientation.

Hypothesis 3: Leverage is positively related to long-term orientation.

Regression data

Leverage will be regressed on several variables according to the following regression model:

∑ ∑

∑

The sample consists of nonfinancial firms from 50 countries, covering the years 1989-2014. To analyze the effects of culture on capital structure, individualism, uncertainty avoidance and long-term orientation, will be used as the main explanatory variables. These cultural values are time-invariant, gained from the website of Geert Hofstede. Other firm-specific and country-specific control variables are taken into account as well. Firm-specific and leverage data is gained from the Compustat Global database. The whole database in the sample period is used, to mimic the population as good as possible. Country specific data is gained from the database of The World Bank.

(8)

8 Dependent variable:

Long-term debt to equity is used as leverage for the dependent variable. According to de Jong et al. (2008), it is better to use long-term debt and to exclude short term-debt. Short-term debt consists of trade credit, which differ in determinants. The average of the leverage in our sample is 0.194.

Independent variable:

Hofstede’s individualism, uncertainty avoidance, and long-term orientation will be used as proxies for the cultural values we are interested in. In his study from 1967- 1973 he analyzed a large database of employee value scores collected within IBM. The data covered 40 countries. A questionnaire of 30 questions was used to construct the values. These were value based questions, where answers from 1 to 5 were possible, depending on how much value you attach to the specific question. Afterwards, countries were being added through the same way of research, as supported by Hofstede. According to Hofstede, culture changes slowly, so the values can be used today. The values range between 0 and 100, where 100 is the highest and 0 the lowest. The data is included in table 2 in the appendix, sorted on individualism. It shows that developed countries are more individualistic than developing countries.

Control variables, firm-specific

The variables, as discussed in the literature review, used in this study according are defined as in table 3.

Table 3, control variables – firm-specific

Control variable Description

Tangibility Tangibility is computed by dividing

total property, plant and equipment by total assets.

Profitability Profitability is computed by dividing net income by total assets.

Size Size is computed by dividing total assets to

total sales

Liquidity Liquidity is computed by dividing current assets by current liabilities.

(9)

9 Control variables, country-specific

Öztekin also showed that inflation and the quality of the countries’ institutions is also significant correlated to capital structure. Quality of a country’s institution will be measured by their national legal system. To control for his, a dummy for countries based on their national legal system is added. A study done by Franke et al. (1991) shows that cultural values are linked to the wealth of nations. This implies that possible effects of culture on capital structure, comes from the economic condition of countries. To control for this, a dummy for developing countries is added. Also the GDP growth is used to capture financial crises. The variables are defined as in table 4.

Table 4, control variables – country-specific

Control variable Description

Developing country Countries are classified using the classification criteria of The World Bank. The dummy

variable takes a value of one if a country is a developing country, zero otherwise.

GDP Growth Annual GDP Growth is based on constant local

currency. Depreciation of fabricated assets and depletion and degradation of natural resources is not used for the calculation.

Inflation Annual inflation is measured by the annual

growth of the GDP implicit deflator. All is in local currency.

Law Countries’ national legal systems are divided

into two laws. Common and civil law. The dummy variable takes a value of one if a country is based on common law, zero otherwise.

Data acquired through The World Bank database

Industry effect

To control for industry, firms are divided into three industries: the primary, manufacturing and service industry. SIC-codes are used to determine the industry for the firm. Table 5 shows the classifications of the firms.

Tabel 5, SIC-codes

Industry SIC-codes

Primary 0000-1999

Manufacturing 2000-3999

Service 4000-9999

Data acquired through the Compustat Global database

Data on the variables per country, averaged out through the whole sample period is included in table 6 in the Appendix, sorted on leverage. There does not seem to be a relationship with developed or developing countries like individualism.

(10)

10 Regression results

Leverage will be regressed on several variables according to the following regression model:

∑ ∑

∑

To test cultural significance, firm-level regressions are carried out. But only fractions of the sample size and period will be used, due to imbalances. For example, there are nearly 150,000 observations through the period for the US, and only 12,000 for South Korea. Periods and countries are dropped to get a better balanced panel. The final sample size consists of 6 developed and 6 developing countries. Two sample periods will be used to carry out tests. Firstly, between 2004-2006. Secondly, between 2012-2014, with the same 12 countries, but with 1 more developed and developing countries added. Dummy interaction variables are always included in the regression, but insignificant interaction between cultural and non-dummy variables are dropped. The results are presented in table 7 and 9.

Table 7, Firm level panel regression 2004-2006

1 2 3 4 5 6 Individualism -0.048 -0.078 0.022*** 0.036 (0.296) (0.304) (0.000) (0.232) Uncertainy avoidance 0.024 -0.043 0.024*** 0.042* (0.253) (0.343) (0.004) (0.088) Long-term orientation 0.016 0.054 -0.008* -0.052 (0.225) (0.285) (0.088) (0.153)

Firm-specific control No No No No Yes Yes

Country specific control No No No No No Yes

Industry control No No No No Yes Yes

0.0008 0.0001 0.0000 0.0012 0.0028 0.0056 Number of observations 5087 5087 5087 5087 5087 5087

P-values are reported in the parenthesis. *, **, ***, significant at the 10, 5, and 1 percent level, respectively.

According to table 7, the model explains at most 0.6% of difference in leverage. There are also no significant interaction between cultural and other variables. When not controlling for anything, culture does not have influence on leverage at all, according to model 1 to 4. The low confirms this as well. Adding firm-specific and industry control variables, all

(11)

11 three cultural values are significant different from zero according to model 5. If we look at model 6, where all control variables are added and the is the highest, only the

coefficient of uncertainty avoidance is significant at the ten percent level. To look at the economic significance, we look at the change in leverage by increasing uncertainty avoidance from the 10th to the 90th percentile from this sample. The percentile ranks and averages on leverage and uncertainty avoidance are presented in the appendix in table 8. As uncertainty avoidance moves from the 10th percentile, 44, to the 90th percentile, 85, leverage will increase by 1.64. This is above the 90th percentile of leverage, which makes uncertainty avoidance an important determinant of leverage. However, the sign of the coefficient is the opposite of hypothesis 2. Uncertainty does have an effect, but not the effect as expected. This goes into our economic intuition and literature.

Table 9, Firm level panel regression 2012-2014

1 2 3 4 5 6 Individualism 0.004 0.001 0.026* 0.099** (0.108) (0.776) (0.073) (0.050) Uncertainy avoidance -0.005 -0.004 0.035** 0.080*** (0.257) (0.541) (0.005) (0.007) Long-term orientation 0.005** 0.004 0.001 -0.040 (0.050) (0.157) (0.925) (0.285) TANG 2.45*** (0.001) 5.76*** (0.008) TANG*UAI -0.046*** -0.037*** (0.001) (0.009)

Firm-specific control No No No No Yes Yes

Country specific control No No No No No Yes

Industry control No No No No Yes Yes

0.0003 0.0002 0.0002 0.004 0.0192 0.05 Number of observations 6330 6330 6330 6330 6330 6330

According to table 9, the model explains at most 5% of difference in leverage. If we now look at model 6 again in another sample period, individualism and again uncertainty avoidance, are statistically significant from zero. There is also interaction between uncertainty avoidance and tangibility. Table 10 in the appendix presents the percentile ranks and averages of leverage, individualism, uncertainty avoidance, and tangibility on this sample. As uncertainty avoidance moves from the 10th percentile to the 90th percentile,

(12)

12 leverage will increase by 3.25. But depending on tangibility, uncertainty avoidance

reduces the increase in leverage. Holding tangibility constant at the 10th percentile, moving uncertainty avoidance from the 10th to the 90th percentile, reduces leverage by 0.07. Holding tangibility constant at the 90th percentile, reduces leverage by 1.03. Holding tangibility constant, leverage increases between 2.22 and 3.18, depending on tangibility. This is above the 90th percentile, which makes uncertainty avoidance economic significant in this period as well. This sign is again the opposite of our expectation, and goes into economic intuition and literature again.

Individualism is statistically significant from zero as well. When moving from the 10th to the 90th percentile, leverage would increase by 5.445. This is at the 99th percentile, which means that individualism is of economic significance as well in this sample period.

To avoid overrepresentation, only 12 to 14 countries are used in the firm-level regression. This amount may be too small to get robust results. To test robustness, the whole sample size will be used. All variables will be averaged out, and country-level analysis are carried out. 4 different groups are regressed. All firms, manufacturing industry, service industry, and primary industry. Firstly, since the most recent year in the sample period has the most data, that one will be used as the reference year. The results are presented in table 11.

Table 11, Country-level panel regression 2014 1 All 2 Man 3 Serv 4 Prim Individualism 0.026** -0.013 -0.007 0.135** (0.027) (0.272) (0.792) (0.046) Uncertainy avoidance 0.029*** 0.004 -0.034* 0.012 (0.005) (0.575) (0.084) (0.858) Long-term orientation 0.014** -0.014** 0.008 0.070 (0.050) (0.041) (0.693) (0.150) Liquidity 2.086*** (0.001) LIQ*IDV -0.009* (0.069) LIQ*UAI 0.015*** (0.000) LIQ*LTO -0.008*** (0.001)

(13)

13

Tangibility 24.174**

( 0.030)

TANG*IDV -0.312*

(0.088) Firm-specific control Yes Yes Yes Yes Country specific control Yes Yes Yes Yes

Adj. 0.90 0.05 0.56 0.37

Number of observations 50 50 50 50

It seems that in this period, all cultural variables are statistically significant different from zero in the whole industry in model 1. But there seems to be industry effects, since the effects of culture differ across industries, and the adjusted R-squared differs between them as well. Because of the high adjusted , we will look at the first model. Percentile ranks and averages on leverage, cultural values, and liquidity are presented in table 12 in the appendix. Individualism and uncertainty avoidance are the only cultural variables that has been significant once, so only these variable will be discussed. When individualism avoidance moves from the 10th to the 90th percentile, leverage increases by 1.56. The negative interaction between individualism and liquidity however, reduces the increase, depending on the liquidity. If held constant at the 10th percentile, moving from the 10th to the 90th percentile, leverage will be again reduced by 0.78. If held constant at the 90th percentile, leverage will be again reduced by 2.92. Holding liquidity constant, leverage increases between -1.36 and 0.78, depending on liquidity. These results are ambiguous.

When uncertainty avoidance moves from the 10th to the 90th percentile, leverage increases by 1.83. But due to positive interaction, leverage is also reduced by individualism, depending on the liquidity. If held constant at the 10th percentile, moving from the 10th to the 90th percentile, leverage will be again increased by 1.36. Holding liquidity constant at the 90th percentile, leverage will be again increased by 5.11. Holding liquidity constant, leverage increases between 3.19 and 6.94, depending on liquidity. This amount is above the 90th percentile, and thus economic significant. This is according to hypothesis 1. Other cultural variables seem significant as well, but they were not at firm-level so they will be left out at the discussion. These results only come from a one year robustness-check, and to investigate further, the same analysis will be carried out on the whole sample period. Results are presented in table 13.

(14)

14

Table 13, Country-level panel regression 1989-2014 1 All 2 Man 3 Serv 4 Prim Individualism -0.032 -0.01 -0.056 -0.005 (0.251) (0.510) (0.517) (0.638) Uncertainy avoidance -0.014 -0.006 -0.019 0.002 (0.407) (0.588) (0.726) (0.814) Long-term orientation 0.008 -0.001 0.049 -0.005 (0.372) (0.933) (0.337) (0.434) Firm-specific control Yes Yes Yes Yes Country specific control Yes Yes Yes Yes

Adj. 0.19 0.82 0.08 0.24

Number of observations 50 50 50 50

Looking at table 13, culture does not affect leverage at all. There are no significant interaction between culture and non-dummy control variables as well.

Conclusion

This study has gone beyond the mainstream research on capital structure by using culture to explain leverage. 3 cultural values are being investigated, individualism, uncertainty avoidance, and long-term orientation. Behavioral bias is used to explain the link between culture and capital structure. Optimism is used to explain the link between individualism and capital structure. Financial stability and riskiness of debt is used to explain the link between uncertainty avoidance and capital structure. Investor’s preferences is used to explain the link between long-term orientation and capital structure. This lead to the following 3 hypothesis:

Hypothesis 1: Leverage is positively related to individualism

Hypothesis 2: Leverage is negatively related to uncertainty avoidance. Hypothesis 3: Leverage is positively related to long-term orientation.

Country- and firm-level regressions are carried out to investigate the hypotheses.

On firm-level, many countries and periods were dropped due to imbalances. This leaves us with 12 countries in the period 2004-2006, equally divided between developing and developed countries. In the second sample period from 2012-2014, the same 12 countries are used, but with 1 more developing and developed country. Firm-level analyses

(15)

15 only cultural variable that remained statistically and economically significant in both

sample periods. In the most recent period, uncertainty avoidance was more economically significant than the earlier period. However, the sign of the coefficient is positive, opposite of hypothesis 2. This goes into economic intuition and literature, which state that the more one avoid uncertainty, the more financial stability it seeks, and thus the less debt one takes, since debt is associated with risk. It is possible that more debt is not associated with more risk, as assumed. This matter should be investigated upon. But although the sign is different from what he expected, it does not change the fact that uncertainty avoidance affects the choice of capital structure. In the most recent period, individualism is also statistically and economically significant.

We then check for robustness, using country-level regressions of the whole sample period. The first regression is carried out with 2014 as reference year. All 3 cultural values seem to be economic significant, and the sign of uncertainty avoidance is positive as well. However, a regression on the whole sample period indicates that all cultural values are insignificant. These result may indicate that culture may be important determinants of leverage, depending on the period. On firm-level, uncertainty avoidance and individualism are significant in the most recent sample period. On country level, uncertainty avoidance and individualism are significant as well in the most recent period. It is possible that culture is becoming more important in time. The more recent periods had more significant cultural values than the older periods. Shocks, like crises, may have cause the mass to hold more on their morals than before. However, due limitations in this study are the lack of data in earlier years, and in many developing countries, it is still not concluded whether culture is a reliable determinant of leverage. The three hypotheses are rejected. The limited amount of data on developing countries makes it hard to find a balanced panel to analyze. This restricts us to a small sample size and period. The amount of countries for the robustness check can also be doubted to be of use. However, even with the small sample period and size the significant finding tells us that culture can play a role in determining capital structure. Further research should be carried out when more data become available. An event study can be carried out to see whether culture is indeed becoming more involved with the financial world as time passes.

(16)

16

Appendix

Table 1, determinants of capital structure

Determinants BO CN DJ FAN FG GIA KE TW

Bankruptcy probability 0 - 0 0

Fixed assets 0 + + + 0 0

Free cash flow -

Growth opportunities 0 - - - - 0 + 0

Non-debt tax shields + 0 0 0

Profitability - - - -

Size + - + + + 0 - 0

Volatility in cash flow -

Abbreviations used above stand for Booth et al. (2001), Chaplinsky and Niehaus (1990), De Jong et al. (2008), Fan et al. (2013), Frank and Goyal (2009), Giannetti (2003), Kester (1986), and Titman and Wessels (1988).

Table 2, Hofstede’s cultural values

Countries Individualism Uncertainty Avoidance Long-term orientation

Colombia 13 80 13 Indonesia 14 48 62 Pakistan 14 70 50 Peru 16 87 25 China 20 30 87 Singapore 20 8 72 Thailand 20 64 32 Vietnam 20 30 57 Chile 23 86 31 Egypt 25 80 7 Hong Kong 25 29 61 Saudi Arabia 25 80 36 Malaysia 26 36 41 Portugal 27 99 28 Jordan 30 65 16 Mexico 30 82 24 Nigeria 30 55 13 Romania 30 90 52 Philippines 32 44 27 Croatia 33 80 58 Greece 35 100 45 Turkey 37 85 46 Brazil 38 76 44 Russia 39 95 81 Argentina 46 86 20 Japan 46 92 88 Morocco 46 68 14 South Korea 46 92 100 India 48 40 51 Spain 51 86 48 Austria 55 70 60 Poland 60 93 38 Finland 63 59 38 South Africa 65 49 34 Germany 67 65 83 Switzerland 68 58 74

(17)

17 Norway 69 50 35 Ireland 70 35 24 France 71 86 63 Sweden 71 29 53 Denmark 74 23 35 Belgium 75 94 82 Italy 76 75 61 New Zealand 79 49 33 Canada 80 48 36 Hungary 80 82 58 Netherlands 80 53 67 Great Britain 89 35 51 Australia 90 51 21 United States 91 46 26

Table 6, Country level descriptives

Countries Obs Lev Tang Prof Size Liq GDP Inf Law Dev Turkey 2058 -6.97 0.32 0.04 3.01 2.5 4.03 43.03 0 1 France 8899 -1.79 0.18 0.01 7.51 2 1.68 1.56 0 0 Saudi Arabia 828 -0.15 0.47 0.08 7.67 2.41 4.19 4.64 1 0 Ireland 959 -0.05 0.32 -0.02 224.41 2.11 4.93 2.21 1 0 Denmark 1626 -0.02 0.3 -0.02 7.81 2.6 1.44 2.11 0 0 Morocco 343 -0.02 0.31 0.07 1.52 1.93 3.88 2.18 1 1 Jordan 1203 0.11 0.42 0.02 7.15 3.91 4.39 5.23 0 1 Australia 19536 0.14 0.33 0.91 85.64 6.94 3.16 3.07 0 0 Romania 460 0.14 0.52 0.01 2.55 3.05 1.76 54.34 0 1 Singapore 9388 0.18 0.3 -0.06 11.36 2.34 6.47 1.5 1 0 Hungary 268 0.19 0.46 0.04 6.11 2.28 1.84 9.88 0 1 China 35364 0.2 0.34 1.31 32.9 2.24 9.62 5.49 0 1 Nigeria 687 0.23 0.44 0.06 430.32 1.61 5.8 24.04 1 1 Malaysia 14437 0.25 0.36 0.01 5.68 2.93 6.09 3.83 1 1 Colombia 344 0.26 0.43 0.05 3.23 1.83 3.75 13.67 0 1 Poland 4281 0.26 0.32 0.05 39.66 2.81 3.68 11.72 0 0 Egypt 585 0.31 0.41 0.06 72.22 2.77 4.25 9.76 0 1 South Africa 4183 0.31 0.3 0.39 616.47 2.28 2.48 9.16 1 1 Peru 1102 0.32 0.52 0.06 3.28 1.83 3.81 363.27 0 1 Vietnam 2110 0.32 0.29 0.07 1.69 2.09 6.82 17.02 0 1 South Korea 11878 0.35 0.34 0 1.44 1.87 5.34 3.83 0 0 Sweden 5458 0.36 0.17 -0.08 10.64 2.67 2.02 2.48 0 0 Canada 22508 0.39 0.41 -0.21 33.63 3.47 2.3 2.21 1 0 Croatia 522 0.4 0.52 0.02 3.37 1.58 2.04 3.55 0 0 Switzerland 2675 0.46 0.28 0.06 8.01 2.69 1.75 1.15 0 0 Netherlands 2091 0.5 0.24 0.02 -0.31 1.96 2.16 1.94 0 0 Austria 930 0.52 0.29 0.02 8.46 4.68 2.08 1.88 0 0 Belgium 1253 0.53 0.29 0.01 14.92 2.04 1.87 2 0 0 New Zealand 1394 0.53 0.37 0.26 75.29 2.28 2.53 2.31 1 0 Spain 1232 0.53 0.3 0.35 4.95 1.41 2.12 3.3 0 0 Japan 49536 0.54 0.29 -0.11 1.16 2.01 1.27 -0.34 0 0 India 30938 0.56 0.35 0.04 93.81 8.9 6.49 6.9 1 1 Hong Kong 14298 0.58 0.26 -0.28 14.27 3.3 3.83 2.49 1 0 Finland 1714 0.63 0.25 0.05 18.38 1.82 1.79 2.1 0 0 Brazil 4532 0.65 0.37 -0.08 24.91 1.67 2.76 378.65 0 1 Italy 2390 0.67 0.24 0 24.19 1.51 0.8 3.11 0 0

(18)

18 Great Britain 28263 0.75 0.28 -0.11 15.85 2.34 2.09 2.88 1 0 Norway 2841 0.8 0.32 -0.06 60.81 2.8 2.39 3.78 0 0 Chile 1694 0.82 0.46 0.01 17.09 3.02 5.19 7.69 0 1 Mexico 1295 0.82 0.44 0.04 504.03 2.21 2.9 12.6 0 1 Indonesia 4449 0.84 0.4 0.02 4.11 3.73 5.29 12.56 0 1 Philippines 2067 0.88 0.37 0.06 3044.22 5.47 4.21 6.83 1 1 USA 149927 0.91 0.29 -0.96 11.55 3.28 2.51 2.17 1 0 Pakistan 3211 0.92 0.47 0.05 2.68 1.47 4.1 10.53 1 1 Argentina 831 0.96 0.45 0.02 2.99 1.5 3.47 212.65 0 1 Greece 2321 1.11 0.37 -0.01 5.88 2.02 1.06 5.98 0 0 Portugal 596 1.14 0.32 0 1.85 1.05 1.57 4.21 0 0 Thailand 6566 2.18 0.4 0.04 3.4 2.41 4.87 3.62 0 1 Germany 7995 3.1 0.21 -0.03 1.56 3.27 1.71 1.6 0 0 Russia 1440 4.71 0.47 0.05 42.1 6.99 0.8 133.91 0 0

Table 8, percentile ranks and averages, 2004-2006

Leverage Uncertainty avoidance

10th percentile 0 44

90th percentile 1.11 85

Average 0.95 64

Table 10, percentile ranks and averages, 2012-2014

Leverage Uncertainty avoidance Individualism Tangibility

10th percentile 0 44 14 0.024

90th percentile 1.19 85 69 0.678

Average 0.48 64 46 0.323

Table 12, percentile ranks and averages, 2014

Leverage Individualism Uncertainty

avoidance Long-term orientation Liquidity 10th percentile 0.01 20 30 16 1.44 90th percentile 0.79 80 93 82 5.41 Average 0.27 48 64 46 3.47 Reference list:

Adler, N. International diensions of organizational behavior. Ohio: International Thomson Publishing.

Aggarwal, R. “International differences in capital structure norms: an empirical study of large european companies.” Management International Review, 21, 1981, 75-88.

Antonczyk, R. C., and Salzmann, A., J. “Overconfidence and optimism: the effect of national culture on capital structure.” Management

International Review, 54, 2014, 381-415.

Antonczyk, R. C., Brueuer, W., and Salzmann, A., J. “Long-term orientation and relationship lending: a cross-cultural study on the effect of Time preferences on the choice of corporate debt” Research in International Business and Finance, 31, 2014, 132-151

Bancel, F., and Mittoo, U.C. “cross-country Determinants of capital structure choice: a survey of European firms.” Financial Management, 33, 2004,

(19)

19 103-132.

Berk, J., and DeMarzo, P. (2011). Corporate Finance. Harlow, England: Pearson education.

Braum, C.F., Stephan, A., and Talavera, O. “The effects of uncertainty of nonfinancial firms.” Economic Inquiry, 47, 2009, 216-225.

Chui, A.C.W., Lloyd, A.E., and Kwok, C.C.Y. “The determination of capital structure: is national culture a missing piece to the puzzle?” Journal of International Business Studies, 33, 2002, 99-127.

De Jong, A., Kabir, R., and Nguyen, T.T. “Capital structure around the world: the roles of firm- and country-specific determinants.” Journal of

Banking and Finance, 32, 2008,1954–1969.

Frank, M.Z., Goyal, V.K. “Capital structure decisions: which factors are reliably important? Financial Management, 38, 2009, 1-37.

Gao, W., Ng, L., and Wang, Q.“Does corporate headquarters locations matter for firm capital structure? “ Financial Management, 40, 2011, 113-138. Hackbarth, D.. “Managerial traits and capital structure decisions.” Journal of

financial and quantitative analysis, 43, 2008, 843-881

Harris, M., and Raviv, A. “The theory of capital structure.” The Journal of Finance, 46, 1991, 297-355.

Heaton, J., B. “Managerial optimism and corporate finance.” Financial Management, 31, 2002, 33-45

Heine, S., J., and Lehman, D., R. “Is there a universal need for positive self- regard?” Psychological Review, 106, 1999, 766-794.

Heine, S., J., and Lehman, D., R. “Cultural variation in unrealistic optimism: does the West feel more invulnerable than the East?” Journal of personality and social psychology, 106, 1999, 766-794.

Hofstede, G. (2001). Culture and organizations: software of the mind. London: McGraw- Hill

Hofstede, G. (2001). Culture’s consequences: comparing values, behaviors, institutions, and organizations across nations, second edition.

Thousand Oaks CA: Sage Publications.

Hofstede, G. Dimensions of national cultures. Verkregen via: http://geerthofstede.nl/dimensions-of-national-cultures Hofstede, G. Dimensions of national cultures. Verkregen via:

http://geerthofstede.nl/research--vsm

La Porta, R., Lopez-De-Silanes, F., Shleifer, A., Vishny, R. W. “Legal determinants of external finance” The journal of finance, 52, 1997, 1131-1150.

Larwood, L., Whittaker, W. “Managerial myopia: self-serving biases in organizational Planning.” Journal of Applied Psychology, 62, 1977, 194-198.

Malmendier, U., Tate, G., Yan, J. “Overconfidence and early-life experiences: the effect of managerial traits on corporate financial policies.” The journal of the American finance association, 66, 2011, 1687-1733.

Öztekin, Ö. “Capital structure decisions around the world: which factors are reliably important?.” Journal of Financial and Quantitative Analysis, 50,

2015, 301-323.

Rajan, R.G., and Zingales, L. “What do we know about capital structure? some evidence from international data.” The Journal of Finance, 50, 1995, 1421-1460.

The effects of culture on capital structure : a cross-country analysis : is culture a missing factor affecting the level of leverage a firm takes on?

Contents