0
Master Thesis
Cultural Influences on Cash Holdings: Post
Crisis Evidence
Author:
Lukas Ostermeier
Student ID number:
S2732939
Supervisor:
Marnix Reijenga
Date of submission: 13
thJune, 2016
MSc International Financial Management
1
Table of Contents
1. Abstract ... 2
2. Introduction ... 3
3. Literature Review and Hypothesis Development ... 5
4. Data and Methodology ... 11
4.1. Dependent variable ... 12
4.2. Main explanatory variables ... 12
4.3. Firm level control variables ... 13
4.4. Country level control variables ... 18
2 1. Abstract
This research focuses on the cultural influences on cash holdings for public listed firms of 34 global countries from 2009 to 2014. To measure the cultural influences, the scores from Hofstede for individualism and uncertainty avoidance were used. The results are found in this study are: a positive significant influence on cash holdings for uncertainty avoidance and a negative significant influence on cash holdings for individualism. In addition, cultural influences on cash holdings is weaker in multinational firms than in domestic acting firms. At the end of the research, continental differences for cultural influences on cash holdings have been detected as well.
3 2. Introduction
During the last three decades, the level of cash holding has more than doubled on average (Bates, Kahle and Stulz, 2009). In the media, there are reports about the continuous growth of cash holdings within companies, such as Apple and Microsoft (Fernandez and Gonenc, 2015). It is a global phenomenon but in some countries, such as the United States, the rise is way above average (Pinkowitz, Stulz and Williamson, 2012). Fernandez and Gonenc (2015) and Pinkowitz, Stulz and Williamson (2012) states that firms cash holdings are rising in general. From 1990 to 2011, the median percentage of cash and cash equivalent over total assets per firm raised from 8.74% to 11.96%. Especially in the United States, it raised from 5.30% to 13.95% (Fernandes and Gonenc, 2015). How can these differences be explained between countries?
The level of cash holding is an important benchmark for potential investors to see how risk averse firms are. On the other hand, this high amount of cash holdings is very costly. It is costly because instead of holding cash, managers could invest it in projects with potential positive net present values (Al-Najjar, 2013). So, why is the level of cash holding in some countries at a record level? Many researches have focused on different ways to explain this phenomenon. One reason for the increase of cash holding is that idiosyncratic risk has been rising historically, thus leading to higher cash levels for firms (Bates, Kahle and Stulz, 2009). Contrary, Ross (1989) argued that an increase in finance and information technology should lead to lower transaction costs, fewer agency conflicts and lower costs of obtaining external financing and therefore to less need of holding cash.
4
countries. Due to a lack in the prior literature to post crisis data, the time frame between 2009 and 2014 has been chosen. Other research papers about this topic, such as Chang and Noorbaksh (2009) chose the time frame from 1995 to 2004, Ramirez and Tadesse (2009) from 1980 to 2004. Chen, Dou, Rhee, Truong and Veeraghavan (2015) from 1989 to 2009 and Li, Griffin, Yue and Zhao (2013) from 1997 to 2006. Moreover, the papers from Fernandes and Gonenc (2015) and Pinkowitz, Stulz and Williamson (2012) are showing, that the amount of cash holdings is changing relatively fast but not in every country with the same pace. Therefore, it is interesting to know if the findings from Chang and Noorbarkhsh (2009) would be confirmed by this study. Based on these facts, the focus of this paper is the following research question:
Is there a cultural influence on cash holdings in the post crisis time frame?
In a sub question, cultural influences on cash holdings between domestic firms and multinational firms are compared. This sub question has been chosen because in different countries, the variable multinationality had different relationships with cash holdings (see, e.g., Pinkowitz, Stulz and Williamson, 2012; Becker, 2014; Fernandes and Gonenc, 2015). Therefore, it would be interesting to know, if cultural influences in multinationals have more, the same or less influence on cash holdings compared to domestic firms.
Furthermore, Fernandez and Gonenc (2009) have shown in their paper that the level of cash holding differs among geographical areas. But the papers from Chang and Noorbaksh (2009), Ramirez and Tadesse (2009), Chen, Dou, Rhee, Truong and Veeraghavan (2015) and Li, Griffin, Yue and Zhao (2013) were only based on global control groups. In their paper, they did not distinguish geographical areas. It can be assumed that not only the amount of cash holdings differs geographically, but also the cultural impact on cash holdings. For this assumption, four continental subgroups out of Asia, Europe, North America and a global control group are used in this research.
5
detailed overview about the entire data sample and a detailed explanation for every variable used in this research, followed by the regression model development. Next, chapter 5 shows an overview about the regression results and some robustness tests. Finally, in chapter 6 is the conclusion.
3. Literature Review and Hypothesis Development
The level of cash holding is an option to manage risk and is for every firm one of the most important financial strategic decisions. Cash holding is defined with a proxy of cash, which is calculated in the following way: Cash and short-term investments over the difference of total assets minus cash and short term investments, in short cash over net assets (Cash/NA), (see, e.g., Opler, Pinkowitz, Stulz and Williamson, 1999; Dittmar Mahrt-Smith and Servaes, 2003; Pinkowitz, Stulz and Williamson ,2003). According to Ramirez and Tadesse (2009), there are various main reasons why firms are holding cash. One reason is that they can manage risk by holding cash. They can use cash as a buffer for ongoing operations and for undesired states of nature, even when incoming cash flow is declining over time. The next reason is that firms can use liquid assets for speculation. They can for example benefit from exchange rate fluctuation. In addition, when there is asymmetric information about the real value of assets between the owning manager and the equity market, the value of cash becomes a signal of the real value of their assets. In this situation, according to Myers and Majluf (1984), cash holding can be used as financial slack. This theory is based on the assumption that cash will always be more preferred than debt and equity. This assumption is called pecking order theory (Meyers and Majluf, 1984). Another reason for holding cash is related to taxation. Firms, which are generating income international, are trying to avoid repatriated taxes and keep the cash abroad (see, e.g., Foley, Hartzell, Titman and Twite, 2007; Ramirez and Tadesse, 2009).
6
to spend it somehow and when there are no best projects available, they invest it in unprofitable projects. In an efficient capital market, firms would theoretically hold cash only for covering normal operations because when cash is needed for other purposes, it can be raised for free from equity and debt (Ramirez and Tedesse, 2009). In short, agency theory can be explained as a contrary view between managers and shareholders about how much cash would be the most beneficial amount to hold of shareholder wealth (Opler, Pinkowitz, Stulz and Williamson, 1999). Furthermore, Opler, Pinkowitz, Stulz and Williamson (1999) discussed in their research three perspectives why managers tend to hold too much cash. The first one is simply the fact that managers are risk averse and they try to avoid market discipline with a high excess of cash. The second reason can be that managers use cash to get more flexibility with their own projects. Cash can be seen as a free cash flow. Therefore managers can use this cash to invest in projects, otherwise they would not get the permission from the capital market to do it. Thirdly, managers could keep more cash to avoid the payout to shareholders because they want to keep funds in the firm.
Another important impact on cash holdings is the amount of R&D investments. Prior researchers argued that R&D intensive industries have on average more cash holding than other industries. R&D intensive companies have higher costs when they are raising external funds because of higher uncertainty. It is difficult for outsiders to evaluate a project because the projects are mostly tactic knowledge based (Bates et.al. 2009). The second argument is about cash as an operational buffer to manage risk (Brown and Peterson, 2011). They argued that the costs of increasing R&D are very high for a firm. Most R&D expenditures are used to develop tactic knowledge from employees and to increase their output in a team. Because these employees cannot be fired very easily, firms usually cannot reduce this kind of R&D spending. Also, Baum, Caglayan and Talavera (2012) argued that R&D firms have usually less financial flexibility than firms that are mainly investing in physical capital. To be able to finance R&D activities, even in times with low cash flows, firms often hold a larger cash reserve. In 2012, Baum, Caglayan and Talavera (2012) confirmed that financial firms prefer holding more cash over fixed capital investment expenditures when they plan additional future R&D investments.
7
pressure. In 2013, Apple Inc. was urged to buy back a huge amount of their shares after the investor Carl Icahn had accumulated shares for Apple Inc. valued at $1.5 billion. In June 2013, Apple Inc. had $146 billion in cash and cash equivalent. Because of the pressure from Carl Icahn, Apple Inc. agreed to return $100 billion to shareholders by the end of 2015 which would include a $60 billion stock repurchase program (see, e.g., Schloetzer and Lee, 2014; Jiang and Lie, 2016). Also, Jiang and Lie (2016) discovered that self-interested managers are not willing to invest excess cash. According to their results, self-interested managers try to keep their level of cash high until external pressure for their firms is not a problem.
As we have seen there are several aspects, which could influence cash holding and also that there can be big differences between countries. In the following section, this research is looking at country specific aspects of cash holding and the question: Can they explain parts of the differences? One country specific aspect is the level of shareholder protection, because of the agency problematic. According to Pinkowitz, Stulz and Williamson (2006) and Kalcheva and Lins (2007), weak external shareholder protection leads to higher cash holding and lower firm value. They hold more cash because of managerial preferences (Dittmar Mahrt-Smith and Servaes, 2003). Furthermore, the internal governance is an important factor. According to Dittmar and Mahrt-Smith (2007), has one dollar in a weakly governed country only half of the value than one dollar in a strong governed country.
8
H1a: Firms in high avoiding cultures hold more cash than firms in low uncertainty-avoiding cultures.
The other cultural aspect, which is influencing cash holding is individualism (see, e.g., Li, Griffin, Yue and Zhao, 2013; Chen, Dou, Rhee, Truong and Veeraghavan, 2015). Hofstede (2001) defined individualism with people, which tend to hold an independent rather than an interdependent self-image. Managers in individual cultures emphasize individual freedom and avoid strong group cohesion (Griffin, Li, Yue and Zhao, 2009). According to Markus and Kitayama (1991) and DeMooij and Hofstede (2010), managers in individualistic cultures often overemphasize their own ability and this behavior is also seen in financial situations. This means that managers in cultures with high individualism ideally should have less cash holding as they are too optimistic in managing the challenges and do not think they need to hold cash for unpredictable operations in the future. Contrary, managers in a collective culture have more self-monitoring and prefer not to take too much risk (Chen, Dou, Rhee, Truong and Veeraghavan, 2015). Chen, Dou, Rhee, Truong and Veeraghavan (2015) found in their study a negative relationship between individualism and cash holding and argued that one reason for this is that managers in a more collective culture are using cash holding as a signal to the public that their companies are well managed. On the other hand, they agreed that managers in a more individualistic culture are more confident about their financial situation and as a result they tend to underestimate the demand of cash holding. Furthermore, they prefer to engage in acquisitions and capital expenditures when they hold higher amounts of cash because of their self-confidence. Taken the above into account, one could formulate for the cultural factor individualism the following hypothesis:
H1b: Firms in individualistic cultures hold less cash than firms in collective cultures.
9
Their results are based on 160 researches in 62 societies. The research is measured on practices and values. In their questions, they are asking how culture is related to societal, organizational and leadership effectiveness. Overall, they developed nine dimensions of culture (House, Hanges, Javidan, Dorfman and Gupta, 2004).
On the other hand, the GLOBE research is not free from criticism as well. The fact that random people answer the questions can lead to the issues that some of them could have problems to answer it when it gets too complex (Smith, 2006). Furthermore, Hofstede (2006) is criticism GLOBE’s negative correlation between values and practices. He is arguing that the differences between values and practices are not clear. To him, they are not independent, because the correlations of the values and practices in seven out of nine dimensions are negative.
However, after researching the relationship between cultural differences and cash holding, one other question is how multinational acting firms are affected from their cultural backgrounds? In general, multinationality has been an often researched variable in explaining the amount of cash holdings. In the United States, for example, the level of cash holdings of multinational companies increased by 400% from 1988 until 2010. During the same time frame, the assets of the United States firms have only increased by 100%. On the other hand, the cash holdings of domestic firms increased by 66%, whereas the assets rose by 40% (Pinkowitz, Stulz and Williamson, 2012). Pinkowitz, Stulz and Williamson (2012) found a strong relationship between cash holdings and research and development (R&D) for multinationals compared to domestic firms. According to their results, multinationals from the United States have a higher increase in cash holding during the last decades than comparable firms from other countries. This trend is still continuing, because the level of cash holdings in the United States is at the moment at a record level (Platt, 2016).
10
the more advantage it was for multinationals to hold more cash and keep their earnings abroad. Pinkowitz, Stulz and Williamson (2012) partially disagreed and argued that the amount of tax treatment for repatriations could only explain where multinationals are holding their cash but not how much.
On the other hand, Westerman and Von Eije (2005) found in their paper arguments, which support a negative relationship between multinationality and cash holding. They argued that because of the liberalization of the financial market, for example the introduction of the euro, we have a decreasing currency volatility. Less currency volatility means that we need less cash holding as a buffer. Furthermore, Fernandes and Gonenc (2015) have figured out that the strong positive relationship between multinionality and cash holding from Pinkowitz, Stulz and Williamson (2012) is only representative in some countries. They even concluded that the level of cash holding is overall negatively related to the level of foreign sales. Further, they found big differences between multinationals from emerging markets and multinationals from already developed-market firms. Their results show that geographical and industrial diversification plays an important role in the relationship between cash holding and multinationality. According to their results, only multinationals from emerging markets have an increasing amount of cash holding because they need to support their international expansions.
Fernandes and Gonenc (2015) have shown that the relationship between multinationals and cash holding is strongly influenced by geographical and industrial diversification. In this research there is a focus on the question: Can the cultural factors uncertainty-avoidance and individualism partly explain the different relationships between cash holdings and multinationals in different geographical areas? Some researchers argue that cultural differences have less influence on multinationals because they usually have a workforce from different cultures, which can reduce the cultural impact from the home country (Tutar, Altinoz and Cakiroglu, 2014). Furthermore, Ramirez and Tadesse (2009) found in their global research, a lower effect of national culture on cash holdings for multinationals. On the other hand, you can argue that the cultural factors high uncertainty and individualism count especially for multinationals because it is in the most cases more unpredictable to invest in foreign markets than investing in domestic markets. Because the theory from Tutar, Altinoz and Cakirogly (2014) is more settled in the literature, the following hypothesis has been formulated:
11
In the papers from Chang and Noorbaksh (2009), Ramirez and Tadesse (2009), Chen, Dou, Rhee, Truong and Veeraghavan (2015) and Li, Griffin, Yue and Zhao, (2013), the general cultural influence on cash holdings or risk management was researched. All papers have been researched with global data sets, but they did not separate it into local control groups. Furthermore, Fernandez and Gonenc (2009) have shown in their paper that the level of cash holdings differs a lot among geographical areas. According to Chang and Noorbakhsh (2005), firms are holding more cash in more globalized countries and in countries which are more masculine or have a higher level of uncertainty avoidance. One is assuming that not only the amount of cash holdings differs among geographical areas, but also the impact of cultural influences on cash holdings. It is interesting to know if the cultural influences individualism and uncertainty avoidance have significant different influences on cash holdings in different geographical areas. One way to control geographical differences is to separate the data set into continental subgroups. In this research, the data set is separated into four continental control groups: Europe, Asia, North America and a fourth global control group.
H3: The cultural influence on cash holdings has continental differences.
4. Data and Methodology
12
excluded. Overall, data from 89412 firm-year observations from 34 different countries all over the world is selected. From these 89412 firm-year observations are 47927 from international acting firms. Table 6 in the appendix depicts the means from all descriptive statistics for the variables used in this analysis. In this Table, the means of multinationals and of domestic firms are compared. One statistic is, for example, the mean of the dependent variable the Proxy of cash (Cash/NA) and is for multinationals 26.5% and for domestic firms 25.0%. According to these statistics, it seems like that international acting firms hold more cash on average than domestic firms. Also the firm size, the level of research and development and Tobin`s Q is higher in multinationals.
4.1. Dependent variable
The dependent variable is cash (Cash/NA). In constructing Cash/NA, this research is following the researches from Opler, Pinkowitz, Stulz and Williamson (1999) Dittmar Mahrt-Smith and Servaes (2003) and Pinkowitz, Stulz and Williamson (2003). The proxy of cash is calculated in the following way: Cash and short term investments over the difference of total assets minus cash and short term investments. In short, cash over net assets. In the regressions, the natural logarithm of cash over net assets is used.
4.2. Main explanatory variables
13 Table 1
Descriptive statistics for cultural variation.
This table provides a univariate comparison of the mean, median, minimum and maximum for both explanatory variables. The sample period is from 2009 to 2014. IDV and UAI are standing for the cultural scores from Hofstede for Individualism and Uncertainty Avoidance.
N (Countries) Mean Median Std Min Max
IDV 34 55 48 24.5 14 91
UAI 34 54.3 46 25.42 8 112
4.3. Firm level control variables
The following firm level control variables are all collected from Orbis. The first variable is growth opportunities, measured as Tobin´s Q (TOBINQ). Growth opportunities can have a positive impact on cash holding because possible underinvestment problems are mitigated by higher cash holdings (Myers, 1977). Also, firms with high growth opportunities have usually more projects to invest in. Therefore one expects a positive influence from growth opportunities to cash holdings (Ramirez and Tadesse, 2009). Tobin´s Q is calculated by the firms’ total market value divided with the total asset value. To get the total market value, this research uses the sum of the book value of total assets and the market value of common equity, and subtracts the book value of common equity afterwards (Fernandes and Gonenc, 2015).
The second control variable is firm size (Size) as a proxy to capital markets (Opler, Pinkowitz, Stulz and Williamson, 1999). In larger firms, there is usually a lower information asymmetry compared to smaller firms. In larger firms, it is therefore less costly to convey information about the real value of projects and get access to debt and equity markets (Ramirez and Tadesse, 2009). One expects a negative relationship between the firm size and holding cash. It is measured using the natural logarithm of total assets.
14
that much cash anymore. Cash flow is measured by the net cash flow of operating activities over the difference of total assets minus cash and short-term investments.
The next control variable is research and development (R&D). As we have seen in the literature review, it can be assumed that firms with a high amount of R&D spending will have a higher amount of cash holding. The main reason is that they represent investments in assets whose value is hard to measure. R&D is measured as the ratio of research and development expenses over total sales (Baum, Caglayan and Talavera, 2012).
The fifth variable is capital expenditures (CAPEXP). Capital expenditures consume cash and are therefore needed as a control variable (Ramirez and Tadesse, 2009). On the other hand, according to Bates, Kahle and Stulz (2009) capital expenditures are in general easier to finance than R&D expenditures and should therefore have less influence. Myers and Majluf (1984) are even assuming a negative influence from capital expenditures on cash holding because they suggest that firms save financial slack to fund future investment opportunities. Capital expenditures are measured by the ratio of additions to fixed assets divided by total assets (Ramirez and Tadesse, 2009).
Another control variable is a leverage (Leverage) and is important because firms with a high amount of leverage will be stronger monitored from banks and bondholders. Because of agency theory, managers prefer to have a low debt to divert cash in order to derive private benefits (Ramirez and Tadesse, 2009). Therefore, one assumes a negative relationship between high leverage and the level of cash because of managerial preferences. Leverage consists of total liabilities over total assets.
The next control variable is net working capital (NWC). According to the research of Pinkwotitz, Stulz and Williamson (2003) and Dittmar Mahrt-Smith and Servaes (2003), net working capital is used as a proxy for liquidity of the firm and an alternative to holding cash. It is measured with net working capital over total assets minus cash (Ramirez and Tadesse, 2009).
The last firm level control variable is a dummy variable for paying dividends (DIV_PAYER). If a firm paid a dividend in the year, the dummy variable equals 1 and 0 otherwise. According to Fernandez and Gonenc (2015) firms, which are paying dividends, should have the less need in holding cash.
15
(Cash/NA) varies from 7.24% in Turkey to 36.78% in Hong Kong. More than 30% of cash holdings over net assets are seen in the countries: Canada, China, Israel, Japan, Singapore and the United States. These statistics confirm the prior findings from Fernandez and Gonenc (2015) and showing that the level of cash holdings varies highly among geographical areas.
The mean for Tobin`s Q varies from 0.75 in Russia to 2.04 in Saudi Arabia. The firm size mean varies from 3.58 in India to 7.17 in Mexico. R&D varies from 0% in Poland to 3.7% in the United States. The mean leverage varies from 8.32% in Poland to 24.98% in Portugal. The mean for capital expenditures varies from 0.01% the United States to 5.97% in China. According to Orbis, only 0.14% of all public listed firms from the United States paid dividends between 2009 and 2014, while 90.11% in Japan and 95.87% in China.
16 Table 2
Financial descriptive statistics by country
This table provides a univariate comparison of the mean of all financial variables for all researched countries. The sample period is from 2009 to 2014. The variables are defined as follows: The independent variable is Cash/Na, which is Cash and cash equivalent over net assets. CAPEXP is the ratio of additions to fixed assets divided by total assets. CF is measured by the net cash flow of operating activities over the difference of total assets minus cash and short-term investments. DIV_PAYER is a dummy variable, which is equal to one when a firm paid dividends in the measured year. LEVERAGE is measured with total liabilities over total assets. NWC is measured with net working capital over total assets minus cash (Ramirez and Tadesse, 2009). R&D is measured as the ratio of research and development expenses over total sales. SIZE is measured by the natural logarithm of total assets. TOBINQ is measured by the sum of the book value of total assets and the market value of common equity and subtracted with the book value of common equity afterward (Fernandes and Gonenc, 2015).
Mean Mean Mean Mean Mean Mean Mean Mean Mean Mean Mean Mean Country N TA SALES CASH CF CASH/NA TOBINQ SIZE NWC R&D LEVERAGE CAPEXP DIV_PAYER
18 4.4. Country level control variables
The first country level control variable is market capitalization (MKTCAP), which is measured as a ratio of market capitalization over country´s gross domestic product (GDP). It measures the effect of financial market development in each country. Stulz and Williamson (2003) and Anita, Lin and Patzalis (2007) are suggesting that cultural dimensions are influenced by financial market development in each country. The Data for both variables is collected from the World Bank and are from 2009-2014 (Chang and Noorbakhsh 2009).
The next variable is a Governance Index from La Porte, Lopez-de-Silanes, Shleifer and Vishny (1998) which is called Anti-director Right Index (ADRI) and takes a value from zero to five (Chang and Noorbakhsh 2009). The Anti-director Right Index shows that corporate governance can influence minority shareholders to a certain level. According to the agency theory, managers in a country with weak shareholders’ rights prefer to hold more cash to invest it in value destroying projects. A high level of shareholder’s right can control the level of cash better. Therefore one expects a negative relationship to cash holding (Ramirez and Tadesse, 2009).
The next variable is the dummy variable for common law (COMMON_LAW). If a country comes from an English common law country origin, the dummy variable 1 is included and 0 otherwise (La Porte, Lopez-de-Silanes, Shleifer and Vishny, 1998). Dittmar Mahrt-Smith and Servaes (2003) concluded that firms in common law countries are holding less cash than firms out of countries with a civil law system, because they are less susceptible to agency problems (Chang and Noorbakhsh, 2009).
The fourth control variable is the index of economic freedom (IOEF). The index of economic freedom is measuring how much economic freedom people have in each country. The index data are used for the sectors: business, trade, governance and policy. The index is measured by a list of 50 independent variables of country scores and is divided into 10 broad factors of economic freedom (see, e.g., Ramirez and Tadesse, 2009; Heritage, 2016). Based on the agency problematic, more economic freedom should lead to more cash holdings, because managers tend to hold too much cash (Opler, Pinkowitz, Stulz and Williamson, 1999).
19
structure. Included in this index are the size, the activity and the efficiency. Furthermore, they created ratios for stock market and banking sector development. In bank-based systems, banks are playing the leading role in overseeing the investment decisions from corporate managers. In market-based systems, securities markets also play an important role. Pinkowitz and Williamson (2001) argued in their research that Japanese firms are holding on average more cash than firms from the United States, because Japan has a bank-based and the United States a market-based financial system (Chen, Dou, Rhee, Truong and Veeraghavan, 2015). Based on these argumentations, one assumes a positive relationship between the dummy variable and the level of cash holdings.
Table 3 depicts the descriptive statistics for the cultural explanatory variables and all country level control variables for each country. Individualism in Indonesia, China, and Singapore, has the minimum scores of 14, 20, and 20 points respectively. The United States has a maximum score of 91 followed by Australia with 90, and the United Kingdom with 89 points. In Singapore, uncertainty avoidance has a minimum score of 8 points, followed by Hong Kong with 29 points. Greece and Portugal have the maximum scores of 112 and 104 points respectively. The Anti-director right index varies between 0 and 5. Countries with a score of 0 are Belgium, China, Poland, Russia and Saudi Arabia. Countries with a score of 5 are Hong Kong, India, United Kingdom, Canada, United States and South Africa. The mean for market capitalization varies from 0.00 in Denmark, Finland, Italy, Russia and Sweden to 10.79 in Hong Kong. Singapore with 2.49 has the second highest mean. The overall average is 0.85. In Russia and China, the minimum score of IEOF is 50.96 and 51.96 respectively. The maximum score of IEOF is 89.62 and 87.54 points in Hong Kong and Singapore respectively. In this data sample, 48% of all firms are located in countries with a common law history and 57% in countries with bank-based financial systems. China, Poland, Russia and Saudi Arabia are not classified as a bank or market-based financial system.
20
21 Table 3
Country level control variables
This table provides a univariate comparison of all country level variables for all researched countries. The sample period is from 2009 to 2014. The variables are defined as follows: IDV and UAI are standing for the cultural scores from Hofstede for Individualism and Uncertainty Avoidance. COMMON_LAW is a dummy variable, which is equal to one when the home country has an English common law country origin and zero otherwise (La Porte, Lopez-de-Silanes, Shleifer and Vishny, 1998). MKTCAP is measured by market capitalization over country´s GDP. IOEF is published as an annual guide published by The Wall Street Journal and the Heritage Foundation (www.heritage.org). BNKMKT is a dummy variable, which is equal to one when the home country has a bank-based financial system and zero when it has a market-based financial system. ADRI tells me which level minority shareholders can be influenced by corporate governance (La Porte, Lopez-de-Silanes, Shleifer and Vishny, 1998).
Mean Mean
Country N IDV UAI ADRI COMMON_LAW MKTCAP IOEF BNKMKT
China 13058 20 30 0 0 0.54 51.96
Hong Kong SAR, China 623 25 29 5 1 10.79 89.62 0
India 15065 48 40 5 1 0.75 54.71 1 Indonesia 538 14 48 2 0 0.44 57.2 1 Japan 13585 46 92 4 0 0.73 72.37 1 Saudi Arabia 448 38 68 0 0 0.61 63.28 Singapore 2883 20 8 4 1 2.49 87.54 0 Asia 46200 36.95 50.68 3.15 40.20% 0.93 61.76 89.28% Austria 272 55 70 2 0 0.26 71.53 1 Belgium 430 75 94 0 0 0.59 70.05 1 Denmark 542 74 23 2 0 0 77.4 0 Finland 495 63 59 3 0 0 73.64 1 France 2825 71 86 3 0 0.7 63.81 1 Germany 2619 67 65 1 0 0.42 71.78 1 Greece 1007 35 112 2 0 0.24 58.39 1 Ireland 281 70 35 4 0 0.47 78.35 1 Israel 1439 54 81 3 1 0.74 67.82 1 Italy 1030 76 75 1 0 0 60.76 1 Netherlands 499 80 53 2 0 0.79 74.57 0 Norway 512 69 50 4 0 0.52 70.04 1 Poland 1524 60 93 0 0 0.34 64.72 Portugal 243 27 104 3 0 0.32 63.81 1 Russia 1979 39 95 0 0 0 50.96 Spain 546 51 86 4 0 0.79 68.96 1 Sweden 1769 71 29 3 0 0 72.17 0 Switzerland 833 68 58 2 0 1.98 81.02 0 Turkey 99 37 85 2 0 0.32 63.36 0 United Kingdom 4743 89 35 5 1 0.54 75.53 0 Europe 23687 67.20 65.81 2.51 26.10% 0.46 68.50 57.96% Canada 2352 80 48 5 1 1.18 80.18 0 United States 10545 91 46 5 1 1.22 77.31 0 North America 12897 88.99 46.36 5.00 100% 1.21 77.83 0% Australia 3660 90 51 4 1 1.02 82.57 0 Brazil 1124 38 76 3 0 0.54 56.86 0 Mexico 452 30 82 1 0 0.4 66.83 0 New Zealand 428 79 49 4 1 0.24 81.8 1 South Africa 964 65 49 5 1 2.42 62.73 0
22 4.5. Regression Model
To examine the results for the hypotheses H1a and H1b between cash holding, uncertainty avoidance and individualism, this research is using the following pooled ordinary least squares (OLS) regression with t statistics:
Model 1 Cash/NAit = α + β1IDVit + β2UAIit + β4CAPEXPit + β5CFit + β6DIV_PAYERit +
β7LEVERAGEit + β8NWCit + β9R&Dit + β10COMMON_LAWit + β11SIZEit
+ β12TOBINQit + β13MKTCAPit + β14IOEFit + β17BNKMKTit + β16ADRIit +
β17IndDi + β18YearDi + ℮it
The subscript it stands for firm and year. The independent variable is Cash/Na, which is the natural logarithm of Cash over net assets. IDV and UAI stand for the cultural scores from Hofstede for Individualism and Uncertainty Avoidance. Furthermore, all firm level and country level control variables are included. At the end, industry and year dummies for firm i and the error term are included. Industries are classified as two-digit SIC numbers. This regression shows the overall effects, which the cultural explanatory variables have on cash holdings. Because the dummy variable BNKMKT is not available for the whole dataset, the number of firm year observations for Model 1 shrinks to 72403.
For testing Hypothesis 2, this research uses the first introduced main regression (Model 1) with two different control groups. The first control group is with domestic firms (Model 2). The other control group is with multinational firms (Model 3), which have a minimum ownership from at least 50.01% for at least one subsidiary located in another country (Paul, 2008). Because of nearly perfect multicollinearity the control variable COMMON_LAW is excluded in both regression and the control variable BNKMKT is excluded in the regression for the subgroup of multinational firms.
23
comparative reasons. Part of these Control group are: Australia, Brazil, Mexico, New Zealand and South Africa. Same as in (Model 2) and (Model 3), variables that are highly correlated, more than 0.8, are excluded from the regressions. When the cultural explanatory variables individualism and uncertainty avoidance are highly correlated, they are regressing in two different models. Based on these criteria, the Asian subgroup is regressed in (Model 4), European countries in (Model 5a) and (Model 5b), North American in (Model 6a) and (Model 6b) and at least the global control group in (Models 7a) and (Model 7b). The Asian control group is the biggest one and consists of 46200 firm year observations, the European of 23687, the North American of 12897 and the global control group of 6638.
5. Results
Based on the regression results from (Model 1), presented in Table 4 in the appendix, you can see that the explanatory variable UAI has, as predicted in H1a, a positive relationship to cash holdings and is significant at the one percent level. The explanatory variable IDV has, as predicted in H2a, a negative relationship to cash holdings and is also significant at the one percent level. Both findings are consistent with prior findings from Li, Griffin, Yue and Zhao (2013), Chen, Dou, Rhee, Truong and Veeraghavan (2015) Chang and Noorbaksh (2009) and Ramirez and Tadesse (2009). According to these results, one percent change in uncertainty avoidance leads to a 0.0028% change in cash holdings and a one percent change in individualism leads to a change in cash holdings of minus 0.002%. This means for example, if one country score from Hofstede for uncertainty rises one percent in his next survey, the firms in this country will hold on average 0.0028% more cash.
24
on cash holdings. This is consistent with the argument that firms hold more cash as an operational buffer to manage the risk of investing in the unpredictable R&D field (see, e.g., Bates, Kahle and Stulz, 2009; Brown and Peterson, 2011; Baum, Caglayan and Talavera, 2012). The findings for Tobin`s Q are also consistent with prior argumentations. It has a positive influence on cash holdings and is significant at the one percent level. This result was expected because firms with high growth opportunities have usually more projects to invest in, thus more need of cash (Ramirez and Tadesse, 2009). The shareholder protection, measured as Anti-director Right Index (La Porte, Lopez-de-Silanes, Shleifer and Vishny, 1998), has according to the results, a significant positive influence on cash holdings. This is contrary to the expectation, because according to the agency theory, managers in a country with weak shareholders` rights supposed to prefer to hold more cash (Ramirez and Tadesse, 2009). Against the expectations are the findings for the dummy variable for bank vs. market-based financial systems. Pinkowitz and Williamson (2001) argued in their research that Japanese firms are holding on average more cash than firms from the United States, because Japan has a bank-based and the United States a market-based financial system (Chen, Dou, Rhee, Truong and Veeraghavan, 2015). Contrary to the theory, has the regression result of the dummy variable an at the one percent level significant negative influence on cash holdings. The result of IOEF shows that managers, which have more economic freedom, are holding more cash. This is consistent with the agency argumentation, which says that managers tend to hold too much cash when they are free to do it (Opler, Pinkowitz, Stulz and Williamson, 1999). Finally, there is a positive and at the one percent level significant influence on cash holdings for the firm size and for the level of market capitalization.
25
influence in both control groups, while uncertainty avoidance has significantly more influence for domestics. These findings are consistent with the findings from Ramirez and Tadesse (2009) and the arguments that the cultural impact from the home country will be less after becoming multinational. Further, it does not support the theory, that cultural backgrounds especially affect multinationals, because it is usually more risky to invest in foreign markets. Overall, it cannot be a satisfactory part of the explanation why multinationals are holding more cash after becoming multinational. Same, with the geographical differences for multinationals. Individualism and uncertainty avoidance have a significant influence on cash holdings for multinationals, but only a small one. These influences are not big enough to explain the different relationships between multinationals and cash holdings in different countries and researches (see, e.g., Pinkowitz, Stulz and Williamson, 2012; Becker, 2014; Fernandez and Gonenc, 2015).
What about the general geographical differences? As we have seen in Table 2, the average firm level of cash holdings differs from 33.78% in Hong Kong to 7.24% in Turkey. In general, the North American countries and some Asian countries are holding a high amount of cash. To test if the cultural influences on cash holdings also vary geographically, regressions with four different continental subgroups are made. The results from the regression models (Model 4) to (Model 7b) are documented in Table 5 in the appendix. Based on the results, it can be concluded that the explanatory variable uncertainty avoidance has an at the one percent level significant positive influence on cash holdings in Asia, Europe and the global control group. In North America, uncertainty avoidance has no significant influence on cash holdings. Individualism on the other hand has an at the one percent level significant negative influence on cash holdings for the Asian and global subgroup. In Europe, it has an at the one percent level positive influence on cash holdings. In North America, individualism has no significant influence.
26
underestimate the demand of cash holdings (Chen, Dou, Rhee, Truong and Veeraghavan, 2015). In the North American countries, Canada and the United States, both cultural control variables have no significant influence on cash holdings between 2009 and 2014. This means that the high amount of cash holdings in North America cannot be explained with cultural influences.
5.1 Robustness test
According to the results, all hypotheses can be confirmed, but in order to prove the validity of the results, two robustness tests are performed.
5.1.1 Interaction test
As we have seen in the literature review are the level of R&D and the level of shareholder protection two of the most important impacts on cash holdings: According to Dittmar Mahrt-Smith and Servaes (2003), leads weak external shareholder protection to higher cash holding and lower firm value. They hold more cash because of managerial preferences. In this research, shareholder protection is measured with Anti-director Right Index (ADRI) from La Porta, Lopez-de-Silanes, Shleifer and Vishny (1998).
27
between the cultural explanatory variables and cash holding for multinationals compared to domestic firms. These results are consistent with prior findings in this research, that multinationals are less affected by their home country regulations. The interaction test results for the continental control groups are interesting as well. In Europe and the global control group, ADRI has a positive and at the one percent level significant interaction with IDV and UAI. In Asia, the results were opposite to Europe. Both variables have a negative interaction with ADRI. In North America, the test could not be performed because of nearly perfect multicollinearity. In summary, this means that the level of shareholder protection has in most cases a significant influence on the relationship between the explanatory variables and cash holdings, but the influence is sometimes positive and sometimes negative. The positive interaction of ADRI and R&D for IDV in Europe can be part of the reason why IDV has a positive significant influence on cash holdings in the European control group.
5.1.2. Alternative cultural measurements
28
House, Hanges, Javidan, Dorfman and Gupta (2004) only collected the collectivism score, which is the opposite of individualism score. Therefore, a positive influence on cash holdings is expected from both alternative cultural variables, which are SICP and UASP. Based on the results in Table 15, both alternative cultural variables have an at the one percent level significant positive influence on cash holdings and thus it is consistent with the other regression. H1a and H1b are both confirmed.
For the multinational subgroup and the domestic subgroup, SICP has an at the one percent level significant positive influence on cash holdings, but UASP an at the one percent level significant negative influence on cash holdings. An explanation is the strong negative correlation between UASP and ADRI. Overall, hypothesis 2 can be still confirmed, because both alternative cultural variables have a stronger influence on cash holdings in the domestic control group compared to the multinational control group.
The regression results for the continental control groups for hypothesis 3 are partly different than the regression results measured with Hofstede´s cultural scores. Uncertainty avoidance is not significant for the Asian, the North American and the global subgroup. Only in Europe it has an at the one percent level significant positive influence on cash holdings. The other explanatory variable SICP has in Asia an at the one percent level significant positive influence on cash holdings, but in Europe an at the one percent level significant negative influence. In North America and in the global control group, SICP has no significant influence on cash holdings. In summary, these alternative results confirmed the strong continental differences and therefore H3, even when the results are slightly different to the regression results measured with Hofstede´s cultural scores.
6. Conclusion
29
In this study, the results show that the cultural influence can be part of the explanation why the level of cash holdings differ so much between single countries and why the level of cash holding is in some countries on a record level at the moment. But it can be only a small part of the explanation because the influence is too small to explain the whole phenomena.
Furthermore, this thesis investigated the differences of cultural influences on cash holdings for multinationals compared to domestic firms. These two subgroups have been chosen because the connection between multinationals and cash holdings had in prior studies a wide range of results (see, e.g., Pinkowitz, Stulz and Williamson, 2012; Becker, 2014; Fernandez and Gonenc, 2015). Based on the results, there are no differences for the cultural variable individualism on cash holdings between multinationals and domestic firms. The cultural variable uncertainty avoidance has significantly more influence on cash holdings for domestic firms. These findings are consistent with the findings from Ramirez and Tadesse (2009). These results show that the cultural background cannot be a sufficient reason why firms are holding more cash after becoming international. Furthermore, it cannot explain the different results for the relationship between multinationals and cash holdings in prior studies, because the cultural influences on cash holdings are too small for multinationals.
At the end, this research is showing that the cultural influence on cash holdings differs among Europe, Asia, North America and a global control group. In Asia and the global control group, the influences of individualism on cash holdings are negative and of uncertainty avoidances are positive, which is consistent with prior findings. Europe has a positive influence on cash holdings for uncertainty avoidance and individualism. The unexpected positive influence of individualism on cash holdings in Europe can be partly explained by the positive interaction with the level of shareholder protection and the level of R&D investments. In North America, both factors are not significant. This is interesting because especially in North America, the level of cash holdings is extraordinarily high. Again, this is showing that the cultural influence on cash holdings can only be seen as a small puzzle why the level of cash holdings is at a record level at the moment.
30
could compare these results with prior time frames and with other countries. Another limitation is that for testing geographical differences, only continental control groups were used. In future researches other diversifications can be included as well. The last limitation is the quality of the data source. In this Research most of the data were collected on Orbis. According to Orbis only 0.14% of all public listed firms in the United States pay dividends. This is one example why the accuracy for some control variables is doubtful. In future researches, alternative data sources should be used for comparative reasons.
Acknowledgments:
31 7. Reference List:
Anita, M., Lin, B., Pantzalis, C., 2007. Cultural distance and valuation of multinational corporations. Journal of Multinational Financial Management, 17, 365-383.
Al-Najjar, B., 2013. The financial determinants of corporate cash holdings: Evidence from some emerging markets. International Business Review, 22(1), 77-88.
Bates, T. W., Kahle, K. M., Stulz, R. M., 2009. Why do US firms hold so much more cash than they used to? The Journal of Finance, 64(5), 1985-2021.
Baum, C. F., Caglayan, M., Talavera, O., 2012. R&D Expenditures and Geographical Sales Diversification. Boston College Working Papers.
Becker, J., 2014. The impact of R&D on cash holdings in multinationals – Evidence from the UK. University of Groningen, Masterthesis.
Brown, J. R., Petersen, B. C., 2011. Cash holdings and R&D smoothing. Journal of Corporate Finance, 17(3), 694-709.
Chen. Y., Dou, Y., Rhee, S. G., Truong, C., Veeraraghavan, M., 2015. National culture and corporate cash holdings around the world, Journal of Banking & Finance, 50, 1.
Demirguc-Kunt, A., & Levine, R., 1999. Bank-Based and Market-Based Financial System: Cross-Country Comparisons. Working Paper. The World Bank and University of Minnesota.
DeMooij, M., Hofstede, G., 2010. The Hofstede Model: Applications to global branding and advertising strategy and research. International Journal of Advertising, 29, 85–110.
Dittmar, A., Mahrt-Smith, J., Servaes, H., 2003. International corporate governance and corporate cash holdings. Journal of Financial & Quantitative Analysis, 38, 111–133.
Dittmar, A., Mahrt-Smith, J., 2007. Corporate governance and the value of cash holdings. Journal of Financial Economics, 83, 599–634.
Fazzari, S., Hubbard, R. G., Petersen, B. C., 1988. Financing constraints and corporate investment, Brookings Papers on Economic Activity, 19(2), 141-195.
32
Foley,C. F., Hartzell, J., Titman, S., Twite, G., 2007. Why do firms hold so much cash? A tax-based explanation. Journal of Financial Economics, 86(3), 579-607.
Griffin, D., Li, K., Yue, H., Zhao, L., 2009. Cultural Values and Corporate Risk-taking. Working paper.
Hofstede, G., 1980. Culture's consequences: International differences in work-related values Sage Publication, Beverly Hills, CA.
Hofstede, G., 2001. Culture’s Consequences: Comparing Values, Behaviors, Institutions, and Organizations Across Nations. Sage Publication, Beverly Hills, CA.
Hofstede, G., 2006. What did GLOBE really measure? Researchers’ minds versus respondents’ minds. Journal of International Business Studies, 37.
House, R., Hanges, P., Javidan, M., Dorfman, P., Gupta, V., 2004. Culture, leadership, and organizations. The GLOBE study of 62 societies. Sage Publications, Thousand Oaks, California.
Jensen, M. C., 1986. Agency costs of free cash flow, corporate finance, and takeovers. American Economic Review 76(2), 323-329.
Jensen, M., Meckling, W., 1976. Theory of the firm: Managerial behaviour, agency costs, and ownership structure, Journal of Financial Economics, 3, 305-360.
Jiang, Z., Lie, E., 2016. Cash holding adjustments and managerial entrenchment. Journal of Corporate Finance, 36, 190-205.
Kalcheva, I., Lins, K.V., 2007. International evidence on cash holdings and expected managerial agency problems. Review of Financial Studies, 20, 1087–1112.
La Porta, R., Lopez-de-Silanes, F., Shleifer, A., Vishny, R., 1998. Law and finance. Journal of Political Economy, 106, 1113-1155.
Li, K., Griffin, D., Yue, H., Zhao, L., 2013. How does culture influence corporate risk-taking? Journal of Corporate Finance, 23, 1-22.
Li, Y., Zahra, S. A., 2012. Formal institutions, culture, and venture capital activity: A cross-country analysis. Journal of Business Venturing, 27, 95–111.
33
Myers, S., Maljluf, N., 1984. Corporate financing and investment decisions when firms have information that investors do not have. Journal of Financial Economics 5, 147-175.
Myersm S. C., 1977. Determinants of corporate borrowing. Journal of Financial Economics, 5(2), 147-175
Opler, T., Pinkowitz, L., Stulz, R., Williamson, R., 1999. The determinants and implications of corporate cash holdings. Journal of Financial Economics, 52(1), 3-46.
Paul, J., 2008. Business Environment:Test & Cases 2E, Tata McGraw-Hill Education, New Dehli.
Pinkowitz, L., Stulz, R., Williamson, R., 2003. Do firms in countries with poor protection of investor rights hold more cash? NBER working paper.
Pinkowitz, L., Stulz, R., Williamson, R., 2006. Does the contribution of corporate cash holdings and dividends to firm value depend on governance? A cross-country analysis. Journal of Finance, 61, 2725–2751.
Pinkowitz, L., Stulz, R. M., Williamson, R., 2012. Multinationals and the high cash holdings puzzle. National Bureau of Economic Research, (No. w18120).
Platt, E., 2016. US companies’ cash pile hits $1.7tn. Financial Times, May 20, 2016.
Ramirez, A., Tadesse, S., 2009. Corporate cash holdings, uncertainty avoidance, and the multinationality of firms. International Business Review, 18, 387-403.
Ross, S. A., 1989. Institutional market, financial marketing, and financial innovation. The Journal of Finance, 44(3), 541-556.
Saunders, M. N., Lewis, P., Thornhill, A., 2011. Research Methods for Business Students. 5/e. Pearson Education.
Schloetzer, J. D., Lee, R., 2014. The Activism of Carl Icahn and Bill Ackman, Georgetown McDonough School of Business Research Paper, No. 2442317.
Smith, P., 2006. When elephants fight, the grass gets trampled: The GLOBE and Hofstede projects. Journal of International Business Studies, 37(6), 915-921.
34
Tutar, H., Altinoz, M., Cakiroglu, D., 2014. A Study on Cultural Difference Management Strategies at Multinational Organizations. 10th International Strategic Management
Conference 2014, 345-353.
Westerman, W., Von Eije, H., 2005. Multinational Cash Management in Europe Towards Centralisation and Disintermediation: The Philips Case. Managerial Finance, 31(10), 65-74.
Orbis, 2016. Orbis: Company information across the globe. (http://www.bvdinfo.com/en-gb/our-products/company-information/international-products/orbis). Retrieved 05-04-2016
35 8. Appendix
Table 4
Regression results Model´s 1, 2 and 3
This Table reports the pooled OLS regression results with firm and industry dummies. The sample period is from 2009 to 2014. The variables are defined as follows: The independent variable is Cash/Na, which is the natural logarithm of Cash and cash equivalent over net assets. IDV and UAI are standing for the cultural scores from Hofstede for Individualism and Uncertainty Avoidance. CAPEXP is measured by the ratio of additions to fixed assets divided by total assets. CF is measured by the net cash flow of operating activities over the difference of total assets minus cash and short-term investments. DIV_PAYER is a dummy variable, which is equal to one when a firm paid dividends in the measured year. LEVERAGE is measured by total liabilities over total assets. NWC is measured by net working capital over total assets minus cash (Ramirez and Tadesse, 2009). R&D is measured as the ratio of research and development expenses over total sales. COMMON_LAW is a dummy variable, which is equal to one when the home country has an English common law country origin and zero otherwise (La Porta, Lopez-de-Silanes, Shleifer and Vishny, 1998). SIZE is measured by the natural logarithm of total assets. TOBINQ is measured by the sum of the book value of total assets and the market value of common equity and subtracted with the book value of common equity afterward (Fernandes and Gonenc, 2015). MKTCAP is measured by market capitalization over country´s GDP. IOEF is published as an annual guide published by The Wall Street Journal and the Heritage Foundation (www.heritage.org). BNKMKT is a dummy variable, which is equal to one when the home country has a bank-based financial system and zero when it has a market-based financial system. ADRI tells me which level minority shareholders can be influenced by corporate governance (La Porta, Lopez-de-Silanes, Shleifer and Vishny, 1998). Robust standard errors are clustered at the firm level and ***, **; and * stands for the statistically significant at the 1%, 5% and 10% levels. The standard errors are presented in the brackets.
Model 1 Model 2 Model 3 All Domestics Multinationals
36 IOEF 0.013381*** 0.019007*** 0.005132*** (0.00) (0.00) (0.00) BNKMKT -0.05087*** -0.04846*** (0.01) (0.01) ADRI 0.029987*** 0.010098** -0.01455*** (0.00) (0.00) (0.00)
Year Dummy YES YES YES
Industry Dummy YES YES YES
N 72403 29152 46200
R² 0.502828 0.495959 0.470761
Adj. R² 0.502202 0.494416 0.469798
Table 1
Regression results Model`s 4-7
This Table reports the pooled OLS regression results with firm and industry dummies. The sample period is from 2009 to 2014. The variables are defined as follows: The independent variable is Cash/Na, which is the natural logarithm of Cash and cash equivalent over net assets. IDV and UAI are standing for the cultural scores from Hofstede for Individualism and Uncertainty Avoidance. CAPEXP is measured by the ratio of additions to fixed assets divided by total assets. CF is measured by the net cash flow of operating activities over the difference of total assets minus cash and short-term investments. DIV_PAYER is a dummy variable, which is equal to one when a firm paid dividends in the measured year. LEVERAGE is measured by total liabilities over total assets. NWC is measured by net working capital over total assets minus cash (Ramirez and Tadesse, 2009). R&D is measured as the ratio of research and development expenses over total sales. COMMON_LAW is a dummy variable, which is equal to one when the home country has an English common law country origin and zero otherwise (La Porta, Lopez-de-Silanes, Shleifer and Vishny, 1998). SIZE is measured by the natural logarithm of total assets. TOBINQ is measured by the sum of the book value of total assets and the market value of common equity and subtracted with the book value of common equity afterward (Fernandes and Gonenc, 2015). MKTCAP is measured by market capitalization over country´s GDP. IOEF is published as an annual guide published by The Wall Street Journal and the Heritage Foundation (www.heritage.org). BNKMKT is a dummy variable, which is equal to one when the home country has a bank-based financial system and zero when it has a market-based financial system. ADRI tells me which level minority shareholders can be influenced by corporate governance (La Porta, Lopez-de-Silanes, Shleifer and Vishny, 1998). Robust standard errors are clustered at the firm level and ***, **; and * stands for the statistically significant at the 1%, 5% and 10% levels. The standard errors are presented in the brackets.
Model 4 Model 5a Model 5b Model 6a Model 6b Model 7a Model 7b
Asia Europe UAI Europe IDV North
America UAI
North America IDV
Global UAI Global IDV
37 R&D 0.001916 1.307134*** 1.341005*** 1.389224*** 1.389224*** 0.488134** 0.49704*** (0.13) (0.08) (0.08) (0.08) (0.08) (0.19) (0.19) COMMON_LAW -0.10305*** -0.11718*** (0.01) (0.01) SIZE 0.019516*** 0.016875*** 0.016297*** -0.00033 -0.00033 -0.01788*** -0.01856*** (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) TOBINQ 0.032828*** 0.054086*** 0.052394*** 0.052699*** 0.052699*** 0.080554*** 0.080689*** (0.00) (0.00) (0.00) (0.00) (0.00) (0.01) (0.01) MKTCAP -0.00089 0.047018*** 0.08699*** -0.02549 -0.02549 -0.00046 -0.03376*** (0.00) (0.01) (0.01) (0.08) (0.08) (0.01) (0.01) IOEF 0.010051*** 0.007239*** 0.000612 0.005788 0.005788 (0.00) (0.00) (0.00) (0.01) (0.01) BNKMKT -0.25118*** -0.3027*** (0.03) (0.03) ADRI -0.05453*** 0.047621*** 0.037324*** (0.00) (0.00) (0.00)
Year Dummy YES YES YES YES YES YES YES
Industry Dummy
YES YES
YES YES YES YES YES
N 46200 23687 23687 12897 12897 6628 6638
R² 0.591168 0.42223 0.422379 0.528745 0.528745 0.425418 0.425744
Adj. R² 0.590396 0.420075 0.420225 0.525951 0.525951 0.418575 0.418904
Table 6
Overall descriptive statistics.
This Table provides univariate comparison of the mean of variables for the subsamples multinationals and domestics firms. Multinationals are companies, which have a minimum ownership from at least 50.01% for at least one subsidiary located in another country (Paul, 2008). The sample period is from 2009 to 2014. The variables are defined as follows: The independent variable is Cash/Na, which is the natural logarithm of Cash and cash equivalent over net assets. IDV and UAI are standing for the cultural scores from Hofstede for Individualism and Uncertainty Avoidance. CAPEXP is measured by the ratio of additions to fixed assets divided by total assets. CF is measured by the net cash flow of operating activities over the difference of total assets minus cash and short-term investments. DIV_PAYER is a dummy variable, which is equal to one when a firm paid dividends in the measured year. LEVERAGE is measured by total liabilities over total assets. NWC is measured by net working capital over total assets minus cash (Ramirez and Tadesse, 2009). R&D is measured as the ratio of research and
development expenses over total sales. COMMON_LAW is a dummy variable, which is equal to one when the home country has an English common law country origin and zero otherwise (La Porta, Lopez-de-Silanes, Shleifer and Vishny, 1998). SIZE is measured by the natural logarithm of total assets. TOBINQ is measured by the sum of the book value of total assets and the market value of common equity and subtracted with the book value of common equity afterward (Fernandes and Gonenc, 2015). MKTCAP is measured by market capitalization over country´s GDP. IOEF is published as an annual guide published by The Wall Street Journal and the Heritage Foundation (www.heritage.org). BNKMKT is a dummy variable, which is equal to one when the home country has a bank-based financial system and zero when it has a market-based financial system. ADRI tells me which level minority shareholders can be influenced by corporate governance (La Porta, Lopez-de-Silanes, Shleifer and Vishny, 1998)..
Variable N/Gesamt N/Dom. N/Mult. Mean/Gesamt Mean/Dom. Mean/Mult.
TA (mil.U$) 89412 41485 47927 1388.39 427.90 2219.79
Cash (mil. U$) 89412 41485 47927 174.75 55.26 278.17
Sales (mil. U$ 89412 41485 47927 1219.26 348.87 1972.65
Cash/NA 89412 41485 47927 25.8% 25.0% 26.5%
UAI 89412 41485 47927 54.53 52.39 56.38
39 Table 2.
Correlation Matrix
This Table displays the correlation coefficients among the main variables for the 89,412 firm-year observations over the period 2009-2014.. The variables are defined as follows: The independent variable is Cash/Na, which is the natural logarithm of Cash and cash equivalent over net assets. IDV and UAI are standing for the cultural scores from Hofstede for Individualism and Uncertainty Avoidance. CAPEXP is measured by the ratio of additions to fixed assets divided by total assets. CF is measured by the net cash flow of operating activities over the difference of total assets minus cash and short-term investments. DIV_PAYER is a dummy variable, which is equal to one when a firm paid dividends in the measured year. LEVERAGE is measured by total liabilities over total assets. NWC is measured by net working capital over total assets minus cash (Ramirez and Tadesse, 2009). R&D is measured as the ratio of research and development expenses over total sales. COMMON_LAW is a dummy variable, which is equal to one when the home country has an English common law country origin and zero otherwise (La Porta, Lopez-de-Silanes, Shleifer and Vishny, 1998). SIZE is measured by the natural logarithm of total assets. TOBINQ is measured by the sum of the book value of total assets and the market value of common equity and subtracted with the book value of common equity afterward (Fernandes and Gonenc, 2015). MKTCAP is measured by market capitalization over country´s GDP. IOEF is published as an annual guide published by The Wall Street Journal and the Heritage Foundation (www.heritage.org). BNKMKT is a dummy variable, which is equal to one when the home country has a bank-based financial system and zero when it has a market-based financial system. ADRI tells me which level minority shareholders can be influenced by corporate governance (La Porta, Lopez-de-Silanes, Shleifer and Vishny, 1998).
40 Table 8
Interaction whole Data
This Table reports the pooled OLS regression results with firm and industry dummies. The sample period is from 2009 to 2014. The variables are defined as follows: The independent variable is Cash/Na, which is the natural logarithm of Cash and cash equivalent over net assets. IDV and UAI are standing for the cultural scores from Hofstede for Individualism and Uncertainty Avoidance. IDVR&D is the intercept term of individualism and R&D. UAIR&D is the intercept term of uncertainty avoidance and R&D. IDVADRI is the intercept term of individualism and ADRI. UAIADRI is the intercept term of uncertainty avoidance and ADRI. CAPEXP for capital expenditures measured by the ratio of additions to fixed assets divided by total assets. CF is measured by the net cash flow of operating activities over the difference of total assets minus cash and short-term investments. DIV_PAYER is a dummy variable, which is equal to one when a firm paid dividends in the measured year. LEVERAGE is measured by total liabilities over total assets. NWC is measured by net working capital over total assets minus cash (Ramirez and Tadesse, 2009). R&D is measured as the ratio of research and development expenses over total sales. COMMON_LAW is a dummy variable, which is equal to one when the home country has an English common law country origin and zero otherwise (La Porta, Lopez-de-Silanes, Shleifer and Vishny, 1998). SIZE is measured by the natural logarithm of total assets. TOBINQ is measured by the sum of the book value of total assets and the market value of common equity and subtracted with the book value of common equity afterward (Fernandes and Gonenc, 2015). MKTCAP is measured by market capitalization over country´s GDP. IOEF is published as an annual guide published by The Wall Street Journal and the Heritage Foundation (www.heritage.org). BNKMKT is a dummy variable, which is equal to one when the home country has a bank-based financial system and zero when it has a market-based financial system. ADRI tells me which level minority shareholders can be influenced by corporate governance (La Porta, Lopez-de-Silanes, Shleifer and Vishny, 1998). Robust standard errors are clustered at the firm level and ***, **; and * stands for the statistically significant at the 1%, 5% and 10% levels. The standard errors are presented in the brackets.
