Cultural influence: the reliance of free cash flows on
cash holdings in the EMU
Master Thesis Business Administration Finance
University of Groningen
Faculty of Economics and Business
JEL-classification: G30, G32
Cultural influence: the reliance of free cash flows on
cash holdings in the EMU
Author:
Mr D.M. Anosowitsch
Studentnumber:
1681338
Date :
11-08-2012
City :
Groningen
Adress:
Nieuwstraat 289C
7311BP Apeldoorn
Email:
d.m.anosowitsch@student.rug.nl
Course code:
EBM866A20
First supervisor:
Ms A.G. Schertler
Second supervisor:
Mr A. Plantinga
ABSTRACT
TABLE OF CONTENTS
SECTION 1: INTRODUCTION ...4
SECTION 2: THEORY AND EMPERICAL HYPOTHESES ...6
SECTION 3: METHODOLOGY ... 13
SECTION 4: DATA AND DESCRIPTIVE STATISTICS ... 16
SECTION 5: RESULTS ... 21
SECTION 6: DISCUSSION ... 30
SECTION 7: REFERENCES ... 32
SECTION 1: INTRODUCTION
cash and liquid balances in countries where people tend to avoid uncertainty, are more masculine, and have a long term orientation. Chang and Noorbakhsh (2009) state that one has to control for the differences in corporate governance and development of the financial market. To cope with these two issues, this study focuses on the countries in the EMU excluding Ireland1. Ramirez & Tadesse (2009) examined the relationship between national cultures, the nationality of the firm and its cash holdings. They found that firms in countries with high uncertainty avoidance hold more cash.
This paper discusses several topics which are not included in previous relevant papers. Chang and Noorbakhsh (2009) and Ramirez and Tadesse (2009) focus on the effect of culture on cash holdings. This study analyses the effect of culture on the dependence of free cash flows on cash holdings. Furthermore, a clear separation is made between countries with strong and weak cultural identities. The liquidity issue is underlined by adding an extra liquidity variable in the basic cash holdings regression model, namely the market capitalization ratio (EQL). Besides, the focus of this study is on EMU countries, where Chang and Noorbakhsh (2009) base their research on international
sovereignties. This paper will also consider the influence the financial crisis has had on the reliance of the free cash flows on cash holdings, and it will discuss whether there is a difference in the cultural influence on the reliance of free cash flows on cash holdings between years of crisis and non-crisis. Another addition to the existing literature is that this paper analyses whether there is a difference between southern and other EMU countries in the sample in the extent to which their free cash flow relies on cash holdings. This is another cultural separation within the EMU, different from that of the cultural framework of Hofstede based on credit ratings.
The existence and increase of large fractions of liquid assets at management disposal, was a cause for concern among researchers and shareholders. The agency problem, that managers influence the level of cash and the dependence of free cash flows on cash holdings, is still a complicated puzzle for shareholders. This study will provide more insight in the way the cultural identity, and therefore the manager’s behaviour, influences to what degree free cash flows rely on the cash holdings.
The outline of the research paper is as follows. After the introduction, the theoretical foundation will be explained in the second section. Here, the hypotheses will be outlined based on different prevalent financing theories. In section three and four, the methodology, data and descriptive statistics will presented. The results are discussed in section five. In the sixth section, the paper concludes with the discussion and recommendations of this research.
1
SECTION 2: THEORY AND EMPERICAL HYPOTHESES
Section 2 will provide a theoretical overview of the optimal level of cash based on the trade-off theory, pecking order theory and the free cash flow theory. It will also analyse the anticipated relationships between the cash holding determinants and the cash holding with the help of these theories. Moreover, the probable relationship between the cultural identity interactions terms and the dependence of free cash flows on the cash holding is explained, according to the three above mentioned theories. Based on these anticipated relationships, the research questions will be formulated.
First, we will take a look at how the three fundamental finance theories used in this study determine the optimal level of cash in a firm. The trade-off theory argues that a firm’s optimal level of cash and cash equivalents is a trade-off between the marginal costs and marginal benefits, which determines the suitable level of cash a firm should be holding (Myers 1984). The marginal cost of the cash holding is the opportunity cost of the low return on liquid assets. In perfect capital markets, where there is no premium for liquidity or taxes, this would mean that holding cash has no opportunity costs or fiscal advantages. Therefore, under these circumstances, holding liquid funds would not affect shareholders’ wealth (Opler et al., 1999). The trade-off theory predicts that highly constrained firms hold more cash than firms which have easier access to the capital markets. D’Mello et al. (2008) argues, in line with the trade-off theory, that allocating more cash reserves makes the firms less on external debt and provides funds to hedge against future setbacks. The rationale is that firms with a probability of bankruptcy presumably face higher financial distress costs and, as a result, larger costs of debt. Miller & Orr (1966) & Keynes (1934) suggest that the optimum liquidity level is preferred by transaction costs motives. By holding cash excessively, transaction costs for the raising of external funds are minimized, and the level of liquidity forms a buffer. Myers (1984), Myers & Majluf (1984), and Almeida et al. (2004) argue that costly market capital makes it important for firms to maintain a cash reserve that provides sufficient liquidity, so that positive net present value projects can continue to be funded even when internal cash flows decline. Almeida & Campello (2002) proposes a theory of corporate liquidity where the optimal level of cash is dependent on the accessibility of the capital markets and the prominence of future investment opportunities.
The free cash flow theory implies that managers may prefer to hold excess cash because they are averse to risk, or want to follow their own interests. By holding more cash, less external financing is required and it prevents them from having to provide the stakeholders with detailed information about projects. To control for this state of affairs, outside monitoring is needed (Opler et al., 1999). On the other hand, Easterbrook (1984) and Jensen (1986), among others, argue that agency costs of managerial entrenchment will result in the fact that shareholders would prefer it if the firm did not contain a significant cash balance. Potentially, it could lead to overinvestments in negative net present value projects that aid managers better than it does shareholders. Firms that are characterized by high asymmetry between managers and debt holders, concerning the value of assets in place and the future growth opportunities, face high adverse selection costs by raising external funds. Therefore, holding a higher cash level would reduce the need of issuing securities, and a positive relation between information symmetry and cash ratios is expected (D’Mello et al., 2008). Moreover, agency problems like underinvestment, debt overhang and asset substitution, may suffer additional financing costs (Myers, 1977; Jensen & Meckling, 1976). Consequently, managers that attempt to avoid these kinds of costs maintain a high level of internal financing flexibility and a higher level of cash holding (Ozkan & Ozkan, 2004).
The next step is to analyse how the relationship between the cash holdings determinants and the level of cash is based on the finance theories in this study. Kim et al. (1998) claims that the relation between free cash flows and the optimal cash level is negative, which is based on the trade-off theory. The study claims that cash flows represent an additional source of liquidity for the firm, and can therefore substitute the level of cash. In contradiction, Myers & Maljuf (1984) argue that the presence of information asymmetry will create a hierarchy in the use of funding resources for the firm. The hierarchy theory prefers internal funding before going to the market for external financing. Therefore, higher cash flows will result in higher cash levels (Feirreira & Vilela, 2004; Opler et al., 1999; Ozkan & Ozkan, 2004). So when raising external funds is more expensive, cash holdings depend more strongly on free cash flows.
more on their growth opportunities than their tangible assets. As a result, firms will keep higher cash levels to avoid costs of financial distress (Harris and Raviv, 1990; Shleifer & Vishny, 1992). These arguments strengthen the idea that it is likely that firms with higher growth opportunities allocate more internal cash, which is in line with the pecking order and trade-off theory (D’Mello et al., 2008). According to the free cash flow theory, managers of firms with poor investments opportunities are likely to hold more cash to ensure the availability of funds to invest in growth projects, even when the present value of these projects is negative. The result is a decrease in shareholder value and a lower market-to-book ratio. Therefore, a negative relationship between the opportunity set and the cash holdings is probable (Ferreira & Vilela, 2004). The research and development expenditures in particular are a form of investment where information asymmetries are of great influence, because the estimation of the return is difficult to measure (Opler & Titman, 1994; Aboody & Lev, 2000). Consequently, it is likely that firms with higher R&D cost will hold more liquid funds (Opler et al, 1999).
Different empirical studies show that a reduction of cash levels is a consequence of increased financial leverage inside the firm (Kim et al., 1998; Ferreira & Vilela, 2004; Ozkan & Ozkan, 2004). Next to that, Ozkan & Ozkan (2007) have only found a negative relationship between cash holdings and leverage where the levels of leverage are low. This relation becomes positive at higher levels of debt, as the cost of financial distress increases. John (1993) mentioned that a high debt ratio is a proxy for the access to debt markets; a high leverage ratio indicates high financial distress costs and better access to the capital markets. Based on this argument, a negative relationship between leverage and allocated cash ratios is likely. On the other hand, increasing the leverage might result in higher bankruptcy costs. Allocating higher cash levels diminishes distress and maintains borrowing capacity. Opler et al. (1999), provide conflicting predictions on the relationship between the level of cash and leverage. Initially, due to the fact that it is expensive and difficult for highly levered firms to acquire additional funds, they hold more cash, in order to avoid the need of external financing and to be able to take advantage of investment opportunities. On the other hand, they argue that firms with low leverage are less subject to outside monitoring by stakeholders; therefore, cash holdings may be inversely related to leverage. Next to that, poor performance may result in a high leverage ratio and low levels of cash, while good performance could result in the opposite. This argument also suggests a negative relationship between leverage and cash.
The findings of Opler et al. (1999) and Dittmar et al. (2003) demonstrate that, according to the trade-off theory, cash holdings significantly decrease with net working capital, because net working capital is a substitute for cash. Likewise, Lang et al. (1996) say that firms will sell off non-core assets in periods of economic distress, or compete for the available pool of resources (Fazzari & Peterson, 1993).
firms with a greater cash flow uncertainty will hold more cash, since they are more likely to have a sudden need for it (Opler et al, 1999; Colquitt et al., 1999, Kim et al., 1998), according to the trade-off theory.
Firms that pay dividends or can easily sell their assets, hold lower levels of cash (Myers et al., 1977; Kim et al., 1998; Opler et al., 1999; Dittmar et al., 2003). Shleifer & Vishny (1992) states that firms which currently pay dividends, can raise funds at low cost by reducing its dividend payments. In contrast, a firm which does not pay dividends has to use the capital markets to raise funds. Thus, looking at the trade-off theory, firms that pay dividends are likely to hold less cash.
The interest in stocks on liquid stock markets is higher than on non-liquid stock markets. Where the issuance of equity goes easily, this results in more positive investment opportunities. Through these liquid stock markets, investors do not lose access to their savings because they can confidently, cheaply and easily sell their stake in the firm (Levine, 1991; Bencivenga et al., 1996). And although large markets do not necessarily function well and taxes may distort incentives of listed firms, many use the market capitalization ratio as an indicator of stock market development, under the assumption that stock market size is positively correlated with the ability to mobilize capital and diversify risk. The level of cash is determined by a trade-off between the confidence of investors and the growth of the gross domestic product of a country. If the stock markets were less liquid, this would result in higher cash levels, because the firms would face more financing constraints.
take-over due to the amount of resources needed for the bid. Consequently, it is probable that managers of large firms have more discretionary power and therefore hold larger cash balances, based on the free cash flow theory (Ferreira & Vilela, 2004). In conclusion, according to the theory, the relation between size and cash holdings is ambiguous. Table 1 shows a summary of the likely relationship between the cash holdings and the cash holding determinants, based on the three finance theories.
Variables Trade-off theory Pecking order theory Free cash flow theory
Cas h flows Negative Positive
Inves tments opportunities Positive Positive Negative
Leverage Unknown Negative Negative
Liquid as sets substitutes Negative
Cas h flow uncertainty Positive
Dividend payme nts Negative
Equity liquidity Positive
Real size Negative Positive Positive
Table 1: summary of the probable relationships between the cash holdings and the cash holding determinants, according to the trade-of, pecking order- and free cash flow theories. 2
Hofstede (1980, 1991) provides the most comprehensive framework to analyse the effects of cultural values on firms and behavioural decision making, and therefore forms the fundamental framework of this study in finding an answer to the research question. Kirkman et al. (2006) criticized that Hofstede’s model provides an overly simplistic view of culture, as it contains only five concepts, fails to capture the malleability of culture over time, and ignores cultural heterogeneity within countries. Despite the critics, researchers employ the framework due to its clarity, parsimony and resonance with corporate managers (Chang & Noorbakhsh, 2009).
Hofstede (1980) defined power distance as the extent to which inequality among people was considered a natural aspect of social ranking. This inequality is measured by the formal power in different positions. The power of distance is measured by the participation of individuals. With a large power distance level, no close relationships between people can be built. A smaller power distance creates more responsibility towards each other and society (Etemadi et al., 2009). According to House et al. (2001) power distance dimension is the degree to which individuals in a country expect and agree that power is unequally shared. The power distance index shows that the power distance of Western and Northern European countries tends to be lower than that of Southern European countries, with the exception of France and Belgium. France and Belgium scored high, even though they do not necessarily agree with the power distance principle; they maintain it by their desire to avoid conflict with upper management. This behaviour is the reason why these countries continue to score high in
2
this category. Besides, it is interesting to perceive that countries with Roman languages (like France, Italy or Spain) score higher on the index than countries with Germanic languages (like Germany, Austria or the Netherlands). Thus far, no link between the power distance and the influence of the free cash flows on cash holdings has been documented. This study, however, could provide a better notion of whether a higher power distance creates a higher sensitivity between cash flows and the cash holdings.
According to Hofstede (1984), individual cultures primarily look at people’s own interest or the interest of well-defined groups they belong to, such as the firm they work at. In addition to that, Hofstede (2001) argues that there is a distinction between the degrees to which people in a country have an independent rather than an interdependent self-esteem. In individualistic cultures, people consider themselves as autonomous and independent human beings, while in collective environments, individuals see themselves as part of a social group, connected and less differentiated from others (Markus & Kitayama, 1991). Likewise, they discovered a link between individualism, over-optimism and overconfidence, as individualists think too highly of themselves and their abilities. According to Malmendier and Tate (2005), managers overestimate their returns and regard external funds as unduly expensive. Therefore, they overinvest when they have high cash holdings, but curtail investments when they require external funds, which is in line with the free cash flow theory. Since the capital markets limit the extent to which managers can pursue self-interested investments (asymmetric information), the inflow of free cash flows determines the level of the cash holdings and the
investment policy. In general, Southern European countries have a more collective mind-set compared to the individualistic attitude of the northern countries in Europe.
The cultural dimension masculinity measures the degree of cultural toughness versus tenderness in society (Hofstede 1980, 1991). In a masculine society, firms depend strongly on results, challenge and recognition, and rewards individuals on performance rather than equality. Managers succeed or fail and are rewarded on an individual basis. Therefore, managers develop the tendency to hold sufficient cash balances and free cash flows at their disposal, so they do not have to submit themselves to the scrutiny of the financial markets in case of underperformance. In highly masculine countries, corporate cash holdings decisions are likely to be consistent with the free cash flow theory and the pecking order theory. Countries affected by the German culture tend to score very high when compared to other countries.
pecking order theory (Ramirez & Tadesse, 2009). In general, Southern European countries portray a strong uncertainty avoidance culture, especially Portugal and Greece.
Hofstede (1991) argues that the long term orientation approach relates to common values in the oriental cultures dominated by Confucian traditions like persistence, ordering relationships by status and observing this order, thrift, and having patience, self-reliance and a sense of shame. Positive NPV projects are generally evaluated on their long term value generating capacity, instead of short term returns. The environment is equipped to be sustainable and to achieve a stable stream of value for a long period of time. Newman and Nollen (1996) show that long term cultural focus includes providing long term employment. To maintain the long term job security, there must be enough cash balances provided by streams of free cash flows.
Table 2 is a summary of the expected relationships between the different interaction terms and the cash holdings based on the relevant finance theories. Any associations based on the literature are positive in nature. Based on this theoretical framework, this study will analyse whether a strong cultural identity will positively influence the dependence of free cash flows on cash holdings in EMU countries. To be more specific, a high unequally shared power culture, a more individualistic culture, a high masculine culture, a more uncertainty evasive culture, and a more long term orientated culture will positively influence the dependence of free cash flows on cash holdings, according to the different theories.
Inte raction te rms Trade -off the ory Pe cking orde r the ory Fre e cash flow the ory PDI x CF
IDV x CF Positive
MAS x CF Positive Positive
UAI x CF Positive Positive
LTO x CF Positive
SECTION 3: METHODOLOGY
Section 3 will discuss and clarify the regression model, and determine the distinction between strong and weak countries in the EMU based on Hofstede’s model. Secondly, the definitions of the basic cash holdings model will be explained.
To test whether the total reliance of free cash flows on cash holdings is higher at firms with a strong cultural identity instead for a weak cultural identity, a panel least square (PLS) regression has been modelled for ten years for 246 firms. This model consists of the basic cash holding determinants, supplemented by two variables. Because the sample period of this study is 2001-2010, it includes the financial crisis of 2008-2010. Therefore, an extra dummy variable is added to adjust for this period. The year 2008 was depicted as the year of the beginning of the crisis where firms where reacting to the financial crisis and 2010 is designated as the end of the crisis, because the sample period stops at that year. Also, an extra liquidity variable is added, being the market capitalization ratio. To test the research question, the cultural identity of a firm (Y) times the free cash flow ratio is appended.
CASHi, t = β0 + β1CF+ β2MTB+ β3SG+ β4R&D+ β5LEV + β6NWC + β7DIV + β8VOL + β9EQL +
β10SIZE + β11 FIN +β12Y*CF + ε
3
The cultural identity consists of the individualism (IND), power distance index (PDI), masculinity (MAS), the uncertainty avoidance index (UAI) and the long term orientation (LTO). To expand the research in this study, it does not only expose the separated cultural aspects of the model, but also the aggregation of all cultural country variables. A stronger influence of the free cash flows on the cash holdings is anticipated, because according to the literature, all the separate cultural dimensions show a positive relation. Only the first four dimensions are aggregated, meaning the LTO variable is excluded, because the information on Greece and Luxemburg regarding this is incomplete. The cultural identity of countries is translated into a dummy variable to make a difference between strong and weak cultures. To determine the definition of strong and weak cultures, table 3 shows the five different indicators of the Hofstede’s model and the aggregated number of the first four dimensions. The countries are ranked from the highest to the lowest score, based on the Hofstede framework. The dummy variable has a value of one when the selected countries are above the median numeric score of a cultural identity variable, and zero otherwise.
3
Please note that the countries with the highest scores on the power distance index are identical to the best scoring countries on the uncertainty avoidance index. The five countries with scores above average on the two other cultural indexes are not the same, but that is irrelevant for this study.
C ountry PDI C ountry IND C ountry MAS C ountry UAI C ountry LTO C ountry AGG
France 68 Net herlands 80 Austria 79 Greece 112 Netherlands 44 Belgium 288
Belgium 65 It aly 76 It aly 70 Portugal 104 Finland 41 It aly 271
Portugal 63 Belgium 75 Germany 66 Belgium 94 France 39 France 268
Greece 60 France 71 Greece 57 France 86 Belgium 38 Greece 264
Spain 57 Germany 67 Belgium 54 Spain 86 It aly 34 Spain 236
Italy 50 Finland 63 Luxemburg 50 It aly 75 Aust ria 31 Germany 233
Luxemburg 40 Luxemburg 60 France 43 Luxemburg 70 Germany 31 Portugal 225
Net herlands 38 Austria 55 Spain 42 Austria 70 Port ugal 30 Luxemburg 220
Germany 35 Spain 51 Portugal 31 Germany 65 Spain 19 Austria 215
Finland 33 Greece 35 Finland 26 Finland 59 Greece - Net herlands 185
Austria 11 Portugal 27 Net herlands 14 Net herlands 53 Luxemburg - Finland 181
Table 3 Hofstede’s cultural dimensions scores for the five dimensions and the aggregated score for the first four dimensions ranked from high towards low for the countries in the sample.
Beneath, the definitions of the determinants of the basic cash holding model will be explained.
SECTION 4: DATA AND DESCRIPTIVE STATISTICS
In section 4, the sample and data selection is amplified. Next to that, the descriptive statistics are employed for the regression model determinants and will be executed for separate countries or years. The sample includes the twelve member states of the Economic Monetary Union (EMU) before the introduction of the Euro in 2002. All the selected countries in the sample were member of the Euro zone in 1992, when the EMU was established in the treaty of Maastricht, except for Greece. Greece joined the EMU on 1 January 2001. By then, Greece finally met the convergence criteria4. In the complete sample period, all the countries are members of the EMU. Managing and attracting cash and the costs of converting money from one currency to another are not of influence anymore on the level of the cash holding by selecting the Eurozone countries, and therefore the different exchange and interest rate influences on cash holdings are eliminated. Countries that joined the EMU after 2001 can be regarded as less developed markets, because they did not meet the convergence rules of entering the EMU before. According to the World Bank, all the EMU countries in 2001 were denoted as developed markets, because all markets had a gross GDP exceeding the World Bank upper middle income threshold for 3 consecutive years. Furthermore, the ratio of investable market capitalization to GDP was high5 (Financial Structure Dataset from the World Bank). Therefore, selecting only EMU
countries eradicates the problem of the influence which the different financial market development of sovereignties has on cash holdings (Chang & Noorbakhsh, 2009 and Stulz & Williamson, 2003). La Porta et al. (1997) made a distinction by arguing that the legal environment, as described by both legal rules and their enforcement, is important for the size and extent of a country´s capital market. A good legal environment protects the potential financiers against expropriation by entrepreneurs; it increases the willingness for giving up funds in exchange for securities, and thus expands the scope of the capital markets. Civil law countries have both the weakest investor’s protection and the least developed capital markets, compared to common law countries.6 This dissimilarity between the different legal structures can therefore be of influence on the level of external financing and,
consequently, on the level and determination of the cash holding and free cash flows. By eliminating Ireland in the sample, the problem of the influence of different corporate governance systems of sovereignties on cultural values are eliminated (Chang & Noorbakhsh, 2009). Rajan & Zingales
4
The convergence criteria are presented in Article 121(1) of the Treaty establishing the European Community (EC Treaty). The criteria focus is on: price stability, government finances, exchange rates and long term interest rates.
5
The value of listed shares to GDP, is calculated using the following deflation method: {(0.5)*[Ft/P_et + F t-1/P_et-1]}/[GDPt/P_at] where F is stock market capitalization, P_e is end-of period CPI, and P_a is average annual CPI. The data is from the Standard and Poor's Emerging Market Database (and Emerging Stock Markets Factbook). Data on GDP in US dollars is from the electronic version of the World Development Indicators. End-of period CPI and annual CPI are from the IMF’s International Financial Statistics, October 2005.
6
(1998) show that firms in industries which are more dependent on external finance have more cash. Surprisingly, this effect decreases considerably in countries with little shareholder protection. If firms simply hold more cash because it costs more to raise external funds when shareholder protection is weak, one would expect the importance of financing needs to become stronger, not weaker. The final sample includes firms from eleven countries: Greece, Italy, Spain, Portugal, the Netherlands,
Germany, France, Austria, Finland, Belgium and Luxemburg. The firms in the main index of each country in the sample have been used7. The firms constituted in the main indexes are the largest firms in the countries based on their market capitalization. The focus will be on these large firms, because they are the most internationally oriented and strongly depend on access to the capital markets. The firms in the sample are those firms that are in a main indexes of the selected countries in 2010. Fluctuations of the composition of the indexes will not be taken into consideration. Financial related institutions have been excluded from the sample because of their capital requirements following the implementation of Basel II (or the proposed Basel III) for financial intermediaries or Solvency I for insurance firms or pension funds. The reasoning behind excluding financially related firms from the sample is that subscribed liquidity targets by regulation artificially influence the level of cash and the influence of free cash flows on cash holdings in contrast to other firms. In total, 73 financial
intermediaries are excluded by this restriction. Subsequently, the firms: Aperam, TNT Express, Enel Green Power, Fiat Industrial and IAG are excluded from the sample due to a lack of data. Only of the last two years a part of the data is available for these firms. The firm ArcelorMittal and GDF Suez have more than one quoting on indexes in the sample; therefore, the listing in other countries than those in which their headquarters are seated, has been erased. Following the same reasoning, it has been checked whether the firms have their main listing in the country in which they are depicted in the sample, and this is in fact true. In other words, no firms have their headquarters in a country outside the sample countries. Furthermore, the Spanish firm Bolsas Mercados, a large outlier, is excluded because of the deviations in the results. The firm showed a cash ratio (cash and cash equivalents to net assets) of more than 135. The total number of firms included in this paper is 246. Mediterranean countries have less data points, due to a lack of available data. This study covers a ten year time period, from 2001 up to 2010. During this time period, influences of the financial and credit crisis are included. All the examined variables are yearly results. The data for the determinants of the cash holdings was gathered using Thomson’s DataStream databases. The macro economic data comes from the economic outlook of the International Monetary Fund.
Table 4 shows the descriptive statistics of the sample for the variables of de basic cash holding regression model and the added dummy variable for the financial crisis years. There are few remarks
7
to make based on the sample statistics. During the sample period 2001-2010, the average cash ratio in the EMU counties was 13.5 %, which is quite low compared to other studies8. The Dutch firm Akzo Nobel was the positive outlier concerning the cash ratio, because it made a firm transition and disinvested many assets in 2007. The difference between the average cash ratio and the median value was large, as some firms had very high cash ratios. Another remarkable observation is that a few firms kept very low levels of cash: the lowest value is 0.01%, which is exceptional. The average free cash flow ratio was relatively steady at 8%, with a low volatility. The average market-to-book ratio shows that there were many growth opportunities in the EMU, since the market averagely overvalued the book values of firms. During the sample period, the firms’ sales increased with 10%, but the fluctuations were strong, mainly due to the financial crisis (standard deviation of 0.94). The sales growth of some firms decreased with more than 80% or, in contrast, grew tremendously. The R&D ratio show a steady average of 3%. The average leverage ratio of the firms was around 30%, and some firms had no debt on their balance sheets. The mean net working capital ratio was approximately 14%. Some firms reached a negative ratio of over 40%, resulting in highly illiquid firms. Despite the
financial crisis, most firms paid out dividends to their shareholders. On average, the firms’ market capitalization consisted of 3% of the GDP of a country.
CASH CF MTB SG R_D LEV NWC VOL DIV EQL SIZE FIN Me an 0.135 0.079 1.478 0.106 0.030 0.300 0.137 0.041 0.935 0.031 16.026 0.330 Me dian 0.082 0.077 1.310 0.052 0.011 0.284 0.101 0.025 1.000 0.008 16.107 0.000 Maximum 1.665 0.806 8.310 23.306 0.285 1.075 4.049 1.619 1.000 1.524 19.289 1.000 Minimum 0.000 -0.423 0.621 -0.896 0.000 0.000 -0.402 0.002 0.000 0.000 9.692 0.000 Std. De v. 0.221 0.078 0.665 0.939 0.047 0.154 0.246 0.069 0.245 0.124 1.670 0.471 Ske wne ss 11.016 -0.259 3.511 21.886 2.565 0.515 6.233 12.775 -3.544 9.026 -0.448 0.722 Kurtosis 186.657 18781 24.282 505.963 10.529 3.451 83.266 265.399 13.562 92.283 3.168 1.521 Jarque -Be ra 1524009 11105 22369 11353131 3782 56 293910 3095932 7206 369581 37 190 Probability 0 0 0 0 0 0 0 0 0 0 0 0
Table 4 descriptive statistics of the basic cash holding model for the years 2001-2010
Table 5, displayed beneath, gives the average values of the variables in the model for the separate countries. One should be very careful when drawing conclusions from this table, because of the small number of firms that have been taken into account for each country. When comparing it to the sample averages, it becomes clear that on average, the Greek firms hold higher cash levels when compared to firms in Luxemburg and Portugal. The free cash flow ratio averages between countries are slightly different. The Dutch firms jump out, because of their many growth opportunities and high sales growth.
8
Country FIRMS* CASH CF MTB SG R&D LEV NWC VOL DIV EQL SIZE Greece 50 0.203 0.084 1.533 0.159 0.014 0.302 0.246 0.051 0.884 0.004 12.800 Italy 27 0.121 0.078 1.502 0.194 0.034 0.325 0.080 0.039 0.861 0.007 15.791 Spain 24 0.119 0.056 1.613 0.165 0.012 0.079 0.024 0.012 0.978 0.026 16.874 Portugal 17 0.074 0.068 1.366 0.116 0.002 0.442 -0.024 0.029 0.887 0.019 15.081 Netherlands 19 0.155 0.061 1.922 0.368 0.043 0.292 0.114 0.067 0.857 0.039 15.740 France 35 0.135 0.065 1.435 0.055 0.040 0.285 0.073 0.027 0.962 0.013 16.984 Germany 23 0.125 0.089 1.541 0.041 0.040 0.276 0.141 0.029 0.957 0.010 16.895 Austria 14 0.126 0.090 1.371 0.083 0.010 0.311 0.129 0.037 0.846 0.012 14.661 Belgium 12 0.098 0.105 1.733 0.106 0.056 0.287 0.063 0.039 0.875 0.019 15.083 Luxemburg 3 0.074 0.050 1.389 0.131 NA 0.076 0.066 0.166 1.000 0.265 12.765 Finland 22 0.132 0.054 1.600 0.152 0.021 0.260 0.158 0.059 0.895 0.106 14.967 EMU 246 0.139 0.076 1.549 0.162 0.030 0.311 0.122 0.040 0.907 0.022 15.220
* number of firms in the s ample of a particulair country
Table 5: average ratios of the regression variables for the basic cash holding model per country.
Firms in countries like Greece, Portugal, Spain, but also Austria invest low amounts on R&D with respect to their sales. Spain and Portugal respectively have a quite low and quite high leverage ratio. This is peculiar, because the countries have much in common in the macro-economical sphere. Firms in Portugal do have a negative average net working capital ratio, which means that their short term assets are lower than their short term liabilities and cash. This could result in large liquidity problems. Further research shows that firms in Spain and Luxembourg have low debt levels. German and France firms pay an average of more than 90% of the time dividend, which is slightly higher than the firms in the other countries.
Figure 1: the average cash-, free cash flow- and volatility of the free cash flows ratios for each year.
volatility of the free cash flows lies around 4%, and remains quite stable. In appendix table 1, all the average values of the regression variables for each seperate year are portrayed. Some remarks: the market-to-book ratio (growth opportunies) naturally made a sharp decline in 2007-2008, due to the financial crisis. After 2008, a negative sales growth was measureable, but in 2010 there was an average sales growth of 27%, which restored the sales growth to the level of 2003, the beginning of the growth path. The R&D, net working capital, dividend, equity liquidity and size ratios remained steady throughout the years. The leverage ratio showed a small increase in 2008, 2009 and 2010. The fact that the leverage ratio and cash ratio increased simultaneously is not in accordance with most of the literature. John (1993) explains this relation by argueing that increasing the leverage is likely to result in higher bankruptcy costs. John’s study stated that allocating higher cash levels diminishes distress and reserves borrowing capacity.
Table 2 in the appendix shows the Pearson’s correlation matrix, which shows the correlations between all determinants with the corresponding standard deviations and significance levels. It can be concluded that there are signs of mutually high correlations. The significant correlations between the cash flow ratio and the cultural interaction terms with the free cash flows are high, but this can is partly explained by the logical interaction effect with the free cash flows. The individuality and long term orientation interactions term show high significant correlation ratios above 0.7. Due to these high correlation levels, it logically follows that the concerning determinants cannot be seen separately. The net working capital ratio and the cash ratio have a highly significant positive correlation; looking at the current literature, this is a new observation. Normally, the two variables are substitutes for each other, which would result in a negative correlation. After checking for definitions and outliers, no clear reason can be found for this unexpected result. When looking at the influence this strange result has on a country specific level, only the firms in Portugal appears have a negative average relationship concerning the net working capital ratio: this counts for 17 of the 246 firms in the sample. The net working capital ratio is positively correlated with the market-to-book ratio with a 10% significance level. The same significance level is measurable between the sales growth ratio and the volatility of the cash flows. The other high correlations in the right tail are correlations between interaction terms, which are not in a panel least square model at the same moment, so of no importance. The high correlation between the interaction term AGG and the other interactions is logical, because the AGG is a summation of the other cultural identities of countries except for the long term orientation dimension.9
SECTION 5: RESULTS
In section 5, the results of this study are presented. First, tests are performed to measure whether the regression data fits the random or fixed effects model. Secondly, this section contains regression tests for the basic cash holdings model supplemented with the crisis dummy. From now on, when the basic cash holding model is mentioned, it means the model which includes the crisis dummy. After this, we will check the results found at the basic cash holding regression model with the relationships found in the current literature. Next, this section will use a regression study to research the basic cash holding model with the added cultural interaction variables to answer the research question. At the end of the section, a few robustness checks will be performed, to clarify the influence of the financial crisis and the differences between south and other EMU countries on the relationship of free cash flows depending on cash holdings.
The panel data employed in the research allows for testing random and fixed effects in the model. When conducting the Hausman test for period and cross section random effects, not all p-values show significant results, and for that reason the random effects model is not appropriate to employ in this panel analysis. Table 3 of the appendix shows the results of redundant fixed effects tests, with cross-section and period fixed effects. Significant results are the sequence, the effects are not redundant. Therefore, the hypothesis for the fixed effects model does not have to be rejected, and the fixed effects model is suitable. The check shows that the fixed effect model is the most fitting in the panel least square analysis. In addition, the fixed effect models is preferred, since the entities in the sample covers a complete population, in particular the main indexes in countries which are not randomly chosen (Brooks, C, 2008). In this paper, only the cross-section fixed effects are used, because period fixed effect are not applicable due to the crisis dummy variable and the different timeframes of the subsamples.
the pecking order theory. Higher investment opportunities would also result in higher cash balances, which is concurrent to the literature (Kim et al., 1998; Opler et al., 1999, Ferreira & Vilela, 2004; Ozkan & Ozkan, 2004) and the trade-off and pecking order theory. Based on this table, one can conclude that when firms in the EMU hold more leverage, they maintain more cash and equivalents inside the firm. According to Ozkan & Ozkan (2007) this positive relationship between cash holdings and leverage exists when there are higher levels of debt, as the cost of financial distress grows. The net working capital ratio is highly significant with a positive coefficient of 0.6204. Surprisingly, the net working capital coefficient is positive instead of negative, according to the literature. This observation is confirmed when analysing the correlations. Thus, more short term liquid assets result in a higher cash holding. Further research could give a more specific idea of this result for EMU countries in this time period. When taking a closer look, only Portugal has a negative coefficient of -0.024. In Portugal, the net working capital is a substitute for cash or cash equivalents in this period. Consequently, firms with a greater cash flow uncertainty will hold more cash, as they are likely to have a more
unanticipated need for it, a conclusion supported by the literature (Opler et al, 1999; Colquitt et al., 1999, Kim et al., 1998) and in line with the trade-off theory. The new liquidity measure added to the cash holding model is not significant, so it cannot be said whether stock markets in EMU countries are liquid and whether firms face borrowing constraints. The control variable size does appear to be significant with a small negative coefficient, in accordance with the trade-off theory. It can be
concluded that smaller firms hold more cash than larger firms, because larger firms have easier access to the capital market. The financial crisis dummy is highly significant, which means that, in
CF 0.154 0.044*** 0.117 0.044*** MTB 0.014 0.007** 0.011 0.006* SG -0.005 0.003 -0.004 0.003 R_D -0.121 0.151 -0.182 0.152 LEV 0.118 0.029*** NWC 0.620 0.021*** 0.633 0.021*** VOL 0.186 0.068*** 0.230 0.068*** DIV -0.012 0.013 EQL 0.102 0.088 0.082 0.089 SIZE -0.039 0.008*** -0.031 0.008*** FIN 0.036 0.005*** 0.034 0.005*** CONSTANT 0.604 0.129*** 0.502 0.128*** R-squared 0.931 0.929 Adjusted R-squared 0.918 0.917 F-statistic 75 75 Prob(F-statistic) 0 0
*** Coefficient is significant at p<0,01 level (two -tailed) ** Coefficient is significant at p<0,05 level (t wo -tailed) * Coefficient is significant at p<0,10 level (t wo -tailed)
Basic Model (1) Reduced form (2)
Table 6: regression results for the cash holding model, with the coefficients (st.dev) and corresponding significances. The first model is the basic cash holding model; the second is the reduced form cash holding model. The regressions are employed with the cross-section fixed effects. It is natural that a firm’s decisions about leverage, cash holding and investment policy are determined simultaneously. This joint determination could create inconsistent estimates. To test the robustness of the results found in model 1, a reduced form regression is tested by omitting the dividend and leverage variable. The results of this check are reported in table 6, model 2. Only some small changes are demonstrable; the signs of the coefficients did not change. The significance levels did not change either, except for the market-to-book ratio, which is now significant at a 10% level instead of a 5% level. The cash flow coefficient decreased and the coefficient of the cash flow volatility increased slightly. In conclusion, the problem of joint determination of leverage, cash holdings and investment policy affect our findings marginally, so it does not result in any biases.
When employing the Durbin Watson-test to check for autocorrelation, it shows value statistics around 1.55. According to the Durbin Watson table, the value is inconclusive for any sign, so no autocorrelation is measured.
due to the numerous variables in the model. The three different types of tests all show highly significant p-values, which indicates that the hypotheses of homoscedasticity is rejected and there is heteroscedasticity. When performing the OLS regressions with the heteroscedasticity-consistent standard error estimates, a potential solution to overcome the heteroscedasticity, this shows no marginal differences in the OLS results than a test without the consistent error. However, the problem remains that the tests are performed in an unstructured dataset instead of the panel dataset. Also, when performing the White-test again with the consistent error factor included, the p-values are still highly significant, and show signs of heteroscedasticity. Transforming all independent and dependent variables into log functions, or reducing them by some measure of “size” to re-scale the data to pull in extreme observations, did not make a marginal difference. The panel least squares estimates is still unbiased and consistent with the heteroscedasticity problem, but they no longer have the minimum variance among the class of unbiased estimators.
In order to answer the research question whether a strong cultural identity, according to the Hofstede’s dimensions, entails that free cash flows rely more on the cash holdings during the sample period 2001 until 2010, six models are depicted in table 7. Each of these models adds a different cultural dimension times the cash flow ratio (interaction term) to the basic model.
CF 0.157 0.053*** 0.275 0.075*** 0.130 0.052** 0.157 0.053*** 0.161 0.088* 0.128 0.060** MTB 0.014 0.006** 0.014 0.006** 0.013 0.006** 0.014 0.006** 0.015 0.006*** 0.013 0.006** SG -0.005 0.003 -0.005 0.003 -0.005 0.003* -0.005 0.003 -0.005 0.003** -0.005 0.003 R_D -0.121 0.151 -0.158 0.152 -0.112 0.151 -0.121 0.151 -0.161 0.158*** -0.116 0.151 LEV 0.118 0.029*** 0.120 0.029*** 0.118 0.029*** 0.118 0.029*** 0.248 0.030*** 0.117 0.029*** NWC 0.620 0.021*** 0.617 0.021*** 0.620 0.021*** 0.620 0.021*** 0.681 0.020*** 0.621 0.021*** VOL 0.187 0.069*** 0.176 0.069** 0.179 0.069*** 0.187 0.069*** 0.081 0.065 0.179 0.069** DIV -0.013 0.013 -0.011 0.013 -0.011 0.013 -0.011 0.013 0.002 0.012 -0.01 0.013 EQL 0.102 0.088 0.080 0.089 0.108 0.089 0.102 0.088 0.103 0.079 0.107 0.089 SIZE -0.039 0.008*** -0.041 0.008*** -0.039 0.008*** -0.039 0.008*** -0.033 0.008*** -0.039 0.008*** FIN 0.036 0.005*** 0.037 0.005*** 0.036 0.005*** 0.036 0.005*** 0.033 0.005*** 0.036 0.005*** CONSTANT 0.605 0.130*** 0.622 0.129*** 0.602 0.129*** 0.605 0.130*** 0.460 0.122*** 0.599 0.130*** CI(#) _CF -0.009 0.080 0.170 0.085** 0.066 0.079 -0.009 0.080 -0.058 0.095 0.049 0.076 R-squared 0.931 0.931 0.931 0.931 0.873 0.931
Adjus ted R-squared 0.918 0.919 0.918 0.918 0.851 0.918
F-statistic 75 75 75 75 41 75
Prob(F-statistic) 0 0 0 0 0 0
CI means cultural identity
*** Coefficient is significant at p<0,01 level (two -tailed) ** Coefficient is significant at p<0,05 level (two -tailed) * Coefficient is significant at p<0,10 level (two -tailed)
CI=LTO (7) CI=AGG (8) CI=PDI (3) CI=IDV (4) CI=MAS (5) CI=UAI (6)
Table 7: regression results of six models for the sample period 2001-2010, with the coefficients (st.dev) and corresponding significances. Each model adds a cultural dimension of Hofstede to the basic cash holding model. The regressions are employed with cross-section fixed effects.
cash flows on the cash holdings, it becomes clear that this increased a lot. Without the interaction term the cash flow coefficient was 0.154 in the basic cash holding model. The cash flow coefficient and the coefficient of the interaction term together makes 0.445. Therefore, one can conclude that with firms with a strong cultural identity, measured by the individual mindset of mangers, the reliance of free cash flows on cash holdings increased positively. When focusing on the determinants of the cash holding model, this shows that the significances differ marginally in contrast to the basic cash holding model, and no sign changes are visible for all variables except for model 7. When adding the long term orientation interaction term, the other determinants in the model change. The cash holding now
strongly depends on investment opportunities. And the cash flow uncertainty ratio and net working capital ratio is of less influence. Consequently, because managers are more focused at the future than the present, they are oriented on long term opportunities more than, for example, short term capital. The only sign that changes in the six models of all determinants, is that of the dividend dummy. The other variables provide the same signs as the basic model. This positive relationship is not in
accordance with the literature, but the size of the coefficient is quite small. One can conclude that only countries with strong individualistic behaviour in the EMU will positively contribute to the extent to which free cash flows rely on the cash holdings. None of the other cultural aspects of Hofstede’s model have any impact on the dependence. Lastly, it is important to note that de correlation between the individuality interaction term and the cash flow ratio is high. When performing a test for
multicollinearity to check whether the founded results are valid, and the correlations are not too high, it shows that the variance inflation factor (VIF) is below the value ten, therefore the results found are legitimate.
To test the acquired results, some robustness checks have to be performed. Initially, we will analyze the differences between the relationship of a strong cultural identity and the dependence of free cash flows on cash holdings during crisis and crisis-free periods, instead of the whole sample period 2001-2010. Therefore, subsamples are created. The years 2001 until 2007 are denoted as the non-crisis years, and 2008-2010 are the financial crisis years. The regressions performed for the subsamples are modelled in the same way as in table 7, including the same interaction terms. Only the dummy variable for the financial crisis in the model is excluded, because of the subsamples of the crisis and non-crisis years. In the whole sample period 2001-2010, only the added interaction term for the individualistic culture is significant at a 5% level and results in a higher dependency of free cash flows on the cash holdings. In the non-crisis years, none of the added interaction terms showed any
significant results10. In other words, in non-crisis years, a strong cultural identity in countries does not significantly influence how much free cash flows depend on the cash holdings. In table 8, the
10
regression models are displayed with the cultural interaction terms for the crisis period. CF 0.117 0.068* 0.077 0.091 0.155 0.071** 0.117 0.068* -0.191 0.109* 0.130 0.076* MTB 0.003 0.010 0.003 0.010 0.004 0.010 0.003 0.010 0.003 0.010 0.003 0.010 SG 0.032 0.013** 0.032 0.013** 0.033 0.013** 0.032 0.013** 0.048 0.014*** 0.032 0.013** R_D 0.257 0.177 0.270 0.178 0.266 0.177 0.257 0.177 0.417 0.250* 0.259 0.177 LEV 0.326 0.049*** 0.330 0.049*** 0.328 0.048*** 0.326 0.049*** 0.264 0.050*** 0.328 0.049*** NWC 0.784 0.030*** 0.782 0.023*** 0.782 0.030*** 0.784 0.030*** 0.849 0.030*** 0.783 0.030*** VOL -0.071 0.150 -0.096 0.153*** -0.068 0.149 -0.071 0.150 -0.033 0.148 -0.071 0.151 DIV -0.003 0.017 -0.003 0.017 -0.002 0.017 -0.003 0.017 0.012 0.018 -0.003 0.017 EQL 0.075 0.083 0.078 0.083 0.066 0.083 0.075 0.083 0.043 0.079 0.072 0.083 SIZE -0.094 0.019*** -0.093 0.019*** -0.095 0.019*** -0.094 0.019*** -0.089 0.019*** -0.095 0.019*** CONSTANT 1.404 0.306*** 1.382 0.307*** 1.414 0.306*** 1.404 0.306*** 1.359 0.308*** 1.410 0.307*** CI(#) _CF 0.029 0.116 0.072 0.102 -0.078 0.098 0.029 0.116 0.314 0.115*** -0.011 0.098 R-squared 0.933 0.933 0.933 0.933 0.941 0.933
Adjus ted R-squared 0.903 0.903 0.903 0.903 0.915 0.903
F-statistic 31 31 31 31 36 31
Prob(F-statistic) 0 0 0 0 0 0
CI means cultural identity
*** Coefficient is significant at p<0,01 level (two -tailed) ** Coefficient is significant at p<0,05 level (two -tailed) * Coefficient is significant at p<0,10 level (two -tailed)
CI=PDI (9) CI=IDV (10) CI=MAS (11) CI=UAI (12) CI=LTO (13) CI=AGG (14)
Table 8: regression results for six models for the sample period 2008-2010, with the coefficients (st.dev) and corresponding significances. Each model adds a cultural dimension of Hofstede to the basic model. All the regressions are employed with cross-section fixed effects.
After employing the panel least squares, it immediately shows that the interaction term with the long term orientation variable is very positively significant at a 1% level. Therefore it can be said that, when mangers do not focus on the short term during the crisis period but look ahead instead, this positively influences the dependence of free cash flows on the cash holdings. This positive significant result is in line with the trade-off theory and pecking order theory and the research of Newman & Nollen (1996). They argue that a cultural focus includes providing long term employment. To maintain the long term job security, the cash balances have to be larger, complemented by free cash flows. On a side note, inside the sample of the long term orientation cultural interaction term the firms of the main indexes of Luxemburg and Greece are excluded due to the lack of data. The large
with a 5% level during the crisis. The variables market-to-book ratio and volatility of the cash flows are no longer significant during the crisis. Moreover, the cash flow ratio has been downgraded from highly significant to a significance of the 1% level. In model 10, the cash flow ratio is not even significant anymore. When focusing on model 13, where the interaction term is significant, the sales growth and R&D expense ratio are significant, although the market-to-book ratio no longer is. This is typical, as the sales growth and the market-to-book ratio boldly measure the investment opportunities. This robustness check pointed out that the results found in table 7 for the whole sample period, are not supported by the sub sample period for the crisis and non-crisis years. Here, it can be said that the results are specific for the sample period 2001-2010. And the significant results for the long term orientation interaction term are only applicable in crisis years. It is also important to realize that the correlation between the long term orientation interaction term and the cash flow ratio is above average, but the variance inflation factor (VIF) is below the value ten, therefore the results found are legitimate.
CF 0.137 0.046*** 0.112 0.049** MTB 0.013 0.006** 0.013 0.006** SG -0.005 0.003 -0.005 0.003* R_D -0.106 0.151 -0.134 0.151 LEV 0.123 0.030*** 0.115 0.029*** NWC 0.621 0.021*** 0.620 0.021*** VOL 0.188 0.068*** 0.175 0.069** DIV -0.011 0.013 -0.012 0.013 EQL 0.010 0.088 0.109 0.088 SIZE -0.042 0.008*** -0.041 0.008*** FIN 0.033 0.006*** 0.036 0.005*** CONSTANT 0.650 0.134*** 0.619 0.129*** FIN_CF 0.066 0.051 AREAS_CF 0.177 0.091* R-squared 0.931 0.931 Adjusted R-squared 0.918 0.918 F-statistic 75 75 Prob(F-statistic) 0 0
*** Coefficient is significant at p<0,01 level (two -tailed) ** Coefficient is significant at p<0,05 level (t wo -tailed) * Coefficient is significant at p<0,10 level (t wo -tailed)
Crisis (15) Areas (16)
Table 9: regression results for two cash holding models, with the coefficients (st.dev) and
corresponding significances. Model 15 includes an interaction term with a dummy variable for the financial crisis times the cash flow ratio, and the other model adds an interaction term of a dummy for Southern European countries times the cash flow ratio. All the regressions are employed with the cross-section fixed effects.
In table 9, model 16, the third robustness check is performed to confirm whether there is a difference between southern and other EMU countries concerning how much their free cash flows depend on cash holdings. In other words, this separation between the EMU countries is made based on the grouping of the United Nations rather than Hofstede’s framework criteria. Southern European countries have lower credit ratings (B or lower) than North-Western European countries (A rating), and therefore face more constraints when accessing the capital market.11 Beck et al. (2007) added that without broad access, firms will have a hard time investing in positive NPV projects due to low credit ratings. According to Cosset & Roy (1991) and Hague et al. (1996), the credit rating largely
determines whether firms are able to access the capital market, due to economic fundamentals of the concerning country, international interest rates, and independent or domestic economic fundamentals.
11
See figure 1 in the appendix, in which all the different credit ratings of European countries are shown. Source:
Brewer & Rivoli (1990) argued that, next to the economic fundamentals, political instability is of great importance when determining the credit ratings. This gives reason to suspect that southern firms are more dependent on free cash flows, because they face more borrowing constraints than other countries in Europe. To test this hypothesis, a new variable is added to the basic cash holding model
SECTION 6: DISCUSSION
This study focused on the dependence of free cash flows on cash holdings and the influence of managers behaviour on that relationship, measured by the cultural dimensions of Hofstede’s
framework for EMU countries. Evidence showed that individual minded mangers in EMU countries retain more excess cash to avoid the scrutiny of the financial markets, due to information asymmetry, which increased the agency problems between shareholders and managers at firms. Chang &
Noorbakhsh (2009) found that long term oriented firms have larger cash holdings. This paper adds that long term oriented firms retain less cash flows for their cash holdings during the financial crisis. Therefore, in contradiction to the literature, firms do not retain more free cash flows in order to ensure their long term job security. A reason for this may be that there are many layoffs and reorganizations during a financial crisis, which lessens the need for a high level of long term job security. The
financial crisis does influence the level of cash a firm holds, but it has no impact on the dependence of free cash flows on cash holdings. Therefore, the argument that firms might face borrowing constraints during the financial crisis and therefore let their cash holdings depend more strongly on free cash flows, does not hold. Furthermore, the difference between southern EMU countries and other EMU countries has been measured using the credit ratings, a way to measure the borrowing constraints firms face. This study found support that Southern European countries face more market uncertainty than other EMU countries, which results in lower credit ratings and higher risk premiums in general. Consequently, firms in southern EMU countries face more borrowing constraints on the capital markets, and let their cash holdings depend more on free cash flows than north-western EMU countries. In other words, firms of the EMU countries in general do not face many borrowing constraints, but there are signs that Southern European countries do. The results found with the basic cash holding model in EMU countries in the period 2001-2010 are in line with the literature, except for the net working capital ratio. The new liquidity measure, the capitalization ratio, does not provide any significant results.
In conclusion, a strong cultural identity can influence the reliance of free cash flows on cash holdings, which is proved in this study by the individuality index, long term orientation and credit ratings in the EMU. This research could help raise more awareness among researchers and shareholders regarding the effect and influence which managers’ behaviour and a firm’s borrowing constraints have on the reliance of free cash flows on cash holdings.
would be to increase the number of observations per country. After all, a country comparison is nearly impossible due to the small number of observations in some of the countries in this sample. A
possibility would be to take all listed stocks of a country instead of just the listed firms of the main index. Besides, a comparison between the EMU countries and other continents or, for example, emerging markets, would certainly be a useful addition to this study. It could provide more insight into the way developed and emerging markets differ when it comes to the influence of the crisis, the sensitivity of their cash flows and their cultural identity. For example, Love (2000) concentrates on the relationship between a country’s financial development and the investment-cash flow sensitivity of the firms. She found that firms in counties with a lower level of financial market development, hold more cash. This could possibly be explained by cultural aspects, and shine a light on the potential
SECTION 7: REFERENCES
Aboody, D. & Lev, B., 2000, Information asymmetry, R&D, and insider gains, Journal of Finance 55(6), 2747-66.
Almeida, H. & Campello, M., 2002, Financial constraints and investment-cash flow sensitivities: new research directions, working paper, New York University and University of Illinois.
Almeida, H., Campello, M. & Weisbach, M., 2004, Cash flow sensitivity of cash, Journal of Finance 59(3), 1777-1804.
Barclay, M.J. & Smith C.W., 1995, The maturity structure of corporate debt, Journal of Finance 50, 609-631.
Baskin, J., 1987, Corporate liquidity in games of monopoly power, Review of Economics and Statistic 64, 312-31.
Baumol, W. J., 1952, The transactions demand for cash- an inventory theoretic approach, Quarterly Journal of
Economics 66, 545-556.
Beck, T., Demirguc-Kunt, A. & Soledad Martinez Peria, M., 2007, Reaching out: access to and use of banking services across countries, Journal of Financial Economics 85, 234-266.
Bencivenga, V.R., Smith, B.D. & Starr, R.M., 1996, Equity markets, transactions costs, and capital accumulation: an illustration, The World Bank Economic Review 10(2), 241-65.
Brewer, T.L., Rivoli, P., 1990, Politics and perceived country creditworthiness in international banking, Journal
of Money, Credit and Banking 33(3), 357-369.
Brooks, C. (2008). Introductory Econometrics for Finance. Cambridge University Press.
Chang, K. & Noorbakhsh, A., 2009, Does national culture affect international corporate cash holdings? Journal
of Multinational Financial Management 19(5), 323-342.
Chui, A., Lloyd, A. & Kwok, C., 2002, The determination of capital structure: is national culture a missing piece to the puzzle? Journal of International Business Studies 33, 99–127.
Colquitt, L.L., Sommer, D.W. & Godwin, N.H., 1999, Determinants of cash holdings by property-liability insurers, The Journal of Risk and Insurance 66(3), 401-415.
Dittmar, A., Mahrt-Smith, J. & Servaes, H., 2003, International corporate governance and corporate cash holdings, The Journal of Financial and Quantitative Analysis 38(1), 111-133.
D’Mello, R., Krishnaswami, S. & Larkin, P.J., 2008, Determinants of corporate cash holdings; evidence from spin-offs, Journal of Banking & Finance 32(7), 1209-1220.
Easterbrook, F., 1984, Two agency-cost explanations of dividends, The American Economic Review 74, 650-659.
Etemadi, H., Dilami, Z.D., Bazaz, M.S., Parameswaran, R., 2009, Culture, management accounting and managerial performance: focus Iran, Advances in Accounting, incorporating Advances in International
Accounting 25, 216-225.
Fazzari, S.M & Petersen, B.C., 1993, Working capital and fixed investment: new evidence on finance constraints. RAND Journal of Economics 24(3), 328-342.
Faulkender, M. & Wang, R., 2006, Corporate financial policy and the value of cash, Journal of Finance 61(4), 1957-1990.
Ferreira, M. & Vilela, A., 2004, Why do firms hold cash? Evidence from EMU countries, European Financial
Management 10, 295–319.
Freedman, D.A., 2005, Statistical Models: Theory and Practice, Cambridge University Press
.
Guney, Y., Ozkan, A. & Ozkan, N., 2003, Additional international evidence on corporate cash holdings, working paper.
Harris, M & Raviv, A, 1990, Capital structure and the informational role of debt, Journal of Finance 45(2), 321-349
Hofstede, G.H., 1980, Culture’ conseuquenses, international differences in work related values, Sage, Beverly Hills, CA.
Hofstede, G. H., 1984, The cultural relativity of the quality of life concept, Academy of Management Review 9(3), 389−398.
Hofstede, G.H., 1991, Culture and organizations, software of the mind. McGraw-Hill Book Company.
Hofstede, G. H., 2001, Culture's consequences: comparing values, behaviors, institutions, and organization across nations, 2nd ed. Sage Publication.
House, R., Javidan, M. & Dorfman, P., 2001, Project globe: an introduction, Apllied Psychology: An
Jensen, M.C. & Meckling, W.H., 1976, Theory of the firm: managerial behavior, agency costs and ownership structure. Journal of Financial Economics 3, 305-360.
Jensen, M.C., 1986, Agency costs of free cash flow, corporate finance, and takeovers. The American Economic
Review 76(2), 323-329.
John, T.A., 1993, Accounting measures of corporate liquidity, leverage, and costs of financial distress, Financial
Management 22, 91–100.
Kalcheva, I. & Lins, K.V., 2003, International evidence on cash holdings and expected managerial agency problem, working paper, University of Utah.
Kaplan, N. & Zingales, L., 1997, Do investment-cash flow sensitivities provide useful measures of financing constraints? The Quarterly Journal of Economics 112(1), 169-215.
Keynes, J., 1934, The general theory of employment, interest and money, New York: Harcourt, Brace.
Khoshtinat, M. & Kazempour, E., 2006, Professionalism in accounting, cultural views in developing countries", Accountant 178, 46-56.
Kim, C.S., Mauer, D.C. & Sherman, A.E., 1998, The determinants of corporate liquidity: theory and evidence,
Journal of Financial and Quantitative Analysis 33, 335–359.
Kirkman, B.L., Lowe, K.B. & Gibson, C.B., 2006, A quarter century of culture’s consequences: a review of empirical research incorporating Hofstede’s cultural values framework, Journal of International Business 37, 285-320.
Lang, L., Ofek, E. & Stulz, R.M., 1996, Leverage, investment and firm growth, Journal of Financial Economics 40, 3-29..
La Porta, R., Lopez-de-Silanes, F., Shleifer, A. & Vishny, R.W., 1997, Legal determinants of external finance,
The Journal of Finance 52, 1131-1150.
Levine, R., 1991, Stock markets, growth, and tax policy, Journal of Finance 46(4), 1445-65.
Levine, R. & Zervos, S., 1998, Stock markets, banks, and economic growth, The American Economic Review 88(3), 537-558.
Malmendier, U. & Tate, G., 2005, CEO overconfidence and corporate investment, Journal of Finance 60(6), 2661-2700.
Markus, H.R. & Kitayama, S., 1991, Culture and the self: implications for cognition, emotion, and motivation,
