The marginal value of cash holdings for financial
constrained and unconstrained firms in common law
and civil law countries: EU15 evidence
Abstract
:This paper examines the the different marginal effects of cash holdings on firm value between civil and common law countries, and between financial constrained and unconstrained firms over the period from 2000 to 2016. We assume that the marginal value per unit of currency is larger in common law countries than it is in civil law countries, since the agency costs are lower in common law countries as these countries have better shareholder protection, and that the marginal value of cash holdings in financial constrained firms is larger than in financial unconstrained firms since information asymmetries make it harder for constrained firms to finance from outside. Our results imply that the overall marginal value of cash holdings is significantly lower than 1. Our results also show that the marginal value at the mean is significantly larger in common law countries than in civil law countries. However, we have no significant evidence of a difference between financial constrained and unconstrained firms.
Key Words: Marginal value, Cash holdings, Common law and Civil law, Financial constraints JEL classification: G31 G32 K15
1. Introduction
How much does one unit of cash holdings contribute to firm value? Is it affected by different law systems or policies?Modigliani and Miller (1958) came up with irrelevancy theory and stated that the capital structure is irrelevant for firm value. One can expand the irrelevancy theory to cash holdings, implying that one dollar of cash contributes to exactly one dollar of firm value in a perfect market. However, in the real world, taxes, information asymmetries, and transaction cost exist, making both external and internal funds costly and thus making cash holdings no longer irrelevant.Cash holdings can be viewed as internal funds which can be used for avoiding transaction cost, and can be used for financing activities or investments when external funds are costly (Opler, Pinkowitz, Stulz and Williamson, 1999). In such a case, cash holdings have marginal benefits as well as marginal costs, which makes the marginal value per unit of currency an appealing topic to study.
3 significantly higher in financial constrained firms. However, they found there is no evidence that the marginal value per unit of currency proves to be significantly different in Chinese and German market for unconstrained firms. Moreover, they found that it is hard to obtain external fund for constrained firms in both Germany and China because the marginal value of cash holdings is calculated larger than 1 in these countries.
However, the association between law systems and the marginal value of cash holdings on firm value is an untapped area and also there is no study for marginal value of cash holdings which focuses on the European market. By forming a hypothesis on the difference in impact of common law and civil law firms on marginal value of cash holdings, we analyze the effect from law systems. Practically, managers might benefit more or less from cash holdings, depending on which law systems they are under. Especially for European firms, this paper may come with new views. In previous studies, the marginal value of cash holdings is proven to be influenced by financial constraints (Faulkender and Wang, 2006; Denis and Sibilkov ,2010; and Zhang, von Eije and Westerman, 2015). However, this conclusion has not been proven yet in the the European market. Moreover, the value per unit of currency may vary over time and therefore we study the recent data from 2000 to 2016, which represents the most recent period for the value of cash holdings. Hence, our main research questions of this study are:
l Is there a difference in the marginal value of cash holdings between common law countries and civil law countries?
l Is there a difference in the marginal value of cash holding of financial constrained firms and unconstrained firms?
study coined by Faulkender and Wang (2006), we study the marginal effect on firm value under these different conditions, with fixed time and cross section effect panel.
According to our results, firstly, the overall marginal value at the mean is 0.710, and this value is significantly smaller than 1. Secondly, the marginal value is significantly higher in common law countries than in civil law countries. Thirdly, in both common and civil law countries, the marginal value in financial constrained firms is larger than in financial unconstrained firms. However, there is no evidence that this difference is significant.
5
2. Literature review and hypotheses development:
Modigliani and Miller (1958) stated the irrelevance between capital structure and firm value in perfect capital market in his irrelevance theory. Modigliani and Miller (1961) also suggest that the dividends payment is also irrelevant with firm value in perfect market. In such a capital market, the firm can always raise cash when cash is needed for investing positive NPV projects (Pinkowitz and Williamson, 2002). We can imply that in perfect market, the amount of cash is also irrelevant with the firm value.
However, in real world, taxes, agency costs and information asymmetry exist. This irrelevant relationship does not hold. On the one hand, there are marginal costs for holding cash. According to Opler, Pinkowitz, Stulz and Williamson (1999), cash can be viewed as the opportunity cost of not investing in high return projects. Additionally, they also stated that cash holdings generate interest, which will be taxed as revenue of firms, and it adds the cost of cash holdings. On the other hand, holding cash can bring marginal benefit as well.Opler, Pinkowitz, Stulz and Williamson (1999) argued that cash reserves could be used to invest in future high-return projects and that the marginal costs and the marginal benefit can be traded-off so that at a certain amount of cash holdings, it maximizes the value of firm.
2.1. Theoretical literature review
In the real world, besides other things, agency cost and information asymmetry do influence the effect of cash holdings to firm value. To figure out how marginal value of cash holdings affect firm value, theories based on agency cost and information cost are applied.
2.1.1. Agency cost
increase managerial discretion: to managers’ interest, managers hope to hoard cash as much as possible, and prefer investing in low return project instead of paying dividends to shareholders. Agency cost arise because of managerial discretion; managers prefer to hoard cash simply for reducing firm risk or for holding excess cash to have more flexibility (Opler, Pinkowitz, Stulz, and Williamson, 1999; Pinkowitz, Stulz, and Williamson, 2006). According to Dittmar and Mahrt-Smith (2007), firms prefer to hold more cash reserves for acquisitions. The center issue of agency conflict is about how to allocate cash holdings (Myers and Rajan, 1998), and it is likely that the agency problem is more severe if the amount of cash reserves is larger.
7 if it is under a country with poor investor protection. Hence, it is likely that marginal value of cash holdings differs because of the law systems under which firms operate and that common law country firms are less vulnerable to agency cost problem. This reasoning leads to the following set of hypotheses:
H10: There is no difference for the marginal value of cash holdings between civil and common
law countries.
H1A: The marginal value per unit of currency in common law countries is larger than it in civil
law countries.
2.1.2. Information asymmetries
As the capital market is not a perfect market, information asymmetries exist. Funding from outside becomes difficult because of information asymmetries. Outside funders, in order to keep themselves from buying overpriced securities, may require buying the securities at a discount. Because outside funders do not know the actual price as well as managers do, they may underprice the securities. Hence, information asymmetry makes it costly to get external funds. (Myers and Majluf, 1984)
asymmetries are severe. These findings contradict to Myers and Majluf (1984)'s views about information asymmetries.
As these discussed results are controversial, to find out the different effect of the marginal value of cash holdings on financial constrained and unconstrained firm, we assume that the marginal value per unit of currency is larger in financial constrained firms. According to our assumption, hypothesis 2 is formed as follows:
H20: There is no difference for the marginal value of cash holdings between financial
constrained and unconstrained firms.
H2A: The marginal value of cash holdings in financial constrained firms is higher than in
unconstrained firms.
9 Figure 1. Graph for marginal value of cash holdings for financial constrained and unconstrained firms in civil and common law countries
2.2. Empirical literature review
There is no barrier between external and internal funds in perfect markets, implying that one-unit currency of cash in the firm is exactly equal to one-one-unit of value of the firm
(
Pinkowitz and Williamson, 2004). However, in the real world, because of capital market imperfections, one unit of cash holdings does not necessarily generate exactly one unit of firm value.Denis and Sibilkov (2010) focusing on U.S. listed firms, investigated the association between the value of cash holdings and financial constraints from 1985 to 2002. They argued that the firm value is positively related to the cash holdings level no matter in financial constrained firms or not and that this positive relationship is significantly stronger under the financial constraints. Furthermore, they found that the change in cash holdings is negatively associated to leverage and the amount of cash, which is consistent with the result of Faulkender and Wang (2006).
11
3. Methodology and data collection
3.1. Methodology
To find out the different marginal effect of cash holdings between financial constrained and unconstrained firms in civil and common law countries, we apply a regression model to test the hypothesis which are formulated previously. As the effect that we study is the marginal value of cash holdings, we initially use the regression model of Faulkender and Wang (2006). Combining with the adjusted model of Zhang, von Eije, and Westerman (2015), we use the regression as following1:
!",$ =&',(&)&',(*+
',(*+ = ,-+ ,/ ∆1',( &',(*++ ,2 ∆3',( &',(*++ ,4 ∆56',( &',(*++ ,7 ∆8',( &',(*++ ,9 ∆:',( &',(*++ ,; 1',(*+ &',(*++ ,<=",$+ ,>&1',(*+ ',(*+∗ ∆1',( &',(*++ ,@ ∆1',( &',(*++ ,/-AB",$)/+ ,//C",$)/+ D",$ (1)
Where !",$=&',()&',(*+
&',(*+ denotes the firm market return, which is also the dependent variable in
this regression. E",$ is the variable for the amount of cash holdings; F",$ denotes the earnings before interest expense; GH",$ denotes net assets; I",$ indicates interest expenses; K",$ means dividends; =",$ refers to market leverage; AB",$)/ shows market to book ratio of last year;
C",$)/ denotes the size of the firm, which is measured by the natural logarithm of total assets of
firm i at the end of year t-1. In this regression, ΔM means the change of a particular variable during the period over the end of year t-1 to the end of year t. For instance, ∆E",$ denotes the change of cash holdings of firm N from the end of year t-1 to the end of year t. Additionally, ,- is the intercept and ,/ to ,// are the coefficients of variables in this regression. D",$ is the error item of the regression.
According to the regression model of Faulkender and Wang (2006), we add interact dummy variables to test the different effect of the financial constraints and the law systems. The interact dummy variable is expressed as KOPPQ ∗ ∆1',(
&',(*+. To solve the previous research questions, we
make two dummies: ERA and ERGCSTHIG. If ERA is equal to 1, it indicates that the firm is in a common law country. In the same way, if ERGCSTHIG is equal to 1, it means that the firm is financial constrained, otherwise not. After applying the dummy variables into the regression, we get new regression:
!",$ =&',(&)&',(*+
',(*+ = ,-+ ,/ ∆1',( &',(*++ ,2 ∆3',( &',(*++ ,4 ∆56',( &',(*++ ,7 ∆8',( &',(*++ ,9 ∆:',( &',(*++ ,; 1',(*+ &',(*++ ,<=",$+ ,>&1',(*+ ',(*+∗ ∆1',( &',(*++ ,@ ∆1',( &',(*++ ,/-AB",$)/+ ,//C",$)/+ ,/2ERA ∗ ∆1',( &',(*++ ,/4ERGCSTHIG ∗&∆1',( ',(*++ D",$ (2)
To test hypothesis 1 of difference in civil and common law systems, we exclude the dummy ERGCSTHIG ∗ ∆1',(
&',(*+ from the regression. After examining the effects of the country dummy,
we include the other dummy variable ERGCSTHIG ∗ ∆1',(
&',(*+ to test hypothesis 2.
13 median value, then we define this firm as a high growth opportunity firm, and these firms are considered to be financial constrained. Similarly, if the market to book ratio of a firm is smaller than or equal to the median value, we regard this firms as a low growth opportunities firm, and this firm is defined as a financial unconstrained firm.
Secondly, we separate financial constrained and unconstrained firms through paying or not paying dividends. Firms with dividends are considered as financial unconstrained while non-dividend-paying firms are regarded as financial constrained. Because the firms with dividends indicates that those firms have sufficient cash flow to pay dividends to their shareholders. So those dividends-paid firms are considered as financial unconstrained firms, and on the contrary the firms without payout ratio are regarded as financial constrained firms.
3.2. Data
The sample in this study includes listed firms in the EU 15 countries over the period from 2000 to 2016. The reason of choosing EU as the sample is that EU countries are constrained by EU regulation. It is clear to compare the difference effect of civil law system and common law system if they are in the same economic community. Among EU 15 countries, except for the United Kingdom and Ireland, which are under common law traditions, others adopted civil law as their law system.
Then we use Datastream to obtain financial data according to the ISIN collected from Orbis. The market return is measured by the market value of equity, calculated by the change of the market value (M) A",$− A",$)/ divided by the market value of last year A",$)/. Cash holdings (C) is measured by cash and cash equivalents. Earnings (E) is defined as the sum of interest expense and debt net income. Net assets (NA) is defined by total assets excludes cash and cash equivalents. Leverage (L) is total debt over the sum of market value of equity and total debt. Dividend (D) is measured as the total dividend paid to shareholders. Size (SIZE) is measured by the logarithm of total assets. Interest expense (IE) and market to book value (MTB), which can be downloaded from Datastream directly, are also used in this regression. Additionally, as the data is shown on the Datastream as local currency, we convert British pounds (GBP), Danish krone (DKK), and Swedish krona (SEK) into Euro (EUR).
Table 2. Datastream code and items
Item name Explanation Datastream Code
Market value Market value of equity MV
Cash holdings Cash and cash equivalents wc02005
Net income Net income before extraordinary items wc01751
Interest Expense Interest expense wc01251
Total assets Total asset wc02999
Dividends Total dividend paid to shareholders wc05376
Market to book value Market to book value MTBV
Total debt Total debt wc03255
Table 3. Descriptive statistics of the full sample from 2000 to 2016
Variable Mean Median Maximum Minimum Std. Dev. Observations
Panel A: Civil law countries
!",$ 0.137 0.058 2.388 -0.714 0.498 22991 ∆&",$ 0.018 0.004 0.523 -0.325 0.114 22703 ∆'",$ 0.032 0.011 0.740 -0.462 0.140 22185 ∆()",$ 0.075 0.035 1.516 -1.011 0.325 22563 ∆*",$ 0.000 0.000 0.048 -0.051 0.012 21966 ∆+",$ 0.001 0.000 0.037 -0.038 0.009 22096 &",$,- 0.225 0.145 1.263 0.003 0.234 22751 .",$ 0.289 0.253 0.824 0.000 0.221 21197 /0",$,- 2.090 1.500 12.410 -0.570 1.930 23058 1",$,- 12.249 12.093 17.095 7.610 2.016 23031
Panel B: Common law countries
!",$ 0.181 0.093 2.389 -0.713 0.551 9852 ∆&",$ 0.015 0.003 0.518 -0.325 0.112 10036 ∆'",$ 0.031 0.012 0.736 -0.460 0.141 9712 ∆()",$ 0.091 0.039 1.517 -1.012 0.308 10174 ∆*",$ 0.000 0.000 0.048 -0.051 0.010 9949 ∆+",$ 0.002 0.000 0.037 -0.038 0.007 10315 &",$,- 0.189 0.114 1.258 0.003 0.216 10000 .",$ 0.211 0.169 0.823 0.000 0.188 7989 /0",$,- 2.394 1.730 12.420 -0.570 2.175 9893 1",$,- 11.427 11.138 17.095 7.609 2.262 9832
Notes: This table describes the fixed effect of descriptive statistics for both common and civil law countries from
2000 to 2016. We winsorized the variables at 3% and 97% tails. The descriptive statistics contain mean, median, maximum, minimum of variables, as well as the number of observations and standard deviation of common law countries and civil law countries separately. !",$ is the annual return of year t. Δ3 means the change of variable X from year t-1 to year t. ∆&",$ denotes the change of cash and cash equivalents from year t-1 to year t. ∆'",$represents the change in earnings before extraordinary items from year t-1 to year t. ∆()",$ is the change in net assets. ∆*",$ denotes the change of interest expense on total debt from year t-1 to year t. ∆+",$ is the change in total dividend paid to shareholders. &",$,- means cash holdings and cash equivalents of last year t-1. /0",$,- is the market to book ratio of last year t-1 and 1",$,- is measured by the natural logarithm of total assets of last year t-1. .",$ refers to the leverage ratio, which is calculated by total debt over by the sum of total debt and market value of equity.
17 and it is largest in Luxembourg. For the change of cash holdings, it is highest in Luxembourg too, with 0.025, and is smallest in Denmark with 0.007. According to the descriptive statistics of table 4, the mean stock return in civil law countries is 0.137. This value is lower than the counterpart in common law countries of 0.181. Moreover, the average lagged cash holding ratio in civil law countries is 0.225, which is larger than the average lagged cash holdings in common law countries of 0.211. Furthermore, other variables in these two panels present a group of relatively similar numbers, implying that no strong systematic difference exists between civil law and common law countries through descriptive statistics.
Additionally, we test the correlation coefficient among the dependent variable stock return, the change in the amount of cash, and the other variables, which is presented in table 4. As is displayed in the correlation matrix, there is no coefficient which is larger than 0.3, indicating that multicollinearity is not likely to be an issue in this regression.
Table 4. Correlation Matrix
19
4. Results
In this section, the research questions are tested and the results are displayed. Firstly, we check the marginal effect of cash holdings in common and civil law countries separately through the regression which is mentioned in section 3.1, to find out how one unit of cash holdings affect firm value. Secondly, we test the hypothesis 1 to estimate whether the law systems can significantly influence the marginal value of cash per unit of currency, by running the regression which includes the interaction of the common law dummy on the marginal value of cash holdings. Thirdly, we test hypothesis 2 to find out the impact of financial constraints, which is classified by growth opportunities. We test this for common and civil law countries respectively. Moreover, the robustness test which classifies financial constrained firms and unconstrained firms by paying and not paying dividends is also included for a second opinion on the correctness of the test for financial constraints.
4.1. The different marginal effect per unit of currency for law systems
21
Table 5. The marginal value of cash holdings regression test for civil and common law
countries
Variables Civil law Common law Overall
Intercept -0.127 0.392 0.078 (0.181) (0.106) (0.091) ∆&",$ 0.774*** 0.939*** 0.774*** (0.118) (0.064) (0.062) ∆'",$ -0.02 0.231 0.045* (0.054) (0.028) (0.025) ∆()",$ 0.241*** 0.333*** 0.264*** (0.024) (0.012) (0.011) ∆*",$ 0.066 0.142 0.026 (0.697) (0.305) (0.283) ∆+",$ 4.675*** 7.478*** 5.273*** (0.778) (0.338) (0.315) &",$,- 0.723*** 0.795*** 0.744*** (0.054) (0.025) (0.023) .",$ -1.131*** -1.346*** -1.184*** (0.058) (0.028) (0.026) /0",$,- -0.068*** -0.061*** -0.067*** (0.004) (0.003) (0.002) 1",$,- 0.041*** 0.001*** 0.026*** (0.015) (0.008) (0.007) &",$,-∗ ∆&",$ -0.313*** -0.325*** -0.313*** (0.252) (0.116) (0.107) .",$∗ ∆&",$ -0.478*** -0.564*** -0.489*** (0.333) (0.137) (0.129) &5/ ∗ ∆&",$ 0.151** (0.066) Adjusted R2 0.411 0.422 0.394
Marginal value at the mean 0.565 a) b) 0.758 b) 0.710 a) b)
Observations 14505 5302 19807
Notes: This table displays the result of regression of market value on cash holdings from 2000 to 2016. This chart
this regression are winsorized at 3% and 97%. Standard errors are in the brackets. *, **, and *** represent that variables are significant at 1%, 5% and 10% of confidence level. Moreover, a) means the marginal value per unit currency is significantly different from 1 at 5% confidence level on Wald test, b) indicates the marginal value per unit currency is significantly different from 0 at 5% confidence level on Wald test.
4.2. The different marginal effect per unit of currency on financial constraints
Table 6 shows the fixed panel regression of market value on cash holdings, which is presented by financial constrained and unconstrained firms separately. In this section, as it is discussed in section 3.3, we define low growth opportunity firms as financial unconstrained and high-growth opportunity firms as financial constrained. In this table, panel A describes regression in civil law countries and panel B displays the counterpart in common law countries.
23 For the interaction term (&",$,-∗ ∆&",$), the estimated coefficient in all columns are negative, indicating that the change in cash holdings is negatively in relation with the amount of cash holdings. Also for the interaction term (.",$∗ ∆&",$), the estimated coefficients are all negative, suggesting that the level of leverage negatively affect the change in cash holdings. These negative relationships are consistent with Faulkender and Wang (2006), Denis and Sibilkov (2010), and Zhang, von Eije and Westerman (2015).
Table 6. Regressions for financial constrained firms and unconstrained firms
Panel A: Civil law Panel B: Common law Variables Unconstrained Constrained Unconstrained Constrained
Intercept -0.774*** 0.081 0.436 0.335 (0.163) (0.164) (0.332) (0.234) ∆&",$ 0.472*** 0.878*** 0.669*** 1.117*** (0.085) (0.107) (0.162) (0.181) ∆'",$ 0.041 0.013 0.11 0.353*** (0.033) (0.053) (0.07) (0.094) ∆()",$ 0.216*** 0.343*** 0.335*** 0.433*** (0.014) (0.024) (0.032) (0.041) ∆*",$ 0.093 1.342** 0.504 0.663 (0.35) (0.604) (0.892) (1.178) ∆+",$ 3.509*** 4.964*** 6.33*** 7.762*** (0.423) (0.546) (1.085) (1.137) &",$,- 0.416*** 0.91*** 0.449*** 0.94*** (0.031) (0.064) (0.077) (0.104) .",$ -1.445*** -1.326*** -1.609*** -1.589*** (0.038) (0.049) (0.086) (0.091) /0",$,- -0.504*** -0.041*** -0.504*** -0.043*** (0.017) (0.003) (0.035) (0.005) 1",$,- 0.145*** 0.017 0.048* 0.001 (0.013) (0.013) (0.028) (0.019) &",$,-∗ ∆&",$ -0.088 -0.07 -0.053 -1.139*** (0.133) (0.282) (0.313) (0.463) .",$∗ ∆&",$ -0.231 -0.756*** -0.213 -0.624 (0.16) (0.282) (0.412) (0.613) P-value(C-U) 0.118 0.768 Adjusted R2 0.504 0.460 0.488 0.416
Marginal value at the mean 0.366 a) b) 0.709 a) b) 0.600 b) 0.872 b)
Observations 7679 6826 2320 2982
Notes: This table displays the result of regression of market value on cash holdings in financial constrained firms
25 in the total dividend paid to shareholders. &",$,- means cash holdings and cash equivalents of last year t-1. /0",$,- is the market to book ratio of last year t-1 and 1",$,- is measured by the natural logarithm of total assets of last year t-1. .",$ refers to the leverage ratio, which is calculated by total debt over the sum of total debt and market value of equity. P-value(C-U) is the p-value of interaction term &5(189)*( ∗ ∆&",$. Variables in this regression are winsorized at 3% and 97%. Standard errors are in the brackets. *, **, and *** represent that variables are significant at 1%, 5% and 10% of confidence level. Moreover, a) denotes that the marginal value per unit of currency at the mean is significantly different from 1 at 5% confidence level based on Wald test, b) indicates that the marginal value per unit of currency at the mean is significantly different from 0 at 5% confidence level on Wald test.
4.3. Robustness test for financial constraints
Table 7. Robustness check test for firms with paying dividends
Panel A: Civil law Panel B: Common law Variables Unconstrained Constrained Unconstrained Constrained
Intercept -1.018*** 0.303* 0.329 0.349 (0.168) (0.172) (0.237) (0.334) ∆&",$ 0.762*** 0.718*** 0.973*** 0.949*** (0.08) (0.11) (0.15) (0.216) ∆'",$ 0.287*** -0.104*** 0.581*** 0.112 (0.055) (0.038) (0.086) (0.086) ∆()",$ 0.295*** 0.179*** 0.364*** 0.313*** (0.015) (0.022) (0.028) (0.05) ∆*",$ 0.069 0.461 0.456 -0.262 (0.394) (0.507) (0.853) (1.331) ∆+",$ 4.794*** 3.303*** 8.049*** 5.884* (0.338) (1.196) (0.762) (3.217) &",$,- 0.611*** 0.962*** 0.667*** 1.032*** (0.031) (0.045) (0.063) (0.116) .",$ -1.295*** -1.215*** -1.504*** -1.296*** (0.037) (0.053) (0.075) (0.121) /0",$,- -0.086*** -0.058*** -0.056*** -0.069*** (0.004) (0.004) (0.005) (0.01) 1",$,- 0.114*** 0.002 0.01 -0.003 (0.013) (0.015) (0.018) (0.03) &",$,-∗ ∆&",$ -0.368*** -0.091 -0.868*** 0.467 (0.138) (0.208) (0.295) (0.507) ∆.",$∗ ∆&",$ -0.455*** -0.343 -0.521 -0.955 (0.176) (0.231) (0.439) (0.606) P-value(U-C) 0.820 0.219 Adjusted R2 0.489 0.378 0.439 0.322
Marginal value at the mean 0.558 a) b) 0.592 a) b) 0.731 b) 0.838 b)
Observations 8842 5663 3533 1769
Notes: This table displays the result of regression of market value on cash holdings for financial constrained firms
5. Conclusions and recommendations
This paper examines the different marginal effect of cash holdings on firm value under various legal systems in Europe from 2000 to 2016. The sample is divided in two panels: civil law and common law countries. To test the different marginal effect of cash holdings, we also classify subgroups for financial constraints. The coefficients of the interaction of leverage and the amount of cash holdings are negative. It suggests that the European firms can increase cash holdings value by reducing the amount of cash holdings or the leverage (or both). This result shows that the marginal value of cash holdings is negatively related to the cash level and the ability to enter debt market. This negative relationship between cash level and value of cash is also consistent with Faulkender and Wang (2006), Denis and Sibilkov (2010), and Zhang, von Eije and Westerman (2015). The overall marginal value of cash holdings in Europe 15 countries area is 0.710 per Euro. We assumed that the marginal value per unit of currency at the mean in civil law countries is smaller than it is in common law countries as the agency cost is lower in common law countries because of better shareholder protection in these countries. The marginal value of cash in civil law countries is 0.565 per Euro and in common law countries is 0.758 per Euro, and the difference proves to be significant. This implies that the first alternative hypothesis holds and that the marginal contribution per unit of currency to the firm value is larger in common law countries. Although the marginal value at the mean for all subgroups are at a low level, we still suggest that investors and managers treat liquidity differently under the different legal systems, since the marginal effect of cash holdings is tested to be significantly different in civil and common law countries. Furthermore, the overall marginal value per unit of currency at the mean is proved to be significantly smaller than 1. This result is different from Zhang, von Eije and Westerman (2015).
29 matter if they are defined by growth opportunities or payout ratio, the marginal value at the mean is larger in financial constrained firms than in unconstrained firms. However, the differences between financial constrained firms and unconstrained firms are insignificant. When it comes to the graph in section 2.1, which shows the relationship between financial constrained firms and unconstrained firms in civil law and common law countries, the gap between the two lines of common law countries and civil law countries is still there since we detect a significant result for the differences in law systems. However, the slopes of both lines become not that steep as we expected, because the differences between financial constrained firms and unconstrained firms for both types of countries are not significant according to our findings.
References:
Denis, D.J., and Sibilkov, V., 2010. Financial constraints, investment, and the value of cash holdings. Review of Financial Studies, 23, 247-269.
Dittmar, A., and Mahrt-Smith, J., 2007. Corporate governance and the value of cash holdings, Journal of Financial Economics, 83, 599-634.
Dittmar, A., Mahrt-Smith, J., Servaes, H., 2003. International corporate governance and corporate cash holdings. Journal of Financial and Quantitative Analysis, 38, 111–133.
Drobetz, W., Grüninger, M., and Hirschvogl, S., 2010. Information asymmetry and the value of cash. Journal of Banking and Finance, 34, 2168-2184.
Faulkender, M., and Wang, R., 2006. Corporate financial policy and the value of cash. The Journal of Finance, 4, 1957-1990.
Fazzari, Steven M., Glenn Hubbard, R., and Petersen, B., 1988. Financing constraints and corporate investment, Brooking Papers on Economic Activity, 1, 141–95.
Jensen, M.C., 1986. Agency costs of free cash flow, corporate finance, and takeovers. The American Economic Review, 76(2), 323-329.
La Porta, R., Lopez-de-Silanes, F., Shleifer, A., and Vishny, R., 2000. Investor protection and corporate governance, Journal of Financial Economics, 58, 3-27.
Miller, M. H., and Modigliani, F., 1961. Dividend policy, growth, and the valuation of shares, Journal of Business, 34, 411-433.
Modigliani, F., and Miller, M. H., 1958. The cost of capital, corporation finance and the theory of investment, The American Economic Review, 48, 261-297.
31 Myers, S.C., and Rajan, R.G., 1998. The paradox of liquidity. Quarterly Journal of Economics, 108, 733-771.
Opler, T., Pinkowitz, L., Stulz, R., and Williamson, R., 1999. The determinants and implications of corporate cash holdings, Journal of Financial Economics, 52, 3-46.
Pinkowitz, L., Stulz, R., and Williamson, R., 2006. Does the Contribution of Corporate Cash Holdings and Dividends to Firm Value Depend on Governance? A Cross-country Analysis, The Journal of Finance, 61, 2725-2751.
Pinkowitz, L., Stulz, R., and Williamson, R., 2002. What is a Dollar Worth? The Market Value of Cash Holdings. Working paper, Georgetown University.
Whited, T. M., and Wu, G., 2012. Financial constraints risk. The Review of Financial Studies 19, 531-559.
