October, 2018
Dividend Payout, Effective Future Firm Value and the Role of
National Governance
Heike Sigrid Hofmann Student No: s3411516
For the joint degree:
MSc International Financial Management, at Groningen University (RUG) MSc Business and Economics, at Uppsala University (UU)
Supervisor (RUG): Dr. Nassima Selmane Second Assessor (RUG): Dr. Egle Karmaziene
ABSTRACT
This research examines the impact of dividend payout (DP) on future effective firm value for a large international sample of publicly listed companies. Besides the known share price drop right after dividend announcements, literature provides a solid basis for argumentation towards a positive as well as a negative relationship on a longer horizon, going beyond this initial market reaction. The underlying research is the first to discuss and directly examine this issue, following the valuation model introduced by Fama and French (1998). The statistically significant and robust results show a positive impact of DP towards future firm value on a one- and two-year horizon. This research further accounts for the effect of National Governance (NG), proxied by the World Governance Index, as country level moderator on the main relationship. The interacting effect on the main variables is statistically significant on horizons beyond one year. Keywords: Firm Value, Dividend Payout, National Governance,
2
1
Introduction
Despite numerous attempts of scholars to clarify the reason for the existence of dividends, it remains a major puzzle in corporate finance (Allen, Bernardo, and Welch, 2000). Previous literature focuses on the roots of dividend payout (DP) decisions (e.g. Lintner, 1956; Brav, Graham, Harvey, and Michaely, 2005), the changes in DP throughout history (e.g. DeAngelo and DeAngelo 1990, Fama and French, 2001), potential substitutes (e.g. Allen et al., 2000) and its relationship with specific variables, such as past, current or expected firm value or past profitability (see e.g. Fama and French, 1998; Fama and French 2001 Eades, 1982). That there is an immediate effect of dividends on firm value at the time of dividend announcement (i.e. a share price drop) has also broadly been evidenced (Fama and French, 1998). The results lead to the supposition that dividends are either corporate value enhancing or value destructive. What is striking through, is that these assumptions are made without having tested dividends’ actual impact on effective future firm value going beyond this initial market reaction on DP. And it is, indeed, a valid question that requires clarification, as all, what business comes down to, is the maximization of value (Jensen, 2010).
3
the chance exists that there is no statistical significant relationship between DP and effective future firm value. However, the upcoming arguments are too convincing to believe that the one or the other does exist. Following the findings outlined above, the relationship between dividend payout and firm value requires investigation and is therefore core of the underlying research.
As has been subject of a plethora of investigations in international finance (see e.g. La Porta, Lopez de Silanes, Shleifer, and Vishny, 1998), the underlying research also acknowledges the great diversity seen across countries. More specifically, the impact national governance (NG) may have on important corporate decisions, one of which is on the DP policy. Therefore, a moderating factor on the relationship between DP and the next period's firm value is taken into consideration, which accounts for the level of governance in the country of the respective sample firm. The idea behind this is that NG may impact the decision on DP, and subsequently the relationship between DP and firm value.
In line with the above, the underlying research addresses the following research question: What
is the relationship between DP and future effective firm value in publicly listed corporations?
4
The remainder of this paper is structured as follows. Section 2 discusses the theoretical foundations for this study, presenting the most prevalent models and their deviating interpretation possibilities. Section 3 presents the data and methodology followed by section 4 on the empirical results and its robustness. Section 5 then concludes the paper and discusses recommendations for further research.
2
Literature review
2.1 Dividend payout
5
communication device to the general public; effects which reason the preference for dividends above other cash-back vehicles (Morck and Yeung, 2005). According to Morck and Yeung (2005), important properties of dividends are, for instance, its signalling power to the general public, agency problem mitigating effects, and attraction of specific investor clientele. These theories will be discussed in more detail in the next sub-section. For now it is to note that despite changing demand for dividends throughout time, the market keeps up its request for DP (Easterbrook, 1984; Morck and Yeung, 2005). This shows dividends' economic importance, and together with the existence of many questions that have not yet been answered, reasons further investigation in this field.
Are dividends solely the result of the past or could they also be the trigger for something in the future, for instance increased firm value? Most research legitimately views DP as a result of well-governance, large past profits, … etc. (see for instance Fama and French, 2001; Kale Kini, and Payne, 2012). According to Fama and French (2001) there are three fundamental factors depending on which a firm pays dividends: profitability, investment opportunities and size. The dividend paying firm is large, profitable and has fewer investment opportunities (Fama and French, 2001). Also Kale et al. (2012) find that a higher quality firm pays higher dividends. In contrast, dividends are less likely to be paid by unprofitable firms (Fama and French, 2001). Following this notion, dividend payments could be a simple side-effect of high-value-firms. Looking further beyond, however, an investor is not interested in the past, but in future opportunities. Perhaps dividends do not only mirror the past but also have an effect on what is yet to come. Therefore, the underlying research takes on a new stand point, questioning dividends propulsive power on effective future firm value.
2.2 Dividend payout and firm value
6
firm value. The logic behind this inference is quite simple: with a given investment policy the firm must decide on funding internally (i.e. reduce DP and utilize retained earnings) and external funding (i.e. increase DP and float new shares). According to the authors, in an ideal world these two parts of the equation cancel each other out. Thus, a firm is indifferent between DP and issuing new shares for funding, or retain earnings. Put differently, DP is irrelevant to current and/or future firm value. These conjectures were subject to a plethora of research (Eades, 1982), and thus also constitute the beginning of the underlying research. Fama and French (1998) test the latter statement made by Miller and Modigliani (1961) under non-perfect market conditions, concluding that “dividends have information about expected profitability missed by current, past, and future values of earnings, investment, R&D, and debt variables” (p.837). Following this finding, the idea that there is an effect of DP on future effective firm value going beyond the initial announcement effect is not too farfetched. Considering the fact that perfect market conditions solely exist in theory, and in extension to Fama and French (1989), I first hypothesize that:
Hypothesis 1. Dividend payout has an impact on effective future firm value.
Next, literature provides a number of legitimate arguments for the direction this relationship may take. However, these arguments go both ways. In the following, I will first elaborate the arguments for a positive relationship and then switch to arguments for a negative relationship between dividend payout and effective future firm value.
7
(1998) go hand in hand with what is called Signalling Theory. Due to information asymmetries between corporate insiders and the rest of the world, dividends are said to convey information to investors (Brav et al., 2005), or more specifically, about corporate prospects (Allen et al., 2000; Brav et al., 2005). These assumptions are backed by empirical patterns. First, dividend initiation is accompanied by an abnormal positive announcement effect (Kale et al., 2012). Secondly, every rise in DP produces positive abnormal returns, measured by the increase in share prices (Allen et al., 2000). Last but not least, and perhaps most interestingly, as shown in Michaely, Thaler and Womack (1995), and consistent with prior literature, dividend omissions or cuts cause a negative reaction in a higher magnitude than its positive counterpart regarding the immediate as well as long term market reaction. While the authors explain this asymmetry for the short term with the changing magnitude of the yield, for the long term they remain silent. Not going any further into detail about this (please find further information in the publication of the authors), I want to stress that in line with these empirical findings the practical applicability of the Signalling Theory seems plausible. Consequently, dividends should positively relate to future firm value.
8
dividends reduces free cash flow (Jensen, 1986). Easterbrook (1984) adds that liquidity shortage and possible external funding inflicts market discipline on the firm. This is exemplified through the monitoring effect of external financing. In case of debt financing, financial institutions monitor the company's well governance, requiring discipline and even better management (Opler et al., 1999). Hence, dividends are considered an agency cost mitigating factor, counteracting firm value destruction through the creation of a tight budget. Following this perspective, again, a positive relationship between DP and future firm value can be expected.
In another vein, future firm value may be affected by the type of investors, which are, beyond other things, defined by the DP policy. Clientele Theory explains that, as long as there are taxes, if some investors are tax-exempt while others are not, different type of investors will have deviating preferences with regard to dividends (Easterbrook, 1984). As a result, dividends are said to attract the often tax-exempt institutional investors (Allen et al., 2000, Morck and Yeung, 2005). Institutional investors are found to have higher expertise in detecting high value firms and have an advantage in ensuring sound management (Allen et al. 2000). Not only can these institutions influence business decisions (Becht, Franks, Mayer, and Rossi, 2008), but also can a stock held by institutional owners overcome the free-rider problem of the investors of widely held firms (Morck and Yeung, 2005) by means of monitoring the listed firm’s business activity (Kale et al., 2012). As such, institutional investors can not only enhance firm value but also validate the firm’s quality towards the general public (Kale et al., 2012). Following this notion, the investor perspective also implies a positive relationship between DP and effective future firm value.
9
Hypothesis 2a. Dividend payout positively impacts effective future firm value.
Despite the above discussed convincing arguments for the positive impact of DP on effective future firm value, there are just as much persuasive counterarguments reasoning the opposite.
Benchmarking to Lintner (1956), who is considered the founder of the modern understanding of the dividend policy, and who declared dividends to be ‘sticky’, Brav et al. (2005) confirm, in their 21st century qualitative study, manager’s strong desire to avoid dividend cuts. They find that managers would sell assets, lay off a large number of employees, borrow heavily, or also bypass positive NPV projects in order to avoid to cut back on dividends. Taking these findings into account, dividends’ stickiness and managements’ fear of share price movements in case of cuts, could eventually result in a negative relationship between DP and future firm value. After all, in the long run, no company can pay dividends higher than their net income.
Further above was explained how DP can function as agency cost mitigating factor and subsequently increase firm value. This conclusion, however, is not as clear-cut as it first seems. If the potential investments that a firm faces convey a higher return than the market, the firm would be well off using its internal funds (Jensen, 1986), instead of paying them out to investors. Hence, depending on the investment opportunities, additional liquidity can be value enhancing (Opler et al., 1999). Unfortunately, there can be incentives for managers, such as career concerns, to avoid additional risk and forgo even positive NPV projects (Brav et al., 2005, Easterbrook, 1984). Following this notion, paying out dividends can be a negative indicator for future firm value, as it refrains from investing the funds, either because it does not see enough investment opportunities, or management finds reasons to act risk-averse.
10
cash holdings can, in turn, function as a buffer for difficult times (Dittmar, 2008). Keeping a tight budget by paying excess cash out to investors, can, thus, endanger sound business (Dittmar, 2008), implying a negative relationship between DP and future firm value.
In sum, acknowledging the stickiness of dividends, DP may not be any sign of the degree of well-being of an operation, and subsequent firm value. On the contrary, the reduction of internal funds disables firms' investment opportunities and, along with Agency Theory, provides management with arguments to not invest in positive NPV projects, even if the true reason might be of personal nature. Last but not least, reduced internal funds can come at a major disadvantage in the case of financial distress, as the firm then has to turn to expensive external funding instead. Following these arguments, it can also be hypothesized that:
Hypothesis 2b. Dividend payout negatively affects future effective firm value.
2.3 National governance, dividend payout and firm value
11
National environment is a very broad term. Investigation of all national factors surrounding corporate establishments would certainly go beyond the limits of the underlying research. Therefore, this paper focuses on a national governance as influential factor on country level.
There are reasons to believe that the effect of dividend payout on effective future firm value should be impacted by national governance, which contains factors such as for instance political stability and governmental effectiveness. In order to provide a solid argumentation for these assumptions, some more explanation is required. On the one hand, environmental stability has widely been evidenced to strongly affect firm performance (Fredrickson, 1984; Fredrickson and Mitchell, 1984; Smart and Vertinsky, 1984). More specifically, the level of strength of a country’s legal framework is shown to influence firm value (Durnev and Kim, 2005). On the other hand, literature depicts a connection between financing preferences and firm environment (Fisman and Love, 2003). For instance, enhanced legal protection is found to be associated with more valuable stock markets (La Porta, Lopez de Silanes, Shleifer, and Vishny, 1997). Also, institutional voids are found to be connected to difficulties in financial funding, suboptimal decision-making due to the fear of government interventions, distorted competition, product market failure due to limited regress possibilities for the customer, as well as regulatory blockages for new business opportunities (El Ghoul, Guedhami, and Kim, 2017). At the same time, investment opportunities and related decisions, together with subsequent cash holding, can impact the firm's payout policy (e.g. Fama and French, 1998; Opler et al, 1999). On the basis of these arguments, a complementary impact of an organization's national governance on the relationship between DP and future firm value is expected. Therefore, it is hypothesized that:
12
Hypothesis 3b. The moderating impact of national development substitutes the relationship between dividend payout and effective future firm value.
3
Methodology
3.1 Data
Annual fundamental data and market data on listed companies is collected from the Wharton Research Data Service (WRDS) Compustat - Capital IQ files. This database provides data for firms of numerous countries, with the exception of the USA and Canada. Country level data is
collected from the World Bank Database. The two data sets from WRDS Compustat are matched
using each firm's ISIN code and the respective year. This is then merged with country data using each firm's ISO code and the respective year. The period of 2002 to 2016 is chosen due to data availability of country level data.
13
In order to analyze the effect of DP on future effective firm value, the underlying research uses year values for the main independent variable. Also, for all firm control variables the one-year changes into the future and into the past are considered. Consequently, while the dependent variable relates to the years from 2003 to 2015, the independent variable spans the years of 2002 to 2014. Thus, it is tested, whether the independent variable of a specific year impact the dependent variable in one year. The final sample with one-year variables consists of 32,577 firms and 40,851firm-year observations, spanning 106 countries from 2003 to 2015.
3.2 Regression model
14
firm value until now, I test past DP on current firm value, as also firm value is not any expected but an true effective value. Despite different terminology, the results remain the same. Finally, the model is expanded by a number of firm and country control variables that have shown to be relevant by research, as will be argued further below.
The following regression model is employed:
TOBINi,t = β0+ β1DPi,t−1+ β2DPi,t+ β3∂DPi,t−1+ β4∂DPi,t+1+
β5∂ATi,t−1+ β6∂ATi,t+1+ β7Ei,t+ β8∂Ei,t−1+ β9∂Ei,t+1+ β10INTEXi,t+ β11∂INTEXi,t−1+ β12∂INTEXi,t+1+ β13LEVi,t+ β14∂LEVi,t−1+ β15∂LEVi,t+1+ β16INVi,t+ β17∂INVi,t−1+ β18∂INVi,t+1[+ β19DPi,t−1∗ NGi,t+ β20DPi,t∗ NGi,t+ β21∂DPi,t−1∗ NGi,t+ β22∂DPi,t+1∗ NGi,t] +
Σ Year_Dummies + Σ Industry_Dummies + εi,t , (1)
where i and t refer to company and year, respectively. ∂Xi,t−1 represents the one-year changes of the respective variable, namely (Xi,t− Xi,t−1)/ATt. Likewise, ∂Xi,t+1 represents the expected one-year changes of the respective variable. 𝜀𝑖,𝑡 represents robust standard errors. All variables are defined in the following sub-section. Through utilizing robust standard errors in the regression, all error terms have a constant variance and are uncorrelated. Consequently, autocorrelation as well as heteroskedasticity issues can be neglected (Brooks, 2014).
15
With regard to hypothesis 1, the level of significance of 𝛽1will deliver the key message on whether there is a relationship between dividend payout and future effective firm value. As indicated further above and in accordance with prior empirical research, the existence of a relationship is expected, which means the p-value should be significant. Given its significance, and in view of hypothesis 2, 𝛽1 will answer whether this relationship is positive or negative. A positive coefficient would indicate that an increase in dividend payout relates to a higher future effective firm value, whereas in case of a negative coefficient an increase in dividend payout would be associated with a lower future effective firm value. Then, in accordance with hypothesis 3, the moderating effect of National Governance is taken into account, whereby the model is extended by the interaction term displayed in square brackets. Here, 𝛽19 shows whether NG compliments the main relationship, or functions as substitute.
3.3 Variables
The dependent variable is firm value. Because the utilization of Tobin's Q has found broad acceptance throughout finance literature (e.g. Isakov and Weisskopf, 2015), this ratio is considered most appropriate for the purpose of the underlying research. Firm Value, TOBIN, is calculated as
TOBIN = (MVE + AT − CEQ)/ AT, (2)
MVE = CSHO ∗ PRCCD, (3)
16
The independent variable is one-year lagged value of Dividend Payout, DP. In line with Fama and French (1998) and Eades (1982), it is measured
DP = DVT / AT, (4)
where DVT is the total annual dividends paid. It scaled by total assets in order to put it in the same units as is the dependent variable.
The country level moderator is called National Governance, NG. For the purpose of the underlying research NG is measured via the World Governance Indicator (WGI) provided by the Wold Bank, which summarizes the overall governance of a large number of countries. In line with Kaufmann, Kraay, and Mastruzzi (2009) and Pinkowitz, Stulz, and Williamson (2015), the variable is defined as follows
NG = (VAA + PSNV + GE + RQ + ROL + COC)/6, (5)
where VAA is Voice and Accountability, PSNV is Political Stability and Absence of Violence/Terrorism, GE is Government Effectiveness, RQ is Regulatory Quality, ROL is Rule of Law, and COC is Control of Corruption, as defined by The Wold Bank Group (2018). NG takes a value between -2 and 2.
In order to control for firm specific attributes, a number of firm level control variables are employed whose omission could otherwise cause deviating results. According to Fama and French (1998) earnings, investments, leverage and interest expense are related to firm value. According to Modigliani and Miller (1961), further, taxation can influence an investor's preference of dividends versus capital gains and vice versa. This is why, with regard to earnings measurement, it is important to exclude local taxation in this international research. Hence, the following variables are considered:
17
INV = Investments = INVC/AT, (7)
LEV = Leverage = (DLC + DLTT)/ AT, (8)
INTEX = Interest Expense = XINT/ AT, (9)
where EBIT is Earnings Before Interest and Taxes, INVC is invested capital, DLC is total debt in current liabilities, DLTT is total long term debt, and XINT is the total of interest and related expenses.
In addition to the above, some country variables are added for control purposes. According to a myriad of research the country's inflation and interest rate are important influential factors for stock prices and, subsequently, firm value (see for instance: Bodie, 1976; Fama and Schwert, 1977; Hong, 1977). Also, GDP per capita is a common measure for economic development, which influences firm value (Pinkowitz, Stulz and Williamson 2006). Consequently, the following additional country level variables are also accounted for:
INFL = annual inflation rate, % (10)
INTR = annual deposit interest rate, % (11)
GPD = ln(GDP per capita, denominated in USD). (12)
For robustness purposes two proxies of the independent variable, namely DP_PROX1 and DP_PROX2, are compounded as follows
DP_PROX1 = (DVC + DVP)/AT, (13)
DP_PROX2 = DVT/PRCCD, (14)
18
In order to mitigate the potential problem with outliers, all variables are winsorized at the 1% level, except of dummy variables and NG, as these are already predefined in range.
4
Results
This section discusses the descriptive statistics on the variables used in the underlying empirical research, the Pearson correlation coefficients matrix as well as the regression results employing OLS regression analyses.
4.1 Descriptive statistics
Table 1 shows the descriptive statistics for all variables used in the underlying research. Additionally, summary statistics per country (table 10) can be found in the appendix.
19 Table 1. Descriptive statistics
Variable N Mean Min Max Std. Dev.
TOBINt 245,537 2.004 0.353 23.701 2.917 DPt 120,131 0.021 0.000 0.181 0.030 DPt-1 99,122 0.021 0.000 0.184 0.031 dDPt-1 88,135 0.002 -0.045 0.058 0.012 dDPt+1 88,121 -0.002 -0.069 0.046 0.014 dATt-1 201,477 0.040 -1.585 0.810 0.288 dATt+1 201,340 -0.131 -2.986 0.672 0.438 Et 244,774 0.016 -1.087 0.321 0.181 dEt-1 200,545 0.004 -0.449 0.481 0.100 dEt+1 200,409 -0.008 -0.549 0.418 0.105 INVt 245,536 0.627 -0.267 0.984 0.220 dINVt-1 201,474 0.023 -1.295 0.699 0.234 dINVt+1 201,337 -0.087 -2.422 0.606 0.351 LEVt 245,302 0.214 0.000 0.894 0.198 dLEVt-1 201,266 0.011 -0.430 0.364 0.102 dLEVt+1 201,129 -0.026 -0.717 0.327 0.131 INTEXt 224,265 0.014 0.000 0.100 0.018 dINTEXt-1 179,017 0.000 -0.042 0.035 0.009 dINTEXt+1 178,912 -0.001 -0.051 0.035 0.010 GDPt 226,899 9.524 6.561 11.291 1.332 INTRt 148,369 3.141 0.080 16.765 3.133 INFLt 226,428 3.012 -1.353 13.648 3.068 WGIt 246,479 0.639 -1.659 1.970 0.837 DPt*NEt 120,129 0.015 -0.054 0.193 0.035 DPt-1*NEt 99,121 0.015 -0.055 0.193 0.035 DPt-2*NEt 80,284 0.015 -0.056 0.195 0.035 dDPt-1*NEt 88,134 0.001 -0.049 0.055 0.011 dDPt+1*NEt 88,120 -0.001 -0.065 0.051 0.013
Note: this table provides information on the descriptive statistics of all variables used in this research. The observation period is 2002 to 2016. The definitions of the variables can be found in section 3
20
3, (both 1.8%) as well as SIC 0 (1.9%). Observing such differences in DP across industries do not come as a surprise, as it has already been documented by prior research (see for instance Huston 2015; Michel 1979).
Table 2. Industry (SIC) distribution
All variables DPt
Industry classification N percent mean
0 - agriculture, forestry, fishing, hunting and trapping 2,333 0.95 0.019
1- mining and construction 26,149 10.61 0.018
2 - light manufacturing 57,125 23.18 0.022
3 - heaving manufacturing 78,190 31.72 0.018
4 - transportations, communications, electric, gas and sanitary services
22,587 9.16 0.024
5 - wholesale and retail 23,445 9.51 0.021
7 - service 29,751 12.07 0.029
8 - other services except public administration 6,908 2.80 0.026
Total 246,488 100 0.022
Note: this table provides information on the industry distribution of all firms in the sample for all variables as well as for Dividend Payout only. The numbers refer to the SIC classification. SIC Codes 6 and 9 were excluded from the sample as reasoned in section 3. The observation period is from 2002 to 2016.
Table 3. Descriptive statistics per year
All variables DPt
YEAR N percent N percent
2002 4,839 1.96 2,873 2.39 2003 7,506 3.05 4,188 3.49 2004 7,484 3.04 4,394 3.66 2005 12,848 5.21 8,303 6.91 2006 14,282 5.79 8,194 6.82 2007 9,090 3.69 3,704 3.08 2008 10,029 4.07 4,578 3.81 2009 20,190 8.19 9,434 7.85 2010 21,145 8.58 10,033 8.35 2011 22,613 9.17 10,888 9.06 2012 23,544 9.55 11,055 9.20 2013 24,036 9.75 11,698 9.74 2014 24,685 10.01 11,251 9.37 2015 25,039 10.16 11,379 9.47 2016 19,158 7.77 8,159 6.79 Total 246,488 100 120,131 100
21
Table 3 provides information on descriptive statistics per year. It clearly shows an overall increase of observations per year, except of the year 2007 and 2008 which display a drop in observations of almost a third. These figures indicate a strong impact by the financial crisis, which is mainly referred to having taken place in 2007 to 2008 followed by a post-crisis period (Claessens and Kose, 2013). The right hand side of the table further informs about the distribution of DP over the years with the respective means. An illustration of this data is shown in figure 1.
Figure 1.
Figure 1 depicts that starting from a value of about 2% in 2002, DPt experienced a surge in 2006, marking its top mid-2007. A subsequent sharp decline until the beginning of 2009 then led DPt to come back at pre-crisis levels. The following years were then marked by values below 2% with a turnaround in 2012, where DPt started to pick up again. In 2016 the mean of DPt in the sample stands at 2.27%. The massive payouts taking place before crisis are in line with Floyd, Li, and Skinner (2015), who extends the prior observations on DP documented by
22
Fama and French (2001). However, it is to note that in the underlying research, and unlike Floyd et al. (2015) share repurchases are not part of the observation.
4.2 Correlation analysis
Table 4 shows the correlation matrix. Upfront it is to note that almost all correlations depict statistical significance, on a 5%-significance level. In line with the expectations of the underlying research, DP shows a positive correlation with TOBIN at all times, and so are the related moderating terms. With regard to the control variables, E, LEV and INV are negatively associated with TOBIN, while INTEX is positively related. Turning to country level controls, GDP and WGI are negatively associated with the dependent variable, while INTR and INFL are positively related.
23
to that is the existence of significant information asymmetries, which makes external finance simply more costly that internal funding.
24 Table 4. Pearson correlation coefficients matrix
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14)
25
Table 4 continues
(15) (16) (17) (18) (19) (20) (21) (22) (23) (24) (25) (26) (27) dLEVt-1 dLEVt+1 INTEXt dINTEXt-1 dINTEXt+1 GDPt INTRt INFLt WGIt DPt*NEt DPt-1*NEt dDPt-1*NEt dDPt+1*NEt dLEVt-1 1 dLEVt+1 -0.071* 1 INTEXt 0.031* -0.033* 1 dINTEXt-1 0.390* -0.117* 0.258* 1 dINTEXt+1 -0.278* 0.449* 0.091* -0.101* 1 GDPt -0.076* 0.063* -0.260* -0.071* 0.083* 1 INTRt 0.051* -0.045* 0.368* 0.077* -0.074* -0.431* 1 INFLt 0.081* -0.052* 0.325* 0.112* -0.098* -0.609* 0.781* 1 WGIt -0.069* 0.053* -0.225* -0.060* 0.065* 0.884* -0.420* -0.528* 1 DPt*NEt -0.049* 0.012* -0.151* -0.049* 0.036* 0.489* -0.185* -0.224* 0.566* 1 DPt-1*NEt -0.010* 0.013* -0.151* -0.033* 0.038* 0.498* -0.190* -0.234* 0.570* 0.922* 1 dDPt-1*NEt -0.009* -0.029* -0.040* -0.003 -0.016* 0.126* -0.043* -0.059* 0.156* 0.398* 0.102* 1 dDPt+1*NEt 0.018* 0.014* 0.031* 0.015* 0.008* -0.140* 0.054* 0.078* -0.182* -0.135* -0.166* -0.042* 1
26 4.3 Regression analyses
Table 5 presents the regression results using OLS regression analysis with entity and time fixed effects and controlling for firm- and country-level characteristics. Autocorrelation and heteroskedasticity robust standard errors are displayed in parenthesis. As more variables are added, the number of observations drops due to varying availability of the variables employed. The results exhibit a statistically highly significant relationship between DPt-1 and TOBINt throughout all columns 1 to 13, on a 99% confidence level. This is in line with hypothesis 1; there is an effect of dividend payout on the effective future firm value, here on a one-year horizon.
Moreover, the coefficient of DPt-1 is positive throughout all regressions. The slope decreases as more explanatory variables are added, and shows a coefficient of 12.26 on a 99% confidence level, when all control variables are added in column 10. This is in line with hypothesis 2, suggesting a positive relationship between the main variables. More specifically, a change in DPt-1 by 1 increases TOBINt by 12.26. Statistical significance of the independent main variable DPt-1 in the regressions with various control variables is in line with the idea that dividends convey information that goes beyond that accounted for by these other control variables, such as investments. As discussed in section 2, this idea is, in general, not new, but roots back to Fama and French (1998). What is new, is that this appears to be still relevant despite the changing characteristics of dividend paying firms as documented by Fama and French in 2001, and most importantly, when controlling for the effect of DP on effective future firm value.
27
underlying sample when accounting for the other control variables. The statistical significance of the other control variables conforms to the expectations of the underlying research, as argued in section 3. Of all firm level controls, the largest impact on the dependent variable has INTEXt (Interest Expense), with a coefficient of 17.45, on a 99% confidence level. In other words, a dollar of more interest expense decreases firm value by 17.45. With regard to INVt (Investments) the negative coefficient in the regression is reasonable in the way that dividend payments decrease the possibility of investments as it reduces cash (Easterbrook, 1984). Therefore, the results imply existent agency benefits in firm value through the agency problem mitigating effect of DP. For LEVt (Leverage) the negative association to firm value is the opposite from the expected. Further above I discussed that additional monitoring through external finance, such as banks, should increase the firm's well-governance (Opler et al., 1999). However, the results show that an increase of LEVt decreases TOBINt by 1.024. As found in Fama and French (1998) this negative association might be due to the negative impact of leverage on proxies of profitability (whereas profitability is computed the same way as Earnings in the underlying research).
28
financing and thus, decreases investments on behalf of companies. GDPt, on the other hand, shows a positive impact on firm value, which goes hand in hand with the notion that firms in an efficient country perform better and consequently depict higher firm values.
29 Table 5. Dividend payout and future firm value
(1) (2) (3) (4) (5) (6) (7)
TOBINt TOBINt TOBINt TOBINt TOBINt TOBINt TOBINt DPt-1 14.70*** 13.53*** 11.84*** 11.02*** 10.88*** 12.26*** 12.70*** (9.74) (8.17) (7.33) (6.93) (6.83) (7.65) (7.86) DPt 0.471 1.946 3.319 4.491** 4.080* 3.924* 5.206** (0.29) (1.09) (1.96) (2.68) (2.42) (2.36) (3.11) dDPt-1 24.21*** 20.80*** 15.24*** 14.08*** 13.76*** 13.70*** 13.18*** (11.37) (8.99) (6.81) (6.39) (6.19) (6.29) (5.99) dDPt+1 -19.77*** -15.45*** -10.81*** -10.90*** -9.936*** -9.796*** -10.32*** (-21.52) (-17.42) (-11.40) (-11.54) (-10.48) (-10.35) (-10.71) dATt-1 -0.0635 -0.116 -0.448** -0.387* -0.142 -0.301 (-0.65) (-1.40) (-2.86) (-2.04) (-0.88) (-1.87) dATt+1 -1.175*** -0.981*** -0.224* -0.568*** -0.503*** -0.424*** (-19.52) (-16.51) (-2.25) (-4.79) (-4.34) (-3.66) Et 0.400 0.494 0.351 0.335 -0.230 (1.07) (1.36) (0.97) (0.89) (-0.61) dEt-1 2.463*** 2.062*** 2.023*** 1.367*** 1.671*** (6.45) (5.70) (5.65) (4.02) (4.95) dEt+1 -2.922*** -2.679*** -2.462*** -1.906*** -1.766*** (-9.85) (-9.27) (-8.48) (-6.68) (-6.24) INVt -0.545*** -0.717*** -0.692*** -0.664*** (-7.67) (-9.45) (-10.79) (-10.44) dINVt-1 0.714** 0.742** 0.542* 0.595** (3.01) (3.11) (2.52) (2.79) dINVt+1 -1.207*** -1.060*** -0.848*** -0.923*** (-7.59) (-6.58) (-5.28) (-5.76) LEVt -0.618*** -0.746*** -0.407*** (-10.73) (-8.33) (-4.44) dLEVt-1 -0.00238 0.0280 -0.161 (-0.01) (0.18) (-1.04) dLEVt+1 0.978*** 0.869*** 0.771*** (7.21) (6.48) (5.73) INTEXt 10.12*** 1.719 (8.08) (1.29) dINTEXt-1 -7.675*** -5.243* (-3.56) (-2.46) dINTEXt+1 -4.110* -0.647 (-2.19) (-0.34) GDPt -0.160*** (-24.21) INTRt INFLt NGt DPt*NGt DPt-1*NGt dDPt-1*NGt dDPt+1*NGt Constant 1.405*** 1.286*** 1.244*** 1.614*** 1.879*** 1.732*** 3.064*** (22.71) (20.49) (19.67) (20.69) (21.83) (23.70) (31.93)
Year FE Yes Yes Yes Yes Yes Yes Yes
Industry FE Yes Yes Yes Yes Yes Yes Yes
Country FE No No No No No No No
Entity FE No No No No No No No
30
Table 5 continues
(8) (9) (10) (11) (12) (13) TOBINt TOBINt TOBINt TOBINt TOBINt TOBINt
DPt-1 10.84*** 11.03*** 12.26*** 13.23*** 14.09*** 6.772*** (6.09) (6.22) (6.88) (5.69) (5.97) (3.76) DPt 5.083** 5.140** 4.732* 0.221 2.779 6.017** (2.64) (2.67) (2.45) (0.09) (1.09) (3.12) dDPt-1 11.46*** 11.56*** 11.50*** 13.41*** 11.73*** 2.621 (4.78) (4.83) (4.80) (4.76) (4.09) (1.20) dDPt+1 -9.060*** -9.198*** -8.394*** -7.415*** -7.883*** -2.734** (-8.58) (-8.72) (-8.01) (-6.75) (-7.24) (-2.72) dATt-1 -0.147 -0.139 -0.245 -0.303 -0.351 -0.700*** (-0.76) (-0.71) (-1.25) (-1.55) (-1.81) (-5.26) dATt+1 -0.435** -0.430** -0.360** -0.340* -0.296* -0.260 (-3.11) (-3.07) (-2.58) (-2.44) (-2.13) (-1.56) Et -1.025* -1.049* -0.782 -0.835 -0.699 -1.209 (-2.16) (-2.21) (-1.64) (-1.75) (-1.45) (-1.15) dEt-1 1.534*** 1.501*** 1.559*** 1.672*** 1.718*** 1.894** (3.77) (3.69) (3.83) (4.13) (4.26) (3.20) dEt+1 -0.871* -0.821* -1.042** -1.123** -1.212*** -0.679 (-2.48) (-2.34) (-2.97) (-3.19) (-3.47) (-1.81) INVt -0.453*** -0.450*** -0.438*** -0.458*** -0.377*** -1.110*** (-5.69) (-5.65) (-5.51) (-5.78) (-4.98) (-5.23) dINVt-1 0.674* 0.675* 0.667* 0.700** 0.649* 0.800*** (2.51) (2.52) (2.48) (2.62) (2.44) (4.12) dINVt+1 -0.985*** -0.993*** -0.977*** -0.986*** -0.963*** -0.547* (-5.10) (-5.15) (-5.07) (-5.13) (-5.04) (-2.47) LEVt -1.018*** -1.028*** -0.967*** -0.847*** -0.838*** -2.113*** (-8.03) (-8.09) (-7.57) (-6.59) (-6.66) (-4.85) dLEVt-1 0.0741 0.0887 0.107 0.0376 0.0756 0.873*** (0.38) (0.45) (0.54) (0.19) (0.40) (3.95) dLEVt+1 0.925*** 0.900*** 0.834*** 0.812*** 0.802*** 0.627*** (5.53) (5.43) (5.03) (4.91) (4.91) (3.56) INTEXt 18.74*** 19.22*** 17.45*** 14.18*** 15.26*** 16.44** (8.09) (8.31) (7.49) (6.03) (6.26) (2.97) dINTEXt-1 -8.346** -8.557** -9.004** -8.291** -9.982*** -8.624** (-2.88) (-2.94) (-3.12) (-2.88) (-3.48) (-3.15) dINTEXt+1 -7.504* -6.955* -5.667 -4.139 -4.167 -5.753 (-2.35) (-2.24) (-1.83) (-1.34) (-1.36) (-1.72) GDPt -0.312*** -0.328*** 0.369*** 0.265*** 0.398*** 0.164** (-25.84) (-27.72) (8.85) (6.18) (7.32) (3.00) INTRt -0.0719*** -0.0691*** -0.0804*** -0.0701*** -0.0380* -0.0543*** (-8.62) (-8.13) (-9.61) (-8.46) (-2.09) (-4.98) INFLt -0.0175* 0.0106 0.00568 0.0174 0.00286 (-2.53) (1.48) (0.81) (1.90) (0.48) WGIt -0.899*** -0.911*** 1.386*** 1.357*** (-18.85) (-19.21) (6.38) (11.40) DPt*NGt 8.572*** 6.788** 3.141 (4.13) (3.28) (1.88) DPt-1* NGt -1.260 -2.069 0.630 (-0.65) (-1.08) (0.44) dDPt-1* NGt -4.202 -2.908 2.588 (-1.78) (-1.23) (1.41) dDPt+1* NGt -3.103** -3.154** -5.090*** (-2.89) (-3.02) (-5.39) Constant 4.260*** 4.394*** -1.168*** -0.320 1.617 0.114 (31.23) (32.67) (-3.39) (-0.90) (1.01) (0.21) Year FE Yes Yes Yes Yes Yes Yes Industry FE Yes Yes Yes Yes No Yes
Country FE No No No No Yes No
Entity FE No No No No No Yes
Observations 40,871 40,851 40,851 40,851 40,851 40,851 Adj. R2 0.206 0.208 0.224 0.232 0.264 0.201
31 4.4 Further investigation
So far the effect of one-year lagged DP on current firm value has been examined. As the theoretical explanations in section 2 are not necessarily limited to one year though, it is interesting to see, whether they also hold on longer horizons. Therefore, I further test the same hypothesis with adding additional DP variables which are lagged by two (three) years. Hence, it is being tested whether the results shown in the section above still hold true when spanning the time effect one (two) more year(s). Consequently, while the dependent variable relates to the years from 2004 (2005) to 2015, the independent variables relate to 2002 to 2013 (2012). The sample for two-(three-)year lagged variables finally consists of 30,824 (22,539) firm-year observations ranging from 2004 (2005) to 2015.
Table 6 presents the regression results using OLS regression analysis with random and time fixed effects as well as robust standard errors.
The result in column 1 of table 6 shows that the statistically highly significant and positive impact of DPt-1 on TOBINt remains and is complemented by a statistical significant impact of DPt-2 on TOBINt, on a 90% confidence level. However, adding DPt-3 in column 3 makes this effect vanish. This implies, that the effect of DP on future firm value remains for two years. Subsequently, this is in line hypothesis 1 and 2a, while providing additional information about the two-year horizon. Furthermore, these findings are in line with the notion market prediction span approximately two years (Fama and French, 1998).
32
33
Table 6. Dividend payout and future firm value - additional analysis
(1) (2) (3) (4)
34 4.5 Robustness test
As discussed further above, China, Japan, and India account for the most firm year observations in the sample with regard to TOBIN as well as DP. For robustness purposes the entire sample is split into these three countries on the one hand and the rest of the world on the other hand. This shall clarify whether the results are driven by these countries. The results in table 7 show that the findings discussed before are not driven by those countries that account for the most observations in the sample, namely China, Japan, and India, as even without them the results remain robust.
The beforehand discussed dividend payout development (please refer back to figure 1), shows clear DP differences leading up to and during the financial crisis period, commonly referred to having taken place between 2007 and 2008. However, it can be considered general knowledge that the effective financial crisis period was different from nation to nation. Consequently, the subsample range is decided upon based on figure 1: 2003 to 2005, from 2006 to 2009, and 2010 to 2016. The regression results can be seen in table 8. The regression results in table 8 clearly display that the findings discussed in section 4 are driven by a development starting with the crisis period and increasing in statistical significance in the period post-crisis, herein defined as period from 2010 to 2016. In contrast, the pre-crisis period up shows no statistical significance.
35
Table 7. Dividend payout and firm value - regional subsample analysis All- (CHN+JPN+IND) CHN+JPN+IND
(1) (2) TOBINt TOBINt DPt-1 11.96*** 45.21*** (3.81) (5.29) DPt 3.087 -21.50** (0.90) (-2.59) DPt-2 1.691 -11.17*** (1.17) (-5.14) dDPt-1 13.71*** 52.67*** (3.65) (5.53) dDPt+1 -8.481*** -17.10*** (-4.42) (-6.30) dATt-1 -0.0964 -0.828* (-0.77) (-2.03) dATt+1 -0.431** -0.311 (-3.01) (-1.80) Et 1.520** -3.482*** (3.24) (-3.57) dEt-1 0.0255 2.874*** (0.08) (3.33) dEt+1 -0.368 -1.910** (-0.75) (-2.69) INVt -0.543*** -0.230 (-6.48) (-1.74) dINVt-1 0.0737 1.387* (0.46) (2.51) dINVt+1 -0.412* -1.223*** (-2.40) (-5.14) LEVt 0.0588 -0.689*** (0.41) (-3.36) dLEVt-1 -0.0265 -0.349 (-0.11) (-0.99) dLEVt+1 0.449 0.843** (1.53) (3.19) INTEXt 6.610** 9.000 (3.19) (1.46) dINTEXt-1 -2.360 -19.40* (-0.81) (-2.44) dINTEXt+1 -2.428 -2.441 (-0.77) (-0.31) GDPt 0.0554 0.284* (0.81) (1.99) INTRt -0.00405 -0.380*** (-0.36) (-5.47) INFLt -0.0218* 0.122*** (-2.32) (6.93) WGIt -0.165 -1.265*** (-1.79) (-9.75) DPt*WGIt 2.367 -4.339 (0.86) (-0.40) DPt-1*WGIt -1.847 13.19 (-0.72) (1.13) DPt-2*WGIt 0.259 3.280 (0.21) (1.11) dDPt-1*WGIt -2.873 21.00 (-0.92) (1.77) dDPt+1*WGIt -1.080 -21.29*** (-0.65) (-5.70) Constant 0.514 -0.368 (0.88) (-0.30) Observations 13,498 17,326 Adj. R2 0.242 0.349
36
Table 8. Dividend payment on firm value - periodical subsamples
(2003-2005) (2006-2009) (2010-2015) (2003-2005) +(2010-2015)
(2003-2009) (2006-2015)
(1) (2) (3) (4) (5) (6)
TOBINt TOBINt TOBINt TOBINt TOBINt TOBINt DPt-1 6.621 12.91** 16.32*** 13.92*** 10.17*** 14.69*** (1.85) (2.99) (4.98) (5.14) (3.45) (5.48) DPt 11.64** -2.539 -2.994 0.425 4.062 -2.214 (2.69) (-0.54) (-0.85) (0.15) (1.22) (-0.77) dDPt-1 2.692 12.33* 17.45*** 14.57*** 8.418* 15.11*** (0.58) (2.37) (4.51) (4.45) (2.34) (4.72) dDPt+1 -8.673*** -4.904 -7.143*** -7.960*** -7.834*** -6.541*** (-5.67) (-1.45) (-4.63) (-6.71) (-5.55) (-4.72) dATt-1 -0.213 -0.618 -0.333 -0.273 -0.428 -0.344 (-1.21) (-1.29) (-1.31) (-1.33) (-1.78) (-1.47) dATt+1 -0.397** 0.507 -0.452*** -0.480*** -0.0158 -0.290 (-2.60) (0.84) (-3.52) (-4.56) (-0.05) (-1.73) Et -3.325*** -0.0614 -0.195 -1.033* -1.987** -0.0988 (-3.45) (-0.08) (-0.31) (-1.97) (-2.84) (-0.18) dEt-1 1.981** 1.908 1.523** 1.626*** 1.958** 1.602*** (2.83) (1.91) (2.90) (3.79) (3.22) (3.34) dEt+1 1.148 -2.328** -1.583*** -0.904* -0.199 -1.741*** (1.78) (-3.05) (-3.51) (-2.37) (-0.38) (-4.30) INVt -0.512*** -0.769** -0.408*** -0.416*** -0.617*** -0.457*** (-4.23) (-3.21) (-4.23) (-4.93) (-5.01) (-5.09) dINVt-1 0.307 0.357 0.995** 0.790** 0.327 0.836** (1.35) (0.47) (2.91) (2.85) (0.88) (2.60) dINVt+1 -0.515* -1.821* -0.924*** -0.831*** -1.150** -1.089*** (-2.17) (-2.20) (-5.39) (-5.81) (-2.66) (-4.74) LEVt -0.899*** -1.569* -0.653*** -0.737*** -1.439*** -0.783*** (-3.61) (-2.13) (-4.78) (-6.00) (-3.44) (-5.57) dLEVt-1 0.248 0.947 -0.117 -0.143 0.673 0.0692 (0.66) (1.53) (-0.48) (-0.70) (1.78) (0.31) dLEVt+1 0.477 0.399 0.893*** 0.897*** 0.570* 0.787*** (1.42) (0.86) (4.43) (5.14) (1.96) (4.19) INTEXt 18.16*** 27.32** 9.850*** 11.99*** 25.58*** 12.62*** (3.84) (2.67) (3.75) (5.13) (3.99) (4.88) dINTEXt-1 -3.346 -7.931 -9.851** -7.930* -7.557 -9.556** (-0.61) (-1.20) (-2.78) (-2.55) (-1.69) (-2.98) dINTEXt+1 7.647 -0.180 -7.472 -5.172 1.084 -5.489 (1.68) (-0.02) (-1.81) (-1.52) (0.22) (-1.53) GDPt 0.0926 0.289* 0.177** 0.255*** 0.383*** 0.205*** (1.13) (2.30) (2.64) (5.31) (4.92) (3.75) INTRt -0.0822*** 0.00224 -0.0995*** -0.0898*** -0.00704 -0.0756*** (-3.76) (0.12) (-9.28) (-9.45) (-0.49) (-8.38) INFLt 0.0307* -0.0935*** 0.0207* 0.0277*** -0.0640*** -0.000191 (2.25) (-4.17) (2.39) (3.98) (-4.26) (-0.02) WGIt -0.450*** -0.909*** -0.864*** -0.903*** -0.952*** -0.880*** (-3.46) (-5.18) (-13.57) (-17.66) (-7.99) (-15.91) DPt*WGIt 1.189 9.394* 10.57*** 8.126** 5.777* 10.23*** (0.32) (2.39) (3.45) (3.24) (2.16) (4.27) DPt-1*WGIt 6.514 -2.544 -3.621 -0.631 0.866 -3.215 (1.95) (-0.71) (-1.27) (-0.27) (0.36) (-1.44) dDPt-1*WGIt 3.261 -6.062 -5.199 -3.023 -2.149 -5.950* (0.75) (-1.57) (-1.54) (-1.06) (-0.73) (-2.24) dDPt+1*WGIt -0.845 -2.366 -4.803** -3.623** -1.519 -3.745** (-0.47) (-0.84) (-3.12) (-2.99) (-1.10) (-2.87) Constant 1.043 -0.595 0.472 -0.210 -0.535 0.0895 (1.52) (-0.53) (0.73) (-0.53) (-0.74) (0.18)
Year FE Yes Yes Yes Yes Yes Yes
Industry FE Yes Yes Yes Yes Yes Yes
37
Table 9. Dividend payment on firm value - independent variable proxied
(1) (2) TOBINt TOBINt DP_PROXt-1 32.92*** -6.525*** (4.36) (-4.41) DP_PROXt -23.16** -1.829 (-2.83) (-1.21) dDP_PROXt-1 40.37*** -4.812* (4.78) (-2.00) dDP_PROXt+1 -9.653** -1.285 (-2.91) (-0.72) dATt-1 -1.462** -0.532* (-3.06) (-2.33) dATt+1 -0.0450 -0.233 (-0.14) (-1.63) Et -5.633*** 3.488*** (-5.40) (8.38) dEt-1 4.108*** 0.803 (4.57) (1.83) dEt+1 0.141 -2.961*** (0.20) (-8.11) INVt 0.140 -0.449*** (0.66) (-4.98) dINVt-1 0.884 0.451 (1.33) (1.50) dINVt+1 -1.624*** -0.984*** (-3.65) (-4.96) LEVt -4.035*** -1.356*** (-8.08) (-9.21) dLEVt-1 1.665*** 0.717** (3.49) (3.25) dLEVt+1 1.547*** 0.667*** (4.47) (3.84) INTEXt 51.54*** 17.23*** (6.30) (6.84) dINTEXt-1 -32.79*** -10.77*** (-4.66) (-3.41) dINTEXt+1 -13.08 -4.735 (-1.87) (-1.41) GDPt -0.408 0.330*** (-1.39) (7.76) INTRt -0.208*** -0.0370*** (-4.02) (-4.45) INFLt 0.00906 0.0132 (0.22) (1.68) WGIt -0.0758 -0.823*** (-0.22) (-16.96) DP_PROXt*WGIt -0.272 2.204 (-0.03) (1.56) DP_PROXt-1*WGIt 9.788 5.672*** (1.10) (4.25) dDP_PROXt-1*WGIt 5.947 1.272 (0.69) (0.67) dDP_PROXt+1*WGIt -5.389 -0.0507 (-0.99) (-0.03) Constant 5.626* -0.743* (2.35) (-2.08) Observations 7,540 40,935 Adj. R2 0.328 0.174
38
5
Conclusion
While an immediate share price drop after dividend announcement and its effect on firm value has broadly been evidenced throughout research, the effects on a longer term have not yet been investigated. Literature, however, provides numerous theories (e.g. Signalling Theory, Agency Theory, Clientele Theory,… etc.) that allow to assume effects of dividend payout on firm value which go beyond this initial share price drop after announcement. Arguments for a positive as well as a negative relationship are easily constructed. In view of the economic importance of both firm value and dividend policy, further investigation of the relationship between these variables contributes to recent literature and is of interest to management and investors. The results of the underlying research are in line with the hypotheses that dividend payout has a statistically highly significant positive effect on the effective future firm value. This result maintains on a one- and two-year horizon and is driven by a development starting with the crisis period. The hypotheses that National Governance, proxied by the World Governance Index, may increase or decrease the impact dividend payout has on effective future firm value is further confirmed when accounting for a horizon beyond one year.
This evidence can be interpreted as strongly consistent with the Signalling Theory; dividends convey information to the general public which impacts the next two years firm value in a different manner than the level of earnings and similar variables. The increasing significance of this finding after the year 2006 can furthermore be considered highly consistent with the Agency Theory. The general public seems to value the effect of DP on future firm value even stronger after the occurrences of the financial crisis. It seems natural to account this observation to the agency cost mitigating potential of DP. However, as the evidence provided herein does not rule out alternative explanations, these interpretations remain solely suggestive.
39
effective firm value, which has hitherto been assumed but not evidenced. Secondly, the underlying results strengthen the conclusions of prior research stating that dividends convey information missed by other variables in the way, that it also holds, when accounting for a deferred time axis. The results are also an important take-away for the manager who decides on the firm's payout policy and is interested in the magnitude of the effects of his/her decisions. Following the results presented above managers should cash out free funds to the firm's investors as means of public signalling. Last but not least, investors are well advised to invest into dividend paying stocks as results show that those company's firm value tends to be higher and the effect of dividends on firm value exists not only on payout day but also on a two-year horizon.
40
6
References
Allen, F., Bernardo, A. E., Welch, I., 2000. A theory of dividends based on tax clienteles. The Journal of Finance 55(6), 2499-2536.
Becht, M., Franks, J., Mayer, C., Rossi, S., 2008. Returns to shareholder activism: Evidence from a clinical study of the Hermes UK Focus Fund. The Review of Financial Studies 22(8), 3093-3129.
Bodie, Z. 1976. Common stocks as a hedge against inflation. The Journal of Finance 31(2), 459-470.
Booth, L., Aivazian, V., Demirguc-Kunt, A., Maksimovic, V., 2001. Capital structures in developing countries. The Journal of Finance 56, 87-130.
Brav, A., Graham, J. R., Harvey, C. R., Michaely, R., 2005. Payout policy in the 21st century. Journal of Financial Economics 77(3), 483-527.
Brooks, C. 2014. Introductory Econometrics For Finance. Cambridge University Press, Cambridge.
Claessens, S., Kose, M. M. A., 2013. Financial crises explanations, types, and implications. Unpublished working paper. The Australian National University.
DeAngelo, H., DeAngelo, L., 1990. Dividend policy and financial distress: An empirical investigation of troubled NYSE firms. The Journal of Finance 45(5), 1415-1431.
Dittmar, A., 2008. Corporate cash policy and how to manage it with stock repurchases. Journal of Applied Corporate Finance 20(3), 22-34.
Durnev, A., Kim, E. H. A. N., 2005. To steal or not to steal: Firm attributes, legal environment, and valuation. The Journal of Finance 60(3), 1461-1493.
Eades, K. M., 1982. Empirical evidence on dividends as a signal of firm value. Journal of Financial and Quantitative Analysis 17(4), 471-500.
Easterbrook, F. H., 1984. Two agency-cost explanations of dividends. The American Economic Review 74(4), 650-659.
El Ghoul, S., Guedhami, O., Kim, Y., 2017. Country-level institutions, firm value, and the role of corporate social responsibility initiatives. Journal of International Business Studies 48(3), 360-385.
Fama, E. F., French, K. R., 1998. Taxes, financing decisions, and firm value. The Journal of Finance 53(3), 819-843.
Fama, E. F., French, K. R., 2001. Disappearing dividends: changing firm characteristics or lower propensity to pay?. Journal of Financial Economics 60(1), 3-43.
Fama, E. F., Schwert, G. W., 1977. Asset returns and inflation. Journal of Financial Economics 5(2), 115-146.
Fisman, R. Love, I., 2003. Trade credit, financial intermediary development, and industry growth. The Journal of Finance 58(1), 353-374.
Floyd, E., Li, N., Skinner, D. J., 2015. Payout policy through the financial crisis: The growth of repurchases and the resilience of dividends. Journal of Financial Economics 118(2), 299-316.
41
Fredrickson, J. W., Mitchell, T. R., 1984. Strategic decision processes: Comprehensiveness and performance in an industry with an unstable environment. Academy of Management Journal 27(2), 399-423.
Gao, W., Ng, L., Wang, Q., 2011. Does corporate headquarters location matter for firm capital structure?. Financial Management 40(1), 113-138.
Goll, I. Rasheed, A. M., 1997. Rational decision-making and firm performance: The moderating role of environment. Strategic Management Journal, 583-591.
Hagendorff, J., 2014. Corporate Governance in Banking. In: Berger, A. N., Molyneux, P., Wilson, J. O. S. (Eds.). The Oxford Handbook of Banking. Second Edition. Oxford University Press, New York, pp. 139-159.
Hong, H., 1977. Inflation and the market value of the firm: Theory and tests. The Journal of Finance 32(4), 1031-1048.
Huston, J. L. 2015. The Declaration of Dependence: Dividends in the Twenty-First Century. Archway Publishing, Indiana.
Isakov, D., Weisskopf, J. P., 2015. Pay-out policies in founding family firms. Journal of Corporate Finance, 33, 330-344.
Jensen, M. C., 1986. Agency costs of free cash flow, corporate finance, and takeovers. The American Economic Review 76(2), 323-329.
Jensen, M. C., 2010. Value maximization, stakeholder theory, and the corporate objective function. Journal of Applied Corporate Finance 22(1), 32-42.
Kale, J. R., Kini, O., Payne, J. D., 2012. The dividend initiation decision of newly public firms: Some evidence on signaling with dividends. Journal of Financial and Quantitative Analysis 47(2), 365-396.
Kaufmann, D., Kraay, A., Mastruzzi, M., 2009. Governance matters VIII: Aggregate and individual governance indicators, 1996-2008. World Bank Policy Research, Working Paper No. 4978.
La Porta, R., Lopez de Silanes, F., Shleifer, A., Vishny, R. W., 1997. Legal determinants of external finance. The Journal of Finance 52(3), 1131-1150.
La Porta, R., Lopez-de-Silanes, F., Shleifer, A., Vishny, R. W., 1998. Law and finance. Journal of Political Economy 106(6), 1113-1155.
Lintner, J., 1956. Distribution of incomes of corporations among dividends, retained earnings, and taxes. The American Economic Review 46(2), 97-113.
Martins, T. C., Novaes, W., 2012. Mandatory dividend rules: Do they make it harder for firms to invest?. Journal of Corporate Finance 18(4), 953-967.
Michaely, R., Thaler, R. H., Womack, K. L., 1995. Price reactions to dividend initiations and omissions: Overreaction or drift?. The Journal of Finance 50(2), 573-608.
Michel, A., 1979. Industry influence on dividend policy. Financial Management, 22-26. Miller, M. H., Modigliani, F., 1961. Dividend policy, growth, and the valuation of shares. The
Journal of Business 34(4), 411-433.
Morck, R., Yeung, B., 2005. Dividend taxation and corporate governance. Journal of Economic Perspectives 19(3), 163-180.
42
Pinkowitz, L., Stulz, R. M., 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(6), 2725-2751.
Pinkowitz, L., Stulz, R. M., Williamson, R., 2015. Do US firms hold more cash than foreign firms do?. The Review of Financial Studies 29(2), 309-348.
Rajan, R. G., Zingales, L., 1995. What do we know about capital structure? Some evidence from international data. The Journal of Finance 50(5), 1421-1460.
Seifert, B., Gonenc, H., 2018. The effects of country and firm-level governance on cash management. Journal of International Financial Markets, Institutions and Money, 52, 1-16. Smart, C., Vertinsky, I., 1984. Strategy and the environment: A study of corporate responses to
crises. Strategic Management Journal 5(3), 199-213.
The World Bank Group, 2018. World governance indicators. http://info.worldbank.org/governance/WGI/#home ( access 05.08.2018).
43
7
Annex
Table 10. Country distribution and selected variables
All variables TOBINt DPt WGIt
N percent N mean N mean N mean
44
Table 10 continues
All variables TOBINt DPt WGIt
N percent N mean N mean N mean
45
Table 10 continues
All variables TOBINt DPt WGIt
N percent N mean N mean N mean
