THE EFFECT OF COUNTRY CHARACTERISTICS ON THE TIME BETWEEN AN IPO AND THE FIRST DIVIDEND PAYMENT.

* Corresponding author. E-mail address: j.b.m.sassen@student.rug.nl (Jaap Sassen). Student number: s1998684 Key words: Initial public offering, Dividend policy, Institutional factors, Substitution model

THE EFFECT OF COUNTRY CHARACTERISTICS ON

THE TIME BETWEEN AN IPO AND THE FIRST

DIVIDEND PAYMENT.

Abstract

This research paper makes an attempt at closing the gap in literature by testing agency considerations as an influencing factor for dividend policy of IPO firms. Dividend data of 642 firms from seven countries are used to make statistical interferences about the time it takes an IPO to issue dividend, influenced by different institutional factors of countries. An OLS is used to make these interferences. The results presented here indicate that there is a robust effect of legal enforcement on the time it takes before an IPO firm in a certain country pays its first dividend.

J.B.M Sassen1

2

TABLE OF CONTENTS

1. INTRODUCTION ... 3

2. LITERATURE REVIEW ... 4

2.4 Legal enforcement, outside investor rights, taxes and GDP growth ... 7

2.5 Empirical findings ... 9

2.6 Hypotheses ... 11

3. METHODOLOGY AND VARIABLES ... 12

3.1 Cross-sectional data ... 12

3.2 Variable definitions and relations ... 13

3.2.1 Dependent variable ... 13

3.2.2 Independent variables ... 13

3.3.2 Control variables ... 14

4. SAMPLE, DATA AND DESCRIPTION ... 15

4.1 Sample selection ... 15 4.2 Data measurement ... 17 4.3 Descriptive statistics ... 19

5. EMPIRICAL RESULTS ... 22

6. CONCLUSION AND DISCUSSION ... 26

7. REFERENCES ... 28

8. APPENDIX ... 32

3

1. INTRODUCTION

Dividend payments have preoccupied researchers from the time the work of Miller and Modigliani (1961) was first published. They stated that in a world without taxes the dividend pay-out policy of a firm has no effect on shareholders‟ wealth when the investment policy of a firm is held constant. Lintner (1956) has created a model that predicts 85% percent of the change in dividend policy. His proposition, that the size of dividends is mostly determined by the pay-outs being in line with the earnings of the company without making drastic changes to the dividends paid but with small periodic adjustments to a pay-out level the firm can sustain, is widely supported. However, the most drastic change in dividend payments is paying a first dividend. It is interesting to notice the level of disagreement between researchers about the effect several factors have on the dividend policy of firms. This research paper aims to contribute to understanding the influence of several institutional factors (i.e. outside investor rights, legal enforcement and tax rates) on the difference between countries in terms of the time it takes an IPO (initial public offering) firm to pay its first dividend.

There are still no definitive answers to questions about informational functions of dividend, the influence of taxes and behavioral explanations (Frankfurter and Wood, 2002). Lintner (1956) theorized that there are several factors that influence dividend payments. These factors include agency costs (e.g Easterbrook, 1984), asymmetric information (e.g Miller and Rock (1985), taxes (e.g. Baker, Veit & Powell, 2001) and behavioral theories (e.g. Mitton, 2004). An only recently studied and promising factor to explain dividend policy of firms is agency considerations. La Porta et al. (2000) state that there is a two-sided theory that can explain the dividend policy of a firm. On the one hand there is the outcome model of dividends, which assumes that in countries where investors are well-protected, dividend payments are higher and more common because investors have the power to force pay-outs. It would follow that investors that are weakly protected by country institutions, do not have the power to force management to pay dividend and pay-outs will be lower and more seldom. On the other hand, there is the substitute model of dividends that assumes that in countries where investors are weakly protected, the firms need to signal their good intention with high dividends because paying dividends levitates the chance of expropriation by insiders. Dividend pay-outs will be more likely to occur for that reason. When outside investors are well protected by country institutions there is no need for management to restrict their resources by management and dividends will occur less in the country of subject.

4 Any explanation for dividends as the one from La Porta et al. (2000), should become more evident when considering dividend payments of firms that are not assumed to be in the position to pay dividends, but still do so. Such firms are to be found in IPO-firms, which are known to be young and to have lucrative investment opportunities. Another inconceivable wide gap in literature, is to be found between researches about dividend payment and studies about IPO firms. When connecting the two-sided theory of La Porta et al. (2000) and the notion that IPO-firms should have a strong motive to pay dividends when they do so (e.g. because of their growth opportunities), the following question emerges: do the rights that outside investors are subjected to in a country and the level of legal enforcement in a country, influence the time an firm takes to pay out its first dividend? However, those two country institutions are not the only factors influencing dividend policy. Taxes, as found by Masulis and Trueman (1988), are also an incentive for outside investors and firms to have a preference for either retained earnings or for dividend pay-outs. In the literature sought this factor seemingly lacks attention, even more so in combination with the dividend policy of IPO firms. That is why the following and concurrent research question is appropriate: does the difference between capital gains tax rate and the rate dividend pay-outs are taxed against in a country, have an influence on the time it takes an IPO to pay its first dividend?

For the purpose of answering the questions stated above, data from 642 firms in seven different countries has been collected. It focuses on the years between IPO and a first dividend pay-out. This is combined with data on outside investor rights, legal enforcement and difference in tax rates of those countries, to perform an OLS (ordinary least squares) regression. Several control variables are included. This methodology results in an assessment of the differences in time between an IPO and the first dividend payment between the seven countries. The remainder of this paper is organized as follows: section two gives a comprehensive review of relevant theories and empirical results in literature. Expectations about findings are stated. Section three presents the methodology. The fourth section describes the data collection procedure and the data itself. The results are presented in section five where after this research is concluded and discussed in the last section.

2. LITERATURE REVIEW

5 2.1 Signalling Theory

Not only managers of (new) public firms pay a lot of attention to their dividend pay-outs. Corporate dividend policy is a recurring subject for researchers in managerial finance. Miller and Modigliani (1961) are still relevant because they argued that cash flows from dividend are irrelevant. They showed that an investor can create any income pattern by buying and selling shares and that the expected return is invariant to dividend payments. They stand at the base of a thorny discussion between researchers that is still topical after more than six decades. Even though the pay-out of dividend in a world without transaction costs and taxes is deemed irrelevant, they established the notion that dividends have informational value within them. Less informed market participants are apparently signalled information about future company cash flows by managers through changes in the company‟s dividend policy. The intuition behind this is simple: in the real world there are costs attached to paying a dividend (e.g. transaction costs of issuing new common equity and regulatory considerations) that only firms of quality expecting high future cash flows can afford to pay (Bhattacharya, 1979). Bhattacharya (1979) constructed a model that confirms his „bird in the hand fallacy‟ with two results. The first result is that the shorter the horizon of an investor, the greater his urgency to be enriched by cash means. The second result is that lengthening the horizon of the investor diminishes the importance of the share value at the end of that horizon as opposed to periodic payments towards the horizon. With their equilibrium model, John and Williams (1985) agree with the signalling hypothesis. They show that firms that pay dividends have a clientele of investors that are „cash-demanding‟. Miller and Rock (1985) confirm with their model that the companies‟ dividend pay-outs can be manipulative information-signals. For example, a mediocre or bad performing company can pay costly dividends to deceive investors.

By reviewing these theories it seems clear; dividends are paid to signal information to investors. This signal can be about future company cash flows, to attract investors in general, or about the level of dividends these companies will be paying for some time. This attracts a certain group of investors that generally seek companies that will provide them with cash pay-outs. It can be expected that firms use dividend payments to signal potential investors that their firm is a prospering firm, one that provides investors with a positive return. The opposite is valid too, meaning that a declining level of dividends or even elimination of dividend is a sign that diminishing returns may be expected. 2.2 Agency Theory

6 incur the monitoring of capital markets. The debt theories from Jensen (1986) and Hart and Moore (1994) are built on the idea that dividend policies address agency problems between outside shareholders and corporate insiders. This is done by preventing the insiders to invest in unprofitable projects accompanied by private benefits or to make personal use of the resources. When a firm makes its first dividend payment, it is the first restriction of insiders. In this sense, corporate governance is a promising tool as to why firms would start to pay cash dividends to shareholders.

La Porta et al. (2000) offer two hypotheses that link corporate governance of a country and the dividend policy of a firm. According to the outcome model of dividends, the dividend pay-outs of firms should have a positive relation with the corporate governance of a country. In a well governed country, the investors should have stronger protection rights. This is accompanied with greater power and therefore the investors should be able to force the firm to pay dividends instead of giving the managers the chance to use the excess cash for their private benefits (La Porta et al., 2000).

The substitution hypothesis of dividends sees dividends as a substitute for legal protection. In countries with weak investor protection, the firms need to signal their good reputation with high dividends because paying dividends levitates the chance of expropriation by insiders (La Porta et al., 2000). With the signalling of good intention by the insiders, investors will not doubt if their investments are put to the best use possible.

2.3 IPO’s

7 the resources. However, IPO firms that pay dividends do exist, in surprisingly great numbers. Theoretical motives (i.e. signalling, agency considerations and clientele) to pay dividend for an IPO firm should be the same as for every other firm. It can be assumed that an IPO firm that starts to pay dividends has a strong motive to do so.

2.4 Legal enforcement, outside investor rights, taxes and GDP growth

. There are two broad legal traditions from where laws of a country stem. The legal system of a country incrementally develops without losing this traditional essence. These are: common law (originating from the Anglo-Saxon Empire) and civil law (originating from the Roman Empire). After country specific revisions, the influence of the two traditions are still reflected in the legal system as it is today (Watson, 1974). The different types of law are important indicators for the rights of outside investors and the enforcement and protection of these rights, but not a conclusive relation (La Porta et al., 1998). The quality of outside investor rights and legal enforcement is in its turn seen as a proxy for lower agency costs (La Porta et al., 2000). For example, ownership is more concentrated in countries with low outside investor protection (La Porta et al., 1998; La Porta et al., 1999). The capital markets are also more developed in countries with better investor protection (La Porta et al., 1997). The likelihood of expropriation by insiders is significantly bigger in countries characterized by low levels of investor protection (La Porta et al., 2000). This means that as a shareholder of a company issuing its shares in The Netherlands (civil law tradition), you do not have the same rights as a shareholder of a company issuing shares in the United Kingdom (common law tradition). When country governance functions as a substitute for corporate governance (Chen, Chen and Wei, 2009), the chance of getting your investment back as an investor, either in returns or in case of bankruptcy, should vary per country. According to Brockman, Tresl and Unlu (2014), weak insider trading laws are a result of a country‟s weak legal institutions. Managers spend less time and cash flow on establishing a reputation of reliability when the country-level legal institutions serve this function. According to Mitton (2004), outside shareholders might have a preference for dividends over retained earnings. This preference may be even stronger in emerging markets with weak investor protection, if shareholders perceive a greater risk of expropriation by insiders in such countries (Mitton, 2004). It may therefore be expected, that firms issuing shares in countries that are from a tradition of weak shareholder rights should oppose these country-level institutions with strong firm-level governance to attract investors. Hence, dividend policies do differ per country.

8 before an IPO company pays out dividends. In countries with superior legal systems, shareholders can make the corporate insiders pay out dividends at any time, so as soon and as high as possible. But this can be different with IPO firms. These firms are expected to have high growth opportunities, which need resources invested in them. In that way, a vote between shareholders to make an IPO firm pay out dividends will not hold. It will take firms in better governed countries longer to pay out dividends, since investors are relatively certain that their investments will not be misused. But should shareholders decide that the time of dividend payments has come, the first cash payment will not take long.

When the substitution model of dividends is the prevailing motivation, firms in countries with strong legal institutions will not have to pay dividends. Investors do not need confirmation of the good governance of the corporation, because the governance of the country puts them in a relatively safe position. This should be different in weakly governed countries. To get investors assured their invested resources will be put to good use, the managers have to bind themselves by transferring resources back to their investors. The first payment is expected to be made fairly soon after the IPO of a company to win the trust of the outside investors. In other words, the firm is expected to adhere to an effective commitment system. Dividend payments are indeed an effective commitment system since it is known that the direction of dividend changes is positively related to the returns of common stock on the announcement day (e.g. Woolridge, 1983). When a firm starts paying dividends, the only way without shrinkage of equity is up: more and higher dividend payments. Because of the different institutional factors in countries, there is a need for a differentiating commitment system. That is why the time from IPO until the first dividend payment of firms should vary per country.

9 Another determinant of the willingness of a firm to pay its first dividend is the economic growth rate of a country. The reason the growth rates are important, is that economic growth drives aggregate spending. Firms will experience higher sales growth and financial constraints (Higgins, 1981), assuming the sample firms are representative for economic growth in a country. But, when an economy grows, it also provides firms with more growth opportunities. A firm needs more capital to finance their investments, on top of their increased sales. Since dividend payments reduce the resources for firms at hand to invest in value-adding projects provided by the growth opportunities, it would be wise for a firm to hold its first dividend payment (Lintner, 1956).

2.5 Empirical findings

The vast majority of researchers confirms that markets respond positively to changes deemed positive for investors in the dividend policy of a company, whilst in this research paper dividend payments are seen as a signalling tool. Akhigbe and Madura (1996) find a significant difference in future performance of US-firms. On the one hand are firms that initiate dividends (in the long run higher growth, higher capital investment and higher earnings) and on the other hand are firms that omit dividends (in the long run lower growth, lower capital investment and lower earnings). Bajaj and Vijh (1990) confirm the positive market reactions on dividend payments, and that result is accompanied by a clientele effect. They find that the average price reaction to a change in high-yield dividends is greater than to low-yield dividends. Akhigbe and Madura (1996) and Bajaj and Vijh (1990) agree on positive market reactions towards dividend payments.

There are, however, researchers who have doubts about the information link between dividends and earnings. This link was suggested in theoretical signalling models (see Bhattacharya, 1979). Benartzi, Michaely and Thaler (1997) were unable to find any evidence for dividends being a prognosis for future earnings. This is confirmed by Grullon et al. (2005), after reacting to research by Nissim and Ziv (2001), who stated that dividend payments are informative. Li and Zhao (2008) found empirical evidence for a negative relationship between information asymmetry and: one, the number of dividend payments; two, the time it takes a firm to start paying dividends; three, the height of dividends. To summarize, it is evidently not ruled out that dividend payments signal something, but the distribution of cash as a prognostication for future earnings is contradicted by empirical research (Grullon et al., 2005; Li and Zhao, 2008).

10 Wurgler (2004) find evidence that non-dividend payers tend to initiate dividend payments when demand for dividends is high, and that dividend payers omit dividends when demand for dividends is low. It can therefore be expected, that firms that pay dividends, do so to attract certain investors or even investors at all.

It is important to note that dividend policy in itself differs per country. Denis and Osobov (2008) find that some determinants of paying dividend, i.e. size and growth opportunities, are the same in six countries (United States, UK, Canada, Germany, France and Japan). But when testing the signalling theory, the results are mixed among countries. Other studies show that dividend policy is affected by formal institutions, such as investor protection (La Porta et al., 2000), legal enforcement (Brockman and Unlu, 2009) and tax rates (Faccio, Lang and Young, 2001). The manner, in which managers allocate resources, as well as how they wield corporate governance, is facilitated by governance mechanisms of countries.

11 After testing the agency considerations empirically, La Porta et al. (2000) and Mitton (2004) support the outcome model of dividends. However, these papers look at the pay-out ratio. The research of Adjaoud and Ben-Amar (2010) confirms the outcome model even though they restrict their analysis to Canada and firm-level governance. The substitution model of dividends in the sense of pay-out ratio is supported by Aivazian, Booth and Cleary (2003), Lin and Shen (2012). Athari, Adaoglu and Bektas (2016) support the substitution hypothesis after empirical research as well. In general, the absence of research that links the first payment of dividend to the agency theory is noticeable. The above combination of agency and signalling theory should better explain dividend policy than either theory alone.

2.6 Hypotheses

It is clear that IPO firms do pay dividends, sometimes even quite early in their life-cycle. It has likewise been found that firms that pay dividend are signalling some information, however not about future cash flows. A plausible answer to what IPO firms are trying to tell investors with their dividend payments, can be found in the agency theory. By paying dividends, the firm tries to substitute for weak investor protection and to signal its quality to potential shareholders. This is because a rational investor can perceive a higher risk of expropriation in an organisation with weak corporate governance in a weakly governed country. The investor attaches a lower value to its stock and is therefore less willing to supply capital. When this is combined with the fact that firms that pay dividends are also aware of their clientele and the demands of their clientele, the subsequent hypotheses are:

H1a: Firms pay dividend sooner after their IPO in countries that are characterized with weaker outside investor rights than firms in countries that are characterized with stronger outside investor rights.

H1b: Firms pay dividend sooner after their IPO in countries that are characterized with weaker legal enforcement than firms in countries that are characterized with stronger legal enforcement.

12 H2: Firms pay dividend sooner after their IPO in countries where personal income is taxed less heavily in comparison to capital gains, than firms in countries where personal income is taxed more heavily in comparison to capital gains.

3. METHODOLOGY AND VARIABLES

To measure the effects of legal systems and other variables on the time it takes an IPO firm to pay out dividend for the first time in various economies, an OLS regression model is constructed. The hypothesized relations are tested using cross-sectional data. Cross-sectional data contains observations on multiple entities. This research paper observes the time until the first dividend pay-out of 643 IPO firms, divided over seven countries, within 15 years. The simplest way to deal with this data is with a cross-section regression model. It is specified as follows:

yi = α + β‟X + ui (3.1)

where

yi = Time between an IPO and the first dividend payment for firm i α = the constant of the regression

β = the coefficient of the explanatory variable X = vector of explanatory variables

and ui = random disturbance term

The vector of the explanatory variables X contains the outside investor rights, legal enforcement, difference between personal tax rates and the tax rates of capital gains, GDP growth and the firm-level control variables on dividend payment, stemming from other research. A dummy variable is hereby used for industries that are known to pay out higher and more dividends and industries that are known to pay out fewer and lower dividends. This dummy takes the value of respectively 1 and 0. This is to see if the firm‟s respective industry has an effect on dividend payment too:

13 After an extensive analysis of the available literature, there is no reason to expect interaction effects between variables, except between legal enforcement and growth. Knack and Keefer (1995) state in their study that legal enforcement can influence the economic growth rate of a country. This will carefully be checked in section 4.2. Every other variable has its own relation with the time between an IPO and the first dividend payment, without being enforced or weakened by another variable present in the regression.

3.2 Variable definitions and relations

3.2.1 Dependent variable

The dependent variable is the time between an IPO and the first dividend payment. The dividend pay-out of an IPO company has not often been the subject of studies regarding dividend. The time between an IPO and the first dividend payment should vary for firms in different countries because of different country-specific forces at work. The time between a completed transaction (i.e. an initial public offering) and the payment date of the first cash income dividend payment is under influence of the variables described in the following subsections. Table 1 gives an overview of the expected relation of the independent variables with the dependent variable.

Table 1: Expected relations of the independent variables and control variables with the dependent variable (years between an IPO and the first dividend payment).

Dependent variable

Expected relation

Outside investor rights +

Legal enforcement +

Difference between effective tax rates on capital gains and dividend +

GDP Growth +

Industry -

Size -

Note: expected relations are based on the hypothesised signs provided by the reviewed theory in section 2.4 and 2.5

3.2.2 Independent variables

Outside investor rights

14 strong outside investor rights limit the acquisition of private benefits and suboptimal investments by insiders, weak outside investor rights could create the need for a substituting corporate governance tool to attract investors. The relation between investor protection and amount of time it takes an IPO firm to pay dividend is therefore positive.

Legal enforcement

The quality of the legal enforcement of rules regarding investors influences the dependent variable the same way as outside investor rights do. Incorporated measures in the legal enforcement variable are the rule of law, the efficiency of the judicial system, risk of confiscation or nationalization by the government, and corruption in general (La Porta et al., 1998). In a sense this variable assesses how attractive it would be to do business as an outside investor in the rated country by looking at the quality of the legal enforcement. If a country scores low on this variable, there is, as described in section 2.4 and 2.5, a need for a substituting corporate governance tool to attract investors. Therefore the relation between investor protection and the amount of time it takes an IPO firm to pay dividend is positive.

Tax rates

The tax rate variable reflects the difference between the tax rate an investor is subjected to when he or she receives dividends, and the tax rate an investor is subjected to when he or she sells the shares after a period (often longer than a year) of holding and intrinsic growth. Technically, this is the difference between tax rates on income and tax rates on capital gains. Since tax rates on capital gains and income often differ per country, it is likely to have an effect on the time between an IPO and the first dividend payment between countries (see section 2.4 and 2.6). The bigger the difference between the often heavier income tax rate and the often lighter capital gains tax rate, the more reason an IPO has to hold their dividend payments. This implies a positive relationship between the difference in tax rates and the dependent variable.

3.3.2 Control variables

Firm-level effect: Size

15 Industry-level effect

Another control variable is industry. Research by Baker and Powell (1999), among others, suggests that dividend policies vary across industries. Firms in a particular industry tend to make the same decisions around resource allocation. Firms imitate direct competitors more often than other firms outside of their industry (Van Caneghem and Aerts, 2011). For example, utilities represent a high dividend pay-out industry, whereas firms in manufacturing have more moderate dividend pay-out ratios on average. Technology firms represent a low dividend paying industry. Cohen and Yagil (2010) found more common and higher dividend pay-outs in transportation, banking and retail industries as opposed to less common and lower dividend payments in communication, services and construction industries. A dummy variable that takes the value of 0 for firms operating in an industry that has lower and fewer dividend payments is used. Therefore a negative relationship between industry and the dependent variable is expected.

Country level effect: economic growth

Fazzari et al. (1988) argue that financially constrained firms are less likely to pay dividends. Sales growth, mostly due to economic growth, financially constrains firms. Since dividend payments also reduce the resources for firms at hand to invest in value-adding projects provided by the growth opportunities, it would be wise for a firm to hold its first dividend payment when experiencing country-wide economic growth. Therefore a positive relationship is expected between the GDP growth of a country and the time it takes a firm to pay a first dividend in such a country (Lintner, 1956).

4. SAMPLE, DATA AND DESCRIPTION

16 fact that regulators choose to force companies to pay dividends is in itself a sign of importance of agency considerations. The most plausible reason to do this is making sure the investors are not expropriated entirely and to encourage participation of investors in the equity market.

For purposes of scope and data availability, a timeframe has also been selected. The timespan for IPO‟s is from 01/01/2001 until 31/12/2011 and the first dividend payment corresponding with the IPO firm can be done from 01/01/2001 until 31/12/2015.

17 4.2 Data measurement

18 of one euro of profit paid as dividend to an investor is lower than the value of one euro of intrinsic growth. This is because the personal tax rate on dividend receipts is higher than the effective tax rate on capital gains. In this research paper, the countries fall into two broad categories of tax systems (King, 1977). First the classical system (i.e. Italy, India, Singapore and the UK) and second the imputation system (Netherlands, France and Canada). In the classical system, taxes for corporates and investors are separated. The company pays a flat rate on profit and the investor pays taxes over received dividend or over capital gains when selling shares after a period of holding. When subjected to the classical system, the difference between effective tax rates can be formulated as:

((((100 / (1-Tcorporate)) – 100 + Tdividend) / (100 / (1-Tcorporate)))*100) - ((((100 / (1-Tcorporate)) – 100 + Tcapitalgains) / (100 / (1-Tcorporate)))*100) 4.1

where: Tcorporate = corporate tax rate on profits

Tdividend = personal tax rate on distributed profits Tcapitalgains = personal tax rate on capital gains

In the imputation system, a part of the company‟s tax bill is imputed to the stockholders and is seen as a prepayment of personal income tax. When subjected to the imputation system, the difference between effective tax rates can be formulated as:

((((100 / (1-Tcorporate)) – 100 + Timputation) / (100 / (1-Tcorporate)))*100) - ((((100 / (1-Tcorporate)) – 100 + Tcapitalgains) / (100 / (1-Tcorporate)))*100) 4.2

where: Timputation = Tdividend with tax credit and imputation rate taken in account.

The preference for retaining earnings is expressed by subtracting the effective tax rate on capital gains from the effective tax rate on dividend pay-outs. See Appendix: Table A for an overview of corporate taxes and personal tax rates on capital gains and dividends.

The economic growth in the sense of GDP growth is found in The Global Economy database. The growth of the economy of a country is measured as the annual percentage growth rate of GDP at market prices based on constant local currency. For this study GDP growth is measured as the change in the annual GDP growth rate from the previous year, as the effects economic growth has on firms is

19 The size of the firms in the sample is retrieved from Datastream. Firm size is measured by the number of ordinary shares issued, multiplied by the share price (reported in US Dollars and converted from local currency at the exchange rates in effect at the moment of reporting). This way of measuring firm size is similar to a study of Ferris, Sen and Unlu (2009). The moment of measuring corresponds with the moment the firm made its first dividend payment, as this method is used by Denis and Osobov (2008). The variable size was skewed, so the logarithm of the variable is taken.

The industry variable is operationalized as a dummy by categorizing the industry in two categories. One category is characterized by more and higher dividend pay-outs; the other one is for low and fewer dividend payments. The dummy will respectively take the values 1 and 0. The theory and empirical proof for classifying the industries in categories comes from Cohen and Yagil (2010). The construction, energy, communications and technology sectors (0-1999 and 4700-4999 SIC-codes) fall in the low and few dividend pay-outs category. The retail and wholesale, services, manufacturing and transport sectors (SIC-codes 2000-4699, 5000-8999) fall in the high and more dividend pay-out category. As the industries where an operating firm is less likely to pay dividend are valued by 0 and the industries where dividend payments by firms are more common are valued by 1, the relation

between industry and the dependent variable is negative.

4.3 Descriptive statistics

20 discussed studies, for example from Lipson et al. (1998). Apart from the Netherlands, which only contains 8 firm observations, firms in France take the longest to issue dividend after their IPO. This is a sign of contradiction of H1a and H1b, since France scores relatively low on outside investor rights and legal enforcement compared to other countries in the sample. But on average the tax-rates in France are most adverse towards dividend payments. IPO firms in Italy pay dividend the soonest. The country scores indeed the lowest on outside investor rights and second to lowest on legal enforcement. The firms are also on average the largest in Italy. India knows tax rates that should create a preference for dividend payments, but the hypothesized effect on time between an IPO and the first dividend payment can be contradicted by the low scores on outside investor rights and legal enforcement, as well as by the highest mean economic growth. In Singapore, when looking at tax rates, investors should be indifferent towards profits either being paid as dividend or being retained. The difference between the effective tax rates on dividend payments and capital gains is 0.

21 their IPO and that a few other firms take a disproportionately time to do so. The maximum of preference for dividend payments of investors is found in India (-13.600). The maximum of preference for capital gains of investors in France (24.445).

To check the validity of the industry classification that is used in this research paper to construct the dummy variable, additional descriptive statistics are presented in Table 4: Panel C. The mean of the dependent variable is indeed lower (1.702 years) in industries that have higher and more dividend payments. In industries that have lower and fewer dividend payments it takes IPO firms on average 2.204 years to pay their first dividend.

The correlations between all variables are presented in Table 5. Legal enforcement correlates barely but significantly (p = 0.000) with the dependent variable. This sign is in line with hypothesis 1b and analysed literature (e.g. Aivazian et al., 2003). The indication of the relation between the difference in tax rates and the number of years it takes an IPO to pay its first dividend is also significant and in line with hypothesis 2. The indication of firm size has a negative effect (p = 0.039) on the dependent variable, which is consistent with findings of DeAngelo et al. (2006). Industry correlates, as was indicated by Table 4: Panel C, negatively with the time between an IPO and the first dividend payment. Outside investor rights have a positive relation with the dependent variable, however not significant. Economic growth correlates negatively with the dependent variable. This indication is not

22 have an effect in the interpretability of the regression output. The correlation matrix indicates that other correlations are significant, but these numbers on correlations are small and will not affect results.

5. EMPIRICAL RESULTS

In the following sections the regression results and several robustness tests are presented to identify the relations between the explanatory variables and the time between an IPO and the first dividend payment. First a White test is performed, testing for H0: no heteroskedasticity. This is done because standard errors could be inappropriate due to incorporation of several independent variables in the full model. The null hypothesis is rejected (Obs R2 _{= 141.52, Χ}2 _{= 0.000) and the OLS regressions are} therefore estimated with White's heteroskedasticity consistent standard errors. The effect of using White's correction is that in general the standard errors for the slope coefficients increase relative to the usual OLS standard errors. Consequently the hypotheses have to be tested more conservatively and need more evidence before drawing conclusions.

Four regression models are constructed to analyze the variables. The first regression (Table 6: OLS 1) is generated with only the explanatory variables. The influence of outside investor rights, legal enforcement and difference in tax rates on the time between an IPO and the first dividend is tested. The second model (Table 6: OLS 2) also includes the log of the market value of equity. The third model (Table 6: OLS 3) takes the economic growth into account next to the explanatory variables and the measure for firm size. In the fourth model (Table 6: OLS 4) the dummy variable industry is added. The null hypothesis that all independent coefficients that are tested are together equal to zero is rejected for each model. In OLS (1) the probability of the F-statistic is 0.051, in OLS (2) 0.012 and in OLS (3) and (4) it is 0.000. The probability F-statistic decreases and the adjusted R2 _{increases by} gradually adding control variables. This means that the fitness of the model improves after adding the control variables.

23 Table 6: OLS (4) presents the full model. Outside investor rights do not have a significant effect on the number of years it takes an IPO to pay dividends. Other researches show a significant effect of investor rights on dividend policy (Shleifer and Vishny, 1997; La Porta et al., 2000). An explanation for this inconclusive result can be economic and/or institutional heterogeneity between countries. Although this research paper controlled for economic differences by including GDP growth and firm size, variation in profitability or size of firms and ownership structures across countries potentially affects the results (Leuz, Nanda and Wysocki, 2003). This issue is examined in section 5.2 (robustness). The result for H1a remains inconclusive.

24 The hypothesized preference the two ways of profit-sharing are subject to, for either capital gains or dividend payments (determined by different tax rates), finds no empirical grounding. Several studies found a rise in dividend payments when tax conditions became more favourable towards dividend payments within a country (Wu, 1996; Grullon and Michaely, 2002). But on an inter-country level this research paper finds no evidence for such a relation. Remarkably, the OLS (4) even shows a slightly significant negative relation between dividend preference and the time between an IPO and the first dividend payment. H2 is therefore rejected, since there is no positive relation between tax rates differences on an inter-country level and dividend policy differences of firms.

The control variables all have a significant statistical effect on the dependent variable. Economic growth has a significant and negative effect (p = 0.009) level on the dependent variable. This is the opposite of the hypothesized relation. In the existing literature there is no evidence that GDP-growth tempts managers to return resources to the investors. An explanation could be that GDP growth of a country is not representative for the growth of the firms in the sample, as this research paper assumed in the first instance. The variable firm size has a negative effect on the years between an IPO and the first pay-out of dividends (p = 0.003), which is consistent with the results of e.g. DeAngelo et al. (2006) and Deshmukh (2003). The dummy variable Industry has a significant negative relationship (p = 0.000) with the time between an IPO and the first dividend payment. This effect is theorized to exist because firms imitate direct competitors more often than other firms outside their own industry (Van Caneghem and Aerts, 2011).

5.2 Robustness

25 dividend policy. The explanatory power of the OLS regression with the subsample also increases significantly. The independent variables now explain 10.3% of the variation of the dependent variable. Another robustness test is performed. Since the variable GDP-growth does not have the hypothesized effect on the time between an IPO and the first dividend payment, it is probably viable to measure this variable differently. This research paper assumed that the growth opportunities of firms provided by economic growth are lagged by a year. However, it could also be simultaneous (Altman et al., 2005). Therefore, the variable is re-entered in the regression model as the change of the GDP in the year of the dividend payment of a firm, instead of using the GDP-growth of the former year. The result is presented in Table 7: Alt_EC. The results are more indeterminate than the results from OLS (4), hence this paper refrains from elaborating on these results in further detail.

26 is done for the variable size where the top and bottom percentile of firms are excluded. A sample of 618 observations remains. This manner of sampling does not contribute to the explanatory power or significance of the OLS regression in comparison to Table 6: OLS (4), which is a confirmation of the appropriateness of the original sample distribution.

6. CONCLUSION AND DISCUSSION

Dividend policy has been a hotly debated topic for decades now. It is not hard to imagine that the on-going globalization in this era connects more and more investors with foreign financial markets. This is intertwined with a rise in studies around country level institutions and the consequences they have for stakeholders participating in the globalization. Nevertheless, relatively few researchers focus on firms that become available for investors when they make their initial public offering (IPO). This research paper was, therefore, aimed at contributing to the understanding of influence several institutional factors (i.e. outside investor rights, legal enforcement and tax rates) on the difference between countries in terms of the time it takes an IPO firm to pay its first dividend. Signalling theory, behavioural theories and agency theory have, up until today, failed to unambiguously explain the functions of dividend. This study uses a sample comprised of 642 IPO firms, divided over seven countries that paid their first dividend between 2001 and 2015. The cross-sectional data allowed for OLS-interferences and several robustness tests were performed.

27 payments at an inter-country level has a univocal influence on the time between an IPO and a first

dividend payment.

28

7. REFERENCES

Adjaoud, F., & Ben‐Amar, W. (2010). Corporate governance and dividend policy: shareholders‟ protection or expropriation? Journal of Business Finance & Accounting, 37(5‐6), 648-667. Aivazian, V., Booth, L., & Cleary, S. (2003). Dividend policy and the organization of capital

markets. Journal of Multinational Financial Management, 13(2), 101-121.

Akhigbe, A., & Madura, J. (1996). Dividend policy and corporate performance. Journal of Business Finance & Accounting, 23(9‐10), 1267-1287.

Altman, E. I., Brady, B., Resti, A., & Sironi, A. (2005). The link between default and recovery rates: Theory, empirical evidence, and implications. Journal of Business, 78(6), 2203-2228. Athari, S. A., Adaoglu, C., & Bektas, E. (2016). Investor protection and dividend policy: The case of

Islamic and conventional banks. Emerging Markets Review, 27, 100-117.

Bajaj, M., & Vijh, A. M. (1990). Dividend clienteles and the information content of dividend changes. Journal of Financial Economics, 26(2), 193-219.

Baker, H. K., Veit, E. T., & Powell, G. E. (2001). Factors influencing dividend policy decisions of Nasdaq firms. Financial Review, 36(3), 19-38.

Baker, H. K., & Powell, G. E. (1999). How corporate managers view dividend policy. Quarterly Journal of Business and Economics, 38(2), 17-35.

Baker, M., & Wurgler, J. (2004). A catering theory of dividends. The Journal of Finance, 59(3), 1125 1165.

Benartzi, S., Michaely, R., & Thaler, R. (1997). Do changes in dividends signal the future or the past? Journal of Finance, 52(3), 1007-1034.

Bhattacharya, S. (1979). Imperfect information, dividend policy, and “the bird in the hand” fallacy. Bell Journal of Economics, 10(1), 259-270.

Brockman, P., & Unlu, E. (2009). Dividend policy, creditor rights, and the agency costs of debt. Journal of Financial Economics, 92(2), 276-299.

Brockman, P., Tresl, J., & Unlu, E. (2014). The impact of insider trading laws on dividend payout policy. Journal of Corporate Finance, 29, 263-287.

Brooks, C., 2014: Introductory Econometric for Finance. Cambridge: Cambridge University Press. Chen, K. C., Chen, Z., & Wei, K. J. (2009). Legal protection of investors, corporate governance, and

the cost of equity capital. Journal of Corporate Finance, 15(3), 273-289.

Cohen, G., & Yagil, J. (2010). Sectorial differences in corporate financial behavior: an international survey. The European Journal of Finance, 16(3), 245-262.

DeAngelo, H., DeAngelo, L., & Stulz, R. M. (2006). Dividend policy and the earned/contributed capital mix: a test of the life-cycle theory. Journal of Financial economics, 81(2), 227-254. Denis, D. J., & Osobov, I. (2008). Why do firms pay dividends? International evidence on the

29 Deshmukh, S. (2003). Dividend initiations and asymmetric information: A hazard model. Financial

Review, 38(3), 351-368.

Easterbrook, F. H. (1984). Two agency-cost explanations of dividends. American Economic Review, 74(4), 650-659.

Elayan, F. A., Li, J., Donnelly, M. E., & Young, A. W. (2009). Changes to income trust taxation in Canada: investor reaction and dividend clientele theory. Journal of Business Finance & Accounting, 36(5‐6), 725-753.

Faccio, M., Lang, L. H., & Young, L. (2001). Dividends and expropriation. American Economic Review, 91(1), 54-78.

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.

Fazzari, S. M., Hubbard, R. G., Petersen, B. C., Blinder, A. S., & Poterba, J. M. (1988). Financing constraints and corporate investment. Brookings Papers on Economic Activity, 1, 141-206. Ferris, S. P., Sen, N., & Unlu, E. (2009). An international analysis of dividend payment behavior.

Journal of Business Finance & Accounting, 36(3‐4), 496-522.

Frankfurter, G. M., & Wood, B. G. (2002). Dividend policy theories and their empirical tests. International Review of Financial Analysis, 11(2), 111-138.

Grullon, G., & Michaely, R. (2002). Dividends, share repurchases, and the substitution hypothesis. The Journal of Finance, 57(4), 1649-1684.

Grullon, G., Michaely, R., Benartzi, S., & Thaler, R. H. (2005). Dividend changes do not signal changes in future profitability. Journal of Business, 78(5): 483-512.

Hart, O., & Moore, J. (1994). A theory of debt based on the inalienability of human capital. Quarterly Journal of Economics, 109(4), 841-880.

Higgins, R. C. (1981). Sustainable growth under inflation. Financial Management, 6(3), 36-40. How, J. C., Ngo, K., & Verhoeven, P. (2011). Dividend initiations and long-run IPO performance.

Australian Journal of Management, 36(2), 267-286.

Jain, B. A., Shekhar, C., & Torbey, V. (2009). Payout initiation by IPO firms: The choice between dividends and share repurchases. The Quarterly Review of Economics and Finance, 49(4), 1275-1297.

Jensen, M. C. (1986). Agency cost of free cash flow, corporate finance, and takeovers. American Economic Review, 76(2), 323-329.

John, K., & Williams, J. (1985). Dividends, dilution, and taxes: A signaling equilibrium. Journal of Finance, 40(4), 1053-1070.

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.

King, M., (1977). Public policy and the corporation. London: Chapman and Hall.

30 alternative institutional measures. Economics & Politics, 7(3), 207-227.

La Porta, R., Lopez‐de‐Silanes, F., Shleifer, A., & Vishny, R. W. (2000). Agency problems and dividend policies around the world. The Journal of Finance, 55(1), 1-33.

La Porta, R., Lopez-de-Silanes, F., Shleifer, A., & Vishny, R. (1999). The quality of government. Journal of Law, Economics and Organization, 15(1), 222-279.

La Porta, R., Lopez-de-Silanes, F., Shleifer, A., & Vishny, R. (1998). Law and finance. Journal of Political Economy, 106(6), 1113-1155.

La Porta, R., Lopez‐de‐Silanes, F., Shleifer, A., & Vishny, R. (1997). Legal determinants of external finance. Journal of Finance, 52(3), 1131-1150.

Leuz, C., Nanda, D., & Wysocki, P. D. (2003). Earnings management and investor protection: an international comparison. Journal of financial economics, 69(3), 505-527.

Li, K., & Zhao, X. (2008). Asymmetric information and dividend policy. Financial management, 37(4), 673-694.

Lin, K. L., & Shen, C. H. (2012). The impact of corporate governance on the relationship between investment opportunities and dividend policy: An endogenous switching model approach. Asia‐Pacific Journal of Financial Studies, 41(2), 125-145.

Lintner, J. (1956). Distribution of incomes of corporations among dividends, retained earnings, and taxes. The American Economic Review, 46(2), 97-113.

Lipson, M. L., Maquieira, C. P., & Megginson, W. (1998). Dividend initiations and earnings surprises. Financial Management, 27(3): 36-45.

Masulis, R. W., & Trueman, B. (1988). Corporate investment and dividend decisions under differential personal taxation. Journal of Financial and Quantitative Analysis, 23(4), 369- 385.

McCaffrey, K., & Hamill, P. (2000). Dividend initiation announcements effects in initial public offerings. Applied Financial Economics, 10(5), 533-542.

Michaely, R., & Shaw, W. H. (1994). The pricing of initial public offerings: Tests of adverse-selection and signaling theories. Review of Financial studies, 7(2), 279-319.

Miller, M. H., & Modigliani, F. (1961). Dividend policy, growth, and the valuation of shares. Journal of Business, 34(4), 411-433.

Miller, M. H., & Rock, K. (1985). Dividend policy under asymmetric information. Journal of finance, 40(4), 1031-1051.

Mitton, T. (2004). Corporate governance and dividend policy in emerging markets. Emerging Markets Review, 5(4), 409-426.

Nissim, D., & Ziv, A. (2001). Dividend changes and future profitability. Journal of Finance, 56(6), 2111-2133.

31 Skinner, D. L., & Gilster, J. E. (1990). Dividend clienteles, the tax‐clientele hypothesis, and

utilities. Financial Review, 25(2), 287-296.

Van Caneghem, T., & Aerts, W. (2011). Intra-industry conformity in dividend policy. Managerial Finance, 37(6), 492-516.

Watson, A. (1974). Legal transplants: An approach to comparative law. Charlottesville: University of Georgia Press.

Woolridge, J. R. (1983). Dividend changes and security prices. The Journal of Finance, 38(5), 1607- 1615.

32

8. APPENDIX