Dividend Decisions, Economic Growth and Firm’s Value of Firms Listed at Nairobi Securities Exchange Kenya

The declining and highly volatile firm value observed in the NSE over the last decade has raised concern among scholars and financial practitioners. A declining and turbulent firm value implies lost and unstable shareholders wealth which in turn increases risk to the stock holders. It is therefore important to ensure that the firm value is enhanced to ensure growth and stable wealth of the shareholders. The study was carried out to determine the effect of dividend decisions, economic growth and firms’ value of selected firms listed at Nairobi securities exchange Kenya. The target population was the 46 non-financial companies listed in the NSE. A census of all non-financial firms listed in the Nairobi Securities Exchange was done. The study utilized secondary data from financial reports as published in the NSE handbook and Kenya National Bureau of Statistics for the period between 2008 and 2016. Panel regressions analysis and Pearson’s product moment correlation analysis were used for inferential analysis while means and standard deviations were used for purposes of descriptive analysis. Feasible Generalized Least Square (FGLS) regression results indicated that dividend decisions (p=0.012, <0.05) had a statistically significant positive effect on firm value. Whisman test of moderation further indicated that GDP had significant positive moderation effect on the relationship between each of the dividend decisions and the firm value. The study concludes that; dividend yield has a very strong positive relationship with firm value. That is, increases/decreases in dividend yield will be accompanied by increases/decreases in firm value. The study therefore recommends that corporate managers increase the dividend payout in times of profitability Iqbal and Khan, (2012) studied the effect of dividend decision on shareholder’s wealth of companies listed in Karachi Stock Exchange. Multiple and stepwise regression was used to show relationship between variables. Shareholders wealth was used as the dependent variable as indicated by market price per share (MPS), the independent variable was the dividend decision as measured by dividend per share (DPS), Lagged Price earnings ratio (LPER), Retained Earnings and Lagged Market Value of equity. The research showed that there is significant influence of dividend decision shareholders wealth. This study considered only the dividend decision and assumed a direct relationship between the independent and dependent variable. However, the current study considered MPS as the measure of shareholder’s wealth, a more robust result was enhanced in this study by The FGLS results document a significant positive effect of dividend decision on firm value. The regression results documented a significant moderating effect of GDP on the relationship between the various indicators of financial management decisions and firm value.


INTRODUCTION
According to Wahyudi and Pawestri (2006) firm value is the amount the prospective buyers would be willing to pay if the firm was to be liquidated. Sundaram and Inkpen (2004) describe firm value maximization as a decisionmaking rule that brings benefits to all stakeholders. According to the shareholder theory, the primary goal of a company is usually defined as value maximization for the owners. This means maximizing the value of equity and thereby increasing the shareholder's wealth. According to the theory of firm, the main purpose for the existence of a firm is to maximize shareholder's wealth or firm value (Paminto, 2015). By maximizing the firms value, shareholder's wealth is maximized and thereby helping the organization to achieve the main goal of existence (Dan, 2002). Ganesh et al. (2013) contends that the value maximization objective is considered superior to profit maximization objective. Maximizing the firm value is the basis upon which wealth maximization objective is based. Unlike the traditional profit maximizing goal, value maximization goal considers time value of money and is objective.
According to Thavikulwat (2014), the firm value can be obtained through diverse measures; the values obtained from each measure are likely to be different from each other. The most common measure is the intrinsic values of all the outstanding shares of the firm. Its application, however, requires an efficient real market for shares. Other indicator is capitalized value of its projected future performance (Dolenco et al., 2012). However, this may require accurate projection of future cash flows and discounting rates. To eliminate problems related to estimation, Tobin's Q invented by James Tobin in 1969 as a measure of the firm value was used as a proxy for firm value. Thavikulwat (2014) observed that Tobin's Q eliminates the difficult problem of estimating both cash flow and the rate of return or marginal costs. The ratio uses market value of equities which can accurately be estimated.
The firm value of companies listed in the NSE has faced periods of decline as evidenced by NSE 20 share index and market capitalization rate. The NSE share index declined significantly in the period between 2007 and 2009 from a high of 6161 points to a low of 2474.75 points, this saw investors lose Kshs 80 billion. A further decline in the NSE 20 share index was experienced between 2010 and 2011 from 4559.56 points to 3155.00 points (NSE Monthly Market Statistical Bulletins, 2012). Macroeconomic variables include elements that indicate the aggregate performance of the economy. They focus on aggregate changes in the economy such as, employment, Gross domestic product, political risk and inflation (Kirui, Wawire & Onono, 2014). Macroeconomic variables and in particular GDP and Political risk are likely to affect the relationship between financial management decisions and firm value either positively or negatively as documented by literature for instance Bokpin (2009) and Elkahaldi and Daadaa (2015). However, it remains unclear whether they positively or negatively affect such relation and whether such relation is significant or insignificant.
Macroeconomic variables include elements that indicate the aggregate performance of the economy. They focus on aggregate changes in the economy such as, employment, Gross domestic product, political risk and inflation (Kirui, Wawire & Onono, 2014). Macroeconomic variables and in particular GDP and Political risk are likely to affect the relationship between financial management decisions and firm value either positively or negatively as documented by literature for instance Bokpin, 2009 andElkahaldi &Daadaa (2015). However, it remains unclear whether they positively or negatively affect such relation and whether such relation is significant or insignificant Gross domestic product is the total market value of all final goods and services produced within a given time period from the various factors of production (Mudida, 2003). It is the overall productivity of a specific country in a specific period given her resources (Shahzad & Al-Swidi, 2013). According to Ongore and Kusa (2013), it is the sum total value added of all production units including all taxes and subsidies on products which are not included in the valuation of output. Ndunda (2016) indicates that GDP is a basic economic indicator and measures the level of total economic output relative to the population of a country. GDP affects the overall performance of the economy and is regarded external to the organization and as such, financial managers do not have control over it (Humpe & Macmillan, 2009).
GDP is likely to affect the relationship between the financial management decisions and firm value either positively or negatively. This is largely attributed to macroeconomic variable's effect on overall economy which includes the corporate sector (Muchiri, 2016). Correspondingly, GDP affects the firm's expected future cash flows which in turn affects the firm's capacity to borrow, pay dividends, raise equity and invest in long term assets and thereby affecting the its market valuation (Maina & Ishmail ,2014). The GDP growth rate was utilized in the study to determine the moderating effect of GDP on the relationship between financial management decisions and firm value. As utilized in other studies (Tonmoy & Sadia, 2017, Musau et al, 2018. The Kenya economy has faced cyclical movements with some periods showing high growth rate and others showing dismal growth rate (Kirui, Wawire & Onono, 2014).
Dividend decision is the exercises of dividend payout decisions made by the managers of the organization (Baker, Powell & Veit, 2002). Dividend decisions are the choices regarding the amount of dividend paid and amounts retained, that is, choices about the proportion of dividends to be paid as a proportion of the net income made by the firm (Kapoor, Mishra & Anil (2010). These decisions include, how much to pay, when to pay, how to pay and why pay the dividends, thereby broadening the aspects of the dividend decision variables (Al-Twaijry, 2007).
Numerous finance scholars have focused on the dividend decisions because of the relative importance they play in enhancing profitability and firms value. Dividend decision not only determines the firm value but also affects the shareholder wealth (Zainuddin, 2015). Michaely and Roberts (2007) indicate that the dividend decisions affect the value of company either positively or negatively and therefore very essential. The decisions regarding the optimal dividend decisions are amongst the hardest financial management decisions to make hence presenting financial managers with dividend puzzle. This has led to empirical studies focusing more on effects of dividend decisions on firm value yielding contradictory results. Al-Twaijry (2007) indicates that the management should be careful in their dividend decisions since it has a critical bearing towards profitability and the hence the firms value.
Abdullah, Ahmad and Roslan (2012) advocates for more empirical studies to resolve dividend puzzle. Accordingly, scholars and finance practitioners in different empirical studies adopt different measures of dividend decisions. Velnampy, Nimalthasan and Kalaiarasi (2012) adopt ratio of dividend to net income paid as a measure of dividend decisions with a focus on how much to pay. Obaid (2016) utilize EPS and dividend payout ratio as the key indicator of dividend decision. The study adopted dividend yield as the indicator of dividend decisions as adopted by Kapoor et al. (2010) due to the ability of the ratio to relate the dividend paid with market price per share and hence the tradeoff between capital growth and dividend According to Ferris, Sen, and Unlu (2009), there has been a downward trend from 72% to 55% of companies paying dividends globally. Similarly, both academic and professional literature indicates that the dividend payment has declined significantly in in Kenya for the period between 2008 and 2016. According to NSE statistical bulletin (2017), Nairobi Securities Exchange (NSE) was trading at eight-year lows with companies not declaring dividends. This has been associated with structural adjustments and inability of companies to generate adequate profits to distribute to the shareholders as dividends. Most businesses retain most of their profits in order to take advantage of profitable investment opportunities (Khan & Shamin, 2017).
The ultimate survival of a firm depends on its ability to utilize the resources provided by the shareholders in order to increase the firm value and hence their wealth (Yartey & Adjasi, 2007) In Kenya firm value as indicated by the market capitalization and NSE 20 Share index is highly volatile and thereby increasing the risk of holding such securities (Nyasha & Odhiambo, 2014). In a market where the firm value is not stable, shareholder's wealth is at higher risk. The firm value of companies listed in the NSE has faced periods of decline as evidenced by NSE 20 share index and market capitalization. The market capitalization declined significantly in the period between

LITERATURE REVIEW Theoretical Review
Signaling was introduced by Akerlof and Arrow (1970). Spence (1974) further developed it into signal equilibrium theory, which states that a good firm can separate itself from a bad one by sending superior signals to the capital market. The signal will be convincing only if the bad firm is unable to copy the good firm by sending similar signal. The bad firm will not find it necessary to mimic the signal of the good firm if the cost of the signal is costlier. This theory was further developed by Ross (1977) who suggested that dividend decision will send a signal to the investors; either positive or negative. Bhattacharya (1979) asserts that good companies that pay dividends are perceived to have better value than those that pay lesser dividend.
Allen, Bernardo, and Welch (2000) support signaling theory; in their research, they found that theories regarding information asymmetry model indicates that declaration of dividend payout different from expectation of owners contain information on future earnings. This is further supported by Grullon, Michaely, Benartzi and Thaler, (2005) who indicates that dividend decision changes shows information about future earnings. The theory is relevant in explaining the link between dividend decisions and firm value. Evidence has shown that market reacts positively when dividends increase and react negatively with decrease in dividends in support of signaling theory (Amihud & Li, 2002). Dividend decision changes convey news regarding future cash flows which affects the firm value (Gabillon & Gabillon, 2012).
Shareholder Value Theory was advanced by Friedman (1970). Shareholder value theory is the most prominent economic theory in use by business. It proposes that the main purpose of the business is maximizing shareholder wealth through improving the firm value. The theory strongly argues in favor of maximizing the firm value in order to increase the wealth of shareholders. The firm should be operated in such a way as to cater for the interests of the shareholders. According to Jensen (2001), the management should carry out the operations of the business with the aim of increasing the shareholders' value through maximizing the revenue, minimizing costs and reducing the risks.
Margolis and Walsh (2003) contend that shareholders have primary position over all the other stakeholders, accordingly, the board has a primary duty and responsibility to increase the shareholders' value (Sharfan, 2014). Saint and Tripathi (2006) indicates that the fact that the business exists to maximize the interests of shareholders is so socially incorporated into the financial community that the business community believes it's the truth and therefore the value maximization motive is propagated in and applied in practice, and is justified through other sources. This theory was utilized in the study to support the value maximization motive which forms the basis of the dependent variable in the current study. It also explained the importance of maximizing the value of companies as a primary goal of any financial management decision making unit. Velnampy, Nimalthasan and Kalaiarasi (2014) sought to establish the link between dividend decision and company's financial performance. The findings of the study indicated that the dividend decision do not affect company's return on equity and return on asset. This study considered dividend decision as the only key driver of the company performance. The study also ignores moderating variables which may affect the direct relation between the explanatory variable and dependent variable. There is need to incorporate other variables including intervening variables which may affect the firm value besides dividend decision.

Empirical Review
Abdullah Al Masum (2014) estimated excess stock market returns for the entire population of thirty banks listed in Dhaka Stock Exchange (DSE) for the period between 2007 and 2011. The objective of the study was to determine the relationship between stock market returns of private commercial banks in Bangladesh and their financial performance. The study also intended to determine the degree the returns on stocks can be explained by the dividend decision for similar period of time. The findings of the research indicated that dividend Decisions had significant positive effect on Stock Prices. However, the study considered the dividend decisions as the only key driver of the firm value. The findings of the current study were enhanced by considering other financial management decisions. To enhance the robustness of the research findings, Tobin's Q was utilized in the study as opposed to MPS. Gul, Sajid, Razzaq, Iqbal and Khan, (2012) studied the effect of dividend decision on shareholder's wealth of companies listed in Karachi Stock Exchange. Multiple and stepwise regression was used to show relationship between variables. Shareholders wealth was used as the dependent variable as indicated by market price per share (MPS), the independent variable was the dividend decision as measured by dividend per share (DPS), Lagged Price earnings ratio (LPER), Retained Earnings and Lagged Market Value of equity. The research showed that there is significant influence of dividend decision shareholders wealth. This study considered only the dividend decision and assumed a direct relationship between the independent and dependent variable. However, the current study considered MPS as the measure of shareholder's wealth, a more robust result was enhanced in this study by Research Journal of Finance and Accounting www.iiste.org ISSN 2222-1697 (Paper) ISSN 2222-2847 (Online) Vol.11, No.18, 2020 considering Tobin's q considered superior to MPS. Uwuigbe (2012) studied the association between the financial performance and dividend decision. The study concentrated on listed companies in Nigeria. The study also looked at the association of ownership structure and the dividend payouts. The findings of the study suggested significant positive relationship between the dividend decision and the performance of the companies listed in Nigeria. The study also showed that ownership structure and company's size has a significant impact of the dividend decision of the company. However, this study solely considered dividend decision and ignored the other financial management decisions which may affect the company performance. Further, the robustness of the result of this study were enhanced by use of firm value as the dependent value as opposed to traditional measures of company performance.

Conceptual Framework
A conceptual framework is an investigative tool with several disparities and contexts. It is utilized to make conceptual differences and organize research ideas, it shows the interconnection between variables under consideration and provides researcher with guideline on the research methodology to be applied. Figure 2.1 shows a conceptual framework showing the relationship between dividends decisions and the value of non-financial companies listed in Nairobi securities Exchange. The explanatory variable is dividends decisions and the dependent variable is firm value. The moderating variables include GDP

RESEARCH METHODOLOGY
The study adopted explanatory non experimental research design. Corresponding to Kerlinger and Lee (2000), the design was more suitable in identifying factors that cause change in the value of companies, without influencing the financial management decisions variables. The data was obtained from the company's financial reports as published in the NSE handbook and Kenya National Bureau of Statistics for the period between 2008 and 2016. The study utilized document review guide in appendix II and III. The study utilized both descriptive and inferential statistics to analyse data. Panel multiple regressions were used in the study since the data had time and cross section dimensions. Pearson's Product Moment Correlation analysis was used to determine the relationship between the variables. Panel data was used in the study because it has the advantage of providing better insights compared to time series and cross-sectional data since it is possible to put aside the theoretical effects and enhances comparison of industries over a period of time (Kothari, 2004 Where FVit is the firm value i at time t proxied by Tobin's Q. i denote the company while t is the time period running from 2008 to 2016. DDit denotes a vector of Dividend decisions β1 is coefficient to be estimated while β0 is a constant term. GDP= Gross Domestic product, εit is the composite error term.

RESULTS AND DISCUSSIONS Descriptive Analysis
Results in Table 1 show the summary of the descriptive statistics of dividend decision, GDP growth rate.

Dividends Decisions
Dividends Yield (Dividends per share to market price per share)

Firm Value
Tobin's Q (market Capitalization/Book Values of Assets)

Economic Growth GDP Growth Rate
Research Journal of Finance and Accounting www.iiste.org ISSN 2222-1697 (Paper) ISSN 2222-2847 (Online) Vol.11, No.18, 2020  14.29001 5933.285 Observations 360 360 Dividend decision yield had a mean of 0.5298 indicating that on average companies with lower market price per share (MPS) were paying high dividends per share (DPS) that is more than 50%. Bergmann (2016) found similar trends by documenting increased trend in small cap companies paying higher dividends compared to large cap companies globally. The variability of dividend yield as indicated by standard deviation was at lower of 0.1995 indicating that there was no high variations in dividend yield among selected listed companies. The minimum and maximum of dividend yield was 0.0000 and 2.0767 which indicates that dividend yield was positive at both the lower level and higher level. The results further indicate that dividend yield was skewed to the right indicating that more companies were having a higher dividend yield.
The mean GDP growth rate of 0.1724 implies that GDP was increasing over the study period. The standard deviation of GDP growth rate was 0.4065 indicating that variability in the GDP growth rate was low. However, Obere, Thuku, and Gachanja (2013), found a cyclical movement in GDP growth rate. The difference could be traced to the length of study period (1963 to 2009) which was fairly longer than the period of the current study (2008 to 2016). The minimum and maximum of GDP growth rate was -0.272600 and 13.2500. The negative observation implies that in certain years there was negative growth rate in GDP. GDP growth rate was skewed to the right as shown by positive sign on the skewness which indicates that the GDP growth rate was increasing in the period 2008 to 2016.

Correlation Analysis
The study findings revealed that dividend decisions adopted by listed firms in Kenya had weak but positive strong correlation with firm value (r=0.14). The findings implied that dividend decisions positive influenced firm value of listed firms in Kenya. This finding concurs with those of Abdullah Al Masum (2014), where the study indicated a significant positive relationship between dividend decisions and Stock Prices. Similarly, Gul et al. (2014) indicated a positive relationship between company's decisions and firm value. GDP Growth Rate was found to have correlation of 0.36 with firm value which implied that GDP growth rate had positive and weak association with firm value. 0.14 0.36 1 Correlation coefficients, using the observations (2008 -2016) 5% critical value (two-tailed) = 0.0064 for n = 9

Regression Analysis
The p-value of dividend decisions was at (p=0.034, <0.05) indicating a statistically significant relationship between dividend decisions and firm value of selected firms listed in the NSE (see Table 3). Hence the study failed to reject H02 at α=0.05 and concluded that there is a statistically significant relationship between dividend decision and firm value of selected firms listed in the NSE. The findings of the study were consistent with signaling theory attributed to Akerlof and Arrow (1970) and further developed by Spence (1974) into signal equilibrium theory. The study findings are also supported by Bird in hand theory proposed by Myron Gordon (1963) andJohn Linter (1964). This theories advocate for payment of dividends in order to increase firm value. In contradiction the findings of the study are not supported by dividend irrelevant theories including Modigliani and Miller (1963) who indicated that the value of company is not responsive to the dividend decisions.
The findings of the study compliment the findings in Abdullah Al Masum (2014), where the study indicated Research Journal of Finance and Accounting www.iiste.org ISSN 2222-1697 (Paper) ISSN 2222-2847 (Online) Vol.11, No.18, 2020 a significant positive relationship between dividend decisions and Stock Prices. Similarly, Gul et al. (2014) indicated a positive relationship between company's dividend decisions and firm value. To support the findings in the current study, Uwuigbe (2012) suggested significant positive relationship between the dividend decision and the performance. However, the findings of the study contradict the findings in Velnampy et al. (2014) who found a negative relationship between dividend decisions and firm value. The difference could be traced to the comparing number of variables, measures of variables, and presence of moderating variables in the current study.

Test for Moderating Effects of GDP growth Rate
The study sought to determine the moderating effect of GDP on the relationship between dividends decisions and the firm value. The test applied Whisman and McClelland (2005) which involves two steps; in the first step, each moderator is ran as an independent variable in addition to other explanatory variables. If the coefficient is insignificant at 0.05, then the variable in question becomes an independent variable. If insignificant, then second step is necessary to include interaction effect of the moderator with each of the variable in consideration.
Moderation test was carried out to determine the moderating effect of GDP on the relationship between dividend decisions and firm value. The p value of (p=0.000, <0.05), indicates that GDP had a significant positive moderating effect on the relationship between dividend decisions and firm value. Yensu and Adusei (2016) concur with the findings and indicate that GDP changes are likely to affect the relationship between dividend decisions and firm value in a positive way. However, Farrukh, Irshad, Khakwani, Ishaque and Ansa (2017) refute the findings by documenting an insignificant positive effect. The difference could be traced to the context of the study and period of study under consideration. 0.000 The objective was to determine the effect of dividend decisions on the firm value of selected firms listed in the NSE, Kenya. Dividend Decisions in this case was represented by dividend payout while firm value was represented by Tobin's q. To enable conceptualization and development of hypothesis for testing such relationship, signaling theory was applied. The findings of correlation analysis indicated that dividend decisions had a weak and positive correlation with firm value of selected listed firms in NSE. The FGLS results document a significant positive effect of dividend decision on firm value. The regression results documented a significant moderating effect of GDP on the relationship between the various indicators of financial management decisions and firm value.

CONCLUSIONS
The study documents an upward trend in dividend yield. The study therefore concludes that; firms are preferring payments of dividends as opposed to retaining funds for expansion. Further, the study indicates a positive significant effect of dividend decisions proxied by dividend yield on firm value. The current study concludes that increase in the dividend payout leads to increase in firm value. Moreover, this is supported by various empirical literatures for example the signaling theory of dividends. The study also documents a strong positive correlation between dividend decision and firm value. The study concludes that; dividend yield has a very strong positive relationship with firm value. That is, increases/decreases in dividend yield will be accompanied by increases/decreases in firm value. In regard to the moderation effects of GDP on the relationship between various dividends decisions variables and firm value, the study documents that GDP had a significant positive moderating effect on the relationship between dividends decisions and firm value. This study therefore concludes that GDP is important in such relations and once enhanced the independent variables effect on firm value will also be enhanced.

RECOMMENDATIONS
Dividend decisions were found to have a significant effect on the firm value. Particularly, dividend payout ratio was found to have a positive effect on the firm value. The study therefore recommends that corporate managers increase the dividend payout in times of profitability. This sends pleasant information to the market which in turn increases the market valuation. Capital Markets Authority (CMA) should come up with the necessary regulations as to the minimum amount of dividends a profitable company should pay in order to augment investors' wealth. The study recommends that the corporate managers be a lot more proactive in global and country specific factors that cause alterations in GDP. Application of economic analysis tools is highly recommended so that the companies can make deliberate actions to take cautionary actions to mitigate the risk associated with deviations in GDP growth rate. The government should ensure a proper check in the fiscal and monetary policies to ensure stable economy.