Firm Ownership Characteristics and Long-run Return on Equity Issued: A Case of the Nairobi Securities Exchange

This investigation looked at the link between firm ownership characteristics and long-run return on firms that issued equity at the Nairobi Securities Exchange (NSE) in Kenya. The study covered 12 firms that issued shares in the NSE market from 2006-2008. Ownership characteristics included (state ownership, institutional Ownership, foreign Ownership, big five shareholders, market capitalization, age of the firm and Leverage of the firm) in relation to the average return. The study tested whether each of the firm ownership characteristics influenced long-run performance. Annual return for these companies was based on market return for five years after the firm’s equity shares were issued. The long-run performance was compared with three benchmarks, namely, NSE index, CAPM and Matching firms. Seven hypotheses were developed for the study. Simple-liner and multi-linear regression analyses based on panel data were carried out to relate the extended run return on shares issued. The result of the survey showed that issuing firms performed better than non-issuing firms. These issuing firms also performed better in comparison to CAPM. However, the issuing firms performed worse than NSEI. In conclusion, the long-run performance of equity issued at the NSE does not necessarily underperform relative to non-issuing establishments .

contend that an establishment having several ownership structures can secure the balancing set of crucial economic and political resources needed for better performance. Previous studies have treated ownership categories separately (Choi, Park, & Hong, 2012). Such treatment is likely to ignore the potential advantage of variables that may complement each other in a study (Chen, Li, Shapiro, & Zhang, 2013). This investigation seeks to bridge these apparent gaps in the literature. The remaining part of the study is divided as follows: section two covers the literature review. Section three tests a set of hypotheses. Section four discusses empirical results. Finally, section five concludes the study with recommendations on areas for further research.

Literature Review
The ownership structure can be decomposed into several units. Some of them include State ownership, institutional Ownership, foreign Ownership and big five shareholders. The level by which controls is exercised may be termed as either ownership concentration or ownership identity. The former is the proportion of shares a single owner holds concerning the aggregate firm's shareholding (Anstoniadi, Lazarides, & Sarrianides, 2010). In contrast, the latter is the actual names of crucial shareholders (Grossman & Hart, 1986). Recent studies have revealed that a concentrated ownership structure is commonly found in developing countries (Claessen, Djankov, & Lang, 2000). Ownership structure revolves around agency theory, Jensen and Meckling (1976) and corporate governance theory (Thomsen & Conyon, 2012). The central premise of these theories is that managers can engage in decision-making behavior contrary to the expectations of shareholders.
One type of equity ownership is where the state owns shares in a firm. State ownership arises when the state has control and management of a firm. Some scholars have asserted that State ownership is inefficient and bureaucratic. Villalonga (1999) claims that managers are rarely fired for nonperformance in state-controlled firms. If there is any firing of management, it is not related to its performance (Cragg & Dyck, 1999). Therefore managers have little incentive to focus on the financial or operational performance of the firms they manage. Porta, Silanes and Shleifer ((2002) find that greater state ownership of establishments is linked with lower subsequent financial development and lower economic growth. Iannota, Nocera and Sironi (2007) find that state-owned banks have lower returns than those privately owned despite lower costs associated with their operations. Furthermore, Gursoy and Aydogan (2002) find that banks owned by the state rarely take risk analysis in their operations and many people believe that doing business with state banks would help their establishments when they are in trouble. This has led to many state-owned banks end up with a large amount of bad debt, which is eventually written off (Gursoy & Aydogan, 2002).
The second type of Ownership in a firm is Institutional Ownership. Zhang and Gimeo (2016) identify institutional investors as financial investors like pension funds, mutual funds, hedge funds, banks, insurance establishments, endowments, and foundations holding a substantial amount of equity in publicly traded establishments. Institutional Ownership is regarded as a case where an institution has www.scholink.org/ojs/index.php/jepf Journal of Economics and Public Finance Vol. 7, No. 3, 2021 134 Published by SCHOLINK INC.
share ownership, especially when the state has privatized its' holding. Institutional shareholding is legal Ownership since a legal person owns shares in the name of an institution (Wei & Varela, 2003;Wei, Xie, & Zhang, 2005). Large equity ownership by institutions in an establishment is assumed to encourage stakeholders in monitoring managers' undertakings, stop them from involving in moral hazard activities and focus on shareholders' interest (Belkhir, 2005;Cornett, Marcus, Saunders, & Tehranian, 2007). Institutional investors are more focused on profit, hence have more inducements to scrutinize the establishment's activities. Large shareholders in the form of institutions effectively enforce their rights and can control managers' excesses (Morck, Nakamura, & Shivdasani, 2000). Yuan, Xiao and Zou (2008) have raised two issues related to firm performance and institutional Ownership.
These are; enhanced performance argument and reduction performance argument. Where there is performance enhancement, it is due to the introduction of good corporate governance. Institutional investors aim at best returns. Good performance is also associated with active monitoring. Performance reduction is associated with investors who require quick returns in a short time (Appel, Gormely, & Keim, 2015;Drucker, 1986). This may be detrimental to the organization's performance.
The third form of equity ownership is concentrated Ownership. This type of Ownership means large shareholding in an organization is held by a few shareholders (Appel, Gormley, & Keim, 2015). Some scholars refer to them as the large shareholders or the big five shareholders (Rokwaro, 2013). The majority of shareholdings assert influence on management and control in the firms. These large shareholders may oversee management and intervene when they feel things are not going in the right way (Shleifer & Vishny, 1997). Grossman and Hart (1986) contend that large stakeholders have a high stake in these firms; therefore, they are more willing to involve themselves in decisions relating to the firm actively. However, large shareholders may have divergent views from minorities and, in some instances, expropriate their interests. Berger, Clarke, Cull, Klapper and Udell (2005) posit that concentrated ownerships may bring with them a negative impact on performance in that their behaviors may lead such firms to fall into financial distress and crisis. This is because large shareholding with high authority will control management and create moral hazard behavior. These large shareholdings are often referred to as the big five shareholders (Wahla, Shah, & Hussian, 2012). Big five shareholders are the majority shareholders where they own at most seventy-five percent shareholding (75%). They reflect dominance in the management of firms and, in many instances, are family members or government (Soon & Koh, 2007;Khanna & Palepu, 1999 The leverage level also sways the extent to which firms raise funds in securities exchange it desires to achieve. The utilization of high Leverage helps when an establishment is making gains. Contrariwise, an establishment that is highly levered may be troubled if its profitability is declining and may face a high risk of default compared to an unlevered or less levered establishment in a similar situation. The leverage ratio can be indicated in the following way: Debt/ Equity; or Total debt/Total capital. The leverage ratio is the level at which an establishment is using the funds that are borrowed. It assesses the establishment's solvency and capital structure. Modgiliani and Miller (1958) argue that under capital structure theory, if financial markets are efficient, then debt and equity financing will essentially be substitutable and that the other aspects will point out the ideal capital structure. The function model for Leverage can be expressed in the following way: Market value = f (Capital structure) Market value = f (EqC, DeC) Whereas the obvious form in first difference is; Where, MvF = Market Value of firm, EqC = Equity Capital, DeC = Debt Capital, et−1 = Idiosyncratic terms. Market capitalization is the firm's value of shares The second last independent variable is firm age. Firm age can be a proxy for risk. Old establishments are more expected to be stable, mature and may have more skills because they have been operating for a long time (Liargovas & Skandalis, 2008). A firm's age is associated with experience, knowledge intensity, and entrepreneurial flexibility (Chen, Li, Shapiro, &Zhang, 2013). Age can be a measure of both uncertainty and investor optimism (Ritter, 1991). The age of a firm is evaluated by the day and date before IPO. An establishment that has been in operation for many years can sustain risk. A firm that has been in business for a long time is well known, and there is a small element of uncertainty (Lowry, Officer, & Schewert, 2008;Alvarez, 2015). Ritter (1991), Khurshed (1999), Belghitar and Dixon (2012)

Hypotheses Development
The study developed a set of benchmarks to determine whether firms that issued equity underperform or over-perform these sets of benchmarks. The returns from these firms were calculated first. Changes in share prices determined the returns from these firms during each year plus any dividends paid during the year. After that, the returns were compared with the relevant benchmarks used to evaluate the average return. Concerning this, several null hypotheses were established to test if the average return was statistically unequal to zero. Several other null hypotheses were developed to determine the link between the average return of establishments that issued equity and firm ownership structure characterized by state share ownership, institutional share ownership, foreign share ownership, the big five-share Ownership, Leverage of the firms, age of the firms and market capitalization. All these were based on a 5% level of significance. These hypotheses were aimed at justifying study objectives: How do these microeconomic variables perform compared to the three benchmarks in the long run?  average return for each firm, the study applied three benchmarks to assess the abnormal returns on firms that issued equity from 2006-2013 in Nairobi Securities Exchange. These benchmarks were: Nairobi Securities Market Index based on 20 share index, Capital Asset Pricing Model and Matching Firms. The study did not rely only on the stock market index for comparing net returns because relying on this can yield biased results (Lyon, Barber, & Tsai, 1999). The study used panel data and applied the following diagnostic tests; test of stationarity of the data and co-integration test to ensure a long-run association between the output and predictor variables; granger causality test to establish if one time series is significant in predicting another. Empirical data sets were used to find patterns of correlation.
Four other diagnostic tests were carried out, and these included normality test, multi-co-linearity, auto-co-linearity and homoscedasticity. The study tested the normality of the data using Shapiro-Wilk test. A Multi-co-linearity test was done using the variance inflation factor (VIF Under this benchmark, all annual equity returns have to be evaluated during the period of investigation. The average-annual risk-free rate (RFR) represents the return on Central Bank of Kenya's Treasury bills. This is averaged to give an annual interest-free rate since the treasury bills rates are for 91 days.
An equation represents this: Where: R it = return of firm that issued equity in period t R mt = market return in year t as measured by NSE market index.
R ft = 91 days Treasury bill return in calendar year t Β = beta coefficient of CAPM is determined by using correlation coefficient;  Loughran and Ritter (1995) measure long-run return by matching each issuer with a non-issuing firm closest in size. This study based matching firms and issuing firms on market value (market price per share multiplied by outstanding shares) to determine their sizes. The average return (AR) according to the benchmark is shown below: Where: AR it = average return for matching firm.
R it = return of firm that issued equity i in event year t.
R MF = return of the control portfolio in the event year t under this benchmark. The Matched firm's portfolio returns are equally-weighted average returns on a portfolio of every firm.

Data Analysis, Presentation and Results
The study used the Ordinary Least Square (OLS) regression to test the link between state ownership, institutional Ownership, foreign Ownership and big five shareholders in relation to firm performance in the long run following the equity share issue. The study used the following specification model to test the theory:

Descriptive Statistics
Firms that issue shares at the stock markets are generally regarded as growth firms. They need funds to expand their businesses. Generally, the average raw return on new issues is low; therefore, many studies have found such firms underachieve none issuing establishments in the same market (Loughran & Ritter, 1997 Table 3 shows that the mean average return is .293. This is lower than NSEI, which is1.1364 but higher than CAPM, -4.4663 and Matching firms, with a mean of -2.6782. This means that companies that issued equity during the investigation period performed better than those that did not issue equity. Data were normally distributed as shown by skewness, whose figures were around 0. Similarly, kurtosis values were less than three except for Average return, CAPM and Matching firms (Gujarati & Porter, 2009).

Stationarity Test
Stationary is the statistical properties of a process generating a time series that never change with time (Gujarati & Porter, 2009). All the seven (7)

Diagnostic Tests
The study undertook the following diagnostic tests; Normality test, Multicollinearity test, Auto collinearity test and Homoscedasticity test. The results follow below.

Normality Test
The data had a normal distribution as indicated by skewness and kurtosis in the descriptive statistics (Table 3). Under skewness, no figure for all independent variables was above one, whereas for kurtosis, all statistics were below 3 except for average return, CAPM and Matching firms.   Table 4 above shows that all independent variables have a VIF of less than 10 and the tolerance values are above 0.1. These results confirm that the data had no multicollinearity problem. The data distribution shows that it has a normal distribution and that outliers are few and scattered on both the upper and lower part of the graph.

Hypotheses Results
First, we calculated the average return against the benchmarks that were employed (NSE Index, CAPM The return from NSEI is .587; this is more than firms that issued equity whose return is.233. Thus it shows that the Nairobi Securities Exchange Index return was more than the return from the firms that issued equity. Nairobi Securities Exchange 20 share Index used in this case consists of 20 major firms in the market. This may have resulted in a higher record performance than those that issued shares during the study period.   Loughran and Ritter (1997), Panagiotis (2009), Paskelian and Bell (2010).
However, the results of this study support results by Thomas and Yawen (2011), Dang and Yang (2007). These results in Table 9 show that certain variables were statistically significant at a 5% level regarding measurement models. Institutional Ownership was significant at 0.046 as compared to matching firms.

Regression Model
The study used panel data analysis to establish a relationship between average return and the microeconomic determinants.  Testing the overall level of significance at 5% using ANOVA shows that, in general, all the independent variables put together are insignificant at 0.613 because this is greater than 0.05. There is no statistical substantial link between the big five equity ownership and long-run return.

Market
Capitalization H 08 There is no statistical substantial link between the market capitalization of an establishment and its long-run return.

Not Rejected
Capital Leverage H 09 There is no statistical substantial link between firm leverage and its long-run return.

Not Rejected
Age H 010 There is no statistical substantial link between the age of an establishment and its long-run return.

Summary
The study found the following results; first, from the independent variables, one independent variablethe big five had a significant effect on long-run return for the issuing firms. Secondly, from the benchmarks used, The NSEI performed better than the firms that issued equity. Thirdly, the issuing firms performed better than matching firms. Fourthly, using CAPM as a benchmark, the issuing firms performed better than this benchmark.

Conclusion
This study investigated the long-run return on equity shares issued at Nairobi Securities Exchange from 1 st January 2006 to December 2008. These firms had five years of operation since issuing the shares to make a comparison on their returns to the three benchmarks: Nairobi Securities Exchange Index (NSEI), Capital Asset Pricing Model (CAPM) and Matching Firms (MF). Several studies have concluded that the firms that issue equity in the stock markets underperform non issuing similar firms with the same size for at least five years (Loughran & Ritter, 1995;Barber & Lyon, 1997). Brav (2000) argues that low returns experienced by issuing firms, in the long-run, are because of measurement When we compare NSEI returns with returns from issuing firms, Loughran and Ritter (1995); Barber and Lyon (1997) may be justified. However, this study finds support for studies by Thomas and Yawen (2011); Dang and Yang (2007) that found firms issuing equity perform better in the long-run than those that do not issue equity. This is so when we compare the returns of issuing firms with matching firms and CAPM. Consequently, the study further supports the assertion by Brav (2000) that low returns in the long-run found in IPOs and SEO by other scholars could be measurement errors.

Area for Further Research
There is a need for a more extended period of study to be undertaken in this area. This may shed more light on the performance of shares issued at the Nairobi Securities Exchange.