ASSESMENT OF THE EFFECTS OF HOUSE COUNTRY MACROECONOMICS ON PERFORMANCE CROSS LISTED FIRMS IN EAST AFRICA

This study makes assessment of the effects of host country macroeconomic factors on the performance of cross listed firms in East Africa. It uses quarterly time series from March 2010 to December 2020. The study used stock prices of five Kenyan companies that are cross listed in the stock markets of Uganda and Tanzania. The overall model capturing the impact of the macroeconomic factors on the average prices of stocks shows that interest rate exerts positive and significant impact on the stock market performance, while, inflation has negative and significant effects on the stock market performance in both Uganda and Tanzania. GDP is statistically not significant in influencing the stock prices in both countries. On the other hand, the model that captures individual effect provides mixed result in both Uganda and Tanzania, with the negative effect of inflation reported in three firms in Uganda and two firms in Tanzania; while, a positive effect is found in one firm listed in Uganda and two firms listed in Tanzania. Similarly, the effect of interest rate is positive for three firms in both Uganda and Tanzania; and it is negative in one firm listed in both USE and DSE. GDP negatively affect stock prices of two firms in Tanzania and positively affect the prices of one firm in Uganda.


Introduction
Stock market performance has been an issue of discussion in financial sector for both developed and developing economies. Stock market is considered to be an important component in development of the economy, since it creates integration for users and suppliers of financial resources for investment purposes. The performance of the firms listed in stock markets are affected by macroeconomic variables as it was earlier asserted by Ross (1976) in the Arbitrage pricing theory.
Stocks can be listed in a single market or in multiple markets. Increasing capital market integration due to deregulation and technological advancement in equity trading across the globe has resulted to a surge in cross border listing. The preferred listing destinations for foreign firms include American Stock exchange (NYSE and NASDAQ) and the European exchanges (Dodd, 2013). The practice is also common in East Africa, where, Cross listing of equity is associated with improving stock liquidity, overcoming investment barriers, expansion, reduction of information asymmetry and improving investor protection (Dodd, 2013;Temouri, Driffield, & Bhaumik, 2016). Cross listing also exposes the firm into domestic and foreign macroeconomic fundamentals (such as inflation, interest rate, economic growth and exchange rate) that affect the returns of the firms. Majority of studies have analysed the impact of macroeconomic factors on the stock returns without considering whether the firms are cross listed or not (Barakat et al, 2016;Ndlovu et al. 2018;Abbas et al. 2014;Epaphra & Salema, 2018). While there are limited studies on cross listings and macroeconomics such as those by Muhammad et al (2018), Gupta et al (2016) and Nikkinnen & Salhstrom (2004), however, specific focus on host country macroeconomics received less attention, despite its importance. This study aims to fill this gap.
Host country macroeconomics are significant to be studied as cross listed firms are considered to be affected by the macroeconomic variables of the countries where they trade (Muhammad et al, 2018). The magnitude of the effects depends on the extent to which international markets are efficient and integrated with each other (ibid). Thus, macroeconomic fundamentals of Tanzania and Uganda may contribute to movement in returns of Kenyan firms that are cross listed in the respective countries. This is also supported by Nikkinen & Sahlström (2004), asserting that firms operating in multiple markets are concerned with the economic conditions of those markets since the decision making and profitability of such firms are influenced by the economic conditions in the respective markets.
This paper aims at analyzing the effects of macroeconomic fundamentals of Tanzania and Uganda on the stock returns of the Kenyan firms that are cross listed in DSE and USE respectively. Specific macroeconomic determinants include, the inflation, interest rate and GDP. The paper contributes to the existing literature in the following ways: First the study analyse the impact of host country macroeconomic variables on stock returns of cross listed firms as opposed to many studies that focus on effects of home macroeconomics. Second, the study contributes in the literature relating to regional integration and cross listings. The stock markets in Tanzania and Uganda are very similar while the fluctuations and some of the economic policies in those economies differ.
An understanding of the macroeconomic determinants in regard to stock returns involving regional cross listing is significant in the quest of understanding risk factors as full regional integration brings with it contagion and spill over risks. The rest of the paper is organized as follows. Section 2 reviews the existing literature on macroeconomic fundamentals and stock returns. Section 3 describes the data and the research methodology used Section 4 provides the empirical results. Finally, Section 5 presents summary and conclusions.

Review of Literature
There are various empirical studies undertaken to explain about the effect of macroeconomic factors on stock returns. Through these studies, different data types, research methods and techniques were used to analyze these factors. Gupta et al (2016) analysed the linkages between the American Depository Receipts (ADR) market and home country macroeconomic fundamentals. They found (ADR) prices have a significant short and long run relationship with the economic indicators of the home country. Their results support the assertion that if stock returns accurately reflect the expectation about underlying fundamentals in the future, then they should be used as leading indicators of future economic activities. Narayan & Narayan (2012)  returns was found to be positive and significant in the short run only one country. Apart from being conducted in the more advanced markets, the study used only exchange rate and interest rate leaving out other variables that could potentially affect the returns. But also, their study did not consider the impact at the firm level. Nikkinen & Sahlström (2004) analysed the domestic and US macroeconomic variables on stock market volatilities in Germany and Finish Markets and found US macroeconomic announcements to have significant effects on the volatilities of the European markets. Their study however, did not consider the listing status of the stocks which the current study is addressing. Choi and Kim, (2000) found underlying stock returns, world market returns and US market factor to be the major determinants of ADR returns. However, the study used US market factor instead of specific economic variables. These specific economic variables are very significant in establishing the variations in returns of the listed stocks.
US macroeconomic variables are mentioned to be among important determinant of ADR prices (Muhammed et al, 2018;Acquino and Poshakwale, 2006). In this regard, ADR are considered as extension of the underlying stock market, hence macroeconomic factors such as inflation, interest rate and GDP, that affect the underlying stock market will also affect the ADR (Gupta et al, 2016).

Hypothesis
Inflation is hypothesized to have negative relation with stock returns. An increase in inflation increases the nominal risk-free rate, raising the discount rate hence lowers the stock prices. More so in a competitive economy, higher inflation raises input costs faster than the output prices, making stocks less attractive. Several empirical studies, including Kitati et al (2015) and Zaheer & Rashid (2014) support the negative relation between inflation and stock returns. Interest rate is hypothesized to have negative relation with stock returns. Interest rate affects the cost of finance as well as firm's value of financial assets and liabilities. An increase in interest rate is associated with reduction in stock prices following the substitution effects hypothesis that people will move their funds from the stock markets to interest paying With regard to GDP growth rate, higher rate of economic growth attracts investors since in such an economy there is cost efficiency of production, as well as availability of market (Muli et al, 2017). More so, a higher GDP growth is a measure and a signal of market demand and potential for growing profit, which further attracts investors. It is therefore hypothesized that there is a positive relation between GDP growth rate and stock returns.

Research Method
The data comprises of quarterly values of inflation, interest rate and GDP growth rate Five different companies that include EABL, JHL, KCB, KA and NMG were studies separately. Six different models were then estimated, in which, models 1 to 5 analyses the effect of macroeconomic variables on the stock prices for the five selected companies while model 6 examines the effect of the same macroeconomic variables on the overall stock price.
The macroeconomic variables selected were inflation rate, interest rate and GDP growth rate.
Operationalization of the variables is as provided in table 1.

Theory and Model framework
The theoretical framework of the model is based on Arbitrage Pricing Theory (APT) that was developed by Ross (1976). The theory argues that, the expected return of a financial asset can be modeled as linear function of various factors or theoretical market indices, where sensitivity to changes in each factor is represented by a factor-specific beta coefficient.
In such regard, the asset is mispriced if its current price diverges from the price predicted by the model. The theory asserts that if there are no arbitrage opportunities, the expected return on the assets can be expressed as a linear function of its exposure to the risk factors. That means there is a linear relationship between the expected return on assets and their covariance with other random variables. In that case, the covariance represents systematic risk factors and the slope coefficient between expected returns and covariance is interpreted as a risk premium. The theoretical model can be written as; ( )-Expected return on assets; -Sensitivity of asset price to a factor n -Risk free rate -Risk premium associated with a factor n According to this theory, expected returns of the financial asset is a function of various dynamic factors which might be subjected to sensitivity changes. By incorporating the macroeconomic variables used in the study, the following model framework is developed; Model 1: Model 2: Model 3: Model 4: Model 5: Model 6: ( ) = 0 + 1 1 + 2 2 + 3 3 + 6 … … … … … . . … 6

Empirical Results
Analysis of the data involved the following steps; first, the unit root test was performed by employing the Augmented Dick Fuller test (ADF) to establish if the variables are stationary.
Then diagnostic tests were performed on data to test for multicollinearity. The Auto-Regressive Distributed Lags (ARDL) regression was then employed to investigate the relationship between variables as specified by the model. were stationary at level. The remaining variables became stationary when converted to the first difference I(1) at 5% level of significance.

Auto-Regressive Distributed Lags (ARDL)
To examine the effect of macroeconomic variables on the stock prices, the study employed Auto-Regressive Distributed Lags (ARDL) regression approach. The use of ARDL is backed up with a number of advantages including, its suitability even with small sample size unlike other regression approaches that are suitable with large sample size. Also, ARDL approach produces estimates for both short run and long run, and it accounts for potential endogeneity (Udoh et al., 2015). around 1 percent per quarter but not statistically significant. In principle, the higher the coefficient, the more stable the long run relationship. Thus, the implication of the ECM results is such that for NMG, it will take the least amount of time to entirely return to longrun equilibrium if there is a shock to the macroeconomic variables. The study also checked for the heteroscedasticity and autocorrelation in all six models estimated and findings revealed that there is neither heteroscedasticity problem nor autocorrelation problem for this study.

Results for Tanzania
Findings indicate the existence of significant effects of the selected macroeconomic variables on stock returns in the long run in Tanzania. The long run effect of inflation on stock returns using model 6, show that the average prices of the stock are negatively influenced by the inflation rate. Inflation rate coefficient indicates that a 1% increase in inflation causes the decrease in the stock price by 8.6% when average stock price of the cross listed firms is used. The results may be associated with the assertions that higher inflation is associated with increases in production cost, reduction of economic activities and reallocation of resources from investment to consumption, which will ultimately reduce the price of stocks.
There seem to be variations in the effects when specific firms are analysed using models 1-5. For instance, the inflation rate in Tanzania has a negative effect on stock returns of EABL and KCB in the long run. This means in the long run, the increases in the rate of inflation in Tanzania will lower the stock prices of the two firms supporting the Fama run. If the average prices of the cross listed firms are use, the 1 percent increase in the previous quarter prices would result to the decline in the average prices by 42 percent. Results show that inflation has a negative effect on the overall stock returns in the short run. An increase in inflation by 1% results to a decline in the average stock prices by 3.3 percent in the short run. The results are in conformity with the Fama's hypothesis that in periods of higher inflation, consumers tend to invest less but also firms running costs becomes greater causing share prices to drop. Analysis of the individual stock reveals that, an increase in the inflation rate by 1 percent results into decline in the KCB stock prices by 15 percent in Tanzania in the short run, while it is not statistically significant for the remaining 4 regression models.
The short run effect of interest rate on the overall stock prices is found to be not significant. However, when individual stocks are examined, some of the stocks seem to be significantly affected by the changes in the interest rate. An increase in the interest rate by 1 percent results into decline of EABL, JHL and NMG stock prices by around 3, 4 and 14 percent respectively in Tanzania in the short run. Such results suggest possibility of investors to switch to higher interest paying and less risky investments in the short run when interest rate raises. On the other hand, a 1 percent increase in interest rate result into increase in the KCB stock prices by around 6 percent in Tanzania in the short run.
Similar to the long run results, GDP does not seem to have a significant effect on the overall stock prices in the short run as well. However, it is found to have significant effect on some of the firms in the short runs. An increase in the GDP growth rate by 1 percent results into increase in EABL stock prices by around 5 percent and decline in the JHL and KCB stock prices by around 4 and 14 percent respectively in Tanzania in the short run.

Results for Uganda
Findings from Table 4.4 reveal that there is adjustment back to equilibrium by 36, 9, 34, 34, and 42 percent per quarter for the EABL, JHL, KCB, KA and NMG models respectively in Uganda. The R squared obtained show that the independent variables explain the dependent variable by 48, 44, 54, 53, and 63 percent for the EABL, JHL, KCB, KA and NMG models respectively. Furthermore, findings show that there is neither heteroscedasticity nor autocorrelation problems for this study since the p-vales obtained for these tests are not significant hence the study fails to reject the null hypotheses of constant variance and of no serial correlation.
The overall results as analysed in model 6 indicate that inflation has a negative effect on the stock returns in the long run. As shown in table 4.4, a 1 percent increase in inflation results to 11.6% decline in the stock price. This suggests that at higher inflation rate, stocks become less attractive which may be caused by higher production cost, decline of economic activities and reallocation of resources from investment to consumption. Such results are similar to the ones obtained in Tanzania.
More specifically, there is a negative and significant effect of inflation on the returns of EABL, KCB and NMG while a positive effect is observed for the returns of KA. The negative effect implies that increases in inflation in Uganda would result to decreases in the prices of shares of those cross listed firms. show that in models 1, 2, 3 and 4 the coefficient on GDP growth rate is statistically not significant but in model 5, the coefficient is statistically significant at 5 percent level.
Results in table 5 show that the inflation has a negative effect on the overall price of the stocks of the cross listed firms in the short run. Inflation influences value of financial assets determined by the firm's net earnings. Thus, in a highly inflationary economy stock prices are adversely affected due to decline in investment as a result of higher prices of factor input. Only one firm (NMG) is found to be negatively and significantly affected by the inflation. Similar to the case in Tanzania, the short run effect of interest rate on the overall stock prices is found not to be significant. However, when individual stocks are analysed, JHL is found to be significantly affected by the changes in the interest rate. An increase in the interest rate by 1 percent is observed to results into a decrease of 1.8 percent in price of the JHL shares.
GDP growth rate has no significant relationship with the stock returns at 1%, 5% and 10% significant levels when the average prices of the cross listed firms is used. The results mean that changes in the growth rate do not have any impact on the stock prices. This could be due to the premise that the most of the cross listed companies in Uganda are not in the sectors that contributes more to the GDP. The leading contributor of GDP growth in Uganda is service industry with wholesale and retail services leading in terms of percentage share to GDP. The findings of this study are consistent with some empirical studies by Patatoukas, and Kirui et al, (2014), which also found no significant relationship exists between GDP growth rate and stock returns. This further suggests that firms are more related to the particular stock market where they are listed and less in the general economy.

Conclusion
The paper made attempt to determine the effect of the selected macroeconomic factors of the host country on the stock market performance of the cross listed firms in East Africa. This was achieved first by determining the impact of macroeconomic factors on the returns of individual firm and then establishing such an impact on the combined average prices of those firms. The study found that inflation has both short run and long run negative effect on the overall stock returns in both markets. Contrary to the general belief, the study found interest rate has positive impact on the combined average stock returns in both markets. However, analysis at individual firm level showed both inflation and interest rate, had both positive and negative effect on stock returns. GDP growth rate is found not statistically significant factor that affect stock returns.
The study concludes that macroeconomic variables exert effect on stock returns of the cross listed firms in East Africa. There are mixed results in terms of magnitude and direction of effect. By treating the selected macroeconomic variables as risky factors, the paper is in support of the APT. This study can have valuable contribution to the body of knowledge especially on the nexus between macroeconomic variables on stock returns. It also contributes to the players in the stock market including investors, companies, supervisors and regulators by shading light on the effects of the selected macroeconomic variables on performance of cross listed firms. The results obtained in this study can be used by all the group of players depending on their decision needs. Future studies may consider modeling the home country macroeconomics and stock returns. Also future studies may consider studying analyzing the home country versus host country macroeconomics on stock returns of the cross listed firms.