The Determinants Of Stock Returns In The Emerging Market Of Kenya: An Empirical Evidence

This study examined the determinants of stock returns for the emerging market of Kenya: The Nairobi Securities Exchange.The specific objectives were to establish if certain selected macroeconomic variables affect stock prices in Kenya,to examine the validity of the Capital Asset Pricing Model(CAPM)in the Kenyan market and to determine the sources of risk factors priced in stock returns in the Kenya market.Based on this,the study then documents the risk premiums(betas)for firms listed at the Nairobi Securities Exchange.To answer the study objectives,the thesis is organized into three empirical parts and constitutes five chapters in total.Chapter one introduces the subject under study in a broad perspective,deduces the study problem and/or research questions and offers the motivations for carrying out this research.Chapter two constitutes the first part of the study and addresses the first objective outlined for study.The chapter identifies the selected macroeconomic factors that affect stock prices in Kenya,and establishes their long run equilibrium relationships.The content of this chapter is a published article in the International Journal of Economics and Finance(IJEF),available online from December 2016 as part of the background to the study as a whole.Chapter three constitutes the second part of the study.The chapter presents exploratory empirical tests on the factors identified in chapter two.The chapter also introduces an additional factor,the market premium to the macro variables identified in chapter two.The chapter examines if the widely acknowledged sentiments of “abnormal profits” for the banking sector in Kenya are reflected in stock returns at the Nairobi Securities Exchange.Again,the content of this chapter is a published article in the International Journal of Economics and Financial Issues(IJEFI),available online from January 2017.The study utilizes time series data organized into monthly periods.Daily trading data was obtained from the Nairobi Securities Exchange for the period April 1996: December 2016.The NSE 20 share index(proxy for the market factor)and the Consumer Price Index data was obtained from the monthly economic indicator reports available online at the Kenya National Bureau of Statistics.Monthly average exchange rates and the 91-day Treasury bill rates(proxy for short term interest rates)were obtained from the Central Bank of Kenya,available online.In all the three parts of the study,the time series data set has been checked for stationary and other time series technicalities.Five approaches/methods of research have been utilized in total to complete the data analysis and answer the study objectives.In the first part,the Johansen Co-integration technique and the Vector Error Correction Model are used to establish the long run equilibrium relationships between the selected macroeconomic variables.A multifactor regression model is used to check the robustness of the co-integration test.In the second part,the CAPM model is first applied on portfolios constructed along the NSE sector classification.The CAPM is then extended into a multifactor model by adding the identified macroeconomic factors from chapter two.The third part utilizes first the CAPM using individual stock returns to further examine its validity in the Kenyan market.In the second section the Fama and Mac Beth(1973)twostep procedure is applied to identify the sources of risk factors priced in the market using individual stocks as opposed to portfolios.A robustness check is done by constructing six portfolios based on size(market capitalization).Further checks are done by testing the between estimator regression for the individual stocks.The factors examined in the three parts of the study are the excess market premium(proxy the NSE 20 share index)and four macroeconomic variables including unexpected changes in inflation(?CPI),changes in exchange rates(?EX),changes in broad money supply(?Ms)and changes in short-term interest rates(?ATB).The data was obtained from the Nairobi Securities Exchange,the Kenya National Bureau of Statistics and the Central Bank of Kenya.Monthly data is used for the analysis.The study finds evidence to suggest that there indeed exist a long run equilibrium relationship between the selected macroeconomic variables and the stock prices.Unexpected inflation is found to have a negative,but statistically insignificant long run equilibrium relationship with stock prices in Kenya.Money supply has a negative and statistically significant long run equilibrium relationship with stock prices.The exchange rates and short term interest rates are both found to have positive and statistically significant long run relationships with stock prices.The findings of this part of the study contradict most of the documented evidence available for the Kenyan market,especially on the direction of the effects.In the second part,the study finds that CAPM cannot be rejected in the Kenyan market.Further,predicted returns from the multifactor model suggest that indeed the sentiments of the “above normal profits” for the Kenyan banking sector are reflected in the stock market.The evidence shows that an all bank portfolio outperforms portfolios constructed along the other sectors of the Kenyan economy.The findings of the third part also suggest that the CAPM cannot be rejected for the Kenyan stock market using the individual stock returns.Further,the study finds evidence to suggest that excess market premium(?)is statistically significant as a source of risk factor for the Kenyan market.The macro variables are found to be statistically insignificant.This study concludes that in the absence of the intervening industrial production variable,there exists a negative,but insignificant long run equilibrium relationship between inflation and stock prices in Kenya.Money supply has a statistically significant negative long run equilibrium relationship with stock prices through the intervening variable of inflation.Exchange rates and short term interest rates have both positive and statistically significant long run equilibrium relationships with stock prices.This study reverses the direction of the effects of macroeconomic variables documented in previous studies on the Kenyan market.Based on this,the study concludes that those earlier evidence could be attributed to spurious problems as a result of non stochastic processes in the models used.On the validity of CAPM,the study concludes that the CAPM model can be used to predict expected stock returns for the Kenyan market.Thus,the coefficients of the CAPM estimates could be presumed to be the excess market premiums(betas)for the individual stocks.Further,the study concludes that the excess market risk premium(?)is the most important factor in determining stock returns in the Kenyan stock market.The empirical evidence from this study provides insight into the determinants of stock returns in Kenya.The study proffers particular innovation in the sense that it utilizes an “out of the sample/box” data set to contribute into the ever evolving debate on the determinants of stock returns around the various markets in the world.Also,this study contributes new evidence into the body of knowledge especially from the least known Sub-Saharan Africa region.African markets offer opportunities as the future frontiers of growth and risk diversification for global investors as the Western and Asian developed markets,and East Asia and South American emerging markets get integrated due to globalization.