Causal and cointegrating relationship between macroeconomic variables on stock market prices in Nairobi securities exchange

Abstract

The study of stock market price movements and macroeconomic indicators has been imperative in view of the country’s economic growth because the most sensitive segment of any developing economy is its stock market. The buy and sell decision rules are affected by the investors’ psychology which exerts influence on the macroeconomic events. The critical question when it comes to this is that how instantaneous the information is transferred to the investors and market analyst and in return, reflects on stock market prices. Therefore, the purpose of this research was to analyze causal and cointegrating relationship between macroeconomic indicators and the stock market prices in the context of Nairobi Securities Exchange, Kenya. The study’s specific objectives were; to determine the relation between inflation, exchange rate, interest rate, nominal gross domestic product and stock market prices. Further, the study aimed at investigating the bidirectional Granger causal effect between the selected variables in this study. Efficient Market Hypothesis, Arbitrage Pricing Theory and Capital Asset Pricing Models theories guided this study. The study used longitudinal research design and employed monthly secondary data for the period 2005 - 2018. The data was obtained from Nairobi Stock Exchange, Kenya National Bureau of Statistic and Central Bank of Kenya. Descriptive statistics such as mean, minimum, maximum and standard deviation were computed to understand the nature of data and other general characteristics. Augmented Dickey Fuller, Philip Perron and Clemente-Montañés-Reyes tests confirmed the presence of unit root at levels, and all the variables attained Stationarity after first difference. The Optimum lag length selected was 2. Johansen’s cointegration test showed that the variables were cointegrated thus Vector Error Correction Model was estimated. The error correction term was −1.1804 and significant at p − value 0.000 which indicated a long-term relationship among. Jarque-Bera test showed the residuals followed normal distribution. There was no serial correlation among the variables as per Durbin Watson statistic(DW − Stat 2.022 < 4). Inflation and interest rate was found to negatively and significantly affect stock market prices with coefficients of −0.8371 (p − value 0.005) and -4.0876 (p − value 0.000) respectively. However, exchange rate and nominal gross domestic product had positive and significant effects on stock prices at 0.0001 (p − value = 0.012) and 0.00002 (p − value = 0.000) respectively. It is recommended based on the findings that the government should adopt expansionary monetary policy to enhance credit creation and curb interest rate by stabilizing exchange rate changes thus promoting investment in stocks and shares. There is need for the government to encourage activities that increases gross domestic product since it is an important macroeconomic indicator for health economy.