Drivers of horticultural exports in Kenya: co- integrating vector error correction approach

Abstract

Kenya’s economy is dependent on horticultural exports. The stability of these horti- cultural exports, however, is not stable in terms of their markets and it is not guaran- teed, as it is highly unstable. The main objective of this study was to find out the ma- jor drivers of horticultural exports in Kenya by utilizing monthly time series data ob- tained from IMF, Food and Agriculture Organization Statistics and Central Bank of Kenya for the period 2005 - 2017. It was hypothesized that inflation, exchange rate and interest rate have no effect on horticultural exports. Conventional unit root tests were performed to test for unit root using Augmented Dickey-Fuller and Philips- Perron and further, Kwiatkowski-Phillips-Schmidt-Shin unit root test was performed. Zivot-Andrews test was applied to test for unit root with one structural break and fur- ther and Clemente-Montañés-Reyes unit root test was used to test for unit root with multiple structural breaks. Unit root tests indicated that all the variables had unit root at levels and after first difference, they were stationary and thus integrated of order one I (1). Johansen’s test for cointegration was carried to test for cointegration and the results indicated the variables were cointegrated. VECM model was estimated to de- termine the long run relationship with respect to each of the variables. Diagnostic tests such as Jarque-Bera test for normality and it indicated data was normally distributed, Lagrange Multiplier test for serial correlation showed no serial correlation. Breusch– Pagan/Cook–Weisberg test for heteroscedasticity indicated that errors are homoscedasic and variance inflation factor showed no multicollinearity. Model stabil- ity was carried out and it was found that the model was stable hence the model was suitable for analysis and making statistical inferences. The results indicated that the variables were cointegrated at r = 2 of 11.37280 < 15.41 and greater than its crit- ical value at 5 percent level of significance and that there existed a long-term relation- ship between inflation rate, exchange rate and interest rate. VECM model showed that the error correction term of −0.0853 , which was statistically significant (p − value 0.0000 < 0.0500). Error correction term of −0.0853 showed that 8.53 per- cent of the adjustments are made in the first month and it takes approximately 11.72 months for the system to return to its long run equilibrium path. Inflation (p - value 0.00 < 0.05), exchange rate (p − value 0.03 < 0.05) and interest rate (p − value 0.0207 < 0.05) showed that they significantly affect horticultural exports). It is recommended that the government should intervene with commercial banks to fur- ther reduce interest rate. There is need to design policy aimed at stabilizing macroe- conomic environment to increase horticultural exports such as targeted exchange rate through application of foreign reserves adjustments.