Supply Response of Rice in Ghana: A Co-integration Analysis

Abstract

This study presents an analysis of the responsiveness of rice production in Ghana over the period 1970-2008. Annual time series data of aggregate output, total land area cultivated, yield, real prices of rice and maize, and rainfall were used for the analysis. The Augmented-Dickey Fuller test was used to test the stationarity of the individual series, and Johansen maximum likelihood criterion was used to estimate the short-run and long-run elasticities. The land area cultivated of rice was significantly dependent on output, rainfall, real price of maize and real price of rice. The elasticity of lagged output (12.8) in the short run was significant at 1%, but the long run elasticity (4.6) was not significant. Rainfall had an elasticity of 0.004 and significant at 10%. Real price of maize had negative coefficient of -0.011 and significant at 10% significance level. This is consistent with theory since a rise in maize price will pull resources away from rice production into maize production. The real price of rice had an elasticity of 2.01 and significant at 5% in the short run and an elasticity of 3.11 in the long run. The error correction term had the expected negative coefficient of -0.434 which is significant at 1%. It was found that in the long run only real prices of maize and rice were significant with elasticities of -0.46 and 3.11 respectively. The empirical results also revealed that the aggregate output of rice in the short run was found to be dependent on the acreage cultivated, the real prices of rice, rainfall and previous output with elasticities of 0.018, 0.01, 0.003 and 0.52 respectively. Real price of rice and area cultivated are significant 10% level of significance while rainfall and lagged output are significant 5%. In the long run aggregate output was found to be dependent on acreage cultivated, real price of rice, and real price maize with elasticities of 0.218, 0.242 and -0.01 respectively at the 1% significance level. The analysis showed that short-run responses in rice production are lower than long-run response as indicated by the higher long-run elasticities. These results have Agricultural policy implications for Ghana.

Key Words: Supply response, Rice, Error Correction Model, Co-integration Analysis, Ghana