The asymmetric J-curve phenomenon: Kenya versus her trading partners

Abstract

This paper examines asymmetries in the J-curve effects of real exchange rate on Kenya’s trade balance by using panel data for bilateral trade with 30 trading partners. The data covers the period from 2006q1 to 2018q4 and the Pooled Mean Group (PMG) estimation technique, under both the linear and nonlinear ARDL frameworks, is applied. This paper departs from previous studies by using a modified version of the standard trade balance model, which is more suited for bilateral trade analyses, and by incorporating nonlinearities. The findings of the PMG estimation based on the assumption of symmetric exchange rate effects reveal J-curve effects in only 7 bilateral trade relations. However, when the estimation is performed assuming asymmetric effects, the J-curve effects are evident in 13 cases. Long-run and short-run asymmetries are also confirmed and it is established that a simultaneous bilateral real depreciation of the exchange rate boosts the long-run trade balance. The implication of these findings is that a devaluation policy can be used to raise competitiveness of Kenya’s exports in the long-run.

Keywords

• Bilateral trade balances
• Exchange rate
• Asymmetric effects
• J-curve
• Kenya
• Panel data