Exchange Rate and Trade Balance; J Curve Effect in Ethiopia

Abstract

More recent   studies   have   used   bilateral   data   in   testing   the J curve    phenomenon. However, The  paper  has  attempted  to  estimate  the  J-curve  phenomenon  for  Ethiopia  using  a yearly data ,aggregate data  over the period 1974-2010 in a  multilateral trade model  basis.

A  recent   technique   in  co integration, ARDL approach , more importantly the Hendry’s general to specific approach   has  been  employed,  which  allowed  for  a  simultaneous  short-run  and  long-run  analysis  of  the  trade  balance  model.  Empirical  results  suggest  there  is  strong evidence for the fulfillment of the Marshall-Lerner condition, as the trade balance improves  in the  long  run  in  response  to  a  currency  devaluation, while giving rise to a J-curve effect in the short   run.      Both   Johansen’s   and   autoregressive   distributed    lag approach     are respectively     used.    Co integration  test results  showing   that   LREERI ,LTB LRGDP and LIPI move all together in long run.   Corresponding   error- correction models as well as impulse response functions indicate that, following currency devaluation trade balance first deteriorates before it later improves, i.e. exhibiting the J-curve pattern. These results are relevant for policy making instruments in Ethiopia. It is likely that such policy may be able to produce the desired outcome—i.e., the trade balance could improve. The       policy   issue   that arises from    these observations is the usefulness of the real exchange rate as a tool for correcting trade imbalance.

The Granger Causality test suggests that REER, LRGDP, LIPI does Granger causes trade   balance. As the research works comprise two regimes, the stability analysis was checked.

Key words: Trade Balance, Real Exchange Rate, Autoregressive Distributed Lag (ARDL), Error Correction (ECM)