بررسي تاثير تكانه هاي نرخ ارز رسمي بر اضافه قيمت بازار سياه ارز در ايران : يك الگوي دو مرحله اي

Impact of Shocks to Official Exchange Rate on Black Market Premium in Iran: A Two-Stage Model

Using an Autoregressive Distributed Lag approach to cointegration analysis, this paper examines the impacts of anticipated and unanticipated shocks to official exchange rate on black market exchange rate premium in Iran for the period 1980:1 - 2001:1. Following Barro (1977), Hoffman et al. (1984) and Pozo and Wheeler (1999) a two-stage model is developed to study the long-run movement of black market premium. In the first stage, we construct a forecasting equation for the official exchange rate. The predicted values of official exchange rate obtained from this equation are used to measure the anticipated shocks. The residuals of the estimated forecasting model are used as a measure of unanticipated shock to official exchange rate. In the second stage, we regress the black market premium on anticipated and unanticipated shocks derived in the previous stage. The results indicate that both anticipated and unanticipated shocks to official exchange rate have a significant negative impact on the black market premium in Iran. It is found that an unanticipated increase in the official exchange rate has a greater impact on the black market premium thap the anticipated one. Using dummy variable technique to study the possibility of structural break in the premium movement, the paper finds that the policy of exchange rate unification of 1993 has changed the intercept and the slope of the premium equation. This means that this policy has caused a structural change in the premium movement. More specifically, the intercept has decreased and the slope has increased dramatically after the unification. Indeed, the anticipated devaluation of official exchange rate has had a positive impact on the black. market premium between 1993 and 1995. Finally, the cointegration tests verify the existence of a long-run equilibrium relationship among variables.