Carlos A. Ibarra
currency.5 Note, however, that if today’s currency depreciation leads
to a lower expected depreciation rate, then interest rates (except
perhaps those for very short-term instruments) will fall, tending to
offset the contractionary output effect of the capital account shock.
The automatic output stabilizing mechanism will not work, of
course, if after a currency depreciation the interest rate differential
fails to decline (e.g., because of an adjustment in risk assessments or
expectations of future public policies —as explored in Section 4); it
may even increase, in which case the output effect of the capital account
shock would be reinforced.6 In such case, the conduct of monetary
policy would also be made more complicated. In particular, if a currency
depreciation is not able to restore asset market equilibrium (Equation
1) at a lower peso interest rate, then a counter-cyclical policy response
to the capital account shock could destabilize financial markets.
II. Empirical analysis
Having briefly discussed the macroeconomic role of the interest
rate-exchange rate link in financially open economies, this section turns
to an empirical analysis of the recent Mexican experience. In particular,
it presents estimation results for the dynamic response of peso-dollar
interest rate differentials to a permanent variation in the peso-dollar
exchange rate. The analysis uses weekly interest rate data on Mexican
and US Treasury bills since June of 1996. The impulse response
function is derived from an error correction model estimated by GMM.
a) Description
Figure 1 shows the evolution of the interest rate differential between
Mexican and US Treasury bills from January 1996 through July 2002,
5 Current work on the conduct of monetary policy and the choice of exchange rate regime
in developing countries is giving increasing attention to this type of shock, prompted by the
substantial increase in the volume and volatility of private international capital flows witnessed
recently See, for instance, Hausmann and Rojas-Suarez (1996) and Eichengreen (2001).
6 Casual observation of this type of relationship led Calvo (1997), for instance, to conclude
that the conduct of monetary policy in Mexico suffered from lack of credibility. Eichengreen and
Hausmann (1999) argue that the positive contemporaneous correlation between interest rates
and the exchange rate in Mexico can be explained by the reluctance of authorities to follow anti-
cyclical policies in a context of currency and maturity mismatches in the banking sector.
10