of key macroeconomic variables in the period leading up to speculative attacks, are still
consistent with first-generation models, but only with a restrictive sub-class in which no hint
of future policy imbalances is contained in past and current policy. But the fact that we find
little evidence of a shift in policy in more expansionary directions in the aftermath of
speculative attacks is difficult to reconcile this view.
A final important finding is that the behavior of macroeconomic variables differs
significantly around the time of speculative attacks on the one hand and realignments and
changes in exchange rate regimes on the other. ERM countries undergoing realignments
have significantly higher inflation rates, interest rates, rates of money and credit growth and
budget deficits, and their trade balances are significantly weaker. None of these statements
is true about the events associated with realignments of non-ERM currencies or with the
collapse of the Bretton Woods, Smithsonian, or Narrow Margin regimes of pegged exchange
rates.
Our investigation has obvious relevance to current policy concerns. 1992 and 1993
saw a series of speculative attacks on European currencies that drove the Italian lira and
the British pound out of the Exchange Rate Mechanism (ERM) of the European Monetary
System (EMS) and challenged the viability of the Maastricht blueprint for European
Monetary Unification (EMU). There remains considerable dispute over why these crises
occurred. One view emphasizes the unsustainable policy stances of weak-currency
countries (Dombusch 1993, Committee of Central Bank Governors 1993a,b, Williamson
1993, Goldstein and Mussa 1994). It blames EMS members whose currencies were
attacked for courting danger by their pursuit of lax monetary and fiscal policies and by