Flood and Garber illustrated how a floating exchange rate will anticipate a `state-contingent' regime change, i.e. a change that depends on the achievement of specified economic conditions, using a simple `monetary model' of the UK's return to the gold standard in the early 1920s. Smith and Smith developed this model to obtain the surprising result that the government's well-known desire to return to gold at the pre-war parity of $4.86 weakened sterling. In Discussion Paper No. 465, Programme Director Marcus Miller and Alan Sutherland modify this model by treating the trend tightening of UK vis-à-vis US monetary policy as endogenous, and they find that it then predicts the expected strengthening of sterling. They also modify the `state-contingent' model to allow events fixed in calendar time to play a role in triggering the return to gold and show that such `time-dependent' solutions add to this strengthening of sterling.

Miller and Sutherland justify these modifications on the grounds that they bring the monetary model closer to historical reality. Its assumption of perfect flexibility of goods prices is a poor representation of reality in the 1920s, when policy-makers are known to have been anxious to secure their adjustment to a level consistent with the `target' exchange rate. Their model with sluggish adjustment of goods prices provides a more plausible interpretation of the state-contingent joining condition: the achievement of equilibrium competitiveness at $4.86. The fact that the UK was still perceived to be uncompetitive when sterling rejoined the gold standard in 1925 indicates the importance of the time-dependent joining condition.

Miller and Sutherland argue that the regime change involved in the UK's entry into the ERM of the EMS can be subjected to a similar analysis, since the state-contingent conditions concerning relative inflation rates had been explicitly laid down, while time-dependent considerations were also influential. They concentrate, however, on the principal difference between the events of late 1990 and of the 1920s, i.e. the lack of credibility attached by the private sector to the pegging of sterling. They show that in a model of this `peg with a peso problem', in which labour market participants take account of the possibility of realignment when forming wage contracts, this lack of credibility offsets some or all of the benefits of pegging against a hard currency. Further, a serious commitment to pegging that lacks credibility may be very costly in terms of lost output.

Britain's Return to Gold and Entry into the EMS: Expectations, Joining Conditions and Credibility Marcus Miller and Alan Sutherland

Discussion Paper No. 465, October 1990 (IM)