The burgeoning academic literature on target zones has developed a strong empirical focus and has not shied away from producing negative results, but it has strayed from the point of the original, policy case for their introduction: to reduce the perceived speculative inefficiency of currency markets and stabilize expectations. The simple model assumes efficient, rational currency markets and shows that a target zone will then be stabilizing rather than destabilizing. In Discussion Paper No. 718, Research Fellows Paul Krugman and Marcus Miller maintain that this excessive rationality on the part of investors is both the most serious flaw of the model and the best justification for instituting target zones in practice. They develop a more sophisticated model with two groups of traders: those who require a risk premium to shift their currency portfolios and do not employ trigger strategies to limit their exposure, and `stop-loss' traders who are less risk averse and exit (and enter) at given trigger points.

Krugman and Miller show that the beliefs that motivate policy- makers to operate target zones contradict the standard model's basic premises but are supported by empirical evidence. They argue that stop-loss trading leads to excessive volatility by making currency markets more sensitive to economic fundamentals than rational expectations models imply. Between these traders' entry and exit triggers, the exchange rate has an inverse S-shaped relationship. If target zones assure informed traders that stop-loss orders will not be triggered, however, speculation instead becomes stabilizing, as the market anticipates central bank intervention, and the inverse S- shaped curve reverts to the familiar target zone solution. This result depends critically on the assumptions that a significant part of the market follows a more-or-less mechanical investment strategy and that the central bank sets its target zone neatly within the range that provokes stop-loss orders. This may limit the model's empirical usefulness, but it demonstrates that the policy case for target zones depends not on an idealized view of efficient asset markets but rather on how to manage markets that may be speculatively inefficient.

Why Have a Target Zone?
Paul Krugman and Marcus H Miller

Discussion Paper No. 718, October 1992 (IM)