June 1, 2006
Juann H. Hung and Young Jin Kim
This paper examines past currency crises to shed light on the likelihood that the adjustment of the U.S. current account deficit will involve a dollar crisis. A currency crisis is narrowly defined to be a depreciation that exceeds a critical threshold, regardless of whether it has an adverse effect on the real economy. The literature suggests that one should not infer from the experience of emerging economies what will happen to the dollar. This paper’s empirical findings lend support to that view. Everything else being equal, currencies are more likely to collapse in emerging economies than in industrial countries. The collapse of a currency in industrial countries actually helped to revive economic growth in some instances, even though it tended to cause severe output loss in emerging economies with a pegged currency.
The estimated probability of a dollar crisis, which contains a significant upward bias, has risen significantly from 2003 (about five percent) to 2006 (about 10 percent), a period during which both the current account deficit and the oil price rose substantially. That rise in the estimated probability, though in part due to the oil price hike, signals that the probability would keep increasing if the U.S. current account deficit continues to grow faster than GDP. Nevertheless, a more thorough analysis taking into account of many factors beyond the regressions suggests that the adjustment of the U.S. current account deficit is likely to be gradual rather than involving a dollar crisis; and, in the unlikely event of a dollar crisis, the U.S. economy should be able to withstand it without suffering a severe recession.