| Currency. Each of this dissertation's chapters has something to do with currency, and yet each is relatively different. The first paper is about currency crises, the second about currency substitution and the last about the likelihood that large holdings of cash are related to tax evasion and the underground economy.; In the first paper we look at how currency crisis and contagion can occur through the revelation of information, as well as through trade linkages. This paper sets up a centre-periphery model where the small probability of the periphery central bank being 'maniacal' allows it to fix its exchange rate to the centre nation. The model yields a 'near unique' equilibrium, where a regular central bank with 'intermediate to bad' fundamentals will fix for quite some time before being subject to the possibility of self-fulfilling attacks. As in previous literature, countries with 'good' fundamentals need not devalue. Contagion can occur when there is a correlation of policy types across nations, with a devaluation in one country yielding information about the rest's likelihood to defend their own pegs.; The second paper estimates the percentage of the currency that is outside the United States, Germany, and Switzerland. In doing so we also estimate currency demand equations using cointegrating vectors for Canada, the Netherlands and Austria. While we find that these equations have the 'expected' signs, the same functions estimated only for small denomination bills have the opposite signs on their coefficients.; Worldwide currency substitution has roughly tripled in the past decade in constant dollar terms. The United States, itself, had about 30% of its currency abroad in 1996. It was falling slowly over the 1960s and was as low as 5% in the early to mid 1970s, only rise again in the late 70s--early 80s and again during the 90s. For Germany the percentage is even higher at 69% in 1996. These estimates are lower than thaw of Porter and Judson (1996) who estimated about 50--70 of the U.S. currency was held abroad over the entire period.; Finally the third paper evaluates the literature on the underground economy by examining the assumption that the tax rate is a predictor of currency holdings. It uses modern cointegrating vectors techniques to estimate currency demand functions for 15 industrialised nations and finds that taxation is, for the most part, not significant or of the wrong sign. This finding implies that estimates of the size of the underground economy using the method proposed by Tanzi (1982) will be less than zero. |
