A Non-Parametric Examination of Stability in the Euro

Abstract

Stage three of the introduction programme for the Euro began on January 1, 1999. This in many ways signalled the true beginning of the currency as thereafter the Euro was allowed for trade in foreign exchange markets and also in the issuance of new public debt. However it is already clear that this new currency will become a serious competitor of the Dollar in international usage. To illustrate, forecasts under a number of economic headings suggest that the Euro will be a closer competitor to the Dollar than having the Deutsche Mark as a vehicle currency (Portes and Rey, 1998; and McCauley, 1997). For example, in terms of all foreign exchange transactions, the Euro will be represented 50 to 60% (Dollar - 92%) of the time and the Euro will account for 25% (Dollar - 59%) of world trade invoices (McCauley, 1997). Other forecasts are even more optimistic suggesting equal status for these two currencies (Portes and Rey, 1998). Thus, the Euro is a very important currency for the fundamental welfare of the global economy and any concern over its stability is paramount. The backdrop for this analysis is that participating national currencies of the Euro were previously part of the Exchange Rate Mechanism that unfortunately suffered considerably from currency traders' speculative attack on a number of occasions. These attacks reached their zenith during the 1992/1993 crises leading to a radical revamping of the currency system. Generally, exchange rate fluctuations are measured unconditionally through an average standard deviation value, or conditionally using a GARCH related specification. However, the objective of this paper is to provide a plethora of information on the specific issue of whether speculative activity would unhinge participating governments' confidence in the currency. To achieve this, the paper concentrates on possible large scale exchange rate movements by applying a methodology specifically set-up to provide accurate measures of values located at the tail of the distribution, namely Extreme Value Theory. The findings are mainly encouraging. In many ways, the Euro shows similar characteristics to exchange rate series in general. For instance, the unconditional distribution of the Euro conclusively rejects the assumption of normality as well as having a unit root property and exhibiting volatility clustering. In terms of its stability, this paper finds there is no reason for concern regarding speculative pressure on the Euro. In fact, for a spectrum of probability and quantile measures, it consistently shows less volatility than the Yen. Moreover, the Euro follows a similar risk pattern to the Deutsche Mark, the currency that it is essentially replacing.