What to Expect of the Euro? Analysing Price Differences of Individual Products in Luxembourg and its Surrounding Regions

The European price landscape is characterised by significant price differences among regions and countries for quasi-identical goods. Possible explanations include transport costs, tariffs and non-tariff barriers, market structure, etc? While the introduction of the single European currency reduces exchange rate volatility and hence price volatility across its participating member countries, it is yet unclear to which extent this also implies reductions in price dispersion across countries. Similarly, the introduction of the euro and the associated increase in price transparency will reduce consumers' transactions costs associated with cross-country price comparisons. It remains to be seen whether these transactions cost reductions are sufficient to trigger a reduction in the band of inaction within which prices can fluctuate without triggering consumer arbitrage and whether cross-country price dispersion will be induced to decrease. The latter has rather frequently been claimed to be one of the main benefits of the euro. The aim of this paper is to analyse to which extent the euro may be expected to contribute to increased regional price convergence - a topic being directly related to the euro cash changeover. It is obviously too early to evaluate the price and convergence effects, not to mention the economic benefits, brought about by the introduction of the euro. The cash changeover transition period has barely ended yet. Moreover, many economic effects will only materialise in the long-term. We use individual product prices from the Luxembourg and the surrounding regions Lorraine, Rhine-Palatinate and Walloon collected on several occasions between October 2001 and April 2002 in order to analyse the factors contributing to regional price differences. Luxembourg and its surrounding regions emerge as natural candidate for such a study. At relatively short distance to each other four countries border each other. Furthermore, these regions are highly integrated with each other, which is also apparent from the about 100 000 cross-border commuters travelling into Luxembourg every day. This reduces the obstacles, which somewhere else may effectively seal off regions against each other. The high degree of regional integration also means that a high share of the population was used to compare and pay prices in different currencies prior to the introduction of euro. If we were able to obtain statistically affirmative results in these almost ideal conditions, then this would nurture our hopes for measurable effects for the whole euro area. If this was not the case, then it may prove rather difficult to find price level convergence effects - even in the long-term. The empirical results obtained in this paper generally support the transactions cost arguments. More specifically, the absolute percentage price difference is increasing in distance, but at a decreasing rate. Similarly, crossing borders increases the price deviations, while membership to the former Belgo-Luxembourg currency union reduces price deviations. These results are rather remarkable given the high degree of homogeneity with regard to the geographical area under investigation. These results raise our hopes that the euro will be able to reduce price dispersion in the euro area in the long-term. The data set also allows us to explicitly analyse how packaging size differences influence the deviations from the law of one price. This is a particular feature of this data set, which many other international data sets do not have. The results are affirmative. The larger the differences in unit value representations the larger are price deviations. Thus, this result supports the statement of the European Commission (2001c) arguing that market structure matters. Furthermore, being part of the same retail chain matters. The results clearly suggest that price deviations are smaller on average, if the prices are observed in the same supermarket chain.