Is there really a "border effect" ?

The observed excess price variability in cross-border city pairs compared to that in within-country city pairs has been defined as the "border effect." We used a unique data set from cities that were in the same country at one time and were in two separate countries later on to examine the effects of the presence of a national border on price variability. Interestingly, a border-like effect was detected even during the period when all the cities were in the same country. We also found a large border effect when cities were in two separate countries. However, we found no change in the price variability at cross-border city pairs during the periods both before and after the cities separated into two different countries. This finding suggests that the observed systematic higher variability of consumer prices in cross-border city pairs might not be due to the presence of a border as suggested in the literature. (c) 2007 Elsevier Ltd. All rights reserved.