اثر مرز در اقتصادهای کوچک باز

This paper examines the importance of the national border in relative price variability in two neighboring, small open economies. Using monthly frequency price data of narrowly defined, homogenous consumer products, it finds that the time-series variation in within-country relative prices is about the same in the two countries. After controlling for distance, relative price variation is significantly higher across than within countries. The border is the dominant determinant of relative prices, even after accounting for nominal exchange rate variability and local culture as represented by language spoken. Our estimates of the border effect are largely immune to the bias identified in Gorodnichenko and Tesar [Gorodnichenko, Y., Tesar, L., 2006. Border effect or country effect? Seattle is 110 miles from Vancouver after all. Unpublished manuscript].

A key issue in international macroeconomics is the response of relative prices and quantities to fluctuations in exchange rates. As a starting point in many New Open Economy Macroeconomics models, prices are fixed in the producer currency, generating Producer Currency Pricing, so that changes in the nominal exchange rate get fully passed through to local prices inducing relative price adjustment, which in turn turns on the ‘expenditure switching’ effect, making monetary policy effective under floating exchange rates. In this world, international markets are integrated, and the Law of One Price (LOOP) holds even across locations in different countries. At the same time, while one observes a general decrease in explicit barriers – for most part quantifiable, such as tariffs, quotas, transportation costs and other physical obstacles to travel – to international trade in recent decades, the fact that international markets are more segmented than intra-national ones seems to prevail. In a seminal paper Engel and Rogers (1996) provide evidence not only on the presence of significant market segmentation as reflected in persistent cross-country price differentials of goods belonging to one product category in the US and Canada, with the volatility of price differences depending on geographical distance, but also on the national border serving as an independent source of segmentation. The findings imply that the LOOP fails both within and across countries, but more strongly so in the latter dimension.1 What is particularly striking in the results of Engel and Rogers, echoed in subsequent work by Parsley and Wei (2001) and Beck and Weber (2003), is the magnitude of the ‘border effect’ in relative price differentials, the latter concept defined as the extra variability in relative prices not explained by distance per se. In particular, Engel and Rogers show that crossing the border between the US and Canada, countries with only minor differences in language and culture, is equivalent to traveling a distance of about 75,000 miles. The purpose of this paper is to quantitatively evaluate the importance of the national border in price setting in two neighboring, small but similar sized economies in Eastern Europe, Hungary and Slovakia.2 We focus on time-series properties of the deviations from the LOOP. After describing unconditional volatilities of good-level price differentials in the two countries, we estimate the extent to which barriers to international trade are important in explaining the relative volatility of cross-country price differentials.3 We also explore some key reasons potentially explaining the size of the border effect. The contributions of the paper are two-fold. First, it investigates the impact of national borders on international price differentials in a novel and unique sample of microeconomic prices. The sample draws on data of actual, monthly frequency transaction prices of 20 very narrowly defined goods and services, observed in a total of 56 locations in two small, neighboring countries over a period of 56 months. Relative to other similar studies seeking to provide evidence of the border effect in microeconomic prices, such as Crucini et al. (2005), Engel and Rogers (1996) and Parsley and Wei (2001), our data is specific in many ways, exhibiting both benefits and drawbacks for the purposes of the investigation. Crucini et al. (2005) study price differentials in a large, balanced, annual frequency panel of prices of 220 goods and 84 services, observed in 122 cities around the globe over an 11-year period. Engel and Rogers (1996) use a monthly and bi-monthly sample of price indices of 14 tradable and non-tradable product categories observed in 23 cities in Canada and the United States between June 1978 and December 1994. Parsley and Wei (2001) study quarterly frequency price observations of 27 tradable products in a total of 96 US and Japanese cities over a period of 88 quarters. Relative to these studies, besides the geographic proximity and macroeconomic similarity of the two countries involved, the main advantage of our data set lies in the fact that the goods and services we examine are fully identical over all locations and time, and that actual transaction prices are observed at a high, monthly frequency.4 These features of the data all contribute to reducing the importance of observations of relative price adjustments (or the lack of them) that are solely due to changes in the identity of products over time, across items or locations, and to alleviating censoring problems potentially present in lower frequency price data. Second, the paper focuses on an episode where the countries involved show very similar within-country variation in price differentials. Gorodnichenko and Tesar (2006) demonstrate that the border effect estimated in previous studies confounds the impact of the true border and the extent of cross-country heterogeneity in relative price variability. They also suggest that altering the specification developed in Engel and Rogers (1996) in a simple way allows one to quantify the border effect relative to country specificity in relative price variability. By focusing on an episode with similar time-series variability in relative prices in the two countries, our analysis is able to get around the country heterogeneity problem in a natural way. The rest of the paper is structured as follows. Section 2 describes the data. Section 3 compares price variability within and across countries, and Section 4 quantifies the border effect in the baseline specification. Section 5 provides information on potential determinants of the border effect. Section 6 concludes.

Using retail prices of 20 individual homogenous items grouped into four product categories (durable goods, meat products, other food products and services) observed at 20 locations in Hungary and 36 locations in Slovakia over a period of 56 months, this study estimated the border effects in two small, neighboring, open economies, Hungary and Slovakia. The main advantage of our empirical approach is the focus on an episode where locations are geographically close to each other, international trade barriers are relatively small and vanishing, the physical characteristics of goods and services are homogeneous over time and across locations, and within-country relative price variability is about the same in the two countries. The overriding message of the paper is that the border does matter. That is, in various specifications based on, and extending the one in Engel and Rogers (1996), we find that the national border has an independent, sizeable, statistically significant impact on relative price variability. At the same time, the impact of transportation costs as proxied by distance between locations is much less pronounced. These results are robust to accounting for nominal exchange rate variability, differences in local culture as represented by language spoken and cross-country heterogeneity in relative price variability.