Why price‐level dispersion went up in Europe after the financial crisis

• Date: 01 March 2018
• DOI: http://doi.org/10.1111/twec.12591
• Author: Ad Stokman,Marco Hoeberichts
• Published date: 01 March 2018

ORIGINAL ARTICLE

Why price-level dispersion went up in Europe after

the financial crisis

Marco Hoeberichts

|

Ad Stokman

Economic Policy and Research Division, De Nederlandsche Bank, Amsterdam, The Netherlands

1

|

INTRODUCTION

Persistent price differences across countries are often seen as an indication of incomplete economic

integration of the countries. Engel and Rogers (1996) show that both borders and sticky prices in

combination with volatile exchange rates are key elements in explaining price differences across

countries. Their approximation of the US–Canadian border effect by a distance equivalent of

75,000 miles has challenged many to investigate the sources of international price dispersion. Gorod-

nichenko and Tesar (2009), for example, demonstrate that commonly used dummies to capture border

effects cannot separate border frictions from economic factors that give rise to different distributions

of price volatility within countries. Others investigated price differentials across regions within a sin-

gle country and thereby circumvent the problems associated with measuring country border effects.

See for a recent example Crucini, Shintani, and Takayuki (2010). Exchange rate volatility and ship-

ping costs are examples of time-varying segmentation effects that are found to be important determi-

nants of price dispersion (see, e.g., Parsley & Wei, 2001). Faber and Stokman (2009) consider long-

term trade developments among European countries as a measure of diminishing border effects.

In this paper, we build on Faber and Stokman (2009) who investigated long-term drivers in

price-level dispersion within Europe from 1960 up to 2003. We extend their analysis with recent

years up to and including 2015. Like Faber and Stokman (2009), but unlike many other studies

that use micro-price data, we use long time-series of consumer price levels. The long time span

puts us in a position to identify the time-varying component of border effects that are hard to dis-

tinguish for shorter episodes, and also to disentangle the long term from the short term. This

enables us to analyse in depth the effects of the Global Financial Crisis and the European Sover-

eign Debt Crisis. The macro-approach entails the risk of an aggregation bias, which may occur

when price levels at a more disaggregated level move in different directions. Faber and Stokman

(2009) demonstrated that price-level dispersion in Europe for seven product categories behaves

very similar. Therefore, aggregation bias is not an issue.

Finally, we compare price-level dispersion developments across European countries with cities

in the US. Although the focus of our research is on price dispersion across European countries, a

comparison with price dispersion across US cities adds an interesting perspective. Despite the limi-

tations of comparing price levels in countries in Europe to cities in the US and the data limitations

we have for the US analysis, the comparison allows us to identify factors that are specifically

related to European integration.

DOI: 10.1111/twec.12591

World Econ. 2018;41:913–925. wileyonlinelibrary.com/journal/twec ©2017 John Wiley & Sons Ltd

|

913