An Empirical Investigation of the Ratchet Effect in the OECD, 1985–2009

Abstract

This paper employs a reduced form structuralist model of inflation in the OECD over the period 1985–2009 to find out whether domestic prices respond symmetrically to rising and falling import prices. We find that the response is asymmetrical: domestic prices rise when import prices rise but they do not fall when import prices fall. Our finding thus confirms the presence of a ratchet effect in the sample countries during the sample period, and implies that factors – such as exchange rate fluctuations and movements in tariff rates – that influence import prices tend to be inflationary.

Keywords:

• Ratchet effect
• exchange rate pass-through
• tariff rate pass-through

JEL Classifications:

• E31
• F41

Notes

¹For a recent survey on ERPT and TRPT, see Ghosh and Rajan (2007).

²Such findings do not suggest that trade liberalization does not benefit the economy. Trade liberalization has been shown to be a necessary (although not a sufficient) condition for economic growth. In the vast literature on this point, see Lindert and Williamson (2003).

³Rassekh and Wilbratte (1990) find that a structural model performs better than a monetary model in explaining the inflation rate.

⁴The countries are Australia, Austria, Belgium, Canada, Denmark, Finland, Germany, Greece, Hungary, Iceland, Ireland, Italy, Japan, South Korea, Mexico, New Zealand, Norway, Portugal, Spain, Sweden, Switzerland, Turkey, the United Kingdom, and the United States.

⁵In fact we calculated the import prices from the data available on the World Bank website. First we downloaded the dataset for imported goods and services in both current and constant local currency. Next we calculated the import prices by dividing the imports evaluated in current prices by the imports in constant prices. 2005 is the base year.

⁶The Monte Carlo simulation results show that LSDVC is suitable for the dynamic panel data technique in small samples.

⁷We will provide the results upon request.

⁸Our results differ from those of Rassekh and Wilbratte (1990) for two reasons: first, our sample countries and sample period are much larger and longer; second, we apply certain econometrics techniques that were developed in the 1990s and thus were not available when Rassekh and Wilbratte did their work.