Que fatores afetam os níveis de preços dos países? um olhar sobre o impacto da produtividade agrícola e dos custos de comércio

Abstract

This article sought, by means of cross-sectional regressions, to analyze deviations by the nominal exchange rates in relation to the theory of purchasing power parity (PPP) – in other words, to explain the differences in the price levels of countries after conversion to a common currency. Following closely the structure of cross-sectional research on this topic – especially the work of Gelb and Diofasi (2016) – we included in the equation of deviations from the PPP some variables little explored in this theme, such as agricultural productivity and trade costs. The estimation results confirmed Hassan‟s theory (2016), which postulates that agricultural productivity has a deflationary potential within the Balassa-Samuelson hypothesis, due to the low commercialization of this sector in most developing countries. Hassan (2016) also predicted a decreasing effect of this variable on the price level as countries develop, due to the decrease in the relative consumption of goods from this sector in relation to those of manufacturing and services. This pattern is also confirmed in the regressions, since agricultural productivity only becomes significant when interacted with such relative consumption, which decreases – on average – in GDP per capita, due to non-homothetic preferences. An alternative approach to capture deviations in relation to the PPP was the inclusion of trade costs in the estimated equations, since they may make price equalization linked to arbitrage more difficult. Here, we sought to measure such barriers by means of the implicit measure of trade costs derived by Head and Ries (2001) and Novy (2013) from gravitational models, by aggregating data on agricultural and manufacturing trade. The results of the regressions, however, showed little to no effect on the price level of countries for this measure of trade costs, probably due to the fact that this measure is restricted to the costs in the intensive margin, ignoring the costs on the extensive margin.