Drivers of Inflation Convergence across Countries: The Role of Standard Gravity Variables
One sentence summary: Having a common currency, a free trade agreement, proximity, a common border, or a
colonial relationship between countries increases the probability and decreases the half-life of
inflation convergence.
The corresponding academic paper by Hakan Yilmazkuday has been accepted for publication at Macroeconomic Dynamics.
The working paper version is available here.
Abstract
Using monthly headline inflation data covering 184 countries for the period between January 1971 and December 2020, this paper investigates the role of standard gravity variables on inflation convergence across country pairs. The convergence analysis by unit root tests is based on ten-year rolling windows to control for potential structural changes over time, whereas the corresponding results are connected to the standard gravity variables in the preceding year to investigate the drivers of inflation convergence and its speed. Regarding the existence of inflation convergence, empirical results show that having a common currency, a free trade agreement, proximity, a common border, or a colonial relationship between countries increases the probability of inflation convergence. Regarding the speed of inflation convergence, the very same gravity variables are shown to reduce the half-life of convergence. In both cases, the effects of having a common currency are shown to dominate those of other gravity variables.
Non-technical Summary
Lifetime welfare costs of inflation are significantly high. Recent studies suggest welfare costs of inflation up to 13% of one-year consumption following a 3% increase in inflation. As year-on-year headline inflation rates are highly different across countries and over time (e.g., Democratic Republic of the Congo having a year-on-year inflation of about 555% in 1994, whereas Equatorial Guinea having a year-on-year inflation of about -38% in 1986), investigating inflation differentials across countries and over time is important to understand the welfare inequality across countries. Accordingly, this paper investigates inflation convergence across bilateral countries to shed light on the relative welfare costs of inflation over time.
The inflation convergence analysis in this paper is achieved by using unit root tests for each country pair. Ten-year rolling windows using monthly data are considered to control for potential structural changes over time. Once inflation convergence is determined for any country pair, the corresponding speed of convergence is estimated based on half-life measures. The key innovation in this paper is to connect the pooled version of inflation convergence results (across country pairs and ten-year windows) to the standard gravity variables in the preceding period that are effective in explaining not only international trade (of goods and services) but also international finance (e.g., bilateral asset holdings).
As country-and-time specific factors are known to be effective in explaining inflation convergence in the literature, they are controlled for during the investigation of this paper to mainly focus on the role of gravity variables on inflation convergence. Moreover, to consider causality through the time dimension, inflation convergence results based on ten-year windows are connected to the standard gravity variables in the preceding year.
The motivation behind considering the standard gravity variables as well as country-and-time specific factors in explaining inflation convergence comes from a simple theoretical model combining the two well-known arbitrage conditions, namely the uncovered interest parity and the relative purchasing power parity. Specifically, it is shown that future (expected) inflation differentials between any two countries depend on the current country-specific nominal interest rates (e.g., reflecting monetary policy, exchange rate regime, or business cycles of countries) as well as the deviations from the relative purchasing power parity that can be captured by the standard gravity variables.
The inflation convergence results based on ten-year windows for the monthly period between January 1971 and December 2020 covering 184 countries suggest that certain country pairs have experienced inflation convergence for each and every ten-year window, whereas certain others have not experienced any inflation convergence in any of the ten-year windows. Conditional on having convergence, the results also suggest that there is significant evidence for heterogeneity across country pairs regarding their speed of convergence (based on half-lives).
The heterogeneity across country pairs regarding their inflation convergence and the corresponding half-lives is further investigated in secondary analyses by estimating the effects of standard gravity variables in the current year on the inflation convergence and its speed within the next ten years. The corresponding results suggest that robust to the consideration of country-and-time fixed effects, having a common currency, a free trade agreement, proximity, a common border, or a colonial relationship between countries increases the probability of inflation convergence. For the speed of convergence (conditional on convergence), the very same gravity variables are shown to reduce the half-life of inflation convergence across countries. When the effects of alternative gravity variables are compared in terms of their magnitude, the effects of having a common currency are shown to dominate those of others.
Regarding policy implications, as inflation convergence across countries is an indicator of welfare improvement, international policies toward having a common currency or a free trade agreement would be beneficial for countries in a significant way. As having a common currency dominates the effects of other gravity variables, policy makers may want to prioritize having common currencies with other countries if they would like to benefit more from welfare-improving inflation convergence with other countries. These implications are robust to the consideration of not only country-and-time specific factors but also certain measurement errors and alternative window lengths used for the investigation of inflation convergence.
The corresponding academic paper by Hakan Yilmazkuday has been accepted for publication at Macroeconomic Dynamics.
The working paper version is available here.