El impacto de los factores macroeconómicos en el tipo de cambio real en América Latina: Un análisis de datos de panel dinámico
| Autor | Carlos Cesar Chávez |
| Cargo | Universidad Nacional Mayor de San Marcos |
| Páginas | 6-31 |
Latin American Journal of Trade Policy 8 (2020) – ISSN 079-9668 – Universidad de Chile
6
The impact of macroeconomics factors on real exchange rate
in Latin America: A dynamic panel data analysis
Carlos Chavez*
Abstract
This paper studies the determinants of the real exchange rate using macroeconomic variables, and
whether they can predict it. A panel data is used, which estimator is system GMM that allows controlling
the endogeneity of the variables. In turn, we transformed the variables with forward orthogonal
deviations (FOD) and first difference (FD), which allows us to eliminate unobserved effects that are
invariant in time. To check the robustness of the estimates, different periods were used, from 1980-2019,
2000-2019 and 2010-2019. For the period 1980-2019, it is found that the past values of the real exchange
rate, the current values of inflation, economic growth, fiscal and monetary policy have positive effects on
the current values of the real exchange rate, while the money supply and the terms of trade have negative
impacts on the real exchange rate. For the period 2000-2019, we had similar results and for the period
2010-2019, we found that economic growth has negative impacts on the real exchange rate. It is also
presented the Arellano-Bond test and the Sargan test to estimate model over-identification. Using the
Pedroni test, we estimated the cointegration of the variables with respect to the real exchange rate, finding
cointegration with inflation in the long run. The originality of this paper is that we focused on Latin
American countries, analyzing short-term relationships with the System GMM estimator and long-term
relationships with the Pedroni Test.
Keywords: Real exchange rate, System GMM, Macroeconomic factors, Developing countries.
Resumen
Este artículo estudia los determinantes del tipo de cambio real utilizando variables macroeconómicas y si
estas pueden predecirlo. Se utilizó un panel de datos cuyo estimador es el sistema GMM que permite
controlar la endogeneidad de las variables. A su vez, se transformaron las variables con desviaciones
ortogonales hacia adelante (FOD) y la segunda es primera diferencia (FD), que nos permite eliminar
efectos no observados que son invariantes en el tiempo. Para comprobar la solidez de las estimaciones,
se utilizaron diferentes períodos, de 1980-2019, 2000-2019 y 2010-2019. Para el período 1980-2019, se
encontró que los valores pasados del tipo de cambio real, los valores actuales de inflación, crecimiento
económico, política fiscal y monetaria tienen efectos positivos sobre los valores actuales del tipo de
cambio real, mientras que la oferta monetaria y los términos de intercambio tienen impactos negativos
sobre el tipo de cambio real. Para el período 2000-2019, obtuvimos resultados similares y para el período
2010-2019, encontramos que el crecimiento económico tiene impactos negativos en el tipo de cambio
real. También se utilizó la prueba de Arellano-Bond y la prueba de Sargan para estimar la
sobreidentificación del modelo. Utilizando la prueba de Pedroni, estimamos la cointegración de las
variables con respecto al tipo de cambio real, encontrando cointegración con la inflación en el largo plazo.
La originalidad de este artículo radica en que se enfoca en los países de América Latina, analizando las
relaciones de corto plazo con el estimador System GMM y las relaciones de largo plazo con el Test
Pedroni.
Keywords: Tipo de cambio real, Sistema GMM, factors microeconómicos, países en desarrollo.
JEL classification: F17, F41, F47, F31
* Research Associate at Universidad Nacional Mayor de San Marcos (Lima-Peru). Research Assistant, International
Monetary Fund. Email: carlos.chavez2@unmsm.edu.pe. Received: April, 24th 2020; modifications: December, 3rd
2020; accepted: December, 21st 2020.
Carlos Chávez
The impact of macroeconomics factors on real exchange rate in Latin America: A dynamic panel data analy sis
7
Introduction
This paper studies the macroeconomic variables that can explain future variations in the real exchange
rate, focused on Latin American countries. The main reason for this research is that the real exchange
rate is an indicator that can measure macroeconomic welfare among countries. Besides, the real exchange
rate can be seen as a measure of competitiveness due to its effects on net exports, of the cost of living
among countries, or also as a measure of dependency among countries (Mesquita et al., 2017).
The sample is comprised by 14 countries with 39 years of data availability from 1980 to 2019. The
methodology used is a data panel with System GMM as estimator, and it is described in section 3. We
control for periods of high inflation, such as Peru in 1988-1990, currency crisis in 1999-2000 and financial
crisis 2008-2010, by adding dummy variables. This paper contributes to the literature because it allows
explaining movements in the real exchange rate from macroeconomic variables in countries suffering
from constant inflation such as Latin American countries. In turn, based on Nikolaou (2006), we add
lagging variables because they have dynamic behavior, which means that their past movements can
explain their current movements. Considering all the periods of the regression, we find that the lagged
values of the real exchange rate have a great positive influence on the current value of that variable, in
that vein, we also find that the inflation, GDP and interest rate have positive impacts on the real exchange
rate. On the other hand, monetary supply and the terms of trade have negative impacts on the real
exchange rate, and the public spending have different effects considering different periods. The results
in this paper are consistent with the literature and the empirical evidence, except for public spending.
Furtermore, the Pedroni test is applied, finding that the inflation has long-term relationship with the real
exchange rate.
The paper is developed as follows: The following section makes a review of the supporting literature,
considering the reasons for the chosen variables and their possible effects on the real exchange rate.
Section 3 describes the economic methodology used to estimate the effects of these variables on the real
exchange rate. Section 4 shows the results and interpretations, and section 5 presents the conclusions of
this paper.
Literature Review
International macroeconomic theory on the real exchange rate began in the 1970s. Frenkel (1976)
developed a model for determining the exchange rate from an asset view, which included the money
supply and inflation. This author emphasizes the importance of the expectations to determine the real
exchange rate. In relation to that research, Dornbusch and Fischer (1980) develop a model in which they
integrate the role of relative prices, expectations and asset markets to take into account the relationship
between the exchange rate and the current account. They emphasize the importance of expectations for
short-term behavior in the exchange rate. Monetary changes in models where asset markets move rapidly
relative to the goods market lead to real exchange rate depreciation. Dornbusch (1980) develops a model
in which inflation is related to the real exchange rate and the real interest rate. Dornbusch (1987) develops
a model on the relationship between exchange rate determination and wage effects. Other papers that
have contributed to the development of the theoretical framework in the determination of the real
exchange rate are Messe and Roggof (1988) and Roggof (2003). Additionally, Devereux (1997) reviews
the theory and evidence of the real exchange rate.
Para continuar leyendo
Solicita tu prueba