Earnings mobility, inequality, and economic growth in Argentina, Mexico, and Venezuela

Abstract

This paper examines changes in individual earnings during positive and negative growth periods in three Latin American economies: Argentina, Mexico, and Venezuela. We ask two major questions. First, do panel income changes favor the income recipients who started at the top of the income distribution ("divergent mobility") or those who started at the bottom ("convergent mobility")? And second, are the groups that are found to gain the most when the economy is growing those that are found to lose the most when the economy is contracting ("symmetry of mobility") or is the pattern asymmetric in the sense that the same groups do best both in times of economic growth and in times of economic decline? We find support for the divergent mobility hypothesis only in scattered years in the cases of Mexico and Venezuela and no support at all in the case of Argentina. Rather, earnings mobility is most frequently convergent or neutral in all three countries. As for the symmetry of mobility hypothesis, we find that it is rejected in nearly all cases; rather, those groups that gain the most when the economy is growing are also the ones that gain the most or lose the least when the economy is contracting. Furthermore, we discuss how the absence of divergence reconciles with rising inequality in the countries under study.

La movilidad económica y el crecimiento de la clase media en América Latina

“…Es aún más importante señalar que estos estudios normalmente observan que los cambios en los ingresos son heterogéneos entre diferentes grupos de ingresos, y que los menos favorecidos son aquellos que en general experimentaron los mayores beneficios en la región. En este sentido, Fields et al (2007) presentan evidencias convincentes de que, dependiendo de las características observables, los hogares con los ingresos iniciales más bajos tienden a ganar relativamente más (ver, por ejemplo, Duval Hernández [2006] para el caso de México).…”

“…• En América Latina en particular, y en el mundo en vías de desarrollo en general, los paneles suelen limitarse a zonas urba-nas o a pequeñas muestras y, por lo tanto, no son representativos del conjunto de la población del país (Fields et al 2007). • Dado que las bases de datos de panel que siguen a los individuos o a los hogares durante largos periodos (10 a 20 años)…”

What’s behind pro-poor growth? An investigation of its drivers and dynamics

“…This approach has been recently applied by Ferreira et al (2019), who estimate a counterfactual GIC to relate the distributional impact of economic growth to changes of the structure of the economy. Even Fields et al (2015), in their analysis of earnings mobility in Argentina, Mexico, and Venezuela, recognize the confounding role of measurement errors and transitory earning shocks to which individuals may be subjected in the short-run. Therefore, they propose a framework in which individuals' earnings are decomposed as the sum of two components, one associated with observable and permanent individual characteristics and another that is related to transitory earning components.…”

“…suggests that the identification of individuals may be based on the ranking in the final distribution, and therefore pro-poorness is evaluated by focusing on the income trajectories of individuals who become poor. Lo Bue andPalmisano (2020) propose a non-anonymous version of GIC to evaluate the patterns of mobility experienced by the chronic and transitory poor, where identification is based either on the initial or the final distribution.3 The concept of mobility is multidimensional(Fields and Ok 1999), as it embodies four different aspects, which are described byJantti and Jenkins (2014): re-ranking within the income distribution, income growth, inequality reduction, and uncertainty.4 In this regard, exceptions are the contributions byBárcena and Cantó (2018) andBresson et al (2019).5 By adopting a two-stage least squares procedure,Fields et al (2015) first estimate the part of individuals' earnings associated with permanent characteristics. Then, the predicted values, which represent a proxy of the initial income of individuals, are used in a second regression as the explanatory variable of the individuals' income changes.…”

Household Earnings Mobility and Inequality: The Role of Heterogeneous Macro Effects

“…For example, Berman [11] investigates how income mobility in the United States from 1962 to 2014, measured over four-year periods, correlates with business cycles and finds a positive correlation between economic growth and income mobility. In addition, Fields et al [12] and Fields et al [13] analyze longitudinal data from several Latin American countries and find that high-income households experience lower income growth during recessions. They refer to the observation of high-income households doing worse than low-income counterparts as "convergent mobility".…”