Structural Breaks and Convergence in the Regions of Peru: 1970–2010

• Author: Augusto Delgado,Gabriel Rodríguez
• Published date: 01 May 2015
• DOI: http://doi.org/10.1111/rode.12146
• Date: 01 May 2015

Structural Breaks and Convergence in the Regions

of Peru: 1970–2010

Augusto Delgado and Gabriel Rodríguez*

Abstract

The aim of this paper is the analysis of stochastic and β-convergence in the relative regional per capita

outputs using different unit root tests both with and without structural breaks and using a further test that is

robust to the presence of I(0) or I(1) errors. It allows robust inference on the estimates of the initial per

capita output (intercepts) and the respective growth rates (slopes). The results of the application of unit

root tests without structural breaks show the absence of stochastic convergence. However, by incorporating

the presence of endogenous breaks, the results are reversed for all regions. In the case of β-convergence,

the results of the robust test WRQF show that all regions have a structural break at some point during the

period 1970–2010. We find different behavior for different regions. There is a catching-up process and a

lagging-behind process for different groups of regions towards more negative or more positive paths.

1. Introduction

Statistics for Peru show a negative relationship between the natural logarithm of per

capita gross domestic product (GDP) in 2010 against the mean rate of growth from

1970 to 2010 for each region.1This suggests the possibility that the regional economies

show a pattern of convergence towards a unique path of national growth where the

poorest regions (low level of per capita GDP) catch up with the richest regions (high

level of per capita GDP). Nonetheless, the regions of Moquegua, Arequipa,

Apurimac, Huánuco, and Lima differ from the trend, showing an economic perfor-

mance that is disconnected from the rest of the country. This implies no convergence

towards a single stationary state for the country. The results also show that regions

that started out with higher levels of income (Madre de Dios and Tacna) and moder-

ately high incomes (Huánuco, Tumbes, and Piura) had the worst economic perfor-

mance in that their average growth rates were not positive for 40 years.

In the past ten years it appears possible to observe significant growth in the more

disconnected regions of Peru, which has prompted authors such as Webb (2013) to

point out that a long-time connection exists between the rural and urban economies.

Alcántara (2001) argues that there was a reduction in inequality over the period

1961–1972, but this trend is reverted for the period 1972–1993. Del Pozo and

Espinoza (2011) suggest that this trend is again reverted for the period 1993–2008. All

this evidence revives the concept of economic convergence.

* Rodríguez, Department of Economics, Pontificia Universidad Católica del Perú, Av. Universitaria 1801,

Lima 32, Lima, Perú. Tel: +511-626-2000 (ext. 4998); Fax: +511-626-2874; E-mail: gabriel.rodriguez@

pucp.edu.pe. Delgado: Department of Economics, Pontificia Universidad Católica del Perú, Lima, Perú.

This paper is drawn from the thesis of Augusto Delgado (2014). An extended version of the paper appears

in Delgado and Rodríguez (2013). The authors are thankful for the helpful comments of Efraín Gonzáles

de Olarte, Mario Tello and participants at the DEGIT XVIII, 26–27 September 2013 in Lima, Peru. They

also thank Rodolfo Cermeño, Manuel Gómez-Zaldívar, Daniel Ventosa-Santaulària and the two anony-

mous referees for their input.

Review of Development Economics, 19(2), 346–357, 2015

DOI:10.1111/rode.12146

© 2015 John Wiley & Sons Ltd