Evaluating Fiscal Performance in Latin America and the Caribbean: Structural Balance Estimates from an Original Dataset

• Martín Ardanaz ^[1] ; Ana Corbacho ^[2] ; Alberto Gonzales ; Nuria Tolsa ^[1]
  1. [1] Inter-American Development Bank (USA)
  2. [2] Fondo Monetario Internacional (USA)
• Localización: Hacienda Pública Española / Review of Public Economics, ISSN 0210-1173, Nº 219, 2016 (Ejemplar dedicado a: Public Economics in Latin America), págs. 67-92
• Idioma: inglés
• Enlaces
  • Texto completo (pdf)
• Resumen

  • In recent years, an increasing number of countries have began anchoring their fiscal policy frameworks in terms of rules that target the cyclically adjusted or structural (as opposed to actual) balance in an effort to overcome problems of procyclicality and fiscal volatility. The logic for doing so is in principle compelling: rule-based fiscal policies allow automatic stabilizers to work freely during the cycle and help accumulate fiscal surpluses in good times. However, the estimation of structural balances is subject to a number of methodological challenges, including the degree of estimation uncertainty. This paper presents a range of estimates of the structural budget balance and uses them to analyze the cyclical behavior of fiscal policy in Latin America and the Caribbean. Based on an original dataset comprising detailed fiscal information from 20 countries across the region between 1990 and 2014, the paper finds that the range of estimates can be large for some countries, especially those that derive substantial fiscal revenue from commodity-related activities. In addition, the evidence shows that on average, the region has followed a procyclical policy pattern: a 1 percent increase in the output gap is associated with up to a 0.55 percentage point deterioration in the structural primary balance. This pattern hides substantial regional heterogeneity: procyclicality is more marked in countries that face large terms of trade shocks, but it can be counteracted by higher institutional quality.