Presidential Approval in Peru: An Empirical Analysis Using a Fractionally Cointegrated VAR

Presidential approval in Peru depends on economic outcomes. However, voters are unable to distinguish between outcomes resulting from economic policies and those caused by exogenous shocks. Estimation results from seven Fractional Cointegrated VAR (FCVAR) models suggest that presidential approval increases with the monetary policy interest rate, the terms of trade, and manufacturing employment; and decreases with the nominal PEN/USD exchange rate and ináation volatility. Additionally, a Principal Components Analysis (PCA) conducted over a large set of macroeconomic indicators points to a greater ináuence of external over domestic factors in explaining presidential approval; i.e., economic outcomes that determine the dynamics of presidential approval are not under presidential control in Peru. It can be argued that these Öndings identify a signiÖcant source of political instability and a considerable challenge to democratic governance. To the authorsíbest knowledge, this is the Örst application of fractional cointegration analysis to political economy in Latin America. JEL Codes: C32, D72. Keywords: Economic Voting, Fractional Cointegration, Political Economy, Macroeconomics, Latin America, Peru. Resumen La aprobaciÛn presidencial en el Per˙ depende del estado de la economÌa. Sin embargo, los ciudadanos no son capaces de distinguir entre los resultados econÛmicos determinados por las polÌticas econÛmicas y aquellos que son consecuencia de choques externos. Los resultados de la estimaciÛn de siete modelos VAR Fraccionalmente Cointegrados (FCVAR) sugieren que la aprobaciÛn presidencial aumenta con la tasa de interÈs de polÌtica monetaria, los tÈrminos de intercambio y el empleo manufacturero; y disminuye con el tipo de cambio nominal PEN/USD y la volatilidad de la ináaciÛn. Adicionalmente, un An·lisis de Componentes Principales (PCA) sobre un amplio conjunto de indicadores macroeconÛmicos sostiene que la ináuencia de los factores externos sobre la aprobaciÛn presidencial es mayor que la de los factores domÈsticos; ie., los resultados econÛmicos que determinan la din·mica de la aprobaciÛn presidencial no est·n bajo el control del presidente en Per˙. Esta es la primera aplicaciÛn del an·lisis de cointegraciÛn fraccional para el estudio de la economÌa polÌtica en AmÈrica Latina, seg˙n el conocimiento del autor. ClasiÖcacion JEL: C32, D72. Palabras Claves: Voto EconÛmico, CointegraciÛn Fraccional, EconomÌa PolÌtica, MacroeconomÌa, AmÈrica Latina, Per˙