The asymmetric evolution of economic institutions: evidence from dynamic panel quantile regression with IV and fixed effects

ABSTRACT

Conditional mean regressions show that economic institutions (insteco) correlate positively to regime durability, the quality of political institutions, human capital and GDP per capita. However, using dynamic panel quantile regressions with fixed effects and instrumental variables, we detect that insteco evolves asymmetrically. Adverse shocks worsen insteco of countries with better political institutions and more durable regimes, while positive innovations cause insteco of countries with better political institutions and longer political regime to improve even further. Among advanced economies, negative shocks cause a stronger effort to ameliorate insteco the higher the human capital and GDP per capita. Our exercises include 129

countries.

KEYWORDS:

• Institutions
• asymmetry
• quantile regression

JEL CLASSIFICATION:

• E00
• O43
• P48

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

¹ Acemoglu et al. (2008) and Cervellati et al. (2014) also use 5 years interval to study the evolution of institutions.

² Data selection is inspired by Baryshnikova, Pham, and Wihardja (2016) and Souza, Gross, and Figueiredo (2017).

³ See Acemoglu, Johnson, and Robinson (2001) and Keefer and Knack (2002).

⁴ Koenker (2004) develops estimators for longitudinal panel quantile with fixed effect, but lags of the dependent variable among the covariates bias the fixed effect if the time dimension is short. Chernozhukov and Hansen (2005) instrumentalize with lags of the independent variables.

⁵ See https://www.imf.org/en/Publications/WEO/Issues/2021/10/12/world-economic-outlook-october-2021. Differently from the publication, we grouped all non ‘advanced’ economies in our ‘not developed’ group.