Institutions and Growth Dynamics in Latin America, 1801–2015

Summary

This paper constructs annual data for seven Latin American countries (Argentina, Brazil, Chile, Colombia, Mexico, Peru and Venezuela) over 1801–2015 and tests how institutional quality affects growth via three channels: innovations (patents), investment (I/Y) and education (secondary and tertiary enrolment). Using new composite institution measures (contract‑intensive money, tax capacity and democracy) and IVs based on immigrant‑origin and genetic proximity, the author finds that institutions are statistically significant drivers of innovation, investment and education. However, counterfactual simulations show that adopting the institutional trajectory of the most successful settler economies (Australia, Canada, New Zealand and the US) would only have closed a fraction of the income gap: institutions raise growth but cannot fully explain the long-run divergence.

Key Points

Unique annual dataset spanning 1801–2015 for seven Latin American countries, compiled mostly from national sources.
Institutions proxied by a principal component of contract‑intensive money (CIM), direct tax share and a democracy index.
Identification uses instruments weighted by immigrants’ countries of origin and by genetic proximity; strong first‑stage F‑statistics reported.
Three transmission channels assessed: innovations (patent intensity), investment (I/Y) and education (GER at secondary/tertiary levels).
Estimated elasticities imply institutions significantly boost patents, investment ratios and enrolment; effects are robust in post‑1870 and panel interval checks.
Growth mapping: capital/output, education and patent intensity all significantly predict labour‑productivity growth in 5‑year intervals.
Counterfactuals: had Latin America followed ACNU institutional paths, cumulative gains through education, investment and patents would add roughly 79% to per capita income since 1801 (26% + 23% + 30.1%), but a large residual gap remains.
Main conclusion: institutions matter and work through clear channels, but they explain only part of the great divergence — complementary factors must be examined.

Context and Relevance

This is a major contribution to the institutional literature because it (1) uses long annual series to trace institutional effects across two centuries, (2) explicitly tests the mechanisms (innovation, investment, education), and (3) quantifies how much institutions can — and cannot — explain about Latin American underperformance relative to settler economies. The findings are relevant to economic historians, development economists and policymakers interested in the limits of institutional explanations and the importance of complementary drivers (demography, resource structure, technology adoption, policy choices).

Why should I read this?

Short version: if you want a rigorous, long‑run check on the oft‑repeated claim that ‘bad institutions explain Latin America’s lag’, this paper does the hard yards. It builds two centuries of data, teases out channels (patents, investment, education), runs IVs and counterfactuals — and then tells you the inconvenient truth: institutions matter a lot statistically, but they don’t tell the whole story. Read it for the methods and the sober, quantified take on institutional impact.

Author (style)

Punchy and empirical: the author pushes beyond cross‑sectional claims, supplies detailed time‑series evidence and stresses mechanisms. If you’re interested in policy implications, the paper is blunt — institutional improvement helps, but won’t single‑handedly close the long‑run income gap.

Source

Source: https://onlinelibrary.wiley.com/doi/10.1111/ecot.12449?af=R