Growth States and Development:
Political Institutions and Their Conditional E
ffects
Jonathan Hanson
University of Michigan
April 13, 2004
Abstract
Empirical testing of propositions developed in the New Institutional Economics has
provided convincing evidence that neoclassical economic growth theory incorrectly
assumes away the role of political institutions in growth and development. Findings
of convergence between poorer and richer countries are stronger after controlling
for indexes of institutional quality. Nevertheless, shortcomings exist with this work.
First, there is an underlying assumption that the same growth model applies to all
countries regardless of their current circumstances. Second, the tendency to measure
institutional quality with broad indexes obscures the possibility that di
fferent facets of
political institutions matter in di
fferent ways depending on country conditions. This
paper adopts a framework proposed by Pritchett (2003) in which countries fall into one
of six di
fferent growth states, each of which is assumed to have a unique underlying
growth model. It tests the proposition that transitions between these growth states
are governed in part by di
fferent aspects of political institutions. The tests support the
notion that the e
ffect of political institutions on economic growth and development is
conditional on a country’s circumstances. In particular, three aspects of state strength
are considered: state legitimacy, administrative capacity, and the ability to contain
civil unrest. State legitimacy is important for the transition away from poverty, while
high levels of administrative capacity are more useful in distinguishing the advanced
industrial economies from the rest. Low levels of civil unrest are associated with
periods of rapid growth. A fourth institutional form, the degree to which a country is
democratic, does not appear to be particularly helpful in supporting growth in poorer
countries in that it is associated with a higher probability of economic implosion.
Paper prepared for presentation at the annual meeting of the Midwest Political Science
Association, Palmer House Hilton, Chicago, IL, April 16, 2004.