adjudication and enforcement of individual rights is not similarly short-
sighted. A further advantage of democracy, according to Olson, is that the
promise of a dictator is not enforceable by an independent judiciary and so is
never completely credible.
Amartya Sen, in numerous works including Development as Freedom
(1999), has argued that democracy has a much broader importance than its
ability to promote economic growth – which he does agree with. For Sen the
expansion of freedoms is not just the principal means of development; it is also
the primary end. Democracy as the ability to engage in free speech, publicly
debate and conceptualize needs and be freed from the fear of arbitrary arrest is
for him an important part of his definition of a good society. To ask the ques-
tion ‘is democracy or dictatorship better at promoting development?’ is
mistaken. Democracy is a crucial part of the definition of development.
Democracy and development: the empirics
Some scholars have tried to construct more sophisticated tests of the relation
between political regime and economic performance. Barro (1996) tested the
relation between democracy and growth. He used a panel of 100 countries for
the years 1960 to 1990 and controlled for the impact of various other proximate
determinants of growth such as human capital and education spending, the
fertility rate, government spending, investment and the terms of trade. To this
regression Barro added an index of democracy based on the ability of individ-
uals to participate meaningfully in the political process. The results show that
the democracy variable has no clear relation with economic growth.
Examining the relationship between the rule of law (instead of economic
growth) and democracy showed that democracy has no significant positive or
negative impact on the rule of law. Barro does find, however, that schooling
(particularly primary) is a strong predictor of a country being democratic; there
are also positive links from income and infant mortality. This finding offers
strong support for the Lipset hypothesis.
Dani Rodrik (1998a) agrees that a democratic political system has little
relation with economic growth but finds instead that democratic countries have
more predictable long-run growth rates and greater stability in economic
performance, handle adverse economic shocks better, and that democracies
pay higher wages. Typically, democracies have a wider range of decision-
makers and democratic debate slows down the process of policy implementa-
tion, so such countries are at less risk of being destabilized by the policy whims
of a dictator. Rodrik (1999) in a sample of over 100 countries finds that in the
1970s those with greater civil liberties and political rights experienced lower
declines in economic growth in response to the sharp increase in the world
price of oil. This finding is surprising as much of the argument in this section
has revolved around a common assumption that a strong, perhaps non-demo-
cratic state is needed to undertake unpopular adjustment policies. The result
can be explained because adjustment to shocks requires managing social
Institutions 211