Democracia, confiança e crescimento econômico

Abstract

This study investigated the effects of democracy on economic performance of countries. Analysis of theoretical and empirical literature revealed that this is an issue still not very well understood. On the one hand, democratic regimes favor the efficient allocation of resources and are more prone to building inclusive economic institutions; on the other hand, they enable the exercise of pressure by social groups that require immediate consumption and distributive policies, which are made at the expense of investments. Thus, although the net effect is uncertain, some authors suggest that it is moderately negative - see, for example, Schweinitz (1959) Clay (2000), Tavares et. al (2001) and Acemoglou et. al (2008). This thesis suggests that the lack of consensus about the relation "growth versus democracy" occurs by negligence given the role of trust which is an important element in generating collective action. A high level of trust in a democratic country can make this regime produce different effects on growth. In other words, a democratic regime allows occur pro-distribution pressures and/or implementation of social welfare policies that can be done at the expense of investments. However, the democracy alone does not ensure that these motions occur. The potential of individuals arranged to act collectively produces these effects. The first chapter has a review the theoretical and empirical literature on "growth versus democracy", and "growth versus trust". The idea was to verify the theoretical arguments that justify (or not) the influence of these two elements in the economic performance of countries and their respective transmission mechanisms. It was noted, however, that they has treated separately in the growth regressions - that is, or study the effects of democracy, or the effects of social capital, measured by the trust level. Given the central role of trust, the second chapter has attempted to investigate its main determinants from hierarchical models. Using a sample of 77 225 observations distributed between 54 countries, it was observed that the probability of responding that "most people can be trusted" is higher among those who i) are satisfied with their family income; ii) did not experience separation or divorce; iii) are not unemployed; iv) participate in any social organization; v) valued friends; vi) value altruism; vii) are less prejudiced; viii) have post-materialist values; ix) trust in the country's justice; and xi) are optimistic. In addition, variables such as latitude, Gini index and post-communist past are also significant. On the other hand, ethnic diversity, as well as the proxy for the guarantee of property rights were not statistically significant. The next step, now in the third chapter, was to estimate a growth regression from panel data on a sample of 79 countries between 1994 and 2013. It was to use the trust level as one of the controls in the growth models, including interactively with the proxy for democracy. The results show that in the absence of control by trust level, the parameter relative to democratic was not significant. However, the conclusions change substantially when this control is established. Contrary to what many authors advocate, can not be said that democracy produces moderately negative effects on growth rates; this only happens when the level of social capital is relatively high, generally greater than 30%. This indicates that it is not democracy per se, which produces negative effects on growth rates. The pressure for redistributive policies may be exercised in democratic societies, but they are more likely to be exercised in fact in contexts of high social trust. This can be seen also in the third chapter, when we investigated the determinants of government spending and the level of gross physical capital.