Political Regime Changes, Crises and Growth in Historical Perspective

One of the central questions in the discipline of economics pertains to the determinants of economic growth. In recent years a large number of theoretical as well as empirical studies have analyzed the complex interactions between the country's institutions and economic growth. In particular, the countries' political governance structures are seen as the crucial components of sustained and stable growth, yet their precise contribution is hard to pinpoint. Despite the difficulties associated with establishing a clear causal relationship between political institutions and growth, uncovering the empirical regularities, the anatomy of the interactions between institutional frameworks and economic outcomes is seen as an essential step toward the deeper understanding of the link between them. Another less-researched area of growth literature focuses on the association between financial crises and the countries' short and the long-term growth outcomes.;My dissertation contributes to the literature on the anatomy of the relationship between institutions and growth by examining the role political regime changes play in predicting the on-start of growth "takeoffs" -- significant turning points in a country's growth history. The dissertation further investigates the link between institutions and growth by examining the cross-country "political spillover" effects and their possible impact on growth. In addition, the dissertation takes a new look at the link between financial crises and the countries' long-term growth outcomes.;For the purposes of the study, I compile a unique set of historical data covering a broad range of economic and political variables. These variables include volume of trade, investment, government expenditure, the timing of financial crises and the systems of governance for a set of 61 countries covering the time period 1820-2003.;Chapter 1 examines whether political regime changes as well as the quality of political institutions help predict the on-start of the periods of sustained and rapid growth in a country's economic history -- periods known as growth "takeoffs". I show that controlling for various economic factors, both democratic and autocratic regime changes are significant predictors of the takeoff initiation. This effect, however, is non-linear as I find evidence that countries with low income per capita levels benefit less from positive regime changes. I employ the Hansen's threshold effect methodology to estimate the level of GDP per capita at which the effect of political regime changes on the probability of a growth takeoff becomes significantly different. I also find similar non-linearity in the coefficients for trade openness and the quality of political institutions. In addition, I find evidence that the model developed in the chapter does significantly better in predicting the on-start of rapid growth episodes than the baseline constant probability forecast.;Chapter 2 of the dissertation investigates why political transitions to more representative forms of government are often associated with vastly different outcomes for different countries. I examine the possibility of cross-country "political spillovers", testing in particular whether countries surrounded by relatively more autocratic neighbors also experience a more difficult economic adjustment process following democratization. The dynamic fixed effects model, estimated using the pooled mean group estimator methodology of Pesaran, Shin and Smith (1999) allows me to account for both short and long-run effects of political spillovers between neighboring countries. In particular, I find evidence that for countries which have experienced democratization, moving away from their neighborhoods on the political spectrum implies slower GDP per capita growth in the long run.;Chapter 3 examines the association between different types of financial crises and the country's long-term growth. I estimate the effect of currency, banking and twin crises episodes on the probability of initiating growth takeoffs as well as growth collapses. I find that currency crises are significant positive predictors of growth takeoffs, especially in the post World War II period. The currency crises episodes, however, were reducing the probability of growth takeoffs during the Gold Standard era. In addition, I find evidence in favor of the hypothesis that the prompt return to the original parity with gold following a currency crises -- the so called "resumption rule" of the Gold Standard era may have contributed to the dampening of the economic activity in the long run.;The results in this chapter confirm the intuition that twin crises are likely to significantly dampen economic activity even in the long run. Twin crises are found to be positive predictors of growth collapses during 1980-2003, and negatively influencing the probability of growth takeoffs during the Gold Standard era.