Economic and political 'institutions' and growth: Some econometrics

To argue that either "dictatorships" or "democracies" better promote economic growth assumes that "institutions" profoundly affect economic outcomes. Given this abstract debate, the relevant component of political regime-type is discretion or freedom of action, which either enhances or inhibits sovereigns' ability to execute benevolent economic policy. Cross-sectional, time-series and panel regressions on a workable sample of 143 countries from 1950 to 1992 render contradictory evidence for these and other related "institutional" arguments.;Disabilities beset the literature and warrant additional econometrics. Empirical assessments do not explicitly account for "stages" of economic development or adequately address whether the causal relationship between regimes and the rate of economic development is simultaneous. With respect to measurement, proxies for the rich concept of discretion are unsatisfactory. Elections do not singularly constrain policy-makers. Rather, political party membership and competition, as well as the constitutional position of the executive vis-a-vis the legislative branches of government both limit sovereign action. Also, other "institutions" besides regimes shape policy-choice; including the "organization of capital," specifically stock market development and the "autonomy" of central banks; the "organization of labor" and "fiscal federalism." Finally, published empirical tests do not satisfactorily evaluate the indirect effects of institutions on growth. Econometric process-tracing regarding investment ratios, income distribution and macroeconomic policy indicators (inflation rates, budget deficits, and black market exchange rate premia) illuminates possible paths through which institutions affect expansion.;The following econometrics do not posit a single model for verification. Rather, the inductive empirical investigation is a "sensitivity" analysis that evaluates many hypotheses under different conditioning sets involving possible specification, sampling and measurement biases. The results indicate few consistently robust findings. "Institutions," insofar as social scientists can define them given the nation-state as the unit of analysis, scarcely impact economic outcomes. Notably, this approach precludes "path dependency" and "political culture" components of "institutions," and thus social scientists cannot readily replace proper names with definitions. This foray of comparative politics into macroeconomic growth theory, and vice versa, is tenuous and ultimately suggests several "routes to the modern world."...