Canadian Journal of Economics 46: 1085-1109, 2013
Abstract. We perform convergence tests on the U.S. states for per capita income from 1930-2009. Cross-sectional tests support overall β-convergence and σ-convergence but may not hold true for the last three decades. Time series tests suggest that about half of the states exhibit stochastic convergence and of these all are also β-converging. Probit regressions reveal that the likelihood a state is converging is a function of changing capital to labor ratios, the size of the agricultural sector, and levels of taxation and tax revenue. Regional disparities in convergence remain among the southern and midwestern states.
Economics of Governance 14: 385-409, 2013
Abstract. Since the 1970s, there have been substantial financial reforms across the world, however little research has been devoted to studying the convergence path of these reforms. While Abiad and Mody (2005) find that there is movement towards a global `norm' of financial reform, their findings are based upon a cross-sectional approach that has been widely criticized in the literature. In this paper we offer new time-series evidence on the convergence of financial reforms both across and within regions, which can also act as a metric to measure just how globalized are countries' financial systems. We find that some regions of the world have fully converged, but the advanced economies and Sub-Saharan Africa are not converging. In addition, while most countries have fully converged within their own region, notable exceptions are also identified. These results suggest that while financial reforms have largely become homogenized, important distinctions still remain.
European Economics Letters 2: 20-23, 2013
Abstract. Financial liberalization indicators are tested for sigma convergence and divergence. Sigma convergence requires a significant reduction in the dispersion across nations over time whereas sigma divergence entails a significant increase in dispersion. Using the standard deviation and a linear trend, sigma divergence is supported an index of capital accounts openness, but sigma convergence is supported for an index of domestic financial sector liberalization. Using instead the coefficient of variation, which accounts for the upward trend in each of the measures, strong evidence is found in support of sigma convergence for both measures. This latter result holds for both advanced and developing nations.
Journal of Regional Analysis and Policy 43: 107-177, 2013
Abstract. Previous studies of U.S. political polarization have examined state-level convergence and divergence. Increasingly, scholars have turned from state-level analyses to regional analyses and advanced claims as to whether regions are becoming more similar or dissimilar. Yet, formal testing has remained largely absent. Using panel unit root tests, we determine whether regional political convergence is occurring between 1970 and 2004. Our results suggest the importance of distinguishing between ideological convergence and partisan convergence. We find the four regions and eight subregions are stochastically converging to a national norm in regard to ideology but not with regard to partisanship. Convergence for partisanship is, however, occurring within three of the regions, but not within the Midwest.
Social Science Journal 47: 689-698, 2010
Abstract. In view of the ongoing debate about the degree and direction of political polarization in the U.S., we assess whether the 50 states are converging or diverging in their behavior in state and federal elections. We find that states are diverging in their behavior in federal elections but converging in their behavior in state elections. Previous scholars have shown the need to distinguish between the degree of polarization of elites and ordinary citizens; our findings demonstrate the further need to distinguish between trends in partisan polarization at the federal and state level.
Review of Regional Studies 38: 251-269, 2008
Abstract. Convergence tests are performed on state level turnout rates for U.S. presidential elections from 1896 to 2004. The degree of dispersion in turnout has steadily declined since 1940, suggestive of general overall convergence taking place. Individually, it is found that 29 of the 48 continental states are stationary in their relative trend levels, and 42 states either do not significantly differ from the national average or are significantly trending toward the national average. In total, 25 of the 48 states pass tests for both stochastic and β-convergence, suggesting that national convergence is being achieved by roughly half the states.