Spurred on by Jared Diamond's Guns, Germs, and Steel (New York: Norton, 1997), economists now seek to examine the economic effects of exogenous geographic forces. In this fashion, Daniel Berkowitz and Karen Clay examine how institutions affect economic performance in the United States during the nineteenth and twentieth centuries. Moreover, they examine how geographic factors (e.g., climate, rivers) and other “initial conditions” create these institutions. Following in the lineage of Daron Acemoglu (e.g., Daron Acemoglu, Simon Johnson, and James A. Robinson, “The Colonial Origins of Comparative Development: An Empirical Investigation.” American Economic Review 91, no. 5 (2001): 1369–1401), Berkowitz and Clay argue that institutions are important and that their origins are also important because of persistence. The U.S. states offer an unexploited laboratory.

The tale they weave is both straightforward and complicated. The luck of geography and legal regimes of colonizers gave birth to judicial and political institutions. The two institutional features they focus on—judicial independence and political competitiveness —persist to this day because of the costs of change and because other underlying forces have also not appreciably changed. These institutions shape economic growth and the creation of wealth which are reflected in differential per capita incomes across the states today. The complicated part is measuring the institutions and their maternal origins, and then carefully establishing the connections. To fully understand, I highly recommend this impressive book.

Chapter 1 explains the rather involved historical chain of events and chain of causation. The next two chapters make the case for persistent and distinct legal and political institutions among the states. In Chapter 4, they reveal a “mechanism” to explain the economic and political forces that create origins and persistence. Chapters 5 and 6 examine how origins affect state courts and political competition. The concluding chapter examines the effect of courts and political competition on current (2003) per capita income and offers thoughts on future work and institutional reform.

Perhaps the most compelling idea is how the authors view and examine the role of “elites” in shaping and maintaining institutions. The basic idea is that those same initial geographic and legal conditions also created elites that dominate politics and determine the structure of institutions. Elites are people at the topic of the income distribution (empirically this is the wealth share of the top 1–10 percent) and these economic elites are assumed to be political elites as well. The interests of the elites are argued to depend on their occupations and the key distinction is between trade and agricultural elites. The composition of the elites, of course, depends crucially on original characteristics and thus the self-interest of elites drives the institutions, their persistence, and their effects. The evidence they present shows considerable variation in the composition of elites across states (basically agricultural elites dominate in the south) and that there is a correlation with “occupational homogeneity” and a more competitive state legislature. The idea of economic elites, such as businessmen and large farmers, driving politics and political institutions is compelling, and another approach would include direct connection to interest group theory and to more subtle distinctions among elites.

Berkowitz and Clay tackle a big topic—“the evolution of a nation”—and provide the reader with both statistical data and detailed historical accounts. The overriding question of the study is whether legal and political institutions affect economic performance and moreover, whether or not these institutions are set in motion and then locked in to place by initial conditions. Berkowitz and Clay say yes and (p. 12) “conclude that initial conditions played early and enduring roles in shaping political and legal institutions in the American states.” In this regard, their work is in line with others that conclude that institutions are important and costly to change. What we don't yet know is much about the specific mechanisms by which relatively small differences institutions persist and have important impacts.

The authors move us in the right direction but are limited by the extent of the evidence. Further, the presentation and analysis of data are not always easy to follow. Sometimes summary statistics and definitions for data are within chapters, others are in appendices, and the rest available in larger available data sets. Overall, the important regression estimates, figures, and tables could use more discussion so that the reader can take full advantage of the impressive data collection and analysis that has been conducted by the authors.

Additionally, although the data in general support the basic line of causation argued by the authors, because they cover so much territory, the authors cannot examine some key institutions in sufficient depth. I mention a few such examples here. They stress geographic features of the states but do not acknowledge that the shapes of the states (and thus the measured state features) are clearly economic and political choices. Virginia, for example, once claimed land west to California and British Columbia, and then made a deal with the fledgling United States government to limit claims to the border of the Northwest Territory, and later during the Civil War lost what became West Virginia. Many other states had similar transformations during the evolution from territory to states. What then is the appropriate measure of geographic features? A second issue is the common problem of omitted variable biases. For example, the authors might consider political “contagion” from neighboring states which could be important for small economically connected states. And there are potential measurement issues regarding state initial conditions and indicators of political competitiveness. Still, when the evidence is tallied up, the authors have a made a strong case and have created a template for examining economic and political institutions over time. This is not an easy task and they should be commended.

The front cover of the book shows a 1733 map of the British Empire in North America indicating French and Spanish colonies as well as British ones. The back cover is a compilation of praising remarks on the book by six outstanding scholars in economics and political scientists. These praises are warranted. The book will reinforce the idea that institutions matter but make you wonder if there is anything that can be done to improve them.