Essays on State Capacity and Local Public Goods

The availability of public goods -- public safety, infrastructure, rule enforcement, education, and others -- is one of the hallmarks of a well-functioning society. These public goods are tightly linked to the state, defined as the institutionalized hierarchy of public decision-making (Brumfiel 1994): only when the state attains sufficient capacity, these public goods are effectively provided. However, while the state plays an indispensable role in the provision of public goods, there is no reason to believe it does so efficiently. Thus, developing new ways to evaluate the role of the state in efficient public goods provision and long-run development is a core theme of this dissertation. Moreover, given the enormous variability across the globe in the capacity of the state to fulfill its core mandate of public goods provision, understanding the development of the state and its linkages to the economic environment continues to pose an important and understudied challenge to scholars in political economy, public economics, economic history and development economics.In the first chapter of this dissertation, in joint work with Calvin Zhang, we study the role of the state in providing public goods through local governments. Across counties, cities, special districts and school districts, the U.S. has almost 100,000 local governments employing more than ten million people and making up more than 10% of the U.S. economy(COG 2013). Strikingly, local state capacity is distributed highly unevenly, especially between those within municipal boundaries and those in unincorporated parts of cities. 36 million people live in unincorporated communities without separate municipal government, instead being served by counties and special districts. These places typically have limited local electoral representation, lower levels of public infrastructure, weaker code enforcement and poorer public safety provision (Anderson 2008). We study whether local state services are underprovided in this context. To do so, we combine administrative boundary changes from Californian cities with the universe of individual real estate sales prices for the state over the years 1988-2013. In this way, we can estimate the change in residential home prices and home construction activity in the aftermath of municipal annexation. We then interpret these changes in a spatial equilibrium model with heterogeneous households to estimate households' willingness to pay for public goods provided by local governments. We find that households value a dollar of public goods expenditures by more than a dollar, despite most of the benefits of annexation accruing to landowners and developers.Given that the state is an important element of public goods provision, it is natural to wonder why the state has developed sooner and more effectively in some places than in others. While there is a large literature across the social sciences on the causes of early state formation, two key puzzles remain: first, why did states first form in peculiar locations like Mesopotamia, the Nile Valley, or the Valley of Mexico and not elsewhere? Second, why did incipient subjects accept the extraction by the state instead of evading its power? In the second chapter of my dissertation, I answer these questions using an old idea from cultural anthropology: states arose in regions that offered no refuge to dissidents, such as lush river valleys circumscribed by deserts, mountains, or the ocean (Carneiro 1970). To evaluate this idea quantitatively, I collect data on archaeological excavation sites relating to early states and combine these sites with a large array of agricultural, climatic and other environmental datasets. I then show that the location of early state sites is closely associated with high land quality but low surrounding land quality. After the initial formation of the state, it began its slow but inexorable conquest of human societies across the globe. This process was rapidly accelerated with the development of the state in Europe in the course of the middle ages. In the third and final chapter of my dissertation, in joint work with Eric Weese, we ask: what was the role of the European state in the economic growth unleashed in the the run-up to the Industrial Revolution? To this end, we employ a newly available dataset showing every single boundary change of all European states between AD 1000-2000, amounting to about 9,000 boundary changes. We combine these boundary changes with data on urban growth across Europe (Bairoch et al. 1988). Doing so, we find that cities that were subject to more changes in sovereigns saw significantly lower population growth than cities that were subject to a more stable state. In counterfactual simulations, we establish that the urban population in Europe would have been around 9% larger if European states had been more stable, offering an environment with more effective public goods and more conducive to long-run growth