رای دهنده میانگین نشان نمی دهد: جلسات پر هزینه و رانت های داخلی

How does changing from an assembly to a town-council form of government affect the way in which cities are run? Previous empirical research on this question has not found much of an impact of assemblies on aggregate outcomes such as local public expenditures or taxation. Nevertheless, the specific role of organized insiders may be important to understand how cities and towns governed by citizens' assemblies work. Existing surveys point to local workers as an important pressure group in local assemblies. Using data from local governments in New England I find that municipalities governed by assemblies pay around 4% to 10% higher salaries to their employees. This wage premium is bigger in assemblies with lower attendance, and increasing with the employees' voting power. I prove my results robust to the inclusion of an exogenous representative-government comparison group: municipalities in New York State that lie within 40 miles of the border with New England. The results demonstrate how insider groups derive some advantages from an assembly form of government. More broadly, the potential capture of assemblies by insider groups can be an important risk faced by municipalities with low citizen participation, which provides a rationale for the widespread adoption of representative government at the local level.

The institutions, policies, and practices of local governance are very important to explain how cities run. An extensive literature in urban economics and local public finance (Rubenfeld, 1987) has studied how taxes, public expenditures, infrastructure provision, school quality, and zoning affect urban outcomes and growth. However, relatively less attention has been devoted to the impact of local political and governmental institutions on town and city management. This paper compares the assembly institution to its democratic counterfactual, representative government, as alternative ways to manage the provision of public goods and services in cities and towns. Previous empirical research has not found much impact of assemblies on aggregate outcomes such as local public expenditures or taxation. However, there may be more subtle ways by which local institutions impact urban governance. I hypothesize that the power of insider groups, such as government workers and local officials, becomes very important in a context of costly attendance to meetings. Empirically, I use data from local governments in New England, where many municipalities are governed by an assembly type of government — the Town Meeting. I compare the wages paid in similar job positions between Town-Meeting municipalities and local representative governments. I find that municipalities governed by assemblies pay around 4% to 10% higher salaries to their employees. I prove these results robust to the inclusion of an exogenous representative-government comparison group: municipalities in New York State that lie within 40 miles of the New England border. For historical reasons that are exogenous to the contemporary political debate, New York never introduced the Town Meeting system. The results of this paper can be used to understand how competing forms of local government affect the administration of cities. Positive theories about representative government (such as Niskanen, 1975) have tended to focus on simplified models of representative democracy and ignore the variety and richness of institutional forms that exist in the field. The paper shows how interest groups can derive advantages from less hierarchical institutions of governance. Previous papers (Osborne et al., 2000, Osborne et al., 2005 and Turner and Weninger, 2005) have argued that attendance will tend to be low in assemblies with costly participation, and that the composition of voters will not generally be representative of the population. Borgers (2004), however, points out to the existence of negative externalities associated with higher attendance, which may actually make the outcome of poorly attended assemblies more efficient. Here I show that, in the context of urban city management, voting institutions cannot be analyzed independently of the existence of organized insider groups. Assembly institutions are susceptible to being captured by insider groups with low attendance costs, coincident and well-defined interests, and effective organization (Olson, 1971). Therefore more representative forms of city government seem more attractive in the presence of extreme preferences or organized insider groups and high participation costs, both in terms of allocative and productive efficiency. This work also enriches the existing empirical literature on assembly local government. Assemblies, direct democracy, and direct participation of the citizens in public decisions — these concepts have exerted a great intellectual attraction for many people, from libertarians to the Soviets.1 Despite this theoretical appeal, the empirical work on the performance of municipal assemblies relative to representative city governments is far from complete. The existing literature has focused on the total size of municipal budgets2 and displays a lack of consensus on the effects of assemblies: there does not seem to be a systematic impact of the form of governance on local expenditures and taxation. However, there may be other dimensions in which the two forms of city management differ such as the allocation of rents to interest groups explored here. Furthermore, previous papers in this literature share common econometric problems. They do not properly address endogeneity issues: omitted variables that are correlated with the propensity of having a representative government may also be correlated with the outcome. Also, representative and assembly-type municipalities are very different in terms of population. The literature has not taken seriously the issues of comparability of treatment and control samples, and the potential existence of nonlinearities in the impact of observable variables on the outcomes of interest. The paper therefore expands the existing literature in five new directions. First, it goes a step further in analyzing the institutional reality of actual assembly city governments, where the issues are not limited to voting preferences and rules. As in the recent corporate finance literature, I show that the existence of entrenched insiders is key to understanding the functioning of local government and the trade-offs of different institutional arrangements. Second, I examine the productive efficiency rather than allocative efficiency of municipal assemblies, focusing on an altogether different outcome — municipal wages. Third, I limit the samples by population similarity between representative and assembly governments to their common support in terms of population to enhance comparability. I also prove the results robust to nonlinearities. Fourth, the paper provides a direct link between attendance rates in local assemblies and outcomes. In so doing, it also presents new results on the determinants of contemporaneous attendance to the institution that can be considered the cradle of American democracy. Lastly, I introduce an exogenous control group: municipalities in New York State, which does not have the assembly institution. The paper proceeds as follows. Section 2 introduces the reader to the institutional features of the New England Town Meeting, and discusses the relation between the different economic theories of government and this work. Section 3 introduces the data and its sources. Section 4 presents the results of the analysis, and tests the robustness of the findings. Section 5 concludes.

How does changing from an assembly to a town-council form of government affect the way in which cities are run? Previous empirical research on this question has not found much of an impact for assembly governance on aggregate outcomes such as local public expenditures or taxation. Nevertheless, there may be more subtle ways in which local institutions may impact urban governance. Concretely, the specific role of organized insiders may be important to understand how cities and towns governed by citizens' assemblies are different from representative forms. Previous surveys by political scientists point to local workers as an important pressure group in local assemblies. The importance of public employees as a voting block in assemblies could be amplified by the low participation rates displayed by other citizens, especially in larger cities and towns. The results of this paper show that assembly local governments in the US pay somewhere between 4 and 10% higher wages than demographically-similar representative democracies. Using an exogenous assignment to the representative government control group, I show that the results are not likely to be a product of omitted variable bias. The wage premia associated with assembly governments are bigger in municipalities with low meeting attendance rates and for the employment functions with a larger number of employees. The results demonstrate how insider groups derive some advantages from an assembly form of government. More broadly, the potential capture of assemblies by insider groups can be an important risk faced by municipalities with low citizen participation, which provides a rationale for the widespread adoption of representative government at the local level.