with Thomas Stratmann

We model and empirically test the effects of citizen monitoring on services provided by bureaucrats. Monitoring by citizens is a public good. Because of collective action problems, monitoring is underprovided, allowing bureaucrats to shirk efforts to provide services. Our model shows that collective action problems in monitoring activities are associated with sub-optimal bureaucratic output. Bureaucratic output is predicted to change with the number of citizens affected and the distribution of bureaucracy-generated benefits. Utilizing income data from leases under the purview of the Bureau of Indian Affairs (BIA), we find broad support for our hypothesis that bureaucratic output is inversely related to collective action challenges of bureaucrats’ clients. These collective action problems vary with the number of owners, interests of the largest shareholder, and variations in monitoring costs due to private vs. institutional ownership.

I advance and test a theory that in sequential auctions price rises with the number of bidders. I allow for stochastically arriving and departing bidders, so the number of bidders changes with every auction round both endogenously through the winner of the previous round dropping from future rounds and exogenously through the bidders’ stochastic arrival and departure. I test the theory on the Mecum auctions for collectible cars using the instrumental variables method. The timing of the car going to auction affects price only through the number of bidders present at the time and the number of cars still left to auction. This allows me to instrument time for the number of bidders. The empirical test shows support for the theory and provides a missing explanation for the declining price anomaly prevalent in sequential auctions.

I estimate the aggregate and industry-specific elasticities of scale and markups for the U.S. economy over the period from 1980 to 2019 using data on publicly traded companies. I apply Olley-Pakes and Ackerberg-Caves-Frazer estimation methods and find that the aggregate elasticity of scale for the U.S. economy is 1.1 and has been rising. The elasticity of scale in turn serves as an input for calculating industry markups. Increasing returns to scale help explain observed increases in markups over the last decades for broad sectors of the economy. My estimate of 1.2 for the aggregate markup is significantly lower than the estimate of 1.6 found in recent literature. The large disparity in markup estimates stems from differences in the treatment of fixed and variable costs and the methodological approach to the calculation of markups.