Academic journal article Atlantic Economic Journal

Does the Median Voter or Special Interests Determine State Highway Expenditures? Recent Evidence

Academic journal article Atlantic Economic Journal

Does the Median Voter or Special Interests Determine State Highway Expenditures? Recent Evidence

Article excerpt

Introduction

What determines public sector output? Is it determined by need? The voters? Policymakers? Bureaucrats? The answer likely depends on what public sector output is under consideration. For example, Hall et al. (2015) find while economic variables, such as median income, explain the amount of subsidies rural airports receive through the Essential Air Service program, having a member of Congress on the House Transportation or Ways and Means Committee explains a significant portion of the biannual subsidy a rural airport receives. Similarly, Garrett and Sobel (2003) find that nearly half of all Federal Emergency Management Agency (FEMA) spending from 1991 to 1999 was motivated by politics, not relief. In the case of FEMA, the influences were largely in states that were politically important to the President of the United States and in congressional districts where a member of Congress served on the FEMA oversight committee and thus had some influence over the agency. In the wake of Hurricane Katrina, the role that politics played relative to need became even more acute (Congleton 2006; Shughart 2006).

Sobel et al. (2007) updated the analysis of FEMA in the mid-2000s following the post-September 11 merger between FEMA and the Department of Homeland Security. Interestingly, they find that while states that are politically important for the President to continue to have a higher probability of disaster declaration, Congressional oversight through the committee system no longer explains the amount of disaster declarations. The authors conclude that the merger appeared to have reduced some of the political pressures within FEMA to allocate resources based on politics, not need. In a similar vein. Beaulier et al. (2011) find that traditional political factors such as Congressional oversight or importance to the President no longer played a role in military base closures following the passage of the Base Realignment and Closure Act. These papers and other updates of historical political economy models such as Hall et al. (2012) and Sobel and Hall (2016) show both the importance of replication (Aim 2010) and how the importance of political economy factors, with respect to the outcome of collective decision-making, can change over time.

In this paper we both replicate and update an important paper in the median voter model literature in order to see how the empirical model holds up nearly thirty years later. Congleton and Bennett (1995) is a valuable paper in the public choice literature for at least two reasons. First, the paper was the earliest to explore the determinants of state road expenditures from the perspective of positive economics--trying to explain what actually exists rather than what should exist. Given that the public sector spends billions to build, operate, and maintain highways in the United States, this was an important oversight in the literature given that normative public finance suggests that highways might be a good underprovided by the market (Oakland 1972). Second, the paper was the first to explicitly incorporate both the median voter model and special interests in the same theoretical and empirical model. It is for that reason that Congleton and Bennett (1995) have been highly cited regarding both the determinants of public highway funding (Goel and Nelson 2003; Gamkhar 2003; Waiden and Eryuruk 2012) and the political economy of state policy more generally (Ahmed and Greene 2000; Sutter and Poitras 2002; Hersch et al. 2004; Hall and Schiefelbein 2011).

We begin by discussing our replication of Congleton and Bennett (1995). To preview our results, we can replicate their data, except for Sierra Club membership by state, which we were not able to obtain for the year of their study. Even missing that variable, our results are quite similar to theirs. We estimate their main empirical specifications (minus Sierra club membership) for 2013. Our findings suggest that the dynamics of state government expenditures have changed considerably in nearly three decades. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.