The political economy of the Essential Air Service program

The transitional gains trap occurs when a government subsidy or regulation is put in place to benefit a particular group and those benefits become capitalized into asset prices. As a result, the current recipients are no better off from the program but will be harmed by its elimination. Two implications follow from the transitional gains trap that are relevant to the EAS program. First, subsidy recipients will be harmed by the loss of subsidies and thus will expend resources to ensure that they continue. Second, resources expended through the political process can bring new one-time benefits if subsidies are increased or new airports are added to the program.

Two exceptions to this are Sobel et al. (2007) and Beaulier et al. (2011). Both papers find no congressional committee influence following reforms, suggesting that in certain circumstances reform may be able to minimize the power of committee oversight. Sobel et al. (2007) find that political influence over FEMA by members of FEMA oversight committees was eliminated through the introduction of additional layers of bureaucracy. Beaulier et al. (2011) show that the base realignment and closure act seems to eliminate political influence by congressional committees overseeing the armed services by bundling together all closures into one up or down vote.

Other studies of the EAS focus primarily on the effects on local communities. Warren (2008) uses a matching methodology combined with a difference-in-differences approach to evaluate the impact of the EAS program on local economies and finds that the EAS program is positively related to county income, employment, and population. Grubesic and Matisziw (2011) argue that other Federal Aviation Administration programs seem to make the EAS program redundant when it comes to improving economic performance.

For instance, Morgantown Municipal Airport (MGW) in West Virginia, with help from the City of Morgantown, offered free flights to Wheeling, WV, in November 2014 to meet the annual passenger requirement of the EAS program (McCloud 2014). However, the airline with which MGW contracted subsidized service is allowed only to serve the route between Morgantown and the nearest hub in Washington, D.C.

See Lowell et al. (2011) for a case study from the coach bus industry about efficiency differences between subsidized air service and private bus service.

The EAS is funded out of the FAA’s Airport and Airway Trust Fund (AATF). Revenues deposited into this fund are obtained from aviation-related excise taxes on cargo, fuel and passengers (Federal Aviation Administration 2015). The AATF therefore falls under the purview of the Ways and Means Committee and we expect the Ways and Means Committee to be more politically important for the EAS than the Appropriation or Rules Committee.

Asking why the rules are less stringent for Alaska may be an interesting political economy question in and of itself, but is beyond the scope of the present article.

Communities in Alaska, Hawaii, and Puerto Rico are exempt from legislation dictating that communities must average ten passengers per day in order to remain eligible for subsidies from the EAS program, for instance.

Revenues to the AATF from excise taxes have risen nearly 40 % since FY 2000, according to the Federal Aviation Administration (2015). While beyond the scope of our paper, explaining the more generous funding of EAS airports relative to other possible budgetary priorities, such as expanding facilities and equipment at major airports, would be a fruitful avenue for future research.

Congressional representation data were retrieved from https://​github.​com/​unitedstates/​congress-legislators/​, while congressional committee membership was found at http://​web.​mit.​edu/​17.​251/​www/​data_​page.​html#0.

For robustness, we also estimated our results by Tobit due to the censoring of subsidies at zero. Doing so results in very similar results in terms of statistical significance, although the rules committee is statistically significant in some regressions and district-level economic variables are not statistically significant. To take into account that less than 10 percent of Congressional districts contain multiple airports and distance to the nearest hub, we also re-estimated Tables 2, 3 and 4 using subsidies per district airport as the dependent variable and average distance to hub as an explanatory variable. The results were quantitatively and qualitatively similar to the results presented in the paper.

In unreported regressions, we include the subsidy during the previous period as an explanatory variable to account for persistence in EAS contracts. For example, a district whose member is on the ways and means committee is replaced. The effect on the level of subsidy is unlikely to be immediate. While the subsidy the previous period is statistically significant, it does not quantitatively or qualitatively change our findings for committee representation.

It may be surprising to some that a House leadership position is not associated with larger EAS subsidies. When we looked at the identities of holders of congressional leadership positions over our sample period, we found that the majority of individuals did not represent EAS-eligible districts. Of the 20 individuals who occupied leadership positions, only six represented communities that were EAS eligible when the program was created. Of those six individuals, three represented districts did not receive subsidies under the EAS program, likely because while the districts may have been eligible when the EAS program was implemented, the area is unlikely to have met the criteria afterwards. For example, Congressman Brady represents part of Philadelphia, which was not an airport hub in the 1970s, but currently is one and, hence, is no longer eligible for the program. Similar considerations and effects are present when we look at committee chairs. We therefore believe that that variable is insignificant because the subsidies no longer are substantially important to the districts of those individuals who occupy these leadership positions, and thus these individuals do not push to receive these resources.

In the interests of space, we do not report results using the two-year-ahead subsidy amount for all our robustness checks in the paper. Those results, however, are quantitatively similar to those reported using the current period subsidy.

We end our analysis here as our sample size is sufficiently small that we have power issues when estimating our coefficients. Entering % Rural in column (1) of Table 4 leaves our findings unchanged, but does not show the rising magnitude of the effect of being a member of the Transportation and Ways and Means Committees as the sample becomes ever more rural.

It is more complicated to control for distance to the nearest hub in the congressional district regressions since several districts contain more than one EAS-eligible airport.

All non-airport, non-political variables are observed at the county level.