The political economy of abandoned mine land fund disbursements

Troyan, Jessi; Hall, Joshua C. Article The political economy of abandoned mine land fund disbursements Economies Provided in Cooperation with: MDPI – Multidisciplinary Digital Publishing Institute, Basel Suggested Citation: Troyan, Jessi; Hall, Joshua C. (2019) : The political economy of abandoned mine land fund disbursements, Economies, ISSN 2227-7099, MDPI, Basel, Vol. 7, Iss. 1, pp. 1-17, https://doi.org/10.3390/economies7010003 This Version is available at: https://hdl.handle.net/10419/256936 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ economies Article The Political Economy of Abandoned Mine Land Fund Disbursements Jessi Troyan 1and Joshua Hall 2,* 1Cardinal Institute for West Virginia Policy, Charleston, WV 25339, USA; jltr[email protected] 2West Virginia University, Chambers College of Business & Economics, Morgantown, WV 26506, USA *Correspondence: [email protected]; Tel.: +1-304-293-7870 Received: 14 December 2018; Accepted: 4 January 2019; Published: 10 January 2019   Abstract: What factors determine federal spending on environmental goods? Is severity of the hazard the only metric of consideration, or do other factors play a vital role in explaining spending? This paper seeks to answer this question and to identify disbursement patterns within the context of the Abandoned Mine Land Fund (AMLF) program, a fund created as an aspect of the Surface Mining Control and Reclamation Act of 1977. We explore whether political factors, as well as environmental and health factors, have an explanatory role in disbursement of AMLF monies. The political factors examined include environmental interest group influence and legislator preferences and/or pressures to fund sites in their home states or districts. The results found here suggest that there exists a mix of public and private interests present in AMLF disbursement decisions during the overall span of the program, and that political influences have gained strength in the decision-making calculus in response to changes in the funding structure of the AMLF. Keywords: public choice; public interest; seniority; mining; political economy JEL Classification: D7; H5 1. Introduction What are the determinants of federal spending on environmental goods? Is severity of the environmental hazard the only metric of consideration, or do other factors play a vital role in explaining how funding is disbursed? In this paper, we aim not only to provide answers to these questions, but also to illuminate disbursement patterns of monies from the Abandoned Mine Land Fund (AMLF) that go toward the reclamation of abandoned mine sites throughout the United States. Though the program itself is small, this analytical setting is interesting because of the limited scope of program objectives and the rigidly defined funding source (at least initially). This suggests that the execution of abandoned mine reclamation projects facilitated by the fund should be difficult to influence politically. Within this setting, we examine whether political factors, in addition to severity of abandoned mine site hazard and general abandoned mine site characteristics, influence the distribution of AMLF monies. This project contributes firstly to the economic analysis of government activity. Broadly, there are two economic strands of thought regarding government action. The first is that government action is motivated primarily by the public interest. Congruent with this perspective, money from the AMLF would go toward sites that pose the most severe environmental risks to the general public. The other perspective sees government action as being subject to the influences of interest groups and politician self-interest. Orthogonal to the vision of a benevolent government, these concentrated interests are seen as the driving influences of fund disbursement. López and Leighton (2013) provide a good overview of both of these theories in their analysis of the role of ideas in political change. Economies 2019,7, 3; doi:10.3390/economies7010003 www.mdpi.com/journal/economies Economies 2019,7, 3 2 of 17 Within the existing literature, little has been said specifically regarding the AMLF as an environmental remediation program. However, work has been done with respect to other similarly intended programs such as Superfund that parallels the questions asked in this paper. Barnett (1985) and Hird (1990,1993,1994) offer support for the public interest perspective with their findings that Superfund monies and efforts are allocated toward remediating the most severe sites. The evidence in these works, however, is not strong enough to reject the hypothesis that there are zero pork-spending influences in Superfund allocations but does suggest that these influences are minor with respect to total disbursements. Nonetheless, McNeil et al. (1988) find evidence for the pork-spending hypothesis with respect to tax implementation and subsequent spending. They found that evidence in EPA data that taxes for Superfund were often collected in certain areas, but the bulk of spending occurred elsewhere. Likewise, the theory of rationally self-interested politicians is likewise supported throughout other works. Stroup (1996) and Yandle (1992) argue that Superfund monies are sought by politicians to be brought back to their home constituencies. Tilton (1995) draws comparisons between Superfund and the AML program under the Surface Mining Control and Reclamation Act (SMCRA) in the ways each of these programs dealt with problems of past pollution and finds that AML fares better in terms of assigning responsibility for the past pollution and mitigating future production cost uncertainty associated with liability costs. Nonetheless, the scope of the analysis does not consider how political factors or pure environmental or human well-being concerns affect the execution of either of these pollution mitigation programs. The paper closest to ours in the Superfund literature is Stratmann (1998), who uses a political economy model similar to ours to look at the geographic disbursement of Superfund expenditures to separate out public interest and public choice influences. We use this literature to inform our study of the Abandoned Mine Land Fund. In Section 2, we provide an overview of the history of the AMLF. We then discuss our theoretical framework in Section 3 and follow that up with information on our data in Section 4. Section 5presents our empirical results, with Section 6concluding. 2. History of the Abandoned Mine Land Fund The AMLF was created as a part of SMCRA in 1977. Bamberger (1997) provides a nice overview of the history of the AMLF through 1997. Yonk et al. (2017) have a more recent overview of the history of SMCRA through a public choice lens. The overarching goal of SMCRA is to establish a federal standard for environmentally responsible surface mining, and restoration of the lands after mining has ceased to ensure mitigation of adverse environmental effects of this method of extraction. Due to lax enforcement of state mining regulations prior to the passing of SMCRA, many smaller sites were subject to “blast and grab” mining techniques where small, independent miners/mining operations, often referred to as wildcatters, would use whatever least cost method was at their disposal to expose and extract coal. Afterwards, they might simply leave the area thereafter with no attempt at reclamation. These practices resulted in degradation of land and environmental quality. While small sites characterized the abandoned mines, the prevalence of these sites was the main contributor to alarm. Consistent with the broader goals of SMCRA, the objective of AMLF is to provide for a general funding pool to be allocated towards reclamation efforts of already existing abandoned mine sites, in addition to SMCRA’s efforts of enforcing reclamation on present and future mine sites. As delineated within SMCRA, monies in AMLF may be used for purposes including, but not limited to, reclamation and restoration of abandoned coal surface mines, processing and disposal areas, sealing and filling of deep mine entries, land restoration to mitigate erosion and sedimentation, waterbed restoration, construction and operation of water treatment plants, pollution mitigation for burning coal refuse disposal, and control of coal mine subsidence. The coffers of the AMLF are provided for via a fee levied on extracted coal as specified in section 403 of SMCRA: Economies 2019,7, 3 3 of 17 All operators of coal mining operations subject to the provisions of this Act shall pay to the Secretary of the Interior, for deposit in the fund, a reclamation fee of 35 cents per ton of coal produced by surface coal mining and 15 cents per ton of coal produced by underground mining or 10 per centum of the value of the coal at the mine, as determined by the Secretary, whichever is less, except that the reclamation fee for lignite coal shall be at a rate of 2 per centum of the value of the coal at the mine, or 10 cents per ton, whichever is less. These fees were initially slated to expire in 1992, but extensions have been passed by Congress to maintain the collection of fees, thereby continuing the reclamation of abandoned mine sites. Out of the monies collected from domestic coal production fees, 50% of those collections are allocated to the states. The remaining half is allocated across three broad objectives and falls under federal discretion and the control of the Secretary of the Interior. Ten percent of these funds are marked for allocation into the Rural Abandoned Mine Program. Twenty percent of the funds are funneled into a pool that is used for supplemental grants going toward remediation of more hazardous sites. The remaining 20% can be described as a portmanteau pool where funds are used for emergency projects, federal administrative costs, projects in states without approved reclamation plans, and the Small Operator Assistance Program. Currently, the Office of Surface Mining (OSM) has collected over $10.1 billion worth of fees toward the AMLF. Out of that total, over $7.6 billion has been distributed. Furthermore, OSM estimates that over $3 billion worth of high priority sites remain to be remediated. Back-of-the-envelope arithmetic suggests that the mission of OSM with respect to reclaiming abandoned mines is nearing its twilight. However, one should be careful to avoid the assumption that the volume and severity of abandoned mine sites exists in a static state. Instead, these should be considered in a dynamic light for reasons such as the potential future hazardous deterioration of presently stable underground mines. In the nearly forty years since the passing of SMCRA and the creation of the AML reclamation program, various changes to SMCRA itself have been implemented and the funding structure of AMLF has likewise been altered. The first structural change relevant to this analysis is the enactment of the Abandoned Mine Reclamation Act (AMRA) of 1990 that provided for the accruement of interest on AMLF balances that were not appropriated. This marks the first codified instance in the lifespan of the program that incentivizes any change in allocation patterns. Specifically, this act provides the incentive to decrease general allocations and hold a balance in the AMLF from year to year in order to grow the fund absent additional taxes collected or to mitigate cycles in funding due to cyclical coal production. The next major restructuring of reclamation funding occurred with the passing of the SMCRA Amendments Act of 2006. First, it incrementally reduces the taxes levied per ton of coal produced through September 2021. However, this rate reduction is offset by requirement of “Treasury payments to certified states and tribes in lieu of payments from AMLF.” A further stipulation, requiring future AMLF allocations to be based upon historic coal production, also shifts the allocation patterns. The final restructurings of the AML program relevant to this study are tucked away within the passages of Public Law (PL) 112-141, the “Moving Ahead for Progress in the 21st Century Act,” and PL 113-40, the “Helium Stewardship Act of 2013.” In short, minutia within these bills cap the amount of money allocated to a given state or tribe out of the in-lieu Treasury funds established in the SMCRA Amendments Act annually at $15M and $28M, respectively. 3. Theoretical Framework This paper’s conceptual framework incorporates tenets of economic theories of regulation, bureaucracy, and interest group influences in order to provide a more cohesive and comprehensive explanation of the disbursement of AMLF monies. Within the economic theory of regulation, legislators maximize their acquisition of support from among competing constituencies. Theoretically and practically, this predicts that legislators do not adopt a corner solution with respect to helping one constituency versus another. Instead, in equilibrium, individual legislators make support trade-offs at the margin in order to attain the optimal amount of support. Within the context of AMLF decision Economies 2019,7, 3 4 of 17 making by politicians, the relevant constituencies to consider are interest groups who may provide financial electoral support and the general public responsible for vote counts. From this scenario, both the severity of environmental hazard posed by the abandoned mine site as well as the interest group strength possess important explanatory power regarding AMLF disbursements. The equilibrium framework of regulation and interest groups implies that bureaucrats are politicians’ pawns in executing legislation (Becker 1983;Peltzman 1976). Therefore, it logically follows that such models have no room for bureaucratic influence in policy decision-making. However, if the bureau has significant enough autonomy in pursuing its own interests, as the Office of Surface Mining, Reclamation and Enforcement generally does, the economic theory of bureaucracy must be introduced into this analysis in order to more accurately explain the patterns of AMLF disbursement. Within the economics of bureaucracy, there are also multiple competing explanations of the autonomy and objectives of bureaucracies. One hypothesis suggests that bureaucrats pursue their own interests within the organization, and those interests do not consistently align with the law’s intent (Niskanen 1971). McCubbins et al. (1989) put forth a hypothesis where bureaucrats and politicians have differing objectives, but that constraints such as budget appropriation, administrative rules, and oversight can effectively curb purely bureaucratically interested actions. Weingast and Moran (1983) suggest an even more constrained theory of bureaucratic action with their congressional control hypothesis. Potentially more important than the theories of regulation and bureaucracy is the relationship between legislators and the implementation of the regulations they are responsible for enacting. Consistent with the public choice view that politicians act in their own self interest—substantively meaning they take actions that are likely to increase their chances of job security by means of re-election—politicians have a vested interest in securing funds for their respective states and/or districts. Doing so increases local aspects such as environmental quality, real estate values, and potential tourism revenues. Likewise, reclamation activities could be expected to provide positive employment effects in the area. While the employment effects may only be relevant in the short run, they also typically fall within the reelection timeframe, thereby further incentivizing politicians to secure this virtually ‘free lunch’ money. As such, this interaction suggests that senior politicians will be more effective in funneling AMLF allocations to their home regions. From a policy perspective, it is important to know the extent to which political factors play a role in the distribution of public funds. This is especially true when politics is not supposed to play a role. If political influence can be identified, then perhaps a change can be made to political institutions to remove politics from the process. For example, Garrett and Sobel (2003) find that Federal Emergency Management Agency (FEMA) disaster expenditures are higher in states where their members of Congress serve on FEMA oversight committees. A post 9-11 reorganization of FEMA, however, removed this form of political influence according to Sobel et al. (2007). Similarly, Twight (1989) highlights how politics prevented the military from closing or realigning any domestic military bases from 1960 to 1988. Reforms in the late 1980s led to political factors no longer playing a role (Beaulier et al. 2011). By looking at all the institutional changes in the AMLF program over time in one paper, we provide insight into how institutional changes may have influenced the role of politics in the allocation of funding. Furthermore, the nature of AMLF disbursements with respect to whether it more adequately can be described as a disguised welfare program, or general spending on environmental goods has important implications for how we predict funds to be allocated. 4. Data and Model Estimation As described in the preceding section, the explanatory factors of AMLF disbursement are many. The theoretical underpinning suggests that funding of abandoned mine reclamation projects is a function of hazard severity, and the characteristics of legislators, interest groups, and bureaucratic agents. This general empirical model has been used to study agency dependency (Anderson and Potoski 2016), federal transportation disbursements, (Bilotkach 2018), airport funding under the Economies 2019,7, 3 5 of 17 Essential Air Service Act (Hall et al. 2015), federal disaster declarations and assistance (Husted and Nickerson 2014), NIH funding (Batinti 2016), antitrust enforcement (Dove and Dove 2014), and even disbursement of the swine flu vaccine (Ryan 2014). Such an empirical model will be used to explain and predict funding tendencies of projects that fall under the umbrella of AMLF reclamation objectives. The empirical model to be estimated is an ordinary least squares model that includes state and year fixed effects in order to account for unobserved variations in political and economic conditions throughout the span of the data set. For policymakers not familiar with the methodology employed in these analyses, the goal is to estimate an empirical model that explains variation in AMLF funding. If done properly, we can isolate the effect of specific factors holding constant other variables that might influence AMLF funding. For example, a positive and statistically significant coefficient on whether a mine was an underground mine strongly suggests that underground mines receive higher levels of AMLF funding because policymakers believe they are more costly to remediate, other things being equal. Similarly, a positive and statistically significant coefficient on any variables measuring political influence suggests that political oversight of the AMLF influences the allocation of funds. The equation we estimate is: AMLF Disbursementsi,t=β0+β1HAppSeni,t+β2SHAppSeni,t+β3GreenIndexi,t+β4Incomei,t +β5Privatei,t+β6Statei,t+β7Pri1i,t+β8Pri2i,t +β9Surfacei,t+β10Undergroundi,t+β11Bothi,t+β12Processingi,t +γi+δt+σi,t. (1) The primary dependent variable is the SMCRA-funded AMLF allocation towards reclamation of abandoned mine sites throughout years spanning from 1984 through 2013. These are presented in thousands of inflation-adjusted (1984) dollars. Additional regressions are estimated with the dependent variable as a standardized measurement of SMCRA-funded AMLF allocation per unit of area on a given site to address variation in the size of abandoned mine sites. The data is collected from the Office of Surface Mining Reclamation and Enforcement’s Abandoned Mine Land Inventory System, e-AMLIS. This database consists of an inventory of land and water impacted by past mining endeavors. It is detailed to the extent of including information regarding location, type, and extent of damages as well as reclamation costs. Data is provided by the states managing their own abandoned mine problems or through the OSMRE office responsible for managing these cases where states do not bear that responsibility. In this analysis, only reclamation sites that have been funded to some extent by AMLF are included. However, there exist other abandoned mine sites within the database that have simply not been allocated funding or they have been completely reclaimed through private efforts and funding. Table 1presents summary statistics for the entire sample. Severity of the environmental hazard is measured by the priority status assigned to each abandoned mine site, per problem type, by OSMRE. There are five tiers of priority assigned to inventoried sites. Within this analysis, priority types are coded as dummy variables, so as to treat each level of hazard independently without assuming a linear scale in the degree of hazard. The most serious abandoned mine land problems are those that pose a threat to health, safety, and general welfare of people. These are assigned either Priority 1 or Priority 2 status, and are the only problems required by law to be inventoried. Within these top two priorities, there are seventeen different problem types accounted for—noted without respect to severity. Those problems that have only environmental impacts are classified as Priority 3 problems and are included in the inventory when reclamation on these sites is funded, in some proportion, out of AMLF. Priority 4 and 5 sites consist of lower severity coal related problems such as public facilities and development of public lands. These lower priority reclamation projects have fewer records kept on them, are less likely to receive AMLF monies, and are not included in this dataset. Intuitively, funding amounts are predicted to align with priority levels; the higher the priority, the greater the funding allocated. (As pointed out by an Economies 2019,7, 3 6 of 17 astute referee, this suggests that Priority 1 sites are more costly to reclaim. We are not aware of any data on the cost of reclamation per site type. Our intuition here is driven by the fact that the difference between Priority 1 and Priority 2 sites is that Priority 1 sites are categorized as such because they pose a higher threat to the health and safety of the general public. To us, this suggests higher costs related to the urgency of the reclamation project, in addition to the difficulty of remedying health hazards.) For purposes of this analysis, only Priority 1 and 2 sites are considered, and take a value of one if applicable and zero otherwise. Table 1. Summary statistics for database of abandoned mine land sites, 1984–2013. Variable N Mean Std Dev Min Max Private Ownership 35,528 66.822 46.192 0 100 State Ownership 34,021 4.062 18.455 0 100 House Appropriations Seniority 36,311 27.715 20.646 0 78 Senate Appropriations Seniority 36,311 7.407 9.653 0 51 Environmental Group Strength 36,314 7.132 2.694 3.400 13.10 Priority 1 36,314 0.208 0.406 0 1 Priority 2 36,314 0.590 0.492 0 1 Underground Site 36,314 0.397 0.489 0 1 Surface Site 36,314 0.345 0.475 0 1 Surface + UG Site 36,314 0.238 0.426 0 1 Processing Site 36,314 0.016 0.125 0 1 AMLF Allocation (1000) 36,314 108.793 352.075 0.0004 13,984 Per Capita (PC) Income (1000) 36,311 26.957 8.447 13.358 68.80 House Appropriations Member 36,311 0.882 0.323 0 1 Senate Appropriations Member 36,311 0.549 0.498 0 1 Along with priority designation, abandoned mine sites can be categorized by the type of mining that occurred on site that now requires reclamation. Here, there are four different mine site types accounted for in the e-AMLIS inventory—surface, underground, both, and processing. Presumably, project sites where only processing mining operations occurred would be predicted to receive larger AMLF allocations since the reclamation project is inherently more involved due to the fact that reclamation activities would predominantly involve cleaning up the chemicals involved in processing coal for use. Processing sites have a higher propensity for causing harm to general health, safety, and human well-being. Surface mines are predicted to receive smaller allocations than processing sites, but the largest allocations with respect to the other extraction sites. This is due to the physical nature of the reclamation project itself. Reclamation of abandoned surface mine sites would involve greater terrain restructuring, re-vegetation, and waterway cleaning and restoration. Purely underground abandoned mine sites would be predicted to have the smallest allocations due to the less involved nature of the reclamation project. In these cases, the reclamation process would be primarily characterized by mineshaft reinforcement to prevent cave-ins, and mine entry sealing to prevent, or at least reduce the risk of people entering abandoned deep mines. Finally, abandoned mine sites where both surface and underground mining occurred are predicted to receive AMLF allocations between the size of purely surface or underground sites receive since costs can be diffused across reclamation of both kinds of operations. In theory, the size of distribution on sites where both surface and underground mines occurred would be on a spectrum from pure surface to pure underground with the amount being a weighted average of the proportional combination of mine types. However, data to this extent of detail is unavailable. As such, the amounts predicted reflect an aggregated average of proportion. All four categories are considered in this analysis. Like the priority sites, the site characteristic variables are coded as dummy variables with a value of one corresponding to the relevant sites and zero otherwise. Additional site specifics are accounted for by ownership characteristics of the land where the given abandoned mine site is located. By the e-AMLIS classifications, there are seven different possible Economies 2019,7, 3 7 of 17 categories of landowners, not all of which hold exclusive ownership rights to the given land area. The seven potential stakes are private, state, tribal, Bureau of Land Management, forest service, national park, and a catchall category of other federally owned lands. These ownership stakes are provided as percentages. Among these ownership stakes, we would expect that proportion of private ownership and size of AMLF disbursement will be inversely related. Conversely, higher proportions of state-owned lands on abandoned mine sites are likely to receive the greater AMLF allocations. The remaining five types of land ownership are not considered for purposes of this analysis due to the fact that they make up a miniscule proportion of abandoned mine site ownership on any given site and correspond to a relatively small number of mine sites in the data set. Nonetheless, each of these would be expected to exhibit similar allocation patterns as state-owned lands since they also fall under the broader category of publicly owned lands. Characteristics of legislators are accounted for by seniority and membership on fiscally relevant congressional committees. The main variables included are cumulative seniority of members on the Senate Appropriations, and House Appropriations committees by year and by state. This data is collected from the respective committee’s history websites. For all of these variables, a positive relationship is expected between committee seniority and AMLF allocations. Relatively higher positive relationships are expected for House Appropriations committee members’ seniority since they represent a smaller constituency relative to senators. Variables concerning interest group strength are collected from the 1991–1992 Green Index (Hall and Kerr 1991) . Specifically, this index considers membership per 1000 state residents in environmental organizations, namely Greenpeace, the National Wildlife Federation, and the Sierra Club in 1990. Ideally, this index would be more current, perhaps updated annually. Nonetheless, this provides the most current and comprehensive measure of environmental interest group presence across states. This variable is predicted to have a positive influence on the AMLF allocation through the mechanism of these environmental interest groups pressuring representatives to secure funding for reclamation sites in their respective states. If politicians are self-interested, they have an incentive to respond to these vocal members of their constituency. Real state per capita income is included to account for constituent demand for environmental goods. In addition to the broad inspection of how these factors influence the allocation of AML funds over the recorded lifespan of the program, each of the legal changes in the funding structure previously mentioned are considered, period-by-period to examine the extent to which these changes alter the respective public interest and political influences on AML reclamation funding. The general predictions regarding these legal changes in the funding structure are simple and intuitive. At the outset of the program, the expectation is that AMLF distribution patterns follow the intentions of the program, to reclaim hazardous abandoned mine sites, without respect to outside political sway. When AML funding expands from being a purely fee-based pool such as with the passing of SMCRA ‘06, the political influences on allocation decisions will gain gravity. Likewise, as Treasury payments to states and tribes are capped, as is the case with the passage of PL 112-141, that same political influence on allocation decisions will at least wane, if not drop completely out of the distribution calculus. For purposes of this analysis, abandoned mine sites on Indian Reservation lands are omitted due primarily to the inconsistencies associated with the political variables in question. Given that the reservations are viewed as sovereign entities within United States territory, there exist no measures of seniority within the House and Senate Appropriations committees or within the Green Index for these territories. The lack of a complete set measures render introduction of analysis of AMLF distribution patters on reservations problematic. 5. Empirical Results Regression results are presented in Tables 2–6. In all tables, specifications (1) and (2) report results with AMLF allocations in inflation-adjusted dollars as the dependent variable, while (3) and (4) Economies 2019,7, 3 8 of 17 give these same results with respect to the funding-per-metric unit standardization as the dependent variable. All results are estimated using ordinary least squares (OLS). Specifications (1) and (3) do not include state and year fixed effects, while specifications (2) and (4) do include state and year fixed effects. 5.1. AMLF 1984–2013 Table 2gives an overview of the AMLF distribution patterns over the entire scope of our data set—consisting of 33,947 mine site observations in specifications (1) and (2) and 33,313 observations in (3) and (4) over a nearly thirty-year span. Across all specifications, sites with a Priority 1 ranking are granted larger AMLF allocations. This can be seen in the positive and statistically significant coefficient on the variable Priority 1 in specifications (1), (2), (3), and (4). This result also holds with respect to site-only considerations for Priority 2 abandoned mine sites, though not in funding-per-unit estimations. Table 2. The determinants of receiving AMLF site funding, 1984–2013. Variable (1) (2) (3) (4) House Appropriations Seniority 0.125 −1.208 *** 0.058 0.263 ** (0.092) (0.197) (0.053) (0.115) Senate Appropriations Seniority 3.542 *** −0.290 1.146 *** 1.217 *** (0.203) (0.356) (0.117) (0.206) Environmental Group Strength 2.639 *** −29.644 * −3.006 *** −39.049 *** (0.785) (16.630) (0.451) (10.894) PC Income (1000) 0.376 −0.249 1.118 *** 1.373 * (0.255) (1.383) (0.147) (0.804) Private Ownership −0.197 *** −0.177 *** −0.028 0.023 (0.042) (0.051) (0.024) (0.030) State Ownership 0.582 *** 0.440 *** 0.520 *** 0.571 *** (0.106) (0.111) (0.061) (0.065) Priority 1 48.108 *** 69.563 *** 38.658 *** 39.596 *** (5.883) (6.266) (3.405) (3.646) Priority 2 72.863 *** 80.359 *** 19.313 *** 17.515 *** (4.700) (4.832) (2.717) (2.811) Surface Site −10.978 28.915 −31.990 ** −4.532 (27.547) (27.872) (16.042) (16.322) Underground (UG) Site −59.629 ** −20.963 −6.573 15.199 (27.449) (27.677) (15.987) (16.211) Surface + UG Site 2.272 46.466 * −18.253 9.503 (27.609) (27.942) (16.078) (16.363) Processing Site 43.544 89.884 *** −11.778 6.480 (30.910) (31.189) (17.959) (18.233) Constant 31.794 687.050 *** 20.547 670.500 *** (28.832) (167.982) (16.759) (108.005) State and Year Fixed Effects? No Yes No Yes Observations 33,947 33,947 33,313 33,313 Adjusted R20.028 0.059 0.016 0.033 Residual Std. Error 332.785 327.470 189.952 188.332 F Statistic 82.279 *** 33.092 *** 46.128 *** 18.065 *** Notes: Dependent variable in specification (1) and (2) is per site AMLF allocations in inflation-adjusted dollars, while the dependent variable in specifications (3) and (4) is per site AMLF allocation per metric unit of the mine site. Numbers in parentheses are standard errors. *** p<0.01, ** p<0.05, and * p<0.1. Priority 2 sites are predicted to receive greater allocations in the site-only specifications. This may be largely accounted for by the fact that Priority 2 abandoned mine sites make up nearly 60% of the full data set, whereas Priority 1 sites make up only slightly more than 20% of the full data set. A snapshot of the distributive patterns suggests that holding other considerations constant, including state and Economies 2019,7, 3 15 of 17 estimates suggest that a Priority 1 reclamation project is expected to receive roughly an additional $224,000 per site, or roughly $146,000 per unit area reclaimed. Priority 2 projects are estimated to receive roughly an additional $245,000 per site, or roughly $79,000 per unit area from AMLF allocations. Table 6. The determinants of receiving AMLF site funding, PL 112-141 to PL 113-40. Variable (1) (2) (3) (4) House Appropriations Seniority −6.191 * −9.586 −3.056 ** 45.847 (3.224) (61.732) (1.527) (29.571) Senate Appropriations Seniority 5.301 ** 1.664 2.028 * 0.161 (2.268) (4.219) (1.078) (2.027) Environmental Group Strength −2.147 −34.625 −5.658 151.990 (10.854) (205.311) (5.178) (98.403) PC Income (1000) 7.622 * 11.931 3.373 * 9.696 (3.952) (12.588) (1.905) (6.039) Private Ownership 0.434 −0.051 0.348 0.412 (0.569) (0.628) (0.271) (0.305) State Ownership 1.927 1.093 0.079 0.100 (1.808) (1.845) (0.879) (0.908) Priority 1 89.946 223.694 *** 123.174 *** 145.645 *** (70.781) (82.097) (33.687) (39.776) Priority 2 185.607 *** 244.542 *** 68.527 ** 78.817 ** (62.766) (64.825) (29.791) (31.242) Surface Site 245.520 324.284 120.021 161.992 (363.392) (363.093) (171.561) (173.803) Underground Site 185.177 250.504 128.050 137.683 (361.812) (359.834) (170.785) (172.166) Surface + UG Site 319.017 382.575 161.958 167.584 (363.769) (361.811) (171.681) (173.102) Processing Site 1736.858 *** 2275.683 *** 206.145 240.879 (411.611) (425.240) (194.294) (203.458) Constant −593.614 −618.558 −270.164 −2090.760 ** (414.049) (2124.185) (196.509) (1016.938) State and Year Fixed Effects? No Yes No Yes Observations 832 832 816 816 Adjusted R20.096 0.117 0.018 0.015 Residual Std. Error 609.675 602.616 287.271 287.738 F Statistic 8.389 *** 4.448 *** 2.243 *** 1.382 * Notes: Dependent variable in specification (1) and (2) is per site AMLF allocations in inflation-adjusted dollars, while the dependent variable in specifications (3) and (4) is per site AMLF allocation per metric unit of the mine site. Numbers in parentheses are standard errors. *** p<0.01, ** p<0.05, and * p<0.1. Further bolstering the suggestion that AMLF distribution patterns returned to intended purposes are the estimates on the ownership characteristic, political, and economic variables. In none of the four specifications is there evidence that either private or state proportions of land ownership have any influence on funding decisions. With respect to seniority of House and Senate Appropriations committee members, there is evidence in specifications (1) and (3). However, when the state and year fixed effects are introduced into the specifications, all statistical significance drops from the coefficient estimates, thus suggesting that these influences are moot. A similar story can be told with respect to the estimates on the environmental interest group variable. Again, the pattern holds with respect to per capita income in a state; positive statistical significance is estimated on the coefficients in (1) and (3), but significance drops in (2) and (4) once the fixed effects are included. Once again, there is little evidence in the general trend that mine site type is an influential factor in funding decisions. Out of the four mine types, only processing mines were estimated to receive larger AMLF allocations. Coefficient estimates were found to be positive and statistically significant only at the site level, at 1736.858 and 2275.683 in specifications (1) and (2), respectively. These estimates suggest that processing sites during the July 2012–December 2013 time frame received upwards of two Economies 2019,7, 3 16 of 17 million dollars in reclamation funding. However, when considering funding per unit area of a site, there is no discernible relationship to be found for any of the potential abandoned mine types. 6. Conclusions This paper examines the question of what determines the size of disbursements from AMLF to support reclamation projects on abandoned mine sites. Specifically, it examines if the severity of environmental hazard is solely responsible for AMLF allocations and the magnitude thereof or whether other political and economic forces influence the federal funding of abandoned mine site reclamation. Overall, the evidence suggests that funding for abandoned mine reclamation is a mixture of the products of public and political interests. With the exception of the 1984–1991 time span, sites designated as Priority 1 or 2 consistently are predicted to receive larger disbursements. However, in these time spans, political influences—especially through Senate Appropriations committee tenure and state-ownership of lands—wield consistently strong and significant weight on allocation decisions. This political influence is most pronounced in the years after the AMLF coffers are provided for with Treasury funds in addition to fees levied on domestically extracted coal. After the allocations out of Treasury funds are capped to states, the political influence wanes and the hazard level of sites again becomes the primary influential factor in funding receipts—further bolstering a public interest view of the AML program in total. From a policy perspective, the biggest takeaway from our findings are that political institutions can be changed to remove politics. While the AML program, and the mission of reclaiming abandoned mine lands in total is but a small mission in the scope of federal activities, examination of the funding distribution trends in light of differing institutional contexts analyzed here shed light on how similar programs can be more effectively implemented under the federal umbrella. In short, a program with minimal scope of objectives and funded via taxes/fees—implying a hard budget constraint—limit the extent to which political influences can sway the decision-making calculus of monies allocated through the program—furthermore, supposing the introduction of a softer budget constraint through federal appropriations toward a given project in question, capping the distributions out of that portion of funding likewise limit the extent of political sway. In this sense, our finding contributes to similar papers in public choice showing how institutional reforms can reduce or remove the influence of politics (Beaulier et al. 2011;Hall and Williams 2012;Sobel et al. 2007). Author Contributions: Writing—Original Draft, J.T.; Writing—Review and Editing, J.H. Funding: This research received no external funding. Acknowledgments: The authors acknowledge and appreciate the feedback of Thomas Stratmann. Conflicts of Interest: The authors receive no funding for this paper. During his career, Joshua Hall has received funding from the Charles Koch Foundation, the Thomas Smith Foundation, the Alliance for Markets Solutions, the Institute for Humane Studies, Liberty Fund, and over 30 colleges and universities. References Anderson, Sarah E., and Matthew Potoski. 2016. Agency structure and the distribution of federal spending. Journal of Public Administration Research and Theory 26: 461–74. [CrossRef] Bamberger, Robert. 1997. The Abandoned Mine Land Fund: Grants Distribution and Issues. Washington: Congressional Research Service. Barnett, Harold C. 1985. The allocation of superfund, 1981–1983. Land Economics 61: 255–62. [CrossRef] Batinti, Alberto. 2016. NIH biomedical funding: Evidence of executive dominance in swing-voter states during presidential elections. Public Choice 168: 239–63. [CrossRef] Beaulier, Scott A., Joshua C. Hall, and Allen K. Lynch. 2011. The impact of political factors on military base closures. Journal of Economic Policy Reform 14: 333–42. [CrossRef] Becker, Gary S. 1983. A theory of competition among pressure groups for political influence. Quarterly Journal of Economics 98: 371–400. [CrossRef] Economies 2019,7, 3 17 of 17 Bilotkach, Volodymyr. 2018. Political economy of infrastructure investment: Evidence from the economic stimulus airport grants. Economics of Transportation 13: 27–35. [CrossRef] Dove, John A., and Laura R. Dove. 2014. An examination of multistate antitrust enforcement by us state attorneys general. Journal of Applied Economics & Policy 32: 45–74. Garrett, Thomas A., and Russell S. Sobel. 2003. The political economy of FEMA disaster payments. Economic Inquiry 41: 496–509. [CrossRef] Hall, Bob, and Mary Lee Kerr. 1991. 1991–1992 Green Index: A State-by-State Guide to the Nation’s Environmental Health. Washington: Island Press. Hall, Joshua, and Michael Williams. 2012. A Process for Cleaning Up Federal Regulations: Insights from BRAC and the Dutch Administrative Burden Reduction Programme. Arlington: Mercatus Center. Hall, Joshua, Amanda Ross, and Christopher Yencha. 2015. The political economy of the essential air service program. Public Choice 165: 147–64. [CrossRef] Hird, John A. 1990. Superfund expenditures and cleanup priorities: Distributive politics or the public interest? Journal of Policy Analysis and Management 9: 455–83. [CrossRef] Hird, John A. 1993. Environmental policy and equity: The case of Superfund. Journal of Policy Analysis and Management 12: 323–43. [CrossRef] Hird, John A. 1994. Superfund: The Political Economy of Risk. Baltimore: John Hopkins University Press. Husted, Thomas, and David Nickerson. 2014. Political economy of presidential disaster declarations and federal disaster assistance. Public Finance Review 42: 35–57. [CrossRef] López, Edward J., and Wayne A. Leighton. 2013. Madmen, Intellectuals, and Academic Scribblers: The Economic Engine of Political Change. Palo Alto: Stanford University Press. McCubbins, Matthew D., Roger G. Noll, and Barry R. Weingast. 1989. Structure and process, politics and policy: Administrative arrangements and the political control of agencies. Virginia Law Review 75: 431–82. [CrossRef] McNeil, Douglas W., Andrew W. Foshee, and Clark R. Burbee. 1988. Superfund taxes and expenditures: Regional redistributions. Review of Regional Studies 18: 4. Niskanen, William A. 1971. Bureaucracy and Representative Government. Chicago: Aldine-Atherton. Peltzman, Sam. 1976. Toward a more general theory of regulation. Journal of Law and Economics 19: 211–40. [CrossRef] Ryan, Matt E. 2014. Allocating infection: The political economy of the swine flu (H1N1) vaccine. Economic Inquiry 52: 138–54. [CrossRef] Sobel, Russell S., Christopher J. Coyne, and Peter T. Leeson. 2007. The political economy of FEMA: Did reorganization matter? Journal of Public Finance and Public Choice 17: 49–65. Stratmann, Thomas. 1998. The politics of Superfund. In Political Environmentalism: Going Behind the Green Curtain. Edited by Terry L. Anderson. Palo Alto: Hoover Institution Press, pp. 239–62. Stroup, Richard. 1996. Superfund: The Shortcut That Failed. Bozeman: Property and Environment Research Center. Tilton, John E. 1995. Assigning the liability for past pollution: Lessons from the US mining industry. Journal of Institutional and Theoretical Economics 151: 139–54. Twight, Charlotte. 1989. Institutional underpinnings of parochialism: The case of military base closures. Cato Journal 9: 73–106. Weingast, Barry R., and Mark J. Moran. 1983. Bureaucratic discretion or congressional control? Regulatory policymaking by the Federal Trade Commission. Journal of Political Economy 91: 765–800. [CrossRef] Yandle, Bruce. 1992. Environment and efficiency lovers. Society 29: 23–32. [CrossRef] Yonk, Ryan, Arthur R. Wardle, and Josh Smith. 2017. A Public Choice Exploration of the Surface Mining Control and Reclamation Act. Logan: Strata. c 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).