Presidential particularism in disaster declarations and military base closures.

Kriner, Douglas L. ; Reeves, Andrew

Do presidents and members of Congress bring fundamentally different orientations to public policy decisions, particularly those concerning distributive politics? The conventional wisdom, both in the punditry and the academy, is clear. Members of Congress are mostly parochial; they often prioritize the needs of their local constituencies over those of the nation as a whole. By contrast, presidents are universalistic. As the sole representatives of all Americans, presidents are freed from parochial concerns and instead incentivized to promote the general welfare. The logic is intuitive; its implications are powerful. But is the claim and the assumptions that underlie it true?

In our book, The Particularistic President (Kriner and Reeves 2015b), we examine the extent to which this oft-invoked conception accurately portrays the policies that presidents pursue, specifically within the realm of distributive politics, broadly defined. We argue that presidents routinely depart from the norms of the universalistic framework, which asserts that presidents are primarily, if not solely, driven to pursue policies that benefit the nation as a whole rather than any specific constituency. Instead, we argue that presidents have strong incentives to be particularistic, that is, to prioritize the needs and desires of some citizens over others when pursuing their agendas. Presidents do not serve as a desperately needed constraint on parochial policies. Rather, presidents inject significant political inequality into the allocation of federal resources across the country, inequalities that rival and may often even far exceed those produced by a parochial Congress.

In this article, we elaborate and expand upon several of the analyses of presidential particularistic behavior developed in the book. The article proceeds in four parts. First, we briefly review two of the most important incentives that drive presidential particularism: electoral forces and partisan pressures. Second, we extend our analysis of presidential particularism in a venue where presidents enjoy unilateral authority and where normatively we would hope that objective need alone, and not political imperatives, would govern the geographic allocation of federal resources: natural disaster declarations. We then build on and extend our analysis of military base closings to examine whether both electoral and partisan incentives influenced presidential actions in the national security sphere that also had implications for distributive politics. Finally, we conclude by discussing several promising avenues of future research that can further our understanding of presidential particularism and its place in presidential politics writ large.

Pork Barreling and American Politics

The pork barreling legislator is an indelible image imprinted upon our national consciousness, and history is replete with examples of members of Congress prioritizing the needs of their local constituencies over the national interest. Senator Robert Byrd of West Virginia was unabashedly proud of his prowess in delivering as many federal dollars as he could to his socioeconomically disadvantaged home state. He wrote in his autobiography (Byrd 2005, 149): "I lost no opportunity to promote funding for programs and projects of benefit to the people back home." So voracious and successful was Byrd in his pursuit of federal dollars that his monikers included the "king" or "pope" of pork. Not to be outdone, Alaska's Don Young used his vantage point atop the House Transportation and Infrastructure Committee to champion the infamous "Bridge to Nowhere", a $25 million project to connect Ketchikan Alaska with sparsely populated Gravina Island.

These and a myriad of similar examples illustrate a larger truism concerning congressional politics. Members of Congress are institutionally compelled to wear two hats simultaneously. On one hand, they are members of the national legislature charged with crafting policies that serve the needs of the nation. Yet, they are also representatives of their narrow geographic constituencies, accountable for meeting the needs of their local constituents. The result is that members of Congress routinely bow to parochial pressures and prioritize the needs and wants of their immediate constituents over those of the country as a whole.

Since the founding of the Republic, presidents have gone to great lengths to contrast themselves with members of Congress. Unlike their counterparts on Capitol Hill, presidents are not beholden to narrow geographic constituencies. Rather, the chief executive is a national representative who cares only about the interest of the nation as a whole. The result is that the presidency, by its very institutional design, is incentivized to view policy not parochially, but universally. In a 1795 letter to the selectmen of Boston, President George Washington wrote, "In every act of my administration, I have sought the happiness of my fellow citizens. My system for the attainment of this object has uniformly been to overlook all personal, local, and partial considerations; to contemplate the United States as one great whole ... and to consult only the substantial and permanent interest of our country" (quoted in Wood 2009, 6). The implication is clear: few if any members of Congress could credibly claim the same.

Similarly, in articulating a normative defense of robust presidential leadership, Woodrow Wilson (1908, 67-68) contrasted members of Congress, whom he labeled "representatives of localities" with the president who "is the representative of no constituency, but of the whole people." When the president "speaks in his true character he speaks for no special interest. If he rightly interprets the national thought and boldly insists upon it, he is irresistible."

More recently, President Barack Obama sang this refrain in his 2013 State of the Union address. The president boldly proclaimed that investing in the nation's aging infrastructure was good for America. However, to heighten his proposal's appeal to reluctant members of Congress, the president playfully and a bit mockingly reminded members of the direct benefits an infrastructure program would yield for their individual constituencies. Investment in infrastructure creates jobs, Obama argued, "And I know that you want these job-creating projects in your districts. I've seen you all at the ribbon-cuttings." (1) President Obama championed infrastructure to promote the general welfare. He appealed to members of Congress to support it because it was in the interest of their districts.

Members of Congress themselves have done much to bolster presidential assertions of their superior and more universalistic approach to public policy. Indeed, when it comes to pork barrel politics, Congress has even sometimes looked to the president to save it from itself. For instance, the Republican-controlled 104th Congress passed the Line Item Veto Act in 1996, giving President Clinton--a Democrat--the authority to strike wasteful spending provisions in legislation. Unable to strip the provisions themselves, members of Congress gave credence to the executive's arguments that it could be a better steward of the public purse. In a similar vein, in 2010 following the midterm elections, the Republican House adopted a ban on earmarks, in effect giving more power to the executive departments and agencies over how federal spending is allocated. (2) Despite a chorus of support for delegating these decisions to the executive branch, a few have voiced concerns over the increased influence of the executive branch in the distribution of federal dollars. Former member of Congress Lee Hamilton (D-IN) complained in 2010 that, in passing earmark reform, Congress had "yet again opt[ed] to diminish itself while strengthening the President." (3)

This explicit contrast between congressional parochialism and presidential universalism is also widely embraced in academic research across disciplines. For example, law professor and Federalist Society cofounder Steven Calabresi (1995, 35) has posited that the presidency serves as "our only constitutional backstop against the redistributive collective action dilemma." Members of Congress face strong incentives to funnel as many federal benefits as possible to their constituencies, while concentrating the costs on other parts of the country. A strong, unitary presidency with its national constituency is the best institutional hope for breaking this cycle and pursuing the collective good. Calabresi is not alone. Summarizing the relevant legal literature, Jide Nzelibe (2006, 1218) concludes, "One of the most widespread contemporary assumptions in the discourse about the separation of powers is that while the president tends to have preferences that are more national and stable in nature, Congress is perpetually prone to parochial concerns."

This explicit contrasting view of the policy orientations of presidents and members of Congress also features prominently in many political science treatments of separation of powers dynamics. For example, in The Wartime President, William Howell, Saul Jackman, and Jon Rogowski (2013) argue that members of Congress are concerned with both the national and the local consequences of policy decisions made in Washington. By contrast, presidents care about the national implications of policy choices. Members of Congress exclusively pursue the national interest only when exogenous shocks, such as major wars, encourage them to abandon their parochial biases and to follow the lead of the president.

The extent to which presidents are universalistic or particularistic is of more than theoretical and academic importance, and it is a central focus of The Particularistic President. It has critically important policy implications. Increasingly a diverse chorus has called for greater delegation of policy-making authority to the executive branch. Members of Congress themselves have made and even acted on these calls. Legal scholars, such as now-justice Elena Kagan, have argued that increased delegation of authority to the executive branch over policy formation and policy implementation would produce better outcomes because of the presidency's universalistic orientation. "Because the President has a national constituency," Kagan (2001, 2335) argued, "he is likely to consider, in setting the direction of administrative policy on an ongoing basis, the preferences of the general public rather than merely parochial interests." Some prominent presidency scholars have also followed suit. For example, William Howell and Terry Moe (2015) have trumpeted expanded presidential leadership as a way out of our current governing malaise. Congress, in their assessment, is institutionally all but incapable of providing the leadership and vision needed to solve exigent national problems. They state, "It should come as no surprise that the recent history of legislative activity is littered with bills that, in name, promise to confront challenges of national importance, but in fact constitute little more than disfigured conglomerations of sectional initiatives" (Howell and Moe 2015, 148). For a variety of reasons ranging from the unitary structure of the office to its universalistic orientation and focus on national rather than parochial interests, presidents are the actors best poised to solve these problems if given the requisite means to do so.

Even some Congress scholars, confronted with clear evidence of institutional paralysis on Capitol Hill, have reluctantly called for additional delegation to the executive. While somewhat ruefully acknowledging that presidential power has already expanded considerably at the expense of legislative prerogatives, Thomas Mann and Norman Ornstein (2013, 166) conclude that further "modest shifts to give more leeway to the executive make sense, given the current and continuing dysfunction."

We have little doubt that greater delegation of policy-making authority to the president could help break the gridlock that has too often stymied urgently needed reforms in recent years. However, we are skeptical that greater delegation to the president will necessarily result in less parochial policies. Rather, we argue that--particularly in the realm of distributive politics--there are strong reasons to believe that increased delegation will continue to produce considerable inequality in policy outcomes. However, the nature of that inequality will even better reflect the president's political imperatives than those of powerful members of Congress.

Particularistic Incentives

At least since Paul Light ([1982] 1998), scholars have acknowledged that presidents pursue multiple goals when seeking to put their stamp on public policy. However, a significant strand of presidential scholarship has argued that perhaps the most important motivating force behind presidential behavior is the desire to build a lasting historical legacy. For example, Moe and Wilson (1994, 11-12) contend, "if there is a single driving force that motivates all presidents, it is not popularity with the constituency nor even governance per se. It is leadership. Above all else, the public wants presidents to be strong leaders, and presidents know that their success in office, along with their place in history hinges on the extent to which citizens, political elites, academics, and journalists see them as fulfilling this lofty expectation." Building on this logic, in their recent overview of presidential politics, Howell and Brent (2013) argue that legacy-maximizing presidents assiduously seek to maximize their power and influence to pursue policies that sever the national interest and benefit the greatest number of Americans. These forces compel presidents to embrace a universalistic orientation toward policy decisions that maximize the national interest.

Undoubtedly, presidents are intensely concerned with power and building a legacy of accomplishment that will stand the test of history by pursuing policies widely perceived to have served the national interest. Yet, presidents must temper these ambitions with other more immediately pressing concerns. For example, Cohen (2006, 541) emphasizes the competing directions in which presidents can be pulled: "On the one hand the president is a symbol, representative, and leader of the entire nation. But the president is also a partisan who seeks benefits for some sectors of the polity, such as his party and those who voted for him. Presidents seek these particularized group-specific benefits as they try to build coalitions in support of their electoral and policy goals." Edwards (2000, 67) warns that presidential policies are not always crafted with the national interest alone in mind, but may often be "designed primarily to benefit the president's electoral coalition." Similarly, in their perceptive assessment of the leadership challenges facing Obama, Milkis, Rhodes, and Charnock (2012, 59) describe the 44th president as "an ambitious politician caught between the conflicting institutional and electoral imperatives of contemporary party politics."

The Particularistic President builds on this prior research; develops a range of theoretical expectations concerning the operation, timing, and relative influence of competing incentives and empirically tests these hypotheses across a range of policy areas using an array of data sources. Here, we briefly review two of the most important incentives that could lead presidents to depart from the norms of universalism and instead to pursue particularistic policies. We will then test for their operation in extensions to our earlier analyses of natural disasters and base closures.

Electoral Incentives

Since Mayhew's (1974) seminal work, few analyses of congressional behavior have failed to pay serious attention to the role played by the electoral connection. It is all but a truism in congressional studies that the drive to secure reelection influences virtually every facet of congressional politics from how the chamber is organized, to how members vote, to how they spend their time both in Washington and back home in their districts.

By contrast, presidency scholarship historically has been agnostic toward the role that electoral incentives play in shaping presidential politics. Some studies consciously downplay the importance of electoral incentives; of course, presidents desire reelection, but electoral motivations are decidedly secondary to presidents' ultimate aim: leadership and legacy (e.g., Moe and Wilson 1994). Other scholars have given greater weight to electoral concerns. However, the conventional view is that, whereas electoral incentives encourage members of Congress to cater to their districts and thereby fuel parochialism, electoral incentives encourage presidents to pursue policies that maximize the interests of the nation as a whole. Voters may expect members of Congress to bring home the bacon for their local constituencies (though empirical analyses of the electoral rewards members of Congress reap from pork is mixed at best; see Lazarus and Reilly 2010 for a review). However, most scholars argue that voters primarily hold presidents accountable for national outcomes and the well-being of the country as a whole (e.g., Abramowitz 2008; Clarke and Stewart 1994; Erikson 1989; Norpoth 1985). If so, then presidents have few electoral incentives to prioritize the needs and wants of some Americans over others. Rather, presidents best serve their electoral interests by pursuing policies that maximize benefits for the greatest number of Americans.

However, a growing body of research suggests that voters hold presidents accountable for both national and local outcomes. Local economic conditions, not just aggregate economic figures, factor into voters' decision calculus (Books and Prysby 1999; Mondak, Mutz, and Huckfeldt 1996, Reeves and Gimpel 2012). Voters judge the commander in chief, in part, based on how a war's costs have affected their local communities (Gartner, Segura, and Barratt 2004; Grose and Oppenheimer 2007; Karol and Miguel 2007; Kriner and Shen 2007). And perhaps most directly, voters reward presidents for increases in federal spending in their local constituencies and punish those who preside over cuts in federal assistance to their districts (Kriner and Reeves 2012).

Of course, this does not necessarily imply a deviation from universalism. If the sole object was to secure the most votes, then presidents should pursue policies that maximize benefits for the greatest number of people, regardless of where they are located. However, voters do not directly elect the next president of the United States. The Electoral College does. This, coupled with the system of winner-take-all apportionment used in all but two states (Edwards 2004), provides a potential incentive for presidents to be particularistic (Hudak 2014). Votes in swing states are of greater electoral value to the president than votes in uncompetitive states. As a result, presidents have strong incentives to engage in what we call electoral particularism, that is, to disproportionately target the benefits of federal policies to constituencies located within electorally competitive states. As we discuss in detail elsewhere (Kriner and Reeves 201 2, 2014, 2015a, 2015b), these incentives should be stronger in election years, particularly when the president himself is on the ballot for reelection.

The stark contrast between electoral incentives and universalistic incentives raises a series of important questions. Will universalistic incentives dampen electoral incentives? Will their relative influence, and thus the resulting size of electoral particularism in policy outcomes, vary across policy areas? Or do presidents pursue particularistic policies to shore up their electoral prospects widely? To begin to answer some of these questions, we look for evidence of electoral particularism in two policy venues where we might expect universalistic incentives to be strongest and electoral ones to be weakest: military base closures and disaster declarations.

Partisan Incentives

Of course, presidents are more than reelection seekers. They are also leaders of their political parties (e.g., Galvin 2009). Wood (2009) challenges the idea that presidents pander to centrist opinion and instead argues that presidents prioritize the interests of the core partisan constituencies that brought them to the White House. This may be increasingly true, as the polarization of our political system in recent decades has led many to decry the emergence of a "partisan presidency" (Cameron 2002; Cohen et al. 2008; Galvin 2013; Milkis and Rhodes 2007; Newman and Siegle 2010; Skinner 2008). Presidents, like members of Congress, must juggle dual roles: national leader and partisan leader. As partisan in chief, presidents have strong incentives to pursue policies that systematically channel federal dollars disproportionately to parts of the country that form the backbone of their partisan base.

If we find evidence of constituencies that solidly backed the president at the polls receiving a disproportionate share of federal policy goods, is it necessarily evidence of partisan particularism? That is, would partisan inequalities in the geographic allocation of federal benefits prove that presidents are prioritizing the needs of their core partisan base at the expense of the national interest? Not necessarily. What we label partisan particularism--the pursuit of policies that concentrate federal benefits in parts of the country that reliably back the president's party at the polls--could be an unintended, though no less tangible result of presidents of different parties possessing competing visions of the polices that best serve the national interest and pursuing different programmatic agendas accordingly. For example, Democratic presidents may prioritize social welfare programs that disproportionately benefit the urban poor, not because such constituencies reliably back the Democrats, but because they believe that doing so is in the best interests of the nation as a whole. By contrast, Republicans may focus on different spending priorities, such as agriculture subsidies or defense spending, which could concentrate benefits in Republican-leaning districts. As a result, presidents may pursue universalistic ends through particularistic means.

In this article, we build on analyses in The Particularistic President that cast doubt on this alternate explanation for partisan inequalities by examining policy venues where competing visions of the national interest cannot explain the emergence of inequalities along partisan lines. It is difficult for a Republican president to credibly argue that closing a military base in a Democratic rather than a Republican district is in the national interest, all else being equal. Similarly, a Democratic president cannot truly believe that responding to a natural disaster in a Democratic constituency, but not to an equivalent scale disaster in a Republican district, serves the greater good. For each of these cases, we have a strong prior that objective measures of need (or lack thereof) should drive presidential policy decisions, independent of the political characteristics of the place affected. Evidence of partisan inequalities in base closures and natural disasters is evidence that a primal desire to reward the party base is fueling observed inequalities in the allocation of federal policy benefits.

Presidential Particularism and Natural Disaster Declarations

Much of the empirical analysis in The Particularistic President focuses on the allocation of $8.5 trillion in federal grants across the country from 1984 through 2008. This staggering amount of money reflects not only the spending priorities of the federal government but also the winners and losers in the context of divide-the-dollar of politics. Our results are striking. In election years, swing states receive billions of dollars more in grant spending than do uncompetitive states (see also Kriner and Reeves 2015a). Moreover, we find that core partisan states routinely secure disproportionately large shares of federal grants. For solidly red and blue states like New York and Texas, who sits in the Oval Office can mean the difference between receiving and losing hundreds of millions of grant dollars every year.

These results are strong evidence for the tangible impact of presidential particularism. Yet it is difficult to trace influence through this process. It is nearly impossible to assign the responsibility for where any single dollar of federal spending lands to any one senator, governor, representative, mayor, town council member, state legislator, or U.S. president. In Chapter 5 of our book, we search for evidence of particularism in the way that a prosecutor might search for evidence of employment discrimination in a large corporation over a long period of time. While we find occasional smoking guns--where presidents admit to favoritism or aides suggest motivations other than good public policy--our evidence focuses on an analysis of the broad contours of federal spending and

examines whether that distribution is consistent with presidential biases. We find compelling evidence that particularism diverts billions of dollars every year.

To lock down the president's direct role in producing particularistic outcomes, we examine other policy areas. One of the most important additional policies that we examine is how particularism factors into presidential disaster declarations. Patterns in presidential disaster declarations provide a different type of evidence. Perhaps most importantly, in this policy venue, presidents wield unilateral authority. Unlike the allocation of federal grants, decisions to declare a natural disaster are not products of the interplay of Congress, the president, and the bureaucracy. Moreover, the response of the president to natural disasters is a case where there is a particularly strong expectation of universalism. Once we account for the severity of the natural disaster and the capability of a locality to address the damage, we might expect that the political characteristics of the place should not matter for its chances of getting federal aid. Indeed, formal accounts of the disaster declaration process highlight the role of need and capacity and either do not allow for or specifically reject the notion that political characteristics are influential in determining the eligibility of a locale for a disaster declaration (McCarthy 2004).

Natural disasters wreak massive damage on the United States every year. For example, 25 storms in 2011 and 2012 caused over $1 billion in damage with a total damage of $188 billion. One study estimates that the federal response amounted to $136 billion or almost $400 per household. This response is initiated by the president of the United States. While governors make the initial request, and presidents are advised by bureaucrats on the merits of the request, presidents alone determine what qualifies for a disaster declaration. The governing act (42 U.S.C. sec. 5122(2) defines a disaster declaration as, "any natural catastrophe ... in any part of the United States, which in the determination of the President causes damage of sufficient severity and magnitude to warrant major disaster assistance" (emphasis added). Disaster declarations activate many types of federal assistance to affected areas including direct cash payments to individuals as well as local governments. The presidential disaster declaration is a rare case where presidents act with only limited influence by other politicians to allocate resources to specific constituencies.

It is also important to note the potential electoral returns from presidential disaster declarations. Anecdotes abound of politicians rising to fame or falling from electoral grace because of a botched response to hurricanes, snow storms, or other natural disasters. When Hurricane Betsy ravaged New Orleans in 1965, the mayor literally came to the rescue of a voter stuck atop her roof (Abney and Hill 1966). In 2004, President George W. Bush visited Florida to comfort the victims of a number of hurricanes in the run-up to the presidential election. Photos of President and Mrs. Bush comforting citizens and delivering aid were widely covered in the media, yet the president listed his administration's response to Hurricane Katrina as one of the biggest regrets of his presidency. In addition to many anecdotes of how the response to natural disasters influences presidential favorability, a number of studies have found evidence in the data. For example, Gasper and Reeves (2011) find that presidents see rewards at the ballot box when they grant presidential disaster declarations.

In the book, we analyze presidential disaster declarations from 1984 to 2008, covering the presidencies of Ronald Reagan, George H. W. Bush, Bill Clinton, and George W. Bush. In this analysis, we find that presidential disaster declarations, first and foremost, are a function of need. Measures of damage and the occurrence of severe weather events are the best and strongest predictors of presidential disaster declarations. Universalism is a powerful force in presidential behavior. But we also found that presidential particularism exerts a substantial influence over the disaster declaration process. Counties in core states and counties in swing states were more likely to receive disaster declarations than counties in other states. Additionally, those counties represented in Congress by a member of the president's party were also more likely to receive a disaster declaration. We also found that the effects were strongest in election years. Counties in swing and core states were both more likely to receive disaster declarations in presidential election years.

Here, we expand our data set to include presidential disaster declarations from 2009 to 2013, the first five years of the Obama administration, including 2012 when President Obama ran for reelection. From this expanded data set, we present one of the first analyses of disaster declarations and presidential particularism in the Obama administration. Our question is whether particularism is as strong today as it was under previous administrations. On one hand, President Obama has railed against congressional pork barrel politics, nearly derailing the economic stimulus bill over earmarks; on the other hand, presidential rhetoric aside, President Obama faced a similar landscape as his immediate predecessors including a geography of well-defined core and swing states.

In expanding the data set, we rely on the same sources described in the book. Presidential disaster declarations are publicly posted by the Federal Emergency Management Agency. We aggregate these data to the level of the county and year to indicate which counties receive any disaster declarations in a particular year. From 2009 to 2013, President Obama issued 414 major or emergency disaster declarations with 78% of counties seeing at least one disaster declaration. Obama's issuance of 83 disaster declarations per year was about 10% higher than President Bush's rate of 75 per year. While Obama issued more disaster declarations on average, he tended to grant them to fewer counties within the declared state.

As we previously described, one advantage to analyzing disaster declarations is the compelling hypothesis that need alone should drive disaster declarations. Evaluating presidential universalism requires measures of actual need, which we obtain through University of South Carolina's Spatial Hazard Events and Losses Database for the United States (SHELDUS), which records the damage sustained by locales as a result of severe weather. We also control for the number of severe weather events that a county sees in a given year by including a count for the number of weather events that fall in the seventy-fifth percentile of damage in a given year. We also include each county's per capita income to measure the county's capacity to respond on its own without federal assistance to severe weather events.

Table 1 presents a model of presidential disaster declarations based on our expanded data set from 1984 to 2013. The results are from a logistic model with county-level fixed effects and indicators for each year (which are not presented in the table). Column 1 presents the exact model presented in the book but with the five years of additional data. The substantive results are the same. Electoral, partisan, and coalitional particularism (i.e., rewarding constituencies represented by presidential copartisans in Congress) are present at similar levels when the five years of new data are included. As in Kriner and Reeves (2015b), electoral particularism is active during election years while partisan particularism is present throughout but especially salient during election years.

The increased intensity of core constituency particularism during election years is worthy of further discussion. It is important to note that, in the context of federal grant ? spending, core particularism was influential throughout the presidential administration with no additional effect during election years. This was not the case with disaster declarations. In this context, presidents were significantly more likely to reward core constituencies in election years. This could have to do with the nature of the declaration process and the numerous supporters within a state who lobby the president and observe his responsiveness to their request. In the budget process, it is much easier to blame members of Congress or bureaucrats when a favorite program gets nixed. But when a disaster strikes an area concentrated with campaign donors and other political allies, it might activate the president to be more responsive to their needs at a time when he is especially in need of their support.

Column 2 of Table 1 presents the same model but includes interactions for those observations that take place during the Obama administration. One of the striking findings is that electoral and partisan particularism were even stronger forces when Obama ran for reelection in 2012 as compared to other election years. For example, in 2012 counties in core states and swing states were approximately 3.7 and 4.5, respectively, times more likely to see a disaster declaration than similar counties in noncore, nonswing states in that election year. During all other administrations, counties in swing and core states saw about twice as many disaster declarations in election years. Outside of the 2012 election year, however, there is scant evidence that Obama engaged in particularism at all. During nonelection years in the Obama administration our model shows that counties in swing and core states actually received fewer disaster declarations than those located in other states, though the substantive differences are relatively small.

That Obama was especially rewarding of swing states in 2012 suggests that presidential electoral motivations and their influence on policy outcomes continue to strengthen. No doubt these results are partly driven by the nature of the 2012 election in that Obama was running for reelection in a relatively tight Electoral College contest that forced Romney and Obama to spend inordinate amounts of time wooing voters in battleground states like Ohio, Pennsylvania, and Florida. The fact that Obama did not appear to favor swing or core states during nonelection years suggests that he was, perhaps, less particularistic than his predecessors. However, when confronted with the exigencies of a hotly contested reelection battle, Obama engaged in particularism with respect to disaster declarations to an even greater extent than his colleagues. It would behoove scholars to further investigate whether this pattern holds in other contexts.

Electoral and Partisan Particularism in Base Closings

In The Particularistic President we also examined a second policy area where presidents, at key moments in history, have had unilateral authority over distributive policy outcomes: military base closures. At least since Eisenhower's Farewell Address and his admonition about the growing power of a military-industrial complex, politicians and scholars alike have acknowledged that national defense policies are fertile grounds for particularistic politics. However, Congress, not the president, has long been seen as the primary culprit in using its influence over defense appropriations for political gain.

The increasing concentration of defense industries and spending in rural congressional districts following World War II created a cadre of congressional representatives who assiduously protected defense from major cuts (Thorpe 2014). Instead of America's traditional massive postwar demobilization, these parochial congressional pressures, Thorpe argues, helped drive and sustain America's unprecedented peacetime military buildup.

Undoubtedly, the multiheaded hydra that is the U.S. Congress injects considerable parochialism into American defense policies. But does the commander in chief also pursue electoral and partisan goals when influencing the allocation of defense-related resources around the country-' Most scholarship argues that presidents, particularly when acting in the defense realm as commanders in chief, take a much more holistic view of policy making, emphasizing the national interest than do members of Congress (Howell, Jackman, and Rogowski 2013).

While members of Congress, many of them from rural constituencies that became heavily reliant on defense industries, may have played a lead role in driving the dramatic expansion of the American warfare state in the early Cold War, presidents for decades took the lead in deciding where retrenchments would be made. The politics of base closures is a classic example of the collective dilemma (Mayer 1995). The entire nation stands to benefit from a reduction in redundant defense expenditures. However, these benefits would be distributed diffusely, while the pain of such reductions and closures would be felt most acutely by the affected communities. Members of Congress from districts directly affected by closures have much stronger incentives to fight those closures tooth and nail than do other members to pursue nebulous savings. More colorfully, former Majority Leader Dick Armey (1988, 71) summarized the prevailing view of defense spending on Capitol Hill: "many members look on the Defense bill the way Jimmy Dean looks on a hog, as a giant piece of pork to be carved up and sent to the folks back home." Most members are concerned with getting a bigger slice of the pie for their districts and keeping it; they are decidedly less interested in making rational cuts to defense in the name of greater economy and efficiency.

Past scholarship emphasizing presidential universalism suggests that the president, by contrast, is uniquely positioned and incentivized to pursue economy and efficiency in this context. The president's national constituency is best served by closing inefficient, redundant, and obsolete military installations and either reallocating the savings to other priorities or easing the burden on the taxpayer.

Base closings, therefore, afford a particularly illuminating window into the extent of presidential particularism. In the realm of national security policy making, perhaps more than in any other policy area, we normatively hope that presidents are universalistic, prioritizing only the needs of the nation as a whole, rather than those of individual constituencies. Given Congress's proclivities, presidents are desperately needed as a counterbalance to congressional parochial impulses. But do they actually provide such a counterbalance?

Presidential Particularism and the 1990 Cheney List

Throughout the 1960s and 1970s, Presidents John Kennedy, Lyndon Johnson, Richard Nixon, and Gerald Ford pursued a number of rounds of base closures, each of which was met with cries of politicization from members of Congress representing the afflicted districts. This outcry over alleged presidential politicking grew so great that by 1976 Congress passed a law all but stripping the president of the power to close bases unilaterally. No major base was closed between 1977 and 1988, despite the Department of Defense recommending hundreds of closures and realignments during the period.

By the late 1980s in the waning days of the Cold War, sufficient momentum finally built on Capitol Hill to allow some closures to proceed. However, mindful of alleged past presidential abuses, Congress in 1988 created the Commission on Base Realignment and Closure (BRAC) to identify a list of bases for closure independent of the Department of Defense. The list would then be submitted to the Department of Defense, which could either accept the list in toto, or reject it. The administration was given no power to amend the list in any way. Finally, if the Department of Defense approved, Congress would also be given an up or down vote on the entire list. Under the BRAC process, 145 bases were identified for closure or realignment, while keeping partisan rancor to a minimum.

Shortly after the BRAC process concluded, the Berlin Wall fell. Confronted with the dramatically changed national security situation that emerged with the end of the Cold War era, the new George H.W. Bush administration sought both to realize a peace dividend and to structurally reform the military for new security threats, in part, through a new round of base closings in 1990. However, rather than asking Congress to reconstitute the BRAC process, the new Secretary of Defense Dick Cheney fervently believed that identifying which bases should be closed was the prerogative of the commander in chief. Having served as chief of staff to President Ford, Cheney returned to the White House anxious to reassert presidential prerogatives and to restore to its rightful place an institution that he believed had become embattled and diminished by an overbearing Congress after Watergate. Toward this end, Cheney and the Department of Defense identified its own list of 35 bases for closure and submitted it to Congress.

The reaction on Capitol Hill from congressional Democrats ranged from exceedingly skeptical to overwhelmingly negative. Of the 35 bases proposed for closure, 29 resided in congressional districts represented by Democrats. Of the 21 major bases targeted for closure, 19 were in Democratic districts. Many Democrats openly charged the administration with playing partisan politics with national security and abusing its influence to concentrate the economic pain of base closings in Democratic constituencies. (4)

In The Particularistic President, we looked for empirical evidence of the Democratic claim that the Bush administration was engaging in what we have called partisan particularism. In this case, rather than channeling federal benefits to core co-partisan constituencies, the administration was accused of concentrating pain in opposition strongholds, while insulating its partisan base from the economic disruptions that follow in the wake of major base closures. Both Secretary Cheney and President George H. W. Bush vigorously denied the claims.

The aggregate comparisons levied by Democrats appeared damning. Eighty-three percent of all base closings were in Democratic districts, and 91% of major bases targeted for closure were in Democratic constituencies. However, congressional Republicans offered an alternate explanation for the seeming disparity: Democrats had successfully used their three decade-long control of the House to grab as many bases for Democratic strongholds as possible. (5)

To account for these claims and counterclaims, we constructed a simple statistical model that modeled the probability of a congressional district being targeted for a base closure as a function of two independent variables: whether or not the district was represented by a Democrat, to look for evidence of partisan particularism consistent with Democratic allegations against the Bush administration; and the number of major military bases in the district, to control for the logical Republican retort that more closures would occur in Democratic districts because Democrats had long used their majority status in Congress to secure a disproportionate share of defense resources for their constituencies.

Consistent with the logic articulated by the Bush administration's co-partisan defenders, we did find that the more major bases were in a district, the greater was the probability of it being targeted for a base closure. However, even after controlling for the supply of bases in a district, we found that Democratic districts were disproportionately targeted. Even in national security policy making, we found strong evidence of partisan particularism.

Because the political debate over the Cheney list focused on these charges of partisan particularism, we did not pay much attention to whether or not the Bush administration also appeared to engage in electoral particularism in the 1990 base closings ploy. If partisan incentives trump universalistic incentives, then the president should seek to concentrate the pain of base closures in opposition strongholds while insulating their core partisan base. We found strong evidence of this dynamic.

However, electoral incentives might also factor into presidential decisions about which bases should close and which should remain open. A major base closure can economically decimate the surrounding community. For example, when President Nixon decided to move the Atlantic Cruiser-Destroyer Fleet from Newport, RI, and to close the base, unemployment in Rhode Island almost tripled, and in some communities most directly affected by the closure unemployment surpassed 30% (Mayer 1982). In previous work, we have shown that voters hold presidents accountable for local economic conditions and for the share of federal resources that they receive (Kriner and Reeves 2012). Presidents surely must anticipate that voters will also hold them responsible at the ballot box for a base closure that devastates the local economy. But are these electoral incentives strong enough for presidents to prioritize the needs of some Americans over others and to shield disproportionately bases in swing states from closure? Or do electoral incentives have little sway over the policy decisions of the commander in chief?

To look for evidence of electoral particularism in base closures, we first constructed a pair of regression models. The first replicates the base model from The Particularistic President, but also includes an indicator variable identifying swing states. We define a swing state as one in which the losing presidential candidate averaged 45% or more of the two-party vote over the preceding three presidential election contests. Our dependent variable is the number of bases targeted for closure in the 1990 Cheney list in each congressional district. Twenty-nine districts possessed one base targeted for closure. Three districts possessed two bases on the closures list. Given the small number of districts with more than one closure, we operationalized our dependent variable as a binary variable coded 1 for districts with one or two bases targeted for closure and coded 0 for districts with no bases targeted for closure. (6)

Our two main independent variables of interest are indicator variables identifying whether a district is represented by a Democrat in the U.S. House of Representatives and whether the district resided in a swing state. If partisan particularism influenced the Bush administration's base closure decision making, we would expect to see districts represented by Democrats to be more likely to experience a base closure, all else being equal. If electoral particularism intrudes even into the national security sphere, then districts in swing states should be significantly less likely to be targeted for closures than districts in uncompetitive states. Finally, the model includes a key control: the number of major military installations in each district, as obtained from Scott Adler's Congressional District Data Base for the 101st Congress (Adler n.d.). This allows us to examine the influence of particularistic forces on the geographic allocation of base closures, controlling for preexisting inequalities in the allocation of bases across the country.

The first column of Table 2 presents the results. The coefficients for all three variables are in the expected direction and statistically significant. Consistent with the basis of the Republican counterclaim, the more major military bases in a district, the more likely that district was to be targeted for at least one base closure. However, even after controlling for this basic dynamic, we see strong evidence of presidential particularism. Districts represented by Democrats were significantly more likely to be targeted for closures than were districts represented by Republicans. And districts in swing states were significantly less likely to be targeted for closures than were districts in uncompetitive states, all else being equal.

Figure 1 illustrates the substantive size of the effects. The horizontal line at a predicted probability of 0.04 represents the predicted probability of a closure in the average district represented by a Republican that is not in a swing state. For each of the three explanatory variables, the dot presents the point estimate for the predicted probability of a closure if that factor was increased from 0 to 1 in the case of the two binary variables or from the mean to one standard deviation above the mean for the number of major military installations in the district variable.

We find strong evidence of partisan particularism. Holding the number of major military bases in a congressional district constant, we find that the average Democratic district was more than three times more likely to be targeted for a base closure than the average Republican district. The predicted probability of a closure increases significantly form 0.04 to more than 0.12. Thus, the stark partisan disparities observed in the aggregate were not simply artifacts of Democratic districts having more major bases. Rather, the data strongly suggest that these districts were targeted to serve partisan political objectives.

Districts in swing states, by contrast, had almost no chance of being targeted for a base closure. The predicted probability of a closure in a district in a swing state was just over 0.01. Put another way, whereas the baseline district had about a 1 in 25 chance of being targeted for a closure, the median district in a swing state had only approximately a 1 in 100 chance of being targeted for a closure. (7) The data strongly suggest that electoral particularism also characterizes important decisions that presidents make within the national security sphere. Mindful of the potential electoral implications of economic disruption, the Bush administration all but completely insulated districts in swing states from base closures, regardless of the number of major military bases residing in the district.

[FIGURE 1 OMITTED]

As a robustness check on these results, we estimated an alternative model specification. In this model the dependent variable is the percentage of major bases in each district targeted for closure. This operationalization of the dependent variable eliminates the need to include a control variable for the number of major bases in a district by building it into the denominator. Moreover, this analysis allows us to focus only on those congressional districts that contained at least one major military base according to the Adler data. Of course, a district without a major base could experience a closure. Not all of the bases on the Cheney list were major installations. However, narrowing our focus to only the most important closures provides an important robustness check on our argument that particularistic incentives significantly skewed presidential base closing decisions. The second column of Table 2 presents the results.

Even in this revised specification, we find strong evidence of both electoral and partisan particularism in base closures. All else being equal, Democratic districts experienced 8% more closures than Republican districts, and districts in swing states received 6% fewer closures than similar districts in uncompetitive states.

Finally we examine what happens when the two particularistic incentives collide. Partisan incentives encouraged the Bush administration to insulate the GOP's partisan base from economically disruptive base closures and instead to concentrate closures in constituencies represented by Democrats. However, electoral forces also encouraged the administration to spare districts in swing states from closures for fear that affected voters might seek retribution at the ballot box. What was the fate of Democratic districts within swing states? Such districts are not part of the president's core partisan base, and they do not elect co-partisan members to Congress that might aid the president's legislative coalition-building efforts. However, punishing a Democratic district within a swing state may be electorally risky. The majority of voters in such districts may be Democrats. However, any drop-off in Republican turnout may undermine the president's prospects of carrying the state as a whole and its Electoral College votes. In such cases, we expect electoral incentives to be dominant. As a result, we expect the relationship between a district being represented by a Democrat and its probability of being targeted for a base closure to be conditional on whether or not that district resides within a swing state. In uncompetitive states, Democratic districts should be disproportionately targeted for closures. In swing states, by contrast, we expect both Democratic and Republican districts alike to be insulated from closures.

To test this final hypothesis, we replicated our analyses from Table 2 but estimated separate regressions for districts in swing and uncompetitive states. The first two columns in Table 3 are logistic regressions of the probability of a district being targeted for a closure. (8) Strongly consistent with our hypothesis, in uncompetitive states (column 1), the coefficient for the dummy variable identifying districts represented by a Democrat in the House is positive and statistically significant. By contrast, in the swing state model (column 2), the coefficient is substantively smaller and statistically insignificant; indeed, the standard error is almost the size of the coefficient.

[FIGURE 2 OMITTED]

Figure 2 illustrates the magnitude of the degree to which partisan-induced targeting is conditional on a state's electoral competitiveness. In an uncompetitive state, Democratic districts are more than three times more likely to be targeted for a base closure than Republican districts. The predicted probability of a closure increases from about 0.04 to 0.12. By contrast, in swing states we see no evidence that Democratic districts are more likely to be targeted for closures than Republican states. Regardless of the partisan orientation of a swing state district's member of Congress, the probability of a base closure is very low. Replicating this analysis with the percentage of bases in each district targeted for closure as the dependent variable (columns 3 and 4) yields virtually identical results.

The result of this combination of electoral and partisan particularistic forces is that opposition party strongholds in electorally uncompetitive states bore a strikingly disproportionate share of the 1990 Cheney proposed base closures. It is difficult to see how genuinely national interests or military necessity could produce such a result. Rather, even in the national security realm we find very strong evidence of presidential particularism. In policies touching on distributive politics, even the commander in chief prioritizes the needs of some Americans over those of others to achieve electoral and partisan political gain.

Conclusion

Although they involved two specific policy areas, our analyses of base closures and natural disaster declarations provide strikingly strong evidence of the breadth and depth of presidential particularism. We purposely chose these issues for two reasons. First, in each, presidents acted unilaterally; as a result, presidential responsibility for distributive outcomes was unclouded by competing congressional influence. Second, in each case, there were strong reasons to believe that in this policy realm, more than almost any other, objective need alone should govern presidential behavior. Military necessity, alone, should influence decisions over which bases should be closed and which should remain open. The scale of a disaster and the objective need of the affected community should be the sole determinants of federal disaster assistance. Normatively we might hope that national security and disaster response, more than any other policy areas, would be insulated from particularistic politics. However, the data unambiguously show that they are not. Even in these spheres, we find that presidents pursued highly unequal allocations of federal resources to serve their electoral interests and partisan political imperatives.

In The Particularistic President we found significant evidence of presidential particularism across policy areas, including over the geographic allocation of more than $8.5 trillion of federal grant spending from 1984 through 2008. However, one limitation of our work is that we have focused exclusively on the relative influence of universalistic and particularistic incentives within the realm of distributive politics, albeit broadly defined.

Future research should explore whether electoral, partisan, and other forces lead presidents to pursue particularistic policies that prioritize the needs of some Americans over others in policy areas without significant, immediate distributive politics implications. Distributive politics may be particularly fertile grounds for electoral particularism because voters directly hold presidents accountable for the share of federal dollars they receive (Kriner and Reeves 2012). In other policy areas, where the electoral connection may be less direct, do presidents routinely pander to the nondistributive policy preferences of electorally important constituencies? Further research along these lines is essential to defining the full extent of electoral particularism.

A second potentially fruitful area of future research is to expand the inquiry into electoral particularism to examine the role of the primary calendar. Do presidents systematically reward early primary and caucus states that wield disproportionate influence over the presidential nomination process with greater levels of federal largesse? As but one anecdotal example, consider federal ethanol subsidies, so critical to many potential caucus-goers in Iowa. By almost any economic and environmental accounting, such subsidies are simply not in the national interest (e.g., Hill et al. 2006). Yet, presidential candidates and incumbent presidents have consistently been among the most inveterate supporters of ethanol tax credits and subsidies. Do primary politics fuel electoral particularism more broadly?

Finally, to test between the various potential forces that might produce partisan particularism--particularly the counterargument that presidents may pursue universalistic ends through partisan means--we relied on analyses of policy areas, such as base closures and disaster declarations, where competing visions of how best to pursue the national interest cannot explain inequalities along partisan political lines. However, this case could be further bolstered through careful archival research that might uncover some of the thought processes behind decisions that produced partisan particularistic outcomes. For example, in 1970, Nixon demanded in writing that his chief domestic policy advisor, John Ehrlichman, intervene in the budgetary formulation process to punish the administration's political enemies by slashing federal funding for specific states. Nixon instructed Ehrlichman: "In your budget plans ... I want Missouri, New York, Indiana, Nevada, Wisconsin, and Minnesota to get less than they have gotten in the past ... Senators are going to get a better audience at the White House than those with Democratic Senators who are constantly chopping at us" (Reeves 2001, 171). Before or after Watergate, future scholars are unlikely to find similarly explicit smoking guns that explicitly lay bare the precise motivations behind particularistic policy outcomes. However, additional historical and archival research may yield valuable insight into the role that electoral, partisan, and other particularistic forces play in shaping presidential thinking, both in the realm of distributive politics and in policy making writ large.

References

Abney, F. Glenn, and Larry Hill. 1966. "Natural Disasters as Political Variables: The Effect of a Hurricane on an Urban Election." American Political Science Review 60 (4): 974-81.

Abramowitz, Alan I. 2008. "Forecasting the 2008 Presidential Election with the Time-for-Change Model." PS: Political Science & Politics 41 (4): 691-95.

Adler, E. Scott, n.d. Congressional District Data File [congressional term]. Boulder: University of Colorado Press, https://sites.google.eom/a/colorado.edu/adler-scott/data/congressional-district-data (accessed July 30, 2015).

Armey, Richard. 1988. "Base Maneuvers." Policy Review 42, 70-75.

Books, John, and Charles Prysby. 1999. "Contextual Effects on Retrospective Economic Evaluations the Impact of the State and Local Economy." Political Behavior 21 (1): 1-16.

Calabresi, Steven G. 1995. "Some Normative Arguments for the Unitary Executive." Arkansas Law Review 48 (1): 23-104.

Cameron, Charles. 2002. "Studying the Polarized Presidency." Presidential Studies Quarterly 32 (4): 647-63.

Clarke, Harold D., and Marianne C. Stewart. 1994. "Prospections, Retrospections, and Rationality: The 'Bankers' Model of Presidential Approval Reconsidered." American Journal of Political Science 38 (4): 1104-23.

Cohen, Jeffrey E. 2006. "The Polls: The Coalitional President from a Public Opinion Perspective." Presidential Studies Quarterly 36 (3): 541-50.

Cohen, Marty, David Karol, Hans Noel, and John Zaller. 2008. The Party Decides: Presidential Nominations before and after Reform. Chicago: University of Chicago Press.

Edwards, George C., III. 2000. "Building Coalitions." Presidential Studies Quarterly 30 (1): 47-78.

--. 2004. Why the Electoral College Is Bad for America. New Haven, CT: Yale University Press.

Erikson, Robert S. 1989- "Economic Conditions and the Presidential Vote." American Political Science Review 83 (2): 567-73.

Galvin, Daniel J. 2009. Presidential Party Building: Dwight D. Eisenhower to George W. Bush. Princeton, NJ: Princeton University Press.

--. 2013. "Presidential Partisanship Reconsidered: Eisenhower, Nixon, and Ford and the Rise of Polarized Politics." Political Research Quarterly 66 (1): 46-60.

Gartner, Scott Sigmund, Gaty M. Segura, and Bethany A. Barratt, 2004. "War Casualties, Policy Positions, and the Fate of Legislators." Political Research Quarterly 57 (3): 467-77.

Gasper, John, and Andrew Reeves. 2011. "Make It Rain? Retrospection and the Attentive Electorate in the Context of Natural Disasters." American Journal of Political Science 55 (2): 340-55.

Grose, Christian R., and Bruce I. Oppenheimer. 2007. "The Iraq War, Partisanship, and Candidate Attributes: Variation in Partisan Swing in the 2006 U.S. House Elections." Legislative Studies Quarterly 32 (4): 531-57.

Hill, Jason, Erik Nelson, David Tilman, Stephen Polasky, and Douglas Tiffany. 2006. "Environmental, Economic, and Energetic Costs and Benefits of Biodiesel and Ethanol Biofuels." Proceedings of the National Academy of Sciences 103 (30): 11206-10.

Howell, William, and David Brent. 2013. Thinking about the Presidency: The Primacy of Power. Princeton, NJ: Princeton University Press.

Howell, William G., Saul P. Jackman, and Jon C. Rogowski. 2013. The Wartime President. Chicago: University of Chicago Press.

Howell, William, and Terry Moe. 2015. "Resolved, Congress Should Be Required to Vote Up or Down on Legislation Proposed by the President. Pro." In Debating the Presidency: Conflicting Perspectives on the American Executive, eds. Richard Ellis and Michael Nelson. Washington, DC: CQ Press, 144-52.

Hudak, John. 2014. Presidential Pork: White House Influence over the Distribution of Federal Grants. Washington, DC: Brookings Institution.

Kagan, Elena. 2001. "Presidential Administration." Harvard Law Review 114 (8): 2245-385.

Karol, David, and Edward Miguel. 2007. "The Electoral Cost of War: Iraq Casualties and the 2004 US Presidential Election." The Journal of Politics 69 (3): 633-48.

Kriner, Douglas L., and Andrew Reeves. 2012. "The Influence of Federal Spending on Presidential Elections." American Political Science Review 106 (2): 348-66.

--. 2014. "The Electoral College and Presidential Particularism." Boston University Law Review 94 (3): 741-66.

--. 2015a. "Presidential Particularism and Divide-the-Dollar Politics." American Political Science Review 109 (1): 155-71.

--. 2015b. The Particularistic President: Executive Branch Politics and Political Inequality. New York: Cambridge University Press.

Kriner, Douglas L., and Francis X. Shen. 2007. "Iraq Casualties and the 2006 Senate Elections." Legislative Studies Quarterly 32 (4): 507-30.

Lazarus, Jeffrey, and Shauna Reilly. 2010. "The Electoral Benefits of Distributive Spending." Political Research Quarterly 63 (2): 343-55.

Light, Paul C. [1982] 1998. The President's Agenda: Domestic Policy Choice from Kennedy to Clinton. 3rd ed. Baltimore: Johns Hopkins University Press.

Mann, Thomas E., and Normal J. Ornstein. 2013. It's Even Worse Than It Looks: How the American Constitutional System Collided with the New Politics of Extremism. New York: Basic Books.

Mayer, Kenneth R. 1995. "Closing Military Bases (Finally): Solving Collective Dilemmas Through Delegation." Legislative Studies Quarterly 20 (3): 393-413.

Mayhew, David R. 1974. Congress: The Electoral Connection. New Haven, CT: Yale University Press.

McCarthy, Francis X. 2004. FEMA's Disaster Declaration Process: A Primer. Washington, DC: Congressional Research Service. http://fas.org/sgp/crs/homesec/R4.3784.pdf (accessed July 17, 2015).

Milkis, Sidney M., and Jesse H. Rhodes. 2007. "George W. Bush, the Republican Party, and the 'New' American Party System." Perspectives on Politics 5 (3): 461-88.

Milkis, Sidney M., Jesse H. Rhodes, and Emily J. Charnock. 2012. "What Happened to Post-Partisanship? Barack Obama and the New American Party System." Perspectives on Politics 10 (1): 57-76.

Moe, Terry M., and Scott A. Wilson. 1994. "Presidents and the Politics of Structure." Law and Contemporary Problems 57 (2): 1-44.

Mondak, Jeffery J., Diana C. Mutz, and Robert Huckfeldt. 1996. "Persuasion in Context: The Multilevel Structure of Economic Evaluations." In Political Persuasion and Attitude Change, eds. Diana C. Mutz, Paul M. Sniderman, and Richard A. Brody. Ann Arbor: University of Michigan Press, 249-66.

Newman, Brian, and Emerson Siegle. 2010. "The Polarized Presidency: Depth and Breadth of Public Partisanship." Presidential Studies Quarterly 40 (2): 342-63.

Norpoth, Helmut. 1985. "Politics, Economics, and the Cycle of Presidential Popularity." In Economic Conditions and Electoral Outcomes, eds. Heinz Eulau and Michael Lewis-Beck. New York: Agathon, 167-86.

Nzelibe, Jide. 2006. "The Fable of the National President and the Parochial Congress." UCLA Law Review 5.3: 1217-73.

Reeves, Andrew, and James G. Gimpel. 2012. "Ecologies of Unease: Geographic Context and National Economic Evaluations." Political Behavior 34 (3): 507-34.

Reeves, Richard. 2001. President Nixon: Alone in the White House. New York: Simon and Schuster.

Skinner, Richard M. 2008. "George W. Bush and the Partisan Presidency." Political Science Quarterly 123 (4): 605-22.

Thorpe, Rebecca. 2014. The American Warfare State: The Domestic Politics of Military Spending. Chicago: University of Chicago Press.

Wilson, Woodrow. 1908. Constitutional Government in the United States. New York: Columbia University Press.

Wood, B. Dan. 2009- The Myth of Presidential Representation. Cambridge, UK: Cambridge University Press.

(1.) Barack Obama, "State of the Union Address," Public Papers of the President, February 12, 2013. http://www.presidency.ucsb.edu/ws/index.php?pid:=102826 (accessed July 17, 2015).

(2.) The ban was adopted following the 2010 midterms and applied to the 112th Congress, http:// www.speaker.gov/general/sotu-fact-republicans-have-already-adopted-ban-earmarks-%E2%80%93-willsenate-democrats- join-us (accessed July 17, 2015).

(3.) http://www.centeroncongress.org/congress-keeps-ceding-power-the-president (accessed July 17, 2015).

(4.) Democratic Study Group, "The Great Base Closing Ploy: Creating A Political Tempest to Shield a Bloated Defense Budget" (No. 101-29), March 24, 1990.

(5.) Hearing before the Military Installations and Facilities Subcommittee of the Committee on Armed Services, House of Representatives (HASC 101-52), March 14, 1990, 1.

(6.) Alternatively, a negative binomial model of a count version of the dependent variable taking on the values 0, 1, and 2, yields virtually identical results.

(7.) In calculating both of these probabilities, the Democratic member variable is set equal to zero and the number of major military installations in the district to its mean.

(8.) Negative binomial event count models also yield similar results.

DOUGLAS L. KRINER

Boston University

ANDREW REEVES

Washington University in St. Louis

Douglas L. Kriner is an associate professor of political science at Boston University. He is the author of After the Rubicon and coauthor of The Casualty Gap, and articles in the American Political Science Review, the American Journal of Political Science, The Journal of Politics, and other outlets.

Andrew Reeves is an assistant professor of political science at Washington University in St Louis. His work has appeared in the American Political Science Review, the American Journal of Political Science, and The Journal of Politics, among others.

AUTHORS' NOTE: We thank George Edwards. Brandon Rottinghaus. and Rick Waterman for helpful comments and feedback.

TABLE 1

A Model of County-Level Presidential Disaster Declarations,
1984 to 2013, Logistic Regression Model.

                                                (1)        (2)

Swing state * Election year (Obama)                      0.813 *
                                                         (0.139)
Core state * Election year (Obama)                       0.624 *
                                                         (0.182)
Swing state (Obama)                                      -0.263 *
                                                         (0.066)
Core state (Obama)                                       -0.283 *
                                                         (0.093)
Obama administration (2009 to 2013)                      -0.847 *
                                                         (0.104)
Swing state * Election year                   0.909 *    0.693 *
                                              (0.060)    (0.069)
Core state * Election year                    0.875 *    0.695 *
                                              (0.069)    (0.078)
Swing state                                    0.055     0.137 *
                                              (0.034)    (0.039)
Core state                                    0.098 *    0.186 *
                                              (0.035)    (0.042)
Member of Congress from presidential party    0.076 *    0.076 *
                                              (0.021)    (0.021)
Disaster damage (logged 2005 dollars)         0.179 *    0.180 *
                                              (0.005)    (0.005)
Severe disaster events                        0.203 *    0.201 *
                                              (0.012)    (0.012)
Personal income (logged 2005 dollars)         0.278 *    0.269 *
                                              (0.138)    (0.139)
County population (logged, in millions)       -1.621 *   -1.606 *
                                              (0.157)    (0.159)
Observations                                   77,182     77,182
Number of counties                             3,008      3,008

Note: Models 1 and 2 are logistic regressions in which the
dependent variable is whether a county received at least one
disaster declaration in a given year. Robust standard errors
in parentheses. All significance tests are two-tailed.
* p < .05.

TABLE 2

Electoral and Partisan Particularism in
Base Closings, 1990 Cheney List

                                    (1)         (2)
                                  Closure    % Closures

MC Democrat                       1.272 *     0.083 *
                                  (0.473)     (0.0.34)
Swing state                       -1.240 *    -0.062 *
                                  (0.562)     (0.036)
# Major military installations    0.247 *
                                  (0.0839)
Constant                          -3.485 *     0.061
                                  (0.430)     (0.023)
Observations                        435         209
[R.sup.2]                                      0.034

Note: Model 1 is a logistic regression in which the
dependent variable is whether at least one military base was
closed within a congressional district. Twenty-nine
districts experienced one closure; three districts
experienced two closures. An event count model yields
identical results. Model 2 is a least squares regression in
which the dependent variable is the percentage of major
military installations that were closed within a
congressional district. Model 2 includes only those
districts that possessed at least one major military
installation as identified by Adler's Congressional District
Data Base for the 101 st Congress. Robust standard errors in
parentheses. All significance tests are one-tailed.
* p <.05.

TABLE 3

Partisan Particularism Mediated by Electoral
Competitiveness, 1990 Cheney List

                                   Closure
                                  Non-swing    Swing

MC Democrat                        1.366 *     0.646
                                   (0.510)    (1.214)
# Major military installations     0.264 *     0.151
                                   (0.104)    (0.131)
Constant                          -3.589 *    -4.096 *
                                   (0.471)    (1.103)
Observations                         298        137
[R.sub.2]

                                  % Closure
                                  Non-swing    Swing

MC Democrat                        0.110 *     0.005
                                   (0.042)    (0.058)
# Major military installations

Constant                           0.045 *     0.046
                                   (0.024)    (0.045)
Observations                         154        55
[R.sub.2]                           0.036      0.000

Nate: Models 1 and 2 are logistic regressions in which the
dependent variable is whether at least one military base was
closed within a congressional district. Models 3 and 4 are
least squares regressions in which the dependent variable is
the percentage of major military installations that were
closed within a congressional district. Robust standard
errors in parentheses. All significance tests are one-
tailed.

* p < .05.