Pork Barrel Politics

How Government Spending Determines Elections in a Polarized Era

Andrew Sidman

• Publisher: Columbia University Press
• Year: 2019
• ISBN:
  9780231550406
• Topic: Politics & International Relations
• Subtopic: American Government
• Index: Politics & International Relations

Conventional wisdom holds that legislators who bring "pork"—federal funds for local projects—back home to their districts go a long way toward fending off potential challengers. For more than four decades, however, the empirical support for this belief has been mixed. Some studies have found that securing federal spending has no electoral effects at best or can even cost incumbent legislators votes.

In Pork Barrel Politics, Andrew H. Sidman offers a systematic explanation for how political polarization affects the electoral influence of district-level federal spending. He argues that the average voter sees the pork barrel as an aspect of the larger issue of government spending, determined by partisanship and ideology. It is only when the political world becomes more divided over everything else that the average voter pays attention to pork, linking it to their general preferences over government spending. Using data on pork barrel spending from 1986 through 2012 and public works spending since 1876 along with analyses of district-level outcomes and incumbent success, Sidman demonstrates the rising power of polarization in United States elections. During periods of low polarization, pork barrel spending has little impact, but when polarization is high, it affects primary competition, campaign spending, and vote share in general elections. Pork Barrel Politics is an empirically rich account of the surprising repercussions of bringing pork home, with important consequences in our polarized era.

1

Incumbents and Pork Barrel Politics

ON MARCH 10, 2010, House Appropriations Committee chair David Obey, a Democrat, proclaimed that the Appropriation Committee would not consider any earmarks directed to for-profit companies requested for the 2011 budget. In a remarkable step, House Republicans decided the following day to ban earmark requests from their conference altogether (Clarke and Epstein 2010).¹ Attacks on government spending, both large and small, would not stop there. Passed as part of the Budget Control Act of 2011, automatic, widespread spending cuts, known more familiarly as sequestration, took effect March 1, 2013. While sequestration resulted from bipartisan compromise, support for its continuation rested almost entirely within the Republican Party, and more particularly within members of Congress associated with the Tea Party. Closer to the hearts of many in my discipline, 2013 is also notable as the year Senator Tom Coburn, a Republican, briefly succeeded in his four-year quest to severely limit funding for political science research through the National Science Foundation. These three episodes have in common Republican-initiated or Republican-supported cuts to federal distributive spending. Empirical examinations of the pork barrel have asserted for two decades that the fiscally conservative, and Republicans generally, receive little to no electoral benefit from this type of spending. Why, then, did it take Republicans so long to start addressing it—or, more appropriately, why now? The quick answer is that the high level of polarization in recent politics is necessary for there to be general, systematic costs of pork barreling for Republicans.

This is not a book about Republicans and the pork barrel, but Republicans do receive more attention in the discussion. The simple reason is that distributive politics has had stronger and more complex relationships with House election outcomes for Republicans than for Democrats in the recent era. In part, this may be due to the lopsided nature of polarization itself: Republicans in Congress have become more extreme than Democrats (Bonica et al. 2015), as have Republican voters (Mann 2015). The first aim of this work is to understand the interplay between the pork barrel and polarization in electoral politics. In all of the work on distributive politics and elections, I have yet to come across a single work that examines the role of polarization. Undoubtedly, this is a consequence of data availability; some of the best data on the pork barrel exist for a period of relatively low polarization. It is only recently, considering the full period for which these data exist, that polarization has exhibited sufficient variation to draw meaningful conclusions about its effects. In studying the relationship between pork, polarization, and House elections, this book advances the study of distributive-electoral politics in two additional ways.

First, this work includes data for a longer period than typical studies. Much of the empirical work on the electoral effects of the pork barrel is limited with respect to the time period studied. Research published throughout the 1990s, for example, considers by necessity a period during which Democrats controlled Congress.² Few works have examined a longer period that includes the recent era of Republican control, and fewer still have looked at this relationship in the first decade of the twenty-first century. Those that have done so tend to be single-Congress studies of the impact of earmarks, a very specific type of spending for which high-quality data are available for the 110th Congress.³ All of the district-level research presented here includes data from 1986 through 2012—a period that saw both parties in control of Congress. This is also a period during which polarization varied from low to high, generally increasing throughout the era. Chapter 3, which includes analyses of aggregate election outcomes, looks at the relationship between polarization and public-works spending over a far longer period, starting from 1876.

The second advancement is attention to the indirect effects of the pork barrel on election outcomes. There has been a lot of work on the direct effects of the pork barrel on election outcomes, but very little on its indirect effects: the way the pork barrel shapes other outcomes of interest leading to the result on Election Day. Securing project and programmatic spending for the district is part of what a representative does, and this behavior shapes the way constituents view their representative and the way strategic actors view the upcoming election. In practically every empirical work on the topic, the dependent variable is some variation of the district-level vote share of the incumbent. A few studies have examined the effects of pork on the entry decisions of experienced challengers. A small number of works have also considered the link between the pork barrel and individual attitudes and, ultimately, voting behavior. Each chapter of this work presents a different facet of the electoral relationship leading up to the final chapter on district-level election outcomes.

In the following chapters, I consider three district-level indirect effects. First, I look at the role of the pork barrel in determining primary competition. Potential candidates are strategic actors. They have been most studied in the context of challenger quality in general elections, but strategic concerns are relevant to challenger entry in primaries as well. Second, general-election challenger experience is analyzed in its own right. Part of my conclusion is that potential candidates with electoral experience are part of the same political system as House incumbents. Many of the latter believe pork barreling to be electorally beneficial, and many of the former likely hold the same view. There is, therefore, a deterrent effect of pork barreling. For primary competition, particularly for Republicans, there is an antideterrent, or what could be called an “encouragement” effect, inviting competition when polarization is high. Because the analysis of primary competition does not consider the experience of challengers, it could be the case that many of these challengers, who are electoral amateurs, do not base their decisions to run on the same assumptions as their experienced counterparts. Considerations of candidate spending comprise the third indirect effect to be discussed. Campaign donors are no less strategic in their decision making, and pork barreling influences their behavior as well.

Returning to the question of “why now,” the average representative is sure to have little knowledge of the academic literature on the pork barrel. To be fair to members and their staffs, that literature taken as a whole has been inconsistent regarding the electoral effects of the pork barrel. Studies conducted prior to the publication of Stein and Bickers’s (1994a) seminal work on the topic tend to find no systematic relationship at all. Even for Stein and Bickers, the district-level results are weak, and the lynchpin of their individual-level analysis is voter awareness of incumbent activities. Stein and Bickers popularized data from the Federal Assistance Awards Data System, or FAADS. Subsequent work, most of which employs the FAADS data, has been more successful at linking the pork barrel to election outcomes, often conditioning the effects of pork on legislator characteristics such as party or ideology. The argument I offer is that high polarization is necessary to observe general, systematic partisan effects. The pork barrel is not salient as an issue. At higher levels of polarization, however, information about the pork barrel is more likely to correlate with broader attitudes on government spending in an ideologically consistent manner, leading to more easily observed partisan effects. In this way, polarization explains some of the inconsistency in the literature. Research looking only at periods of low polarization tends to find null effects. Studies that cover more recent elections include a period during which polarization is increasing and the pork barrel is having effects that are more consistently partisan.

Work on the electoral effects of the pork barrel necessarily involves several themes relevant to our understanding of the legislative branch and American politics broadly. The study of congressional elections is largely the study of incumbents and the advantages they possess. Especially where the pork barrel is concerned, the analysis is of an activity we expect to be tied to incumbents—for good, as much of the literature has assumed, or for bad, as I will demonstrate in later chapters. I argue that polarization is key to understanding the relationship between particularistic spending and elections. Underlying polarization are party, ideology, and, in the American context, the way these identities and ways of understanding the political world have become more entangled over the past two decades. I turn to each of these themes now.

Incumbents and Reelection

The importance of David Mayhew’s (1974) Congress: The Electoral Connection to the study of congressional elections cannot be overstated. Its simple and elegant argument is that members are motivated primarily by the electoral goal and that, in achieving this goal, members have distinct advantages over their competitors, through credit claiming, position taking, and advertising. The prevailing view of congressional elections scholars has been that all three of these activities work together to comprise incumbency advantage. To be sure, over the four decades since publication of Mayhew’s monumental work, we have observed members of Congress taking positions through their statements and rollcall behavior, securing federal dollars for their states and districts, advertising their positions and distributive activities, and generally winning election by safe margins and at very high rates. This view, however, developed during an important transition in congressional politics. The 1970s and 1980s witnessed relatively low polarization, both in Congress and in the public. Congress, for the first time in U.S. history, opted to regulate campaign finance. These new regulations coincided with a growth in the number of new members less beholden to the institutional power brokers of the past and more single-minded in their pursuit of the electoral goal. This seeming free-for-all does not describe the Congress of today. Congress in the twenty-first century is more polarized than in decades past, mirroring the growing polarization among partisans in the mass public. The institutional power wielded by the committee chairs of an earlier era is held, perhaps even more strongly, by today’s party leaders. And while the electoral goal remains paramount, the regularity of “wave” elections signals a much closer relationship between the fortunes of the member and those of the party than existed in the 1970s and 1980s. It is in this context that I question the general consideration of pork barreling as credit claiming.

Mayhew writes, “How much particularized benefits count for at the polls is extraordinarily difficult to say, but it would be hard to find a Congressman who thinks he can afford to wait around until precise information is available. The lore is that they count” (1974, 57). A key assumption of my argument, one that differs from much of the literature, is that the pork barrel is an issue on which significant subsets of the public hold negative opinions. Nearly all of the work on distributive politics starts with a simple assumption based in rational choice theory: constituents prefer receiving more benefits over receiving fewer benefits. The assumption that everyone wants something pervades work on distributive politics, even when much of the empirical work on the intersection between distributive and electoral politics has failed to find that direct link. The effects of distributive benefits in several of these studies are not universal. They are conditioned by factors such as party, ideology, or the type of benefit, recognizing that legislators from different parties tend to benefit from different types of programs. The “more is better” assumption, combined with the observation of these conditional effects, has led to the general conclusion that there is always a systematic benefit of pork barreling, assuming the legislator secures the “right” pork for her party, and rarely, if ever, a cost for other pork. Because voters always want more, regardless of whether the benefits are more traditional grant and direct payment programs or contingent liabilities, this spending is the “stuff” of credit claiming. To the contrary, I argue that the pork barrel is not an apolitical assortment of benefits. Collectively, it is, and always has been, an ideological issue.

Party Politics and Polarization

At around the same time that Mayhew was exploring the electoral motivations of members of Congress, other scholars were investigating a phenomenon that would be called “dealignment,” the detachment of large segments of the public from partisan politics, especially in party identification (Abramson 1976; Norpoth and Rusk 1982). This dealignment of the mass public would also occur during a period marked by weak parties in Congress. As late as 1993, Keith Krehbiel was asking “Where’s the Party?” and concluding that the political parties exerted a relatively weak influence over behavior in Congress, despite the increasingly homogenous preferences of party members. As I will show in the next chapter, polarization was very low in the 1970s and has increased steadily through the present. The year 1993 was also one year before Republicans would regain control over both chambers of Congress in an election year marked by concerted efforts by Republican leaders to campaign on a national message (Jacobson 1996). Fast-forward to the present, and few would argue against the emergence of party as a powerful motivator of behavior in Congress, with party leaders wielding more influence than in the earlier eras. Increasing ideological homogeneity within the parties and growing divisions between them are the conditions assumed to underlie a shift of institutional power to party leaders (Rohde 1991). Conditional party government, as this theory is called, is caused by the same divisions that define elite polarization. Members of each party in Congress have become increasingly unified around their preferred policy outcomes, while the ideological distance between average members of the parties has grown. This has allowed the parties to put forward clear ideological positions to the electorate and has left little space for compromise on most issues facing the nation (Lee 2009, 2016).

Despite the prevalence of elite polarization, there is a debate over whether we observe these divisions in the mass public. Some would argue, perhaps most famously Fiorina, Abrams, and Pope (2006), that the belief we are a 50–50 nation is a myth. Although our elections may be highly contested and results evenly divided, the preferences of the mass public are less divided and more moderate than our election outcomes make them appear (Ellis and Stimson 2012). On the other side, several scholars have observed clearer signs of mass polarization, especially in the partisan public—those individuals who identify with one of the major parties. While increasing partisan rancor in the mass public is a cause for concern (Mason 2015), more relevant to this work is the closer relationship between ideology and party identification (Abramowitz and Saunders 2008; Levendusky 2009). Whether the public as a whole remains moderate and shares preferences on most issues is debatable. The subset of partisans, however, have become more willing to adopt the preferred ideological positions of their chosen party and have become far more ideologically consistent across issues than in the past (Abramowitz 2015). Jacobson (2015) notes that the electorate in 2012 was more polarized than in any election for which we have survey data. In preferences and identification, Democrats were more likely to be liberal, Republicans were more likely to be conservative, and both sets of partisans, by virtue of their political efficacy, were more likely to vote. Polarization, therefore, leads to systematic partisan effects of the pork barrel in elections because of the increased consistency between ideological preferences and party in the most likely voters.

A Brief Introduction to the Pork Barrel

The term “pork barrel” is used in this book in its most general sense: particularistic benefits that can be geographically targeted to congressional districts. The earliest academic work on the subject I could find dates to 1919; it defines the pork barrel as a system born in Congress on May 20, 1826, with the passage of the first omnibus appropriations bill for the improvement of rivers and harbors (Maxey 1919). Into the omnibus bill were placed all of the river and harbor appropriations that members desired for their districts. With nearly every member receiving an appropriation, majority support was guaranteed for passage and, in Maxey’s words, “the good items in such a bill would stand as apologists for the bad” (1919, 692). In this way, the common conception of the pork barrel was easy to understand and comported with the observed behavior of most members. Members worked to add their pet projects to the omnibus, presumably to show constituents that they had “worked” on their behalf and deserved to be reelected. Despite this picture of universalism, such public-works spending was not without opposition. Indeed, as will be detailed in a later chapter, spending on these so-called “internal improvements” was met with heavy resistance from the Democratic Party, starting with President Andrew Jackson. Maxey (1919) notes that between Jackson’s presidency and the end of the Civil War, not a single river and harbor omnibus bill became law except in 1852, the last year (ever) of a Whig administration.

The pork barrel today, and for much of the twentieth century, differs from the omnibus bills of the antebellum era in size, scope, and process, but not in principle. Much of what we consider the pork barrel today comes in the form of projects and programs administered through the vast federal bureaucracy, with funding typically coming through the normal appropriations process in Congress. This is not to say that individual member requests for project funding have disappeared.⁴ They are, however, a small part of a vast system of what the federal government generally calls “domestic assistance” and what the academic literature labels “distributive benefits.” The role of the member in this process is less about adding particular projects to regular appropriations or omnibus spending bills and more about budgetary authority over federal agencies and facilitating or assisting interested parties in securing federal money. Federal projects and grant assistance comprise a significant share of the casework done by members and their staffs (Johannes 1983a).

The Catalog of Federal Domestic Assistance (Executive Office 2012), which lists the programs that comprise much of the data on distributive benefits, includes several types of assistance pertinent here. Although there are many ways to classify these programs, this research requires measures of the pork barrel that comport with partisan and, more importantly, ideological preferences. The core of my argument is that the pork barrel has systematic partisan effects only when high levels of polarization activate ideological sentiment toward these types of spending. One could, for example, distinguish pork by perceived partisan preferences for the target of the spe...

Table of contents

• Cover 
• Title Page
• Copyright
• Dedication
• Contents 
• Acknowledgments
• 1. Incumbents and Pork Barrel Politics
• 2. Pursuing the Pork Barrel
• 3. An Electoral History of the Pork Barrel
• 4. Attitudes, Voting, and the Pork Barrel
• 5. Challenges from Within the Party
• 6. General-Election Challengers and Campaigns
• 7. Election Outcomes
• Conclusion
• Notes
• References
• Index