Emerging Trends in The Social and Behavioral Sciences

Distributive Politics: Federal Outlays

Media

Part of Distributive Politics: Federal Outlays

Title

Distributive Politics: Federal Outlays

extracted text

Distributive Politics: Federal Outlays
SANFORD C. GORDON and WOO CHANG KANG

Abstract
We provide an interpretive review of theoretical and empirical research on the distribution of federal outlays in the United States and their political antecedents. We
argue that the foundational scholarly research on the subject to date can be classified
to a great extent according to the perspectives it takes respecting legislative organization, political parties, and the executive branch. After reviewing theoretical and
empirical research in these areas, we discuss some cutting-edge research in this area
of research. One of the most exciting trends in the study of distributive politics is
a much greater focus on issues of causal identification in empirical tests of theories
concerning the allocation of federal resources. Finally, we describe five questions that
we believe ought to motivate future research in this area. These questions concern
the relationship between entitlement and discretionary spending; how to integrate
research on distributive politics with research on political business cycles; how best
to measure executive discretion in specific assistance programs; how to understand
the relationship between spending in an area and support for that spending; and how
to capture the mechanism relating spending and electoral benefits.

INTRODUCTION
In 2012, the federal government spent about $11,600 per citizen, or roughly
23% of the US gross domestic product. While the governments of many
developed countries spend more per capita [even once one accounts for
state and local expenditures, the United States still ranks 14th among OECD
countries (Central Intelligence Agency, 2011)], the sheer enormity of federal
outlays—totaling $3.6 trillion in 2012—makes their allocation a ripe subject
for systematic analysis.
To what extent are federal expenditures in the United States allocated on
the basis of political considerations? More specifically, what are those considerations, and how do US political institutions moderate or exacerbate their
application? This essay is a reflection on scholarly inquiry into these subjects. We first review some of the foundational research on the subject. We
argue that to a great extent, scholarly work can be classified according to the
perspective it takes on (i) legislative organization; (ii) political parties; and
Emerging Trends in the Social and Behavioral Sciences. Edited by Robert Scott and Stephen Kosslyn.
© 2015 John Wiley & Sons, Inc. ISBN 978-1-118-90077-2.

1

2

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

(iii) executive power. After reviewing some of the key findings in this literature, we proceed to a discussion of some cutting-edge areas of research. In
particular, we focus on how empirical scholars studying distributive politics
in the United States have become much more attuned to issues of research
design and causal identification. We conclude by identifying some key issues
that future scholars working in this area will need to grapple with.
Following Lowi’s (1964) famous typology, we take a signature feature of
distributive policies to be divisibility: namely, the ease with which policy
choices “can be disaggregated and dispensed unit by small unit” (p. 690).
Accordingly, in what follows we will focus on the political economy of discretionary domestic spending rather than entitlement spending; whether the
discretion lies with Congress or the executive branch will be a critical issue
in our discussion.
A BRIEF PRIMER ON FEDERAL SPENDING
It is important to define the object under scrutiny before proceeding to a
discussion of how it has been studied. A full consideration of government
spending in the United States is far beyond the scope of this short essay.
However, understanding how expenditures are allocated requires some
basic knowledge about the federal budget. At the most basic level of
aggregation, federal expenditures may be divided into three categories:
discretionary expenditures, mandatory expenditures, and interest payments
on the national debt. Mandatory expenditures are programs such as Social
Security and Medicare, the funding for which is provided for by standing law
and not subject to the recurrent budgeting process. Discretionary spending,
which itself may be divided into defense and nonrelated areas, is decided
on during the annual appropriation process. Table 1 displays amounts (in
billions of nominal dollars) associated with these major categories.
In addition to funding the operation of government agencies, a large fraction of the discretionary budget goes to fund federal domestic assistance
programs; these include grants, direct payments, insurance programs, loans
Table 1
Federal Spending by Category, Fiscal Year 2012
Category
Nondefense discretionary
Defense-related discretionary
Mandatory
Net interest

Amount ($Billions)
615.6
670.5
2134.2
220.4

Source: Office of Management and Budget, Historical Table 8.1.

Distributive Politics: Federal Outlays

3

and loan guarantees. The recipients of these benefits might be private individuals, companies, or institutions (e.g., universities and other nonprofits);
alternatively, the recipients may be states or local governments. $155.7 billion
went to nonmilitary contracts in 2012, and $361.3 billion to military contracts
(source: usaspending.gov).
A critical feature of these categories of spending is the delegation of authority to agencies in the executive branch. Federal agencies differ markedly in
the degree of effective discretion they possess in disbursing funds—whether
to individuals and private organizations or to state and local governments.
For certain programs, Congress specifies the amounts and recipients in statutory formulae—witness, for example, the formulas governing price supports
for agricultural commodities. For others, Congress grants a much greater
degree of discretion to the agency administering the program: grants from
the National Institutes of Health; mortgage insurance from the Federal Housing Administration; or Homeland Security grants to states and localities, for
example. Finally, “earmarks” are directives from Congress to an executive
branch agency to spend its nominally discretionary budget in specific ways.
While some such directives lie in statutes, legislators may place others in
committee or conference reports.
FOUNDATIONAL IDEAS AND RESEARCH
LEGISLATIVE INSTITUTIONS
As noted in the introduction to this essay, much of the early research on distributive politics in the United States is Congress-centered (not surprising
given Congress’s spending power as defined in Article I, Section 8 of the Constitution), and focused in particular on the historically decentralized nature
of the legislative process in the United States.
The modern study of distributive politics in the United States can be traced
to progress in the institutional analysis of Congress in the 1960s. Fenno’s
landmark (1966) study of Congressional Appropriations, The Power of the
Purse, put the focus of studies of legislative behavior squarely on Congress’s
spending power and in particular on the House Appropriations Committee.
Polsby (1968) described the emergence of universalistic norms in the House
of Representatives: most critically, how the discretion of the Speaker of the
House in selecting committee chairs had been supplanted by reliance on
a seniority norm. Mayhew (1974) argued for the importance of the committee system in allowing reelection-minded legislators to take positions
on, and claim credit for, particularistic policies benefiting their respective
constituencies. In addition, Ferejohn’s (1974) exhaustive study of federally
funded local water projects related the allocation of those projects to the

4

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

institutional position of legislators and their resultant influence over the
Army Corps of Engineers. A series of rational choice accounts of legislative
institutions have placed the substantive intuitions of these earlier works
on firm analytical foundations (e.g., Shepsle & Weingast, 1981; Weingast &
Marshall, 1988).
Three clusters of hypotheses emerge from this theoretical literature. First,
we should not expect federal largesse to be concentrated in a minimum winning coalition. This is overwhelmingly confirmed in the data.
Second, we should see a relationship between pork barrel spending in
a legislator’s district or state and his or her positions in key committee
posts and/or seniority. The empirical record on the committee membership
hypothesis is mixed. Recent research confirming the importance of committee positions for at least some committees includes work by Alvarez
and Saving (1997); Engstrom and Vanberg (2010); and Lauderdale (2008).
Research failing to confirm the committee membership hypothesis includes
work by Evans (2004). With regard to the seniority hypothesis, two mechanisms may be at play: access to committee leadership (Roberts, 1990; but
see Arnold, 1979), and increased skill in the legislative process (Engstrom
& Vanberg, 2010; Sobel & Ryan, 2012; but see Anzia & Berry, 2011; and
Lauderdale, 2008).
Third, to the extent that legislators can devote only so much effort to bringing home the bacon, more legislators per citizen in a state or district (brought
about either by fixed representation of states in the senate or malapportionment in the House) should give rise to greater expenditures in that area (e.g.,
Lee, 2000).
POLITICAL PARTIES
The literature referenced above tends to deemphasize the role of party.
However, to the extent that the majority enjoys controls over the legislative
agenda, this advantage should engender advantages in the allocation of federal resources. A number of studies present evidence that voters represented
by members of a majority party receive more benefits than those represented
by members of a minority party (e.g., Albouy, 2013; Balla, Lawrence, &
Maltzman, & Sigelman, 2002; Berry, Burden, & Howell, 2010).
A related body of theoretical research, less focused on the United States
specifically, seeks to make general claims about how strong parties might target government transfers for political benefit. In the model described by Cox
and McCubbins (1986), a risk-averse incumbent party has incentives to direct
transfers primarily toward core supporters, in order to maintain its existing
electoral coalition. By contrast, Lindbeck and Weibull (1987) describe a model
of two-party competition in which, in equilibrium, both parties primarily target marginal constituencies (in their setup, middle-income voters) rather than

Distributive Politics: Federal Outlays

5

core supporters (low-income voters for one party and high income voters
for the other). In the more general setup of Dixit and Londregan (1995, 1996),
parties tradeoff between the electoral benefits of targeting pivotal constituencies and of satisfying core groups of supporters. Asymmetries between the
parties in the efficiency with which they can target certain groups push the
parties away from targeting marginal and toward core voters.
Both the core and swing voter models can claim their share of empirical
support. Evidence for core voter models often overlaps with evidence for
the majority party advantage because the distribution of core voters is measured with levels of support in the previous election, which is correlated with
majority party status (Balla et al., 2002; Levitt & Snyder, 1995; see also Dynes &
Huber, 2013). On the other hand, Bickers and Stein (1996) suggest that incumbents elected in open seat races receive more federal spending. Other scholars
have found evidence that at least some subsets of incumbents who won with
close margins tend to bring home more federal resources to their districts
(e.g., Berry et al., 2010; Lazarus, 2009; Stein & Bickers, 1994).
THE EXECUTIVE BRANCH
Most recently, scholars have begun to include the executive branch more
explicitly in theoretical models of distributive politics. Kiewiet and McCubbins (1988) argue that a presidential veto threat is only credible in changing
spending on distributive programs if the president prefers less spending than
the legislature. McCarty (2000) develops a model in which the presidential
veto can reduce spending unless the president himself has strong preferences over
its distribution. To the extent that the president’s electoral prospects depend
on delivering benefits to (core or marginal) constituents, he may have good
reason to behave in the ways predicted by the models described above.
To the extent that a president’s goals are distributive, politicization might
thus facilitate presidential influence in the allocation of discretionary federal expenditures. What, then, should that influence look like? A president is
likely to benefit from directing resources to copartisan legislators and other
officials, either in exchange for their support on legislative initiatives or to
enhance their electoral prospects. Berry et al. (2010), for example, demonstrate that legislative districts and counties represented by copartisan legislators of presidents receive significantly more federal spending.
The president seeks reelection and to preserve his party’s control of the
White House. Therefore, he and his subordinates have an incentive to allocate federal resources toward that end. Demonstrating what this means in
practice has led to mixed results empirically. Studies on New Deal spending have found that states with more voting variability received more federal employment and agricultural aid (e.g., Wright, 1974). Hudak (2014) also

6

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

shows that presidents direct more discretionary grants to swing states, especially in advance of presidential elections. In general, however, the strength
of empirical support for the swing voter model is weak in studies on the distribution of federal resources. Both Larcinese, Rizzo, and Testa (2008) and
Berry et al. (2010) find that swing states do not receive more federal outlays.
On the other hand, these studies find a positive and significant association
between the levels of support for the president and the allocation of federal
resources at the state level.
In his seminal work on congressional-bureaucratic relations, Arnold (1979)
notes that “most decisions about geographic allocations are bureaucratic
decisions” (p. 8). Despite this authority, bureaucrats have an incentive
to consider preferences of other political actors to accomplish their own
goals (see also Maass, 1950). For example, bureaucrats have an incentive to
consider preferences of members of Congress because attaining budgetary
stability and growth depends on their decisions. Similarly, Stein and Bickers
(1997) claim that concerns about budgets induce bureaucrats to help not
only legislators but also constituents and interest groups. Bureaucrats also
have an incentive to respond to the president’s political needs to obtain
presidential support (Hudak, 2014).
Obtaining empirical evidence on whether, and to what extent, bureaucrats
allocate federal resources in response to the demands of other political officials is challenging, both because those potential influences are many and
because it is difficult to tell the degree to which apparently politicized allocations reflect discretionary bureaucratic actions as opposed to being hardwired into enabling statutes or directed from the White House. With this in
mind, Bertelli and Grose (2009) find that a greater degree of ideological congruence between a state’s senators and the Secretaries of Defense and Labor
is associated with greater discretionary allocations (in the form of contracts
and grants) to the senators’ respective states. The congruence of the senators
and the president has no apparent effect, however. They interpret these findings as implying an autonomous sphere of bureaucratic influence. Gordon
(2011) demonstrates a linkage between White House directives to political
appointees at the General Services Administration and the allocation of subsequent building contracts.
AREAS OF CUTTING-EDGE RESEARCH
We now turn to some recent developments in the study of the distributive
politics of federal outlays. In our estimation the key development in the
literature over the past decade or so has been a greater focus on causal identification. In an ideal world, we could test our hypotheses about distributive
politics by randomly assigning political significance across districts and

Distributive Politics: Federal Outlays

7

observing differences in spending outcomes. In the real world, of course,
such experimental manipulations are impossible. For an observed relationship between spending in a district or state and the hypothesized measure of
its political significance to be interpreted as a causal, it must be the case that,
the status of a district as important or unimportant is as good as randomly
assigned once one conditions on other observable characteristics of districts.
Otherwise, it could be the case that the measure of influence is proxying the
effect of some unobservable, politically neutral feature of the district.
For example, suppose we observe that districts whose legislators sit on
the Appropriations subcommittee overseeing the Department of Housing
and Urban Development receive a disproportionate share of housing grants.
Is this bureaucratic deference to those legislators? Possibly, but it could
simply reflect the member’s self-selection onto that committee owing to the
large number of eligible beneficiaries in her district. The analyst, hoping to
eliminate this possibility, controls for an observable measure of district-level
demand: for example, percent urban in the district. How sure can we be that
this fully neutralizes the bias? Studies of the effects of malapportionment
suffer from a similar issue: perhaps greater per capita spending in Alaska
and Montana is a consequence of diseconomies of scale in providing comparable levels of services to widely dispersed populations (Elis, Malhotra,
& Meredith, 2009), and not the greater effort or pivotality of their senators.
Unfortunately, nearly all of the empirical studies of distributive politics
described in the previous section are susceptible to this criticism.
Because scholars cannot randomly assign political significance as in an
experiment, scholars have recently turned to innovative research designs to
minimize the ubiquitous threat of confounding. One type of design exploits
the panel structure of expenditure data and examines repeated observations
of the same districts. The presence of multiple observations on the same
cross-sectional unit (district or state) allows the analyst to implicitly control
for all observable and unobservable time invariant features of congressional
districts over their lifespans, as well as year-to-year variation in spending
levels. The basic idea is that rather than compare spending in districts that
differ in their observable political characteristics (i.e., “between” district
variation), the analyst examines the effect of changes in the political status of
specific districts on spending (“within” district variation), and then averages
across the many districts that observe a change. This is the approach taken
in a recent papers by Berry et al. (2010), who demonstrate a relationship
between spending in a district and whether the district is represented by a
copartisan of the president; and by Anzia and Berry (2011), who demonstrate
that female legislators are more successful at securing federal expenditures
in their districts than their male counterparts.

8

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

Second, regression discontinuity designs exploit the fact that the assignment of the “treatment” of political importance may be a consequence of
whether some continuous measure falls above or below some exogenously
given threshold; the analyst then compares the outcomes of observations
falling on both sides of the threshold, under the assumption that in that
vicinity, treatment assignment is effectively random. Albouy (2013) takes
this approach in estimating the effect on various categories of spending of
having a Republican or Democratic legislator, focusing on candidates who
won elections by small margins (i.e., those close to the 50% threshold). Anzia
and Berry (2011) also perform a regression discontinuity analysis to isolate
the effect of female legislators.
A third approach is to exploit some natural experiment provided by a plausibly exogenous shock to district-level circumstances, often in conjunction
with a panel data structure. Elis, Malhotra, and Meredith (2009), for example,
employ the discrete changes in representation created by the decennial census reapportionment process to estimate the effect of changes in congressional delegation size on federal outlays. In addition, Gordon (2011) exploits
a minor scandal occurring during the presidency of George W. Bush: a White
House deputy was found to have given a presentation to political appointees
at the General Services Administration mentioning priority congressional
districts for the upcoming election. By comparing GSA spending in the mentioned and unmentioned districts in the weeks immediately before and following the briefing, Gordon recovers an estimate of the effect of a White
House directive to bureaucrats on subsequent spending.
KEY ISSUES GOING FORWARD
We conclude by identifying five key issues in the study of the politics of federal outlays.
HOW DO THE POLITICS OF DISCRETIONARY EXPENDITURES CHANGE WHEN A DECREASING
FRACTION OF THE FEDERAL BUDGET IS DISCRETIONARY?
Two facts—an aging population and dramatic increases in health care
costs—have led to an explosion in the fraction of federal expenditures
devoted to entitlements. Figure 1 displays this trend graphically: whereas
in 1900, less than 1% of the budget was devoted to welfare, pensions, or
healthcare. By 2012 that number had spiked to nearly 60%.
In considering this picture, it is important to point out that discretionary
spending as a fraction of GDP has remain largely unchanged. However, the
fact that it constitutes a decreasing fraction of overall expenditures means
the role of discretionary expenditures in American politics may change. As

9

0.5
0.4
0.3
0.2
0.1
0.0

Proportion entitlements

0.6

Distributive Politics: Federal Outlays

1900

1920

1940

1960

1980

2000

Year

Figure 1 The fraction of federal expenditures devoted to entitlement spending
has increased dramatically in the last century. Source: Budget data compiled by
usgovernmentspending.com

documented by Poole and Rosenthal (2007), in the past 30–40 years it has
become increasingly easy to describe political conflict within Congress as
occurring on a single “left-right” dimension, where that dimension captures
partisan conflict or disagreement about redistributive policies, for example,
entitlements. Moreover, the ideological distance between the two parties
has increased over this period. Historically, pork barrel spending provided
a means to “grease the wheels” of legislative bargaining by permitting
congressional leaders to buy off reluctant coalition members (Evans, 2004).
As legislative gridlock over budgetary and entitlement reform appears
increasingly inexorable, however, the amount of distributive largesse
required to overcome a legislator’s ideological inclinations on the main
issues of the day is likely to increase. If reform is sufficiently important, this
could increase in the importance of pork in fashioning coalitions; however,
if the cost of pork is prohibitive, that could conceivably produce a decrease
in its importance.
DO THE POLITICAL INCENTIVES GOVERNING THE ALLOCATION OF FEDERAL EXPENDITURES
VARY DYNAMICALLY?
As Harold Lasswell famously argued in 1936, politics is about “who gets
what, when, and how”. The studies on distributive politics cited here
generally focuses on the “what.” Starting with Nordhaus (1975), a related
literature on political businesses cycles (PBCs); addresses the “when,” focusing on temporal variation in aggregate government expenditures before and

10

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

after election. Both literatures have a common interest in understanding how
an incumbent might use government resources for electoral ends. Owing to
the independent development of each literature, however, the possibility of
dynamic changes in the allocation of discretionary spending has received
little attention so far.
Keech and Pak (1989) and Mayer (1995) examine whether Veteran’s transfers or military contracts increase in the preelectoral period respectively.
Keech and Pak (1989) fail to find confirmatory evidence for budget expansion On the other hand, Mayer (1995) shows that the number of civilian
agency contract awards rapidly increases just before presidential elections
and primaries.
Kang (2014), on the other hand, examines the possibility that main beneficiaries of government spending vary with electoral cycles. An opportunistic
incumbent often needs support of swing voters to win an election, which
provides him with an incentive to target these voters before elections. Once
winning an election, however, he has various motivations to cater to his core
constituents.
IS THERE A PRINCIPLED WAY TO QUANTIFY EXECUTIVE DISCRETION OVER EXPENDITURES?
As of May 2013, the Catalog of Federal Domestic Assistance describes
2233 federal assistance programs. As noted above, the amount of effective
discretion that presidents possess in administering these programs varies
markedly. Levitt and Snyder (1995) argue that programs with geographically concentrated benefits are more easily targeted than ones with diffuse
benefits: they calculate coefficients of variation for each program using
district averages, and classify programs as having low or high variation.
The former include entitlement programs such as Social Security, Medicare,
and Veterans’ Benefits. The latter include the lion’s share of discretionary
programs. Anzia and Berry (2011) and Berry et al. (2010) also use the
coefficient of variation methodology.
This approach represents an important first step, but has important limitations. While the allocation of entitlement spending has distributive consequences, it is difficult to argue that in the short to medium term that its
allocation reflects tactical political choices of entrepreneurial politicians (as
opposed to, e.g., demographic changes). Moreover, this approach does not
capture institutional constraints on discretion: as noted above, an agricultural subsidy program may have a high coefficient of variation, but it does
not afford the kind of discretion to the Department of Agriculture that, for
example, the ability to lease, operate, and renovate buildings affords to the
General Services Administration.

Distributive Politics: Federal Outlays

11

Codifying the residual discretion for individual assistance programs in
a systematic manner that takes account of, inter alia, statutory constraints
on discretion, the degree to which the program is insulated via staffing
or position in the bureaucratic hierarchy from overhead political control,
promises to be an enormous undertaking. However, it is one well worth
pursuing. Recent theoretical research (Ting, 2012) describes conditions
under which Congress will delegate funding choices to professionalized
bureaucrats rather than bargain directly over distribution. To the extent that
political scientists wish to subject his argument to care empirical scrutiny,
or, more broadly, to refine arguments about which government expenditures
are most likely to respond to the various political stimuli described above,
this undertaking would seem to represent a critical step.
WHY ARE STATES THAT ARE NET RECIPIENTS OF FEDERAL LARGESSE THE MOST HOSTILE TO
GOVERNMENT SPENDING?
Lacy (2009) observes an intriguing statistical relationship: states that vote
Republican in presidential elections, and thus might be expected to have the
greatest hostility to federal spending, tend to be net recipients of federal dollars, whereas states that tend to vote Democratic tend to be net providers;
moreover, this relationship appears to be increasing over time. What is the
relationship between distributive and ideological politics? Endeavoring to
answer this question would be a fruitful line of future research.
CAN OUR EMPIRICAL ANALYSIS BETTER FLESH OUT THE MECHANISM RELATING POLITICAL
EXPENDITURE WITH ELECTORAL BENEFIT?
Cox (2009) notes the importance of distinguishing between distributive
spending on swing groups of voters on the one hand and swing districts on the
other. Suppose, for example, that a party wanted to maximize its majority in
the legislature. Such a party would do well to target expenditures toward
swing districts. However, within those districts, whether that party should
target swing or core voters depends on the mechanism relating political
expenditures with electoral benefits. The “swing voter model” cited above
relies on the idea that pivotal voters will be persuaded by distributive
largesse to support the candidate credited with providing it. By contrast, the
“core voter” model relies on a logic of mobilization (getting the base to turn
out and contribute to a campaign) and coordination (reducing the number
of internecine competitors, e.g., primary challengers).
If persuasion within a district is more critical than persuasion, then we
would expect to see swing voters in swing districts targeted; if, on the other
hand, mobilization and coordination were critical, we would expect to see

12

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

core voters in swing districts targeted. In the absence of microlevel data on
individual recipients of federal assistance and their voting histories, one
possible approach to pinning down the preponderant mechanism would be
to disaggregate spending by discretionary program where the ideological
affinities of groups of voters are well known. For example, do we see a
disproportionate number of small business loans flowing to swing districts
under Republican administrations?
Even if the core-voter-in-swing-district hypothesis receives support, this is
still not the end of the story. It is possible, for example, that legislators seek
pork barrel spending in their districts to claim credit among voters (Mayhew, 1974). However, it is also possible, if not likely, that spending is a way
of mobilizing elites within a district to support an incumbent. To our knowledge, to date only Mebane and Wawro (2002) have investigated this specific question, although Gordon (2011) argues that the elite mechanism is
more plausible for most GSA public building contracts, owing to their low
visibility.
CONCLUSION
In this essay, we have endeavored to describe the past, present, and, it is to be
hoped, future of empirical and theoretical research on the distribution of federal outlays. In contextualizing past, foundational research, it is helpful to see
this research as echoing advances in the study of US political institutions. Earlier research has focused in particular on the relationship between legislative
organization and distribution, while more recent research has revitalized the
critical role played by political parties, the president, and the bureaucracy.
We have argued that some of the most exciting contemporary research on
distributive politics in the United States is centrally concerned with employing innovative research designs to improve causal identification. In addition, lastly, we have identified a number of useful unanswered questions that
future scholars working in the area might seek to answer. These questions
concern the relationship between entitlement and discretionary spending;
how to integrate research on distributive politics with research on political
business cycles; how best to measure executive discretion in specific assistance programs; how to understand the relationship between spending in
an area and support for that spending; and how to capture the mechanism
relating spending and electoral benefits.
Of course, this list of topics is necessarily incomplete. We hope that scholars
working in the area will expand it, and in so doing, match the creativity associated with asking new questions about the subject with intellectual rigor in
answering them.

Distributive Politics: Federal Outlays

13

REFERENCES
Albouy, D. (2013). Partisan representation in congress and the geographic distribution of federal funds. Review of Economics and Statistics, 95(1), 127–141.
Alvarez, R. M., & Saving, J. L. (1997). Congressional committees and the political
economy of federal outlays. Public Choice, 92(1–2), 55–73.
Anzia, S. F., & Berry, C. R. (2011). The Jackie (and Jill) Robinson effect: Why do congresswomen outperform congressmen? American Journal of Political Science, 55(3),
478–493.
Arnold, R. D. (1979). Congress and the bureaucracy: A theory of influence. New Haven,
CT: Yale University Press.
Balla, S. J., Lawrence, E. D., Maltzman, F., & Sigelman, L. (2002). Partisanship, blame
avoidance, and the distribution of legislative pork. American Journal of Political Science, 46(3), 515–525.
Berry, C. R., Burden, B. C., & Howell, W. G. (2010). The president and the distribution
of federal spending. American Political Science Review, 104(4), 783–799.
Bertelli, A. M., & Grose, C. R. (2009). Secretaries of pork? A new theory of distributive
public policy. Journal of Politics, 71(3), 926–945.
Bickers, K. N., & Stein, R. M. (1996). The electoral dynamics of the federal pork barrel.
American Journal of Political Science, 40(4), 1300–1326.
Central Intelligence Agency (2011). The world factbook.
Cox, G. W. (2009). Swing voters, core voters and distributive politics. In I. Shapiro,
S. C. Stokes, E. Wood & A. S. Krishner (Eds.), Political representation. Cambridge,
England: Cambridge University Press.
Cox, G. W., & McCubbins, M. D. (1986). Electoral politics as a redistributive game.
Journal of Politics, 48(2), 370–389.
Dixit, A., & Londregan, J. (1995). Redistributive politics and economic efficiency. The
American Political Science Review:, 89, 856–866.
Dixit, A., & Londregan, J. (1996). The determinants of success of special interests in
redistributive politics. Journal of Politics, 58(4), 1132–1155.
Dynes, A., and Huber, G. A. (2013). Partisanship and the allocation of federal spending:
Do same-party legislators or voters benefit from shared party affiliation? Typescript, Yale
University.
Elis, R., Malhotra, N., & Meredith, M. (2009). Apportionment cycles as natural experiments. Political Analysis, 17(4), 358–376.
Engstrom, E. J., & Vanberg, G. (2010). Assessing the allocation of pork: Evidence from
congressional earmarks. American Politics Research, 38(6), 959–985.
Evans, D. (2004). Greasing the wheels: Using pork barrel projects to build majority coalitions. New York, NY: Cambridge University Press.
Fenno, R. F. (1966). The power of the purse: Appropriations politics in congress. Boston,
MA: Little Brown & Company.
Ferejohn, J. A. (1974). Pork barrel politics: Rivers and harbors legislation, 1947–1968. Stanford, CA: Stanford University Press.
Gordon, S. C. (2011). Politicizing agency spending authority: Lessons from a Bush-era
scandal. American Political Science Review, 105(4), 717–734.

14

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

Hudak, J. J. (2014). Presidential pork: White house influence over the distribution of federal
grants. Washington, DC: Brookings Institution Press.
Kang, W. C. (2014). Who gets what, when: Electoral cycles in pork barrel politics. Ph.D
Dissertation. New York University.
Keech, W. R., & Pak, K. (1989). Electoral cycles and budgetary growth in Veterans’
Benefit programs. American Journal of Political Science, 33(4), 901–911.
Kiewiet, D. R., & McCubbins, M. D. (1988). Presidential influence on congressional
appropriations decisions. American Journal of Political Science, 32(3), 713–736.
Lacy, D. (2009). Why do red states vote republican while blue states pay the bills? Paper presented at the American Political Science Association Annual Meeting, September
3–6, 2009.
Larcinese, V., Rizzo, L., & Testa, C. (2008). Allocating the US federal budget to the
states: The impact of the president. Journal of Politics, 68(02), 447–456.
Lauderdale, B. E. (2008). Pass the pork: Measuring legislator shares in congress. Political Analysis, 16(3), 235–249.
Lazarus, J. (2009). Party, electoral vulnerability, and earmarks in the U.S. house of
representatives. Journal of Politics, 71(3), 1050–1061.
Lee, F. E. (2000). Senate representation and coalition building in distributive politics.
American Political Science Review, 94(1), 59–72.
Levitt, S. D., & Snyder, J. M., Jr. (1995). Political parties and the distribution of federal
outlays. American Journal of Political Science, 39(4), 958–980.
Lindbeck, A., & Weibull, J. W. (1987). Balanced-budget redistribution as the outcome
of political competition. Public Choice, 52(3), 273–297.
Lowi, T. J. (1964). American business, public policy, case-studies, and political theory.
World Politics, 16(4), 677–715.
Maass, A. (1950). Muddy waters: The army engineers and the nation’s rivers. Cambridge,
England: Harvard University Press.
Mayer, K. R. (1995). Electoral cycles in federal government prime contract awards:
State-level evidence from the 1988 and 1992 presidential elections. American Journal of Political Science, 39(1), 162–185.
Mayhew, D. R. (1974). Congress: The electoral connection-Yale studies in political science
(pp. 26). New Haven, CT: Yale University Press.
McCarty, N. M. (2000). Presidential pork: Executive veto power and distributive politics. American Political Science Review, 94(1), 117–129.
Mebane, W. R., Jr., & Wawro, G. J. (2002). Presidential pork barrel politics. Unpublished
Manuscript: 1–33.
Nordhaus, W. D. (1975). The political business cycle. Review of Economic Studies, 42(2),
169–190.
Polsby, N. W. (1969). The growth of the seniority system in the US house of representatives. American Political Science Review, 63(3), 787–807.
Poole, K. T., & Rosenthal, H. (2007). Ideology and congress. New Brunswick, NJ: Transaction Publishers.
Roberts, B. E. (1990). A dead senator tells no lies: Seniority and the distribution of
federal benefits. American Journal of Political Science, 34(1), 31–58.
Shepsle, K. A., & Weingast, B. R. (1981). Political preferences for the pork barrel:
A generalization. American Journal of Political Science, 25(1), 96–111.

Distributive Politics: Federal Outlays

15

Sobel, R. S., & Ryan, M. E. (2012). Seniority and anti-competitive restrictions on the
legislative common pool: Tenure’s impact on the overall production of legislation
and the concentration of political benefits. Public Choice, 153, 171–190.
Stein, R. M., & Bickers, K. N. (1994). Congressional elections and the pork barrel.
Journal of Politics, 56(2), 377.
Stein, R. M., & Bickers, K. N. (1997). Perpetuating the pork barrel: Policy subsystems and
American democracy. Cambridge, England: Cambridge University Press.
Ting, M. M. (2012). Legislatures, bureaucracies, and distributive spending. American
Political Science Review, 106(2), 367–385.
Weingast, B. R., & Marshall, W. J. (1988). The industrial organization of congress;
or, why legislatures, like firms, are not organized as markets. Journal of Political
Economy, 96(1), 132–163.
Wright, G. (1974). The political economy of new deal spending: An econometric analysis. Review of Economics and Statistics, 56(1), 30–38.

SANFORD C. GORDON SHORT BIOGRAPHY
Sanford C. Gordon is Professor of Politics at New York University. He
received his BA from Cornell and his PhD from Princeton. Before his arrival
at NYU, Professor Gordon taught at the Ohio State University; in 2005–2006,
he was a fellow-in-residence at the Center for the Study of Democratic
Politics at Princeton. His research on the politicization of executive branch
agencies, corporate influence in regulation, the political economy of criminal
justice, electoral accountability, and research methodology has appeared in
the American Political Science Review, the American Journal of Political Science,
the Journal of Politics, the Quarterly Journal of Political Science, Political Analysis,
the Journal of Law, Economics, and Organization, and the New England Journal
of Medicine, among other places.
WOO CHANG KANG SHORT BIOGRAPHY
Woo Chang Kang is a PhD candidate in Politics at New York University.
Before coming to NYU, he received his BA in English Language and Literature and MA in Political Science at Korea University. Mr. Kang’s dissertation,
entitled “Electoral Cycles in Tactical Allocation,” examines the distribution
of government spending in the United States, South Korea, and Japan. His
research interests include distributive politics, elections, and voting behavior,
both in the United States and in comparative contexts.
RELATED ESSAYS
Understanding American Political Conservatism (Political Science), Joel D.
Aberbach

16

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

Political Ideologies (Political Science), Edward G. Carmines and Nicholas J.
D’Amico
Elites (Sociology), Johan S. G. Chu and Mark S. Mizruchi
Financialization of the US Economy (Sociology), Gerald (Jerry) F. Davis and
Suntae Kim
Presidential Power (Political Science), William G. Howell
Political Inequality (Sociology), Jeff Manza
Participatory Governance (Political Science), Stephanie L. McNulty and Brian
Wampler
Money in Politics (Political Science), Jeffrey Milyo
Why Do Governments Abuse Human Rights? (Political Science), Will H.
Moore and Ryan M. Welch
The Politics of Disaster Relief (Political Science), Alexander J. Oliver and
Andrew Reeves
The Future of Class Analyses in American Politics (Political Science), Jeffrey
M. Stonecash
The Institutional Logics Perspective (Sociology), Patricia H. Thornton et al.
Public Opinion, The 1%, and Income Redistribution (Sociology), David L.
Weakliem
Constitutionalism (Political Science), Keith E. Whittington



Distributive Politics: Federal Outlays
SANFORD C. GORDON and WOO CHANG KANG

Abstract
We provide an interpretive review of theoretical and empirical research on the distribution of federal outlays in the United States and their political antecedents. We
argue that the foundational scholarly research on the subject to date can be classified
to a great extent according to the perspectives it takes respecting legislative organization, political parties, and the executive branch. After reviewing theoretical and
empirical research in these areas, we discuss some cutting-edge research in this area
of research. One of the most exciting trends in the study of distributive politics is
a much greater focus on issues of causal identification in empirical tests of theories
concerning the allocation of federal resources. Finally, we describe five questions that
we believe ought to motivate future research in this area. These questions concern
the relationship between entitlement and discretionary spending; how to integrate
research on distributive politics with research on political business cycles; how best
to measure executive discretion in specific assistance programs; how to understand
the relationship between spending in an area and support for that spending; and how
to capture the mechanism relating spending and electoral benefits.

INTRODUCTION
In 2012, the federal government spent about $11,600 per citizen, or roughly
23% of the US gross domestic product. While the governments of many
developed countries spend more per capita [even once one accounts for
state and local expenditures, the United States still ranks 14th among OECD
countries (Central Intelligence Agency, 2011)], the sheer enormity of federal
outlays—totaling $3.6 trillion in 2012—makes their allocation a ripe subject
for systematic analysis.
To what extent are federal expenditures in the United States allocated on
the basis of political considerations? More specifically, what are those considerations, and how do US political institutions moderate or exacerbate their
application? This essay is a reflection on scholarly inquiry into these subjects. We first review some of the foundational research on the subject. We
argue that to a great extent, scholarly work can be classified according to the
perspective it takes on (i) legislative organization; (ii) political parties; and
Emerging Trends in the Social and Behavioral Sciences. Edited by Robert Scott and Stephen Kosslyn.
© 2015 John Wiley & Sons, Inc. ISBN 978-1-118-90077-2.

1

2

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

(iii) executive power. After reviewing some of the key findings in this literature, we proceed to a discussion of some cutting-edge areas of research. In
particular, we focus on how empirical scholars studying distributive politics
in the United States have become much more attuned to issues of research
design and causal identification. We conclude by identifying some key issues
that future scholars working in this area will need to grapple with.
Following Lowi’s (1964) famous typology, we take a signature feature of
distributive policies to be divisibility: namely, the ease with which policy
choices “can be disaggregated and dispensed unit by small unit” (p. 690).
Accordingly, in what follows we will focus on the political economy of discretionary domestic spending rather than entitlement spending; whether the
discretion lies with Congress or the executive branch will be a critical issue
in our discussion.
A BRIEF PRIMER ON FEDERAL SPENDING
It is important to define the object under scrutiny before proceeding to a
discussion of how it has been studied. A full consideration of government
spending in the United States is far beyond the scope of this short essay.
However, understanding how expenditures are allocated requires some
basic knowledge about the federal budget. At the most basic level of
aggregation, federal expenditures may be divided into three categories:
discretionary expenditures, mandatory expenditures, and interest payments
on the national debt. Mandatory expenditures are programs such as Social
Security and Medicare, the funding for which is provided for by standing law
and not subject to the recurrent budgeting process. Discretionary spending,
which itself may be divided into defense and nonrelated areas, is decided
on during the annual appropriation process. Table 1 displays amounts (in
billions of nominal dollars) associated with these major categories.
In addition to funding the operation of government agencies, a large fraction of the discretionary budget goes to fund federal domestic assistance
programs; these include grants, direct payments, insurance programs, loans
Table 1
Federal Spending by Category, Fiscal Year 2012
Category
Nondefense discretionary
Defense-related discretionary
Mandatory
Net interest

Amount ($Billions)
615.6
670.5
2134.2
220.4

Source: Office of Management and Budget, Historical Table 8.1.

Distributive Politics: Federal Outlays

3

and loan guarantees. The recipients of these benefits might be private individuals, companies, or institutions (e.g., universities and other nonprofits);
alternatively, the recipients may be states or local governments. $155.7 billion
went to nonmilitary contracts in 2012, and $361.3 billion to military contracts
(source: usaspending.gov).
A critical feature of these categories of spending is the delegation of authority to agencies in the executive branch. Federal agencies differ markedly in
the degree of effective discretion they possess in disbursing funds—whether
to individuals and private organizations or to state and local governments.
For certain programs, Congress specifies the amounts and recipients in statutory formulae—witness, for example, the formulas governing price supports
for agricultural commodities. For others, Congress grants a much greater
degree of discretion to the agency administering the program: grants from
the National Institutes of Health; mortgage insurance from the Federal Housing Administration; or Homeland Security grants to states and localities, for
example. Finally, “earmarks” are directives from Congress to an executive
branch agency to spend its nominally discretionary budget in specific ways.
While some such directives lie in statutes, legislators may place others in
committee or conference reports.
FOUNDATIONAL IDEAS AND RESEARCH
LEGISLATIVE INSTITUTIONS
As noted in the introduction to this essay, much of the early research on distributive politics in the United States is Congress-centered (not surprising
given Congress’s spending power as defined in Article I, Section 8 of the Constitution), and focused in particular on the historically decentralized nature
of the legislative process in the United States.
The modern study of distributive politics in the United States can be traced
to progress in the institutional analysis of Congress in the 1960s. Fenno’s
landmark (1966) study of Congressional Appropriations, The Power of the
Purse, put the focus of studies of legislative behavior squarely on Congress’s
spending power and in particular on the House Appropriations Committee.
Polsby (1968) described the emergence of universalistic norms in the House
of Representatives: most critically, how the discretion of the Speaker of the
House in selecting committee chairs had been supplanted by reliance on
a seniority norm. Mayhew (1974) argued for the importance of the committee system in allowing reelection-minded legislators to take positions
on, and claim credit for, particularistic policies benefiting their respective
constituencies. In addition, Ferejohn’s (1974) exhaustive study of federally
funded local water projects related the allocation of those projects to the

4

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

institutional position of legislators and their resultant influence over the
Army Corps of Engineers. A series of rational choice accounts of legislative
institutions have placed the substantive intuitions of these earlier works
on firm analytical foundations (e.g., Shepsle & Weingast, 1981; Weingast &
Marshall, 1988).
Three clusters of hypotheses emerge from this theoretical literature. First,
we should not expect federal largesse to be concentrated in a minimum winning coalition. This is overwhelmingly confirmed in the data.
Second, we should see a relationship between pork barrel spending in
a legislator’s district or state and his or her positions in key committee
posts and/or seniority. The empirical record on the committee membership
hypothesis is mixed. Recent research confirming the importance of committee positions for at least some committees includes work by Alvarez
and Saving (1997); Engstrom and Vanberg (2010); and Lauderdale (2008).
Research failing to confirm the committee membership hypothesis includes
work by Evans (2004). With regard to the seniority hypothesis, two mechanisms may be at play: access to committee leadership (Roberts, 1990; but
see Arnold, 1979), and increased skill in the legislative process (Engstrom
& Vanberg, 2010; Sobel & Ryan, 2012; but see Anzia & Berry, 2011; and
Lauderdale, 2008).
Third, to the extent that legislators can devote only so much effort to bringing home the bacon, more legislators per citizen in a state or district (brought
about either by fixed representation of states in the senate or malapportionment in the House) should give rise to greater expenditures in that area (e.g.,
Lee, 2000).
POLITICAL PARTIES
The literature referenced above tends to deemphasize the role of party.
However, to the extent that the majority enjoys controls over the legislative
agenda, this advantage should engender advantages in the allocation of federal resources. A number of studies present evidence that voters represented
by members of a majority party receive more benefits than those represented
by members of a minority party (e.g., Albouy, 2013; Balla, Lawrence, &
Maltzman, & Sigelman, 2002; Berry, Burden, & Howell, 2010).
A related body of theoretical research, less focused on the United States
specifically, seeks to make general claims about how strong parties might target government transfers for political benefit. In the model described by Cox
and McCubbins (1986), a risk-averse incumbent party has incentives to direct
transfers primarily toward core supporters, in order to maintain its existing
electoral coalition. By contrast, Lindbeck and Weibull (1987) describe a model
of two-party competition in which, in equilibrium, both parties primarily target marginal constituencies (in their setup, middle-income voters) rather than

Distributive Politics: Federal Outlays

5

core supporters (low-income voters for one party and high income voters
for the other). In the more general setup of Dixit and Londregan (1995, 1996),
parties tradeoff between the electoral benefits of targeting pivotal constituencies and of satisfying core groups of supporters. Asymmetries between the
parties in the efficiency with which they can target certain groups push the
parties away from targeting marginal and toward core voters.
Both the core and swing voter models can claim their share of empirical
support. Evidence for core voter models often overlaps with evidence for
the majority party advantage because the distribution of core voters is measured with levels of support in the previous election, which is correlated with
majority party status (Balla et al., 2002; Levitt & Snyder, 1995; see also Dynes &
Huber, 2013). On the other hand, Bickers and Stein (1996) suggest that incumbents elected in open seat races receive more federal spending. Other scholars
have found evidence that at least some subsets of incumbents who won with
close margins tend to bring home more federal resources to their districts
(e.g., Berry et al., 2010; Lazarus, 2009; Stein & Bickers, 1994).
THE EXECUTIVE BRANCH
Most recently, scholars have begun to include the executive branch more
explicitly in theoretical models of distributive politics. Kiewiet and McCubbins (1988) argue that a presidential veto threat is only credible in changing
spending on distributive programs if the president prefers less spending than
the legislature. McCarty (2000) develops a model in which the presidential
veto can reduce spending unless the president himself has strong preferences over
its distribution. To the extent that the president’s electoral prospects depend
on delivering benefits to (core or marginal) constituents, he may have good
reason to behave in the ways predicted by the models described above.
To the extent that a president’s goals are distributive, politicization might
thus facilitate presidential influence in the allocation of discretionary federal expenditures. What, then, should that influence look like? A president is
likely to benefit from directing resources to copartisan legislators and other
officials, either in exchange for their support on legislative initiatives or to
enhance their electoral prospects. Berry et al. (2010), for example, demonstrate that legislative districts and counties represented by copartisan legislators of presidents receive significantly more federal spending.
The president seeks reelection and to preserve his party’s control of the
White House. Therefore, he and his subordinates have an incentive to allocate federal resources toward that end. Demonstrating what this means in
practice has led to mixed results empirically. Studies on New Deal spending have found that states with more voting variability received more federal employment and agricultural aid (e.g., Wright, 1974). Hudak (2014) also

6

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

shows that presidents direct more discretionary grants to swing states, especially in advance of presidential elections. In general, however, the strength
of empirical support for the swing voter model is weak in studies on the distribution of federal resources. Both Larcinese, Rizzo, and Testa (2008) and
Berry et al. (2010) find that swing states do not receive more federal outlays.
On the other hand, these studies find a positive and significant association
between the levels of support for the president and the allocation of federal
resources at the state level.
In his seminal work on congressional-bureaucratic relations, Arnold (1979)
notes that “most decisions about geographic allocations are bureaucratic
decisions” (p. 8). Despite this authority, bureaucrats have an incentive
to consider preferences of other political actors to accomplish their own
goals (see also Maass, 1950). For example, bureaucrats have an incentive to
consider preferences of members of Congress because attaining budgetary
stability and growth depends on their decisions. Similarly, Stein and Bickers
(1997) claim that concerns about budgets induce bureaucrats to help not
only legislators but also constituents and interest groups. Bureaucrats also
have an incentive to respond to the president’s political needs to obtain
presidential support (Hudak, 2014).
Obtaining empirical evidence on whether, and to what extent, bureaucrats
allocate federal resources in response to the demands of other political officials is challenging, both because those potential influences are many and
because it is difficult to tell the degree to which apparently politicized allocations reflect discretionary bureaucratic actions as opposed to being hardwired into enabling statutes or directed from the White House. With this in
mind, Bertelli and Grose (2009) find that a greater degree of ideological congruence between a state’s senators and the Secretaries of Defense and Labor
is associated with greater discretionary allocations (in the form of contracts
and grants) to the senators’ respective states. The congruence of the senators
and the president has no apparent effect, however. They interpret these findings as implying an autonomous sphere of bureaucratic influence. Gordon
(2011) demonstrates a linkage between White House directives to political
appointees at the General Services Administration and the allocation of subsequent building contracts.
AREAS OF CUTTING-EDGE RESEARCH
We now turn to some recent developments in the study of the distributive
politics of federal outlays. In our estimation the key development in the
literature over the past decade or so has been a greater focus on causal identification. In an ideal world, we could test our hypotheses about distributive
politics by randomly assigning political significance across districts and

Distributive Politics: Federal Outlays

7

observing differences in spending outcomes. In the real world, of course,
such experimental manipulations are impossible. For an observed relationship between spending in a district or state and the hypothesized measure of
its political significance to be interpreted as a causal, it must be the case that,
the status of a district as important or unimportant is as good as randomly
assigned once one conditions on other observable characteristics of districts.
Otherwise, it could be the case that the measure of influence is proxying the
effect of some unobservable, politically neutral feature of the district.
For example, suppose we observe that districts whose legislators sit on
the Appropriations subcommittee overseeing the Department of Housing
and Urban Development receive a disproportionate share of housing grants.
Is this bureaucratic deference to those legislators? Possibly, but it could
simply reflect the member’s self-selection onto that committee owing to the
large number of eligible beneficiaries in her district. The analyst, hoping to
eliminate this possibility, controls for an observable measure of district-level
demand: for example, percent urban in the district. How sure can we be that
this fully neutralizes the bias? Studies of the effects of malapportionment
suffer from a similar issue: perhaps greater per capita spending in Alaska
and Montana is a consequence of diseconomies of scale in providing comparable levels of services to widely dispersed populations (Elis, Malhotra,
& Meredith, 2009), and not the greater effort or pivotality of their senators.
Unfortunately, nearly all of the empirical studies of distributive politics
described in the previous section are susceptible to this criticism.
Because scholars cannot randomly assign political significance as in an
experiment, scholars have recently turned to innovative research designs to
minimize the ubiquitous threat of confounding. One type of design exploits
the panel structure of expenditure data and examines repeated observations
of the same districts. The presence of multiple observations on the same
cross-sectional unit (district or state) allows the analyst to implicitly control
for all observable and unobservable time invariant features of congressional
districts over their lifespans, as well as year-to-year variation in spending
levels. The basic idea is that rather than compare spending in districts that
differ in their observable political characteristics (i.e., “between” district
variation), the analyst examines the effect of changes in the political status of
specific districts on spending (“within” district variation), and then averages
across the many districts that observe a change. This is the approach taken
in a recent papers by Berry et al. (2010), who demonstrate a relationship
between spending in a district and whether the district is represented by a
copartisan of the president; and by Anzia and Berry (2011), who demonstrate
that female legislators are more successful at securing federal expenditures
in their districts than their male counterparts.

8

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

Second, regression discontinuity designs exploit the fact that the assignment of the “treatment” of political importance may be a consequence of
whether some continuous measure falls above or below some exogenously
given threshold; the analyst then compares the outcomes of observations
falling on both sides of the threshold, under the assumption that in that
vicinity, treatment assignment is effectively random. Albouy (2013) takes
this approach in estimating the effect on various categories of spending of
having a Republican or Democratic legislator, focusing on candidates who
won elections by small margins (i.e., those close to the 50% threshold). Anzia
and Berry (2011) also perform a regression discontinuity analysis to isolate
the effect of female legislators.
A third approach is to exploit some natural experiment provided by a plausibly exogenous shock to district-level circumstances, often in conjunction
with a panel data structure. Elis, Malhotra, and Meredith (2009), for example,
employ the discrete changes in representation created by the decennial census reapportionment process to estimate the effect of changes in congressional delegation size on federal outlays. In addition, Gordon (2011) exploits
a minor scandal occurring during the presidency of George W. Bush: a White
House deputy was found to have given a presentation to political appointees
at the General Services Administration mentioning priority congressional
districts for the upcoming election. By comparing GSA spending in the mentioned and unmentioned districts in the weeks immediately before and following the briefing, Gordon recovers an estimate of the effect of a White
House directive to bureaucrats on subsequent spending.
KEY ISSUES GOING FORWARD
We conclude by identifying five key issues in the study of the politics of federal outlays.
HOW DO THE POLITICS OF DISCRETIONARY EXPENDITURES CHANGE WHEN A DECREASING
FRACTION OF THE FEDERAL BUDGET IS DISCRETIONARY?
Two facts—an aging population and dramatic increases in health care
costs—have led to an explosion in the fraction of federal expenditures
devoted to entitlements. Figure 1 displays this trend graphically: whereas
in 1900, less than 1% of the budget was devoted to welfare, pensions, or
healthcare. By 2012 that number had spiked to nearly 60%.
In considering this picture, it is important to point out that discretionary
spending as a fraction of GDP has remain largely unchanged. However, the
fact that it constitutes a decreasing fraction of overall expenditures means
the role of discretionary expenditures in American politics may change. As

9

0.5
0.4
0.3
0.2
0.1
0.0

Proportion entitlements

0.6

Distributive Politics: Federal Outlays

1900

1920

1940

1960

1980

2000

Year

Figure 1 The fraction of federal expenditures devoted to entitlement spending
has increased dramatically in the last century. Source: Budget data compiled by
usgovernmentspending.com

documented by Poole and Rosenthal (2007), in the past 30–40 years it has
become increasingly easy to describe political conflict within Congress as
occurring on a single “left-right” dimension, where that dimension captures
partisan conflict or disagreement about redistributive policies, for example,
entitlements. Moreover, the ideological distance between the two parties
has increased over this period. Historically, pork barrel spending provided
a means to “grease the wheels” of legislative bargaining by permitting
congressional leaders to buy off reluctant coalition members (Evans, 2004).
As legislative gridlock over budgetary and entitlement reform appears
increasingly inexorable, however, the amount of distributive largesse
required to overcome a legislator’s ideological inclinations on the main
issues of the day is likely to increase. If reform is sufficiently important, this
could increase in the importance of pork in fashioning coalitions; however,
if the cost of pork is prohibitive, that could conceivably produce a decrease
in its importance.
DO THE POLITICAL INCENTIVES GOVERNING THE ALLOCATION OF FEDERAL EXPENDITURES
VARY DYNAMICALLY?
As Harold Lasswell famously argued in 1936, politics is about “who gets
what, when, and how”. The studies on distributive politics cited here
generally focuses on the “what.” Starting with Nordhaus (1975), a related
literature on political businesses cycles (PBCs); addresses the “when,” focusing on temporal variation in aggregate government expenditures before and

10

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

after election. Both literatures have a common interest in understanding how
an incumbent might use government resources for electoral ends. Owing to
the independent development of each literature, however, the possibility of
dynamic changes in the allocation of discretionary spending has received
little attention so far.
Keech and Pak (1989) and Mayer (1995) examine whether Veteran’s transfers or military contracts increase in the preelectoral period respectively.
Keech and Pak (1989) fail to find confirmatory evidence for budget expansion On the other hand, Mayer (1995) shows that the number of civilian
agency contract awards rapidly increases just before presidential elections
and primaries.
Kang (2014), on the other hand, examines the possibility that main beneficiaries of government spending vary with electoral cycles. An opportunistic
incumbent often needs support of swing voters to win an election, which
provides him with an incentive to target these voters before elections. Once
winning an election, however, he has various motivations to cater to his core
constituents.
IS THERE A PRINCIPLED WAY TO QUANTIFY EXECUTIVE DISCRETION OVER EXPENDITURES?
As of May 2013, the Catalog of Federal Domestic Assistance describes
2233 federal assistance programs. As noted above, the amount of effective
discretion that presidents possess in administering these programs varies
markedly. Levitt and Snyder (1995) argue that programs with geographically concentrated benefits are more easily targeted than ones with diffuse
benefits: they calculate coefficients of variation for each program using
district averages, and classify programs as having low or high variation.
The former include entitlement programs such as Social Security, Medicare,
and Veterans’ Benefits. The latter include the lion’s share of discretionary
programs. Anzia and Berry (2011) and Berry et al. (2010) also use the
coefficient of variation methodology.
This approach represents an important first step, but has important limitations. While the allocation of entitlement spending has distributive consequences, it is difficult to argue that in the short to medium term that its
allocation reflects tactical political choices of entrepreneurial politicians (as
opposed to, e.g., demographic changes). Moreover, this approach does not
capture institutional constraints on discretion: as noted above, an agricultural subsidy program may have a high coefficient of variation, but it does
not afford the kind of discretion to the Department of Agriculture that, for
example, the ability to lease, operate, and renovate buildings affords to the
General Services Administration.

Distributive Politics: Federal Outlays

11

Codifying the residual discretion for individual assistance programs in
a systematic manner that takes account of, inter alia, statutory constraints
on discretion, the degree to which the program is insulated via staffing
or position in the bureaucratic hierarchy from overhead political control,
promises to be an enormous undertaking. However, it is one well worth
pursuing. Recent theoretical research (Ting, 2012) describes conditions
under which Congress will delegate funding choices to professionalized
bureaucrats rather than bargain directly over distribution. To the extent that
political scientists wish to subject his argument to care empirical scrutiny,
or, more broadly, to refine arguments about which government expenditures
are most likely to respond to the various political stimuli described above,
this undertaking would seem to represent a critical step.
WHY ARE STATES THAT ARE NET RECIPIENTS OF FEDERAL LARGESSE THE MOST HOSTILE TO
GOVERNMENT SPENDING?
Lacy (2009) observes an intriguing statistical relationship: states that vote
Republican in presidential elections, and thus might be expected to have the
greatest hostility to federal spending, tend to be net recipients of federal dollars, whereas states that tend to vote Democratic tend to be net providers;
moreover, this relationship appears to be increasing over time. What is the
relationship between distributive and ideological politics? Endeavoring to
answer this question would be a fruitful line of future research.
CAN OUR EMPIRICAL ANALYSIS BETTER FLESH OUT THE MECHANISM RELATING POLITICAL
EXPENDITURE WITH ELECTORAL BENEFIT?
Cox (2009) notes the importance of distinguishing between distributive
spending on swing groups of voters on the one hand and swing districts on the
other. Suppose, for example, that a party wanted to maximize its majority in
the legislature. Such a party would do well to target expenditures toward
swing districts. However, within those districts, whether that party should
target swing or core voters depends on the mechanism relating political
expenditures with electoral benefits. The “swing voter model” cited above
relies on the idea that pivotal voters will be persuaded by distributive
largesse to support the candidate credited with providing it. By contrast, the
“core voter” model relies on a logic of mobilization (getting the base to turn
out and contribute to a campaign) and coordination (reducing the number
of internecine competitors, e.g., primary challengers).
If persuasion within a district is more critical than persuasion, then we
would expect to see swing voters in swing districts targeted; if, on the other
hand, mobilization and coordination were critical, we would expect to see

12

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

core voters in swing districts targeted. In the absence of microlevel data on
individual recipients of federal assistance and their voting histories, one
possible approach to pinning down the preponderant mechanism would be
to disaggregate spending by discretionary program where the ideological
affinities of groups of voters are well known. For example, do we see a
disproportionate number of small business loans flowing to swing districts
under Republican administrations?
Even if the core-voter-in-swing-district hypothesis receives support, this is
still not the end of the story. It is possible, for example, that legislators seek
pork barrel spending in their districts to claim credit among voters (Mayhew, 1974). However, it is also possible, if not likely, that spending is a way
of mobilizing elites within a district to support an incumbent. To our knowledge, to date only Mebane and Wawro (2002) have investigated this specific question, although Gordon (2011) argues that the elite mechanism is
more plausible for most GSA public building contracts, owing to their low
visibility.
CONCLUSION
In this essay, we have endeavored to describe the past, present, and, it is to be
hoped, future of empirical and theoretical research on the distribution of federal outlays. In contextualizing past, foundational research, it is helpful to see
this research as echoing advances in the study of US political institutions. Earlier research has focused in particular on the relationship between legislative
organization and distribution, while more recent research has revitalized the
critical role played by political parties, the president, and the bureaucracy.
We have argued that some of the most exciting contemporary research on
distributive politics in the United States is centrally concerned with employing innovative research designs to improve causal identification. In addition, lastly, we have identified a number of useful unanswered questions that
future scholars working in the area might seek to answer. These questions
concern the relationship between entitlement and discretionary spending;
how to integrate research on distributive politics with research on political
business cycles; how best to measure executive discretion in specific assistance programs; how to understand the relationship between spending in
an area and support for that spending; and how to capture the mechanism
relating spending and electoral benefits.
Of course, this list of topics is necessarily incomplete. We hope that scholars
working in the area will expand it, and in so doing, match the creativity associated with asking new questions about the subject with intellectual rigor in
answering them.

Distributive Politics: Federal Outlays

13

REFERENCES
Albouy, D. (2013). Partisan representation in congress and the geographic distribution of federal funds. Review of Economics and Statistics, 95(1), 127–141.
Alvarez, R. M., & Saving, J. L. (1997). Congressional committees and the political
economy of federal outlays. Public Choice, 92(1–2), 55–73.
Anzia, S. F., & Berry, C. R. (2011). The Jackie (and Jill) Robinson effect: Why do congresswomen outperform congressmen? American Journal of Political Science, 55(3),
478–493.
Arnold, R. D. (1979). Congress and the bureaucracy: A theory of influence. New Haven,
CT: Yale University Press.
Balla, S. J., Lawrence, E. D., Maltzman, F., & Sigelman, L. (2002). Partisanship, blame
avoidance, and the distribution of legislative pork. American Journal of Political Science, 46(3), 515–525.
Berry, C. R., Burden, B. C., & Howell, W. G. (2010). The president and the distribution
of federal spending. American Political Science Review, 104(4), 783–799.
Bertelli, A. M., & Grose, C. R. (2009). Secretaries of pork? A new theory of distributive
public policy. Journal of Politics, 71(3), 926–945.
Bickers, K. N., & Stein, R. M. (1996). The electoral dynamics of the federal pork barrel.
American Journal of Political Science, 40(4), 1300–1326.
Central Intelligence Agency (2011). The world factbook.
Cox, G. W. (2009). Swing voters, core voters and distributive politics. In I. Shapiro,
S. C. Stokes, E. Wood & A. S. Krishner (Eds.), Political representation. Cambridge,
England: Cambridge University Press.
Cox, G. W., & McCubbins, M. D. (1986). Electoral politics as a redistributive game.
Journal of Politics, 48(2), 370–389.
Dixit, A., & Londregan, J. (1995). Redistributive politics and economic efficiency. The
American Political Science Review:, 89, 856–866.
Dixit, A., & Londregan, J. (1996). The determinants of success of special interests in
redistributive politics. Journal of Politics, 58(4), 1132–1155.
Dynes, A., and Huber, G. A. (2013). Partisanship and the allocation of federal spending:
Do same-party legislators or voters benefit from shared party affiliation? Typescript, Yale
University.
Elis, R., Malhotra, N., & Meredith, M. (2009). Apportionment cycles as natural experiments. Political Analysis, 17(4), 358–376.
Engstrom, E. J., & Vanberg, G. (2010). Assessing the allocation of pork: Evidence from
congressional earmarks. American Politics Research, 38(6), 959–985.
Evans, D. (2004). Greasing the wheels: Using pork barrel projects to build majority coalitions. New York, NY: Cambridge University Press.
Fenno, R. F. (1966). The power of the purse: Appropriations politics in congress. Boston,
MA: Little Brown & Company.
Ferejohn, J. A. (1974). Pork barrel politics: Rivers and harbors legislation, 1947–1968. Stanford, CA: Stanford University Press.
Gordon, S. C. (2011). Politicizing agency spending authority: Lessons from a Bush-era
scandal. American Political Science Review, 105(4), 717–734.

14

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

Hudak, J. J. (2014). Presidential pork: White house influence over the distribution of federal
grants. Washington, DC: Brookings Institution Press.
Kang, W. C. (2014). Who gets what, when: Electoral cycles in pork barrel politics. Ph.D
Dissertation. New York University.
Keech, W. R., & Pak, K. (1989). Electoral cycles and budgetary growth in Veterans’
Benefit programs. American Journal of Political Science, 33(4), 901–911.
Kiewiet, D. R., & McCubbins, M. D. (1988). Presidential influence on congressional
appropriations decisions. American Journal of Political Science, 32(3), 713–736.
Lacy, D. (2009). Why do red states vote republican while blue states pay the bills? Paper presented at the American Political Science Association Annual Meeting, September
3–6, 2009.
Larcinese, V., Rizzo, L., & Testa, C. (2008). Allocating the US federal budget to the
states: The impact of the president. Journal of Politics, 68(02), 447–456.
Lauderdale, B. E. (2008). Pass the pork: Measuring legislator shares in congress. Political Analysis, 16(3), 235–249.
Lazarus, J. (2009). Party, electoral vulnerability, and earmarks in the U.S. house of
representatives. Journal of Politics, 71(3), 1050–1061.
Lee, F. E. (2000). Senate representation and coalition building in distributive politics.
American Political Science Review, 94(1), 59–72.
Levitt, S. D., & Snyder, J. M., Jr. (1995). Political parties and the distribution of federal
outlays. American Journal of Political Science, 39(4), 958–980.
Lindbeck, A., & Weibull, J. W. (1987). Balanced-budget redistribution as the outcome
of political competition. Public Choice, 52(3), 273–297.
Lowi, T. J. (1964). American business, public policy, case-studies, and political theory.
World Politics, 16(4), 677–715.
Maass, A. (1950). Muddy waters: The army engineers and the nation’s rivers. Cambridge,
England: Harvard University Press.
Mayer, K. R. (1995). Electoral cycles in federal government prime contract awards:
State-level evidence from the 1988 and 1992 presidential elections. American Journal of Political Science, 39(1), 162–185.
Mayhew, D. R. (1974). Congress: The electoral connection-Yale studies in political science
(pp. 26). New Haven, CT: Yale University Press.
McCarty, N. M. (2000). Presidential pork: Executive veto power and distributive politics. American Political Science Review, 94(1), 117–129.
Mebane, W. R., Jr., & Wawro, G. J. (2002). Presidential pork barrel politics. Unpublished
Manuscript: 1–33.
Nordhaus, W. D. (1975). The political business cycle. Review of Economic Studies, 42(2),
169–190.
Polsby, N. W. (1969). The growth of the seniority system in the US house of representatives. American Political Science Review, 63(3), 787–807.
Poole, K. T., & Rosenthal, H. (2007). Ideology and congress. New Brunswick, NJ: Transaction Publishers.
Roberts, B. E. (1990). A dead senator tells no lies: Seniority and the distribution of
federal benefits. American Journal of Political Science, 34(1), 31–58.
Shepsle, K. A., & Weingast, B. R. (1981). Political preferences for the pork barrel:
A generalization. American Journal of Political Science, 25(1), 96–111.

Distributive Politics: Federal Outlays

15

Sobel, R. S., & Ryan, M. E. (2012). Seniority and anti-competitive restrictions on the
legislative common pool: Tenure’s impact on the overall production of legislation
and the concentration of political benefits. Public Choice, 153, 171–190.
Stein, R. M., & Bickers, K. N. (1994). Congressional elections and the pork barrel.
Journal of Politics, 56(2), 377.
Stein, R. M., & Bickers, K. N. (1997). Perpetuating the pork barrel: Policy subsystems and
American democracy. Cambridge, England: Cambridge University Press.
Ting, M. M. (2012). Legislatures, bureaucracies, and distributive spending. American
Political Science Review, 106(2), 367–385.
Weingast, B. R., & Marshall, W. J. (1988). The industrial organization of congress;
or, why legislatures, like firms, are not organized as markets. Journal of Political
Economy, 96(1), 132–163.
Wright, G. (1974). The political economy of new deal spending: An econometric analysis. Review of Economics and Statistics, 56(1), 30–38.

SANFORD C. GORDON SHORT BIOGRAPHY
Sanford C. Gordon is Professor of Politics at New York University. He
received his BA from Cornell and his PhD from Princeton. Before his arrival
at NYU, Professor Gordon taught at the Ohio State University; in 2005–2006,
he was a fellow-in-residence at the Center for the Study of Democratic
Politics at Princeton. His research on the politicization of executive branch
agencies, corporate influence in regulation, the political economy of criminal
justice, electoral accountability, and research methodology has appeared in
the American Political Science Review, the American Journal of Political Science,
the Journal of Politics, the Quarterly Journal of Political Science, Political Analysis,
the Journal of Law, Economics, and Organization, and the New England Journal
of Medicine, among other places.
WOO CHANG KANG SHORT BIOGRAPHY
Woo Chang Kang is a PhD candidate in Politics at New York University.
Before coming to NYU, he received his BA in English Language and Literature and MA in Political Science at Korea University. Mr. Kang’s dissertation,
entitled “Electoral Cycles in Tactical Allocation,” examines the distribution
of government spending in the United States, South Korea, and Japan. His
research interests include distributive politics, elections, and voting behavior,
both in the United States and in comparative contexts.
RELATED ESSAYS
Understanding American Political Conservatism (Political Science), Joel D.
Aberbach

16

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

Political Ideologies (Political Science), Edward G. Carmines and Nicholas J.
D’Amico
Elites (Sociology), Johan S. G. Chu and Mark S. Mizruchi
Financialization of the US Economy (Sociology), Gerald (Jerry) F. Davis and
Suntae Kim
Presidential Power (Political Science), William G. Howell
Political Inequality (Sociology), Jeff Manza
Participatory Governance (Political Science), Stephanie L. McNulty and Brian
Wampler
Money in Politics (Political Science), Jeffrey Milyo
Why Do Governments Abuse Human Rights? (Political Science), Will H.
Moore and Ryan M. Welch
The Politics of Disaster Relief (Political Science), Alexander J. Oliver and
Andrew Reeves
The Future of Class Analyses in American Politics (Political Science), Jeffrey
M. Stonecash
The Institutional Logics Perspective (Sociology), Patricia H. Thornton et al.
Public Opinion, The 1%, and Income Redistribution (Sociology), David L.
Weakliem
Constitutionalism (Political Science), Keith E. Whittington