Congressional Dysfunction and the Decline of Problem Solving

Congressional Dysfunction and the Decline of Problem SolvingJonathan Lewallen, University of Tampa
Sean M. Theriault, University of Texas at Austin
Bryan D. Jones, University of Texas at Austin
Working paper. Please do not cite without authors’ permission.
[email protected]
ABSTRACT
The U.S. Congress has been unable to solve problems both pressing and recurring on a whole
range of issues. Observers have been quick to point to increasing polarization as the culprit. Yet
there is nothing about polarization itself that suggests the kinds of breakdowns in problem solving
that we have seen, let alone government shutdowns. We study a second potential explanation for
congressional dysfunction: a decline in the committee system’s information processing capacity.
In this paper we explore how committees take in and translate policy information through
hearings. In analyzing data from 1971 to 2010, we show that hearings have become more one-
sided. Furthermore, committees are devoting less time to developing solutions to public policy
problems. Information processing has changed more rapidly on some issues than others, and
moreover the clusters of issues that have seen the most consistent changes to committee
information processing do not fall along familiar partisan divisions. Our findings help point the
way towards better understanding where Congress and its committee system have become
dysfunctional in order to suggest appropriate paths forward.
In December 2015, the typically cautious Senate Majority Leader Mitch McConnell (R-
Ky.) made a bold statement: “Dysfunction is over” (quoted in Lesniewski and Bowman 2015). To
support his claim, McConnell pointed to a bipartisan education reauthorization bill, a budget
resolution, and a large number of votes on amendments as signs that Congress had turned a corner
to become a more open and productive institution.
In the months that followed McConnell’s optimistic assertion, the Senate found itself
mired in gridlock over an international nuclear research agreement with Iran and federal
assistance for Flint, Michigan (Lesniewski 2016; Henry 2016). Furthermore, Senate Republicans
refused to schedule a hearing on Merrick Garland’s nomination to the Supreme Court; when it
was officially withdrawn, it had set the record for the longest delay on a nomination to the high
court. The appropriations process continues to rely on short-term, stop-gap funding for
government agencies with then-President Elect Trump supporting another stopgap through the
end of March 2017 (Wasson 2016), while Congress “missed its chance” to adequately address the
Zika virus’s spread into the United States according to some public health experts by not acting
prior to the 2016 August recess (Ordoñez 2016).
Many observers view the above examples and others like them as symptoms of a partisan
divide; we view them as information problems rooted in the committee system. By learning about
certain policy areas and becoming key information sources for their colleagues, committees
contribute to informed debate and overcome many of the coordination and collective action
problems inherent in a large legislature. That such problems have not only returned but
overwhelmed Congress in recent years suggests that committee information processing has
weakened.
2
In this paper we examine how Congress processes information in its committee system as
a means to better understand its current institutional dysfunction. By utilizing a new coding
scheme for committee hearings from 1971 to 2010, we document and analyze three pervasive
trends. First, committees are calling fewer witnesses. Second, the hearings themselves have
become less focused on learning about proposed policy solutions under consideration. Third,
hearings particularly in the House have become more one-sided in the information they acquire
and less reliant on multiple perspectives. These trends in turn contribute to a dysfunctional
lawmaking process; namely, governance via brinkmanship and running up against deadlines.
The paper proceeds in three steps. First, we identify what we see as indicators of a
dysfunctional Congress and existing possible explanations for such dysfunction, and we propose
an additional explanation rooted in changes to congressional information processing. Second, we
describe how we code hearings over time to detect changes to committee information processing
and present data on close to 22,000 committee hearings from 1971 to 2010. We further find that
the committee breakdown in information processing does not affect all issues equally; rather some
issue areas have experienced rapid declines in obtaining “good” information while other issue
areas are much less affected. The final section concludes with some ideas for future investigation
into the issue dynamics of dysfunction.
The Information Dimension to Congressional Dysfunction
Congressional approval is at an all-time low. Citizens, journalists, current and former
members of Congress, and political scientists alike lament the increase in party polarization, the
decline of comity and bipartisanship, the resulting fall-off in legislative productivity, and the
decline of trust in government among voters (APSA 2013; Bipartisan Policy Center 2014;
3
Eldridge 2014; Mann and Ornstein 2012). We aim to uncover an additional dimension of
dysfunction by beginning with the fact that agenda space in Congress is scarce; too many
proposals compete for the institution's limited attention. Managing this competition by privileging
reauthorization legislation, Congress, by requiring a review of programs every few years and
tying these schedules to program funding changes, ensures that different issues receive attention
by imposing precipitous costs on itself for inaction (Adler and Wilkerson 2012; Cox 2004; Hall
2004). Yet the last several years have been filled with examples of Congress rushing at the last
minute to address recurring issues and to solve long-standing public problems, and not always
succeeding. The federal debt ceiling disagreement that precipitated the 2013 shutdown went
through a series of short-term fixes often lasting just a few months at a time before Congress
finally enacted a bill suspending the debt ceiling for one year. Federal programs have repeatedly
been funded through short-term continuing resolutions, forcing Congress to continually avert
additional government shutdowns.
Other examples abound. Funding for multiple agriculture programs expired for several
weeks in 2012, creating fears that policy would revert to a 1949 law and milk prices would surge
to $8 a gallon (Nixon 2012). Congress eventually cleared a series of short-term extensions and
finally adopted a longer-term measure two years later. The Federal Aviation Administration's
authorization expired in 2007 and was extended through a series of short-term measures until it
finally lapsed due to congressional inaction in July 2011. The Elementary and Secondary
Education Act was last reauthorized in 2002 as the No Child Left Behind Act and was scheduled
to expire in 2007, but no reauthorization bill emerged prior to 2015 (a reauthorization bill
eventually became law in 2016).
4
In each of these cases, the legislative process broke down even where it was specifically
designed to overcome partisan and ideological barriers to problem solving. It would be one thing
if committees had actively decided to end these programs, but that is not what is happening. We
thus need a definition of dysfunction that transcends the institutional gridlock built into the
political system. When we refer to congressional dysfunction, we primarily are concerned with
these and other examples of brinkmanship dealmaking, stopgap lawmaking, and breakdowns in
congressional problem solving more generally. Such developments, rather than partisan voting
patterns, likely bring the most harm to Congress’s problem-solving capabilities. Stopgap
measures foster uncertainty about policy implementation and force Congress to address the same
problems over and over without ever really solving them.
To be clear, the type of breakdown that we examine in this paper is not dependent upon
ideology. Neither liberals nor conservatives and neither majority party or minority party members
would prefer that Congress engage in lawmaking that “kicks the can down the road.” These
measures make governing less efficient, which liberals detest because government problem-
solving is impeded and conservatives loathe because tax dollars are squandered. In other words,
neither side prefers these types of outcomes, yet they have become the norm in today’s Congress.
Furthermore, our measure of dysfunction does not have an implicit bias towards action. A
productive Congress may be revered by some, but production against others’ preferences may be
much worse than stalemate. Whether they favor action or inaction on a particular policy, more
consensus among the parties should exist on the importance of obtaining good information than in
other arenas of the legislative process. Even if members of Congress face increasing pressure to
support ideological solutions, those solutions require high-quality information to be effective. For
5
this reason we believe that changes to Congress's information processing capacity—its committee
system—lie at the core of the current congressional dysfunction.
Federal programs are falling victim to the partisan war either directly as previously-
obscure agencies become campaign fodder, or indirectly as they are replaced on Congress's
schedule by legislation meant only to draw sharp distinctions between the parties (Bradner 2014;
Rogers 2012). We identify three existing, related explanations for breakdowns in congressional
problem solving that all may fit under the heading of “partisanship,” but highlight different
dimensions thereof. The first such explanation is a competitive electoral environment. Party
conflict involves “collective team play” in order to build positive public reputations for
governance (Cox and McCubbins 2005; Lee 2009, 48; Sinclair 1983). The two parties in
Congress have never been more polarized (that is, more internally cohesive and externally
divided) and have placed a greater premium on gaining or retaining majority party status (Rohde
and Aldrich 2010). At the same time, the two parties recently have become even more
competitive with one another (Theriault and Lewallen 2012); since the 2000 elections, the Senate
majority has changed hands four times, while the House majority has done so twice. A
competitive electoral environment puts even more pressure on the parties and their members to
deny the other side a legislative victory. Meanwhile, whichever party is in the majority has
increasingly shielded its more vulnerable members from difficult votes by leaving critical
problem-solving legislation off the schedule.
The second explanation for breakdowns in congressional problem solving is stronger
parties. In 1950, the American Political Science Association (APSA) issued a report entitled
“Toward a More Responsible Two-Party System.” The report was a response to loose linkages
between state and national party organizations that made it difficult for whichever party gained
6
control of the federal government to establish and implement a coherent agenda. The APSA
Committee on Political Parties felt at the time that stronger parties would deter elites from
misinforming the public and present distinctive choices at the polls. The report made several
recommendations based on its diagnosis, including “a party system with sufficient party loyalty”
and “tightening up the congressional party organization” (APSA 1950, p. 2-8).
Finally, polarized voting coalitions and a lack of bipartisanship surely make legislating
difficult, but these do not fully characterize the institutional dysfunction currently plaguing
Congress. As Theriault (2014) has put it, “Polarization we can live with. Partisan warfare is the
problem.” Partisan warfare likely contributes to problem-solving breakdowns as members of
Congress, particularly in the minority party, use the procedural tools at their disposal to obstruct
or otherwise delay proceedings, which further leads to brinkmanship (Smith 2014; Theriault
2013). One line of argument posits that congressional Republicans have become more
conservative faster than their Democratic counterparts have become more liberal, and that an
asymmetric shift in tactics has accompanied this difference in voting patterns (Mann and Ornstein
2012; Smith 2014; Theriault 2013).
In many ways we have the party-driven Congress that APSA wanted (Sinclair 2003);
members of the two parties in Congress are voting in patterns that are internally cohesive yet
distinct from one another and are doing so at higher rates than any other point in the institution’s
history (Rohde and Aldrich 2010). Party leaders have more tools at their disposal to enforce
discipline and structure the institution’s agenda (Curry 2015; Theriault and Lewallen 2012). We
do not doubt that polarization and partisan warfare within Congress have contributed to increased
gridlock and breakdowns in the legislative process, but we also believe that the problem of
congressional dysfunction has another dimension: breakdowns in the way that the institution, and
7
particularly its committee system, process information about policy problems and solutions.
Information and analysis are critical to governance. As James Madison writes in arguing against
annual elections in the Federalist: “No man can be a competent legislator who does not add to an
upright intention and a sound judgment a certain degree of knowledge of the subject on which he
is to legislate” (quoted in Kramnick [1788] 1987, 328). The term “information processing”
typically refers to the means by which organizations acquire, synthesize, distribute, and use
information (Cyert and March 1963; Huber 1991). It is the organizational analogue to individual
information processing, which translates inputs into outputs (Simon and Newell 1964).
Our study focuses on committee-level information processing to understand why Congress
often struggles to complete legislation that should, in theory, benefit individual members and their
institution. Committee hearings allow committee members to acquire information and
simultaneously signal that information to the rest of the institution and to other institutions
(Diermeier and Feddersen 2000; Katzmann 1989). By connecting outside expertise to the
members who actually make the decisions, committees are critical stages in the flow of
information within the institution (Krehbiel 1991; Porter 1974; Sabatier and Whiteman 1985).
Congress’s vast analytical bureaucracies (the GAO, CBO, and CRS) also work well in producing
reports that are respected and utilized by members on both sides of the aisle, yet reports do not
mobilize the attention of members in the same way that committee hearings have at least the
potential to do.
Charles O. Jones (1975) rightly identifies the institutional necessity to gather information
and identify and define problems as a means of meeting the American public's policy needs.
While Republicans and Democrats in Congress may not always agree on matters of governance,
more consensus should exist on the importance of obtaining good information. Without good
8
information, the parties offer ideologically opposed ideas that only with luck actually solve the
problems those solutions are meant to address and potentially frustrate their initial efforts. With
good information, the parties can still offer ideologically opposed ideas to present voters with
distinct agendas, but they can do so with solutions – either from the left or right – that might
actually solve the problems they have identified and in turn lead to more favorable evaluation
from constituents.
For decades, members of Congress took their cues on how to vote from a variety of
sources, including party leaders, issue specialists from the relevant committees, and constituents
(Kingdon 1973; Matthews and Stimson 1975; Panning 1983; Whiteman 1995). As the policy
environment became more complex and competition between the parties grew, the two chambers
have faced increased uncertainty about which bills would reach the agenda and how much time
would be spent on them in the face of dilatory amendments and demands from multiple
committees to have a say in shaping the chamber's legislative product (Bach and Smith 1988).
Congress has responded with a series of rules changes that have made the legislative process more
“unorthodox” by giving party leaders more responsibility for establishing lines of legislative
authority, negotiating deals, and determining which problems and solutions the institution
addresses. Committee influence in informing members’ vote choices has waned in favor of
increasing the likelihood of passing major legislation (Sinclair 2007).
The 1950 APSA report and other political scientists in its wake wanted a stronger voice
for party leaders to counteract the growing power of presidents (Dodd and Schott 1979). Today
that voice tends to speak loudest when engaging in the kind of brinkmanship that led to a debt
ceiling crisis and government shutdown as well as last-minute, lame-duck dealmaking that creates
uncertainty for numerous federal agencies and state governments.
9
The House Republican discontent over their inability to offer amendments to high-profile
legislation that led to Speaker John Boehner’s resignation in 2015 speaks precisely to
congressional leaders’ current control over the legislative schedule. Even after Paul Ryan (R-
Ohio) replaced Boehner with promises of a more open legislative process, rank-and-file
Republicans largely opposed passing individual appropriations bills for fear of taking tough votes
on proposed amendments (Sherman and Palmer 2015). While members of Congress may lament
their inability to offer amendments on the chamber floor, they also rely on party leaders to inform
them about (and increasingly to directly shape) the solutions they adopt.
Party leaders work to “structure the choice situation” for rank-and-file members, often by
providing information designed to structure a binary choice and highlight the dimension most
amenable to partisan advantage (Curry 2015; Sinclair 1983; Twight 1998). In the extreme, such
an approach to informing members’ vote decisions can lead to situations such as occurred in
September 2016, when Republican leaders had to consider changing the Justice Against Sponsors
of Terrorism Act (JASTA) due to potential “legal ensnarements,” even though such concerns
formed the basis of Obama’s veto message which Congress overrode to make the bill law as well
as a letter written by 28 senators. As Senate Majority Leader Mitch McConnell (R-Ky.) put it,
“Everybody was aware of who the potential beneficiaries were but nobody had really focused on
the potential downside in terms of our international relationships” (Dennis and House 2016).
Even in less extreme cases, Krutz (2001) points out that omnibus legislation allows party leaders
to direct rank-and-file member attention towards certain information and away from other
aspects; the larger bill itself becomes the focus rather than its details, which puts members again
at a disadvantage regarding a bill’s consequences and implementation.
10
Committees thus are a worthwhile place to look for changes in information processing as
their hearings help influence member perceptions and attitudes and provide a focal point where
legislative staff get up to speed on a policy problem. We do not argue that hearings are the only –
or even the most important – place that committees can get information. Interest groups, think
tanks, and the executive branch also have lots of information upon which committee members can
rely. Our claim in this paper is that committee hearings are one of those routinized and regularly
occurring events on Capitol Hill that can offer us insight into how committees are behaving.
While these other sources may supplant hearings, we think it is more likely that members use
these other sources in a similar way to which they use hearings. The difference between hearings
and these other sources is that there is a written record of the hearings while these other sources
simply cannot be counted or analyzed.
Partisan warfare and committee information processing are undoubtedly related;
committees often respond to the partisan environment in which they operate (Fenno 1966). We
further believe that breakdowns in the committee process feed back into the partisan war. If and
when committees restrict their attention to focus on a particular purpose or receive slanted
testimony, then the information available to members of Congress becomes further limited, which
hinders effective problem solving no matter how they try to solve a particular problem.
Consider a human trafficking bill taken up in the Senate in March 2015. Just as the bill
was scheduled for debate, Senate Democrats noticed a provision that limited spending on abortion
services in other countries; the provision always had been in the bill since its introduction two
months prior, but Democrats had not asked whether the bill addressed abortion funding and
Republicans were not forthcoming with that information. The anti-trafficking bill finally passed
the Senate by a 99-0 vote, but not before the Senate experienced heated rhetoric and a largely
11
partisan series of procedural votes. The debate and gridlock over this bill even spilled over into
other institutional responsibilities, as it delayed a vote on Loretta Lynch's nomination to be
attorney general.
We can trace this particular breakdown in congressional problem solving back to
the Senate Judiciary Committee. While we should not necessarily expect a committee to
search high and low for pro-human trafficking advocates, the committee’s hearing
featured four senators, including Democrats Barbara Mikulski and Kirsten Gillibrand, and
four anti-trafficking advocates, all of whom expressed their support for the idea behind the
bill but none of whom addressed the specific legislation in much detail. Two Politico
reporters (Everett and Kim 2015) covering the bill add:
Democrats appeared to have ample changes to spot the language. For example, the
abortion language is on pages 4 and 5 of the bill. But when it came up in committee, the
top Democrat, Vermont Sen. Patrick Leahy, offered an unrelated amendment on the same
page—apparently not noticing the abortion provisions. The bill was approved
unanimously by the judiciary panel. And on Monday, a full two weeks later with still no
ire over the abortion language, Democrats agreed to move ahead.
Had the abortion restriction provision been identified earlier in the process and Democrats'
objections been raised during the committee’s hearing, a floor fight – and a lot of embarrassment
– could have been avoided. Inadequate information processing in this case fed back into the
partisan war and added to Congress's image as a dysfunctional institution.
To that end, we have devised a new scheme for systematically coding congressional
hearings based on several information-gathering dimensions. The next section describes our
coding scheme and then uses data collected on nearly 22,000 hearings from 1971 to 2010 to
evaluate how committee information processing has changed over time and contributed to a
dysfunctional Congress.
Our study focuses on committee-level information processing to understand why Congress
often struggles to complete legislation that should, in theory, benefit all members. Our
explanation is rooted in the idea that members no longer have access to good information about
the solutions under consideration because the (public) process by which that information is
generated and disseminated throughout the institution has changed. We have echoed others in
discussing polarization and partisan warfare as direct causes of breakdowns in the legislative
process, but we also believe that these conditions exert an indirect influence through committee
information processing breakdowns and that those in turn feed back into the partisan war. If
committees restrict their attention to focus on a particular purpose or receive slanted testimony,
then the information available to members becomes more limited, which hinders effective
problem-solving no matter its partisan or ideological character. In this section we describe how
we code congressional hearings to capture how committees acquire information and present data
on changes to information processing over the past four decades.
We start our coding of committee hearings by examining the hearing and testimony
summaries published by the Congressional Information Service (CIS) as well as the Policy
Agendas Project's Congressional Hearings dataset. In addition to the issues they address and the
types of witnesses testifying, we code information-gathering in committee hearings along two
dimensions. The first dimension we use to describe committee information processing is a
hearing's purpose: whether it addresses a problem, policy implementation, or a proposed solution.
The problems and solutions discussed in these committee hearings may not be new; what is
“new” in this context is the relative attention they receive. Problem-focused hearings are those
asking if a particular issue needs to be addressed and how. They tend to address recent studies,
13
policy trends (such as an increase in childhood obesity), natural disasters, and national or
international events. Implementation-focused hearings ask whether the government's current
approach to addressing a particular problem is working or even appropriate. A solution-focused
hearing addresses the benefits or costs of a particular proposal; the problem is taken as given. Not
every hearing purpose is obvious, especially as the “implementation” code could conceivably
describe either a problem or a government solution associated with the federal bureaucracy. The
important distinction is whether or not the bureaucratic solution already has been adopted. If so,
the hearing tends to assess how an agency is carrying out that solution, and so the implementation
code is most appropriate. If the agency has not yet acted on a proposal, then the hearing focuses
on the “solution” and whether the proposal is appropriate. The CIS summaries of each hearing
can guide coders with such language as: “Hearing to review financial problems of Baltimore
residents and related community assistance programs and needs” (problem) or “Hearing to
examine concerns about DOD design and implementation of a force-wide anthrax vaccine
immunization program, including concerns about vaccine safety and efficacy” (implementation).
The second dimension we code is a hearing’s stance. We find that a hearing can take one
of two stances: positional or exploratory. Positional hearings typically hear from only one side of
the debate. All of the witnesses may praise or, alternately, criticize a program or idea, or the
hearing itself may focus only on the positive (or negative) aspects. Exploratory hearings, by
contrast, are those in which the committee hears from both sides of a particular debate, or receives
testimony that imparts information and analysis without also including a witness's personal
opinion. Some language in the CIS summary that would indicate an exploratory hearing or
individual's testimony includes: discusses, explanation of, analysis of, views on, briefing on,
status of, and differing (or conflicting) views on. Positional language includes: objections to, need
14
for, importance of, preference for, negative impact of, charged inadequacy of, and disagreement
with. Per our coding rules, only one witness needs to have provided a view that differs from other
witnesses in order for a hearing stance to qualify as exploratory.
We first obtained our sample of hearings from the Policy Agendas Project’s Congressional
Hearings dataset, which uses a topic coding scheme to trace issue attention in Congress across
time. Our own data collection efforts began in the first congress after the passage of the 1970
Legislative Reorganization Act (1971-1972) and concluded with the hearings that took place in
the 111th Congress (2009-2010), the most recent congress for which the Policy Agendas Project
had data. We gathered data by committee, initially following Smith and Deering’s (1990) findings
on perceptions of conflict in different committees' environments. While we did not subsequently
build on their analysis, collecting data this way leaves us with a broad representation of issues
(see tables 1 and 2). We include several additional committees, such as the House and Senate
Intelligence and Joint Economic Committees, that serve vital information gathering and
processing roles. Our dataset encompasses 21,830 hearings, which represents more than one-third
of the total number of hearings held by all congressional committees during this time period. We
have also coded the number of witnesses that appeared at each hearing to assess the volume of
information gathered in these fora. Our dataset excludes hearings on nominations.
[Tables 1 and 2 about here.]
We focus on three measures in the following analysis: the average number of witnesses
per hearing in a given congress, the percentage of hearings that attend to proposed solutions, and
the percentage of exploratory hearings. We highlight solution-focused hearings rather than either
problem or implementation hearings, though we note that information processing patterns on the
three different hearing purposes undoubtedly are connected. Yet effective problem solving
15
(however defined) requires good information about the solution, legislative or otherwise, under
consideration to address that problem. A decrease in attention to proposed solutions specifically
thus would suggest that committees no longer are “lay[ing] an intellectual and political
foundation” for good problem solving (Kaiser 2013, 27).
The average committee hearing from 1971 to 2010 called on 11 witnesses, while 44
percent of hearings addressed a proposed policy solution and 69 percent of all hearings were
exploratory (see table 3); this latter finding should be encouraging to those who believe hearings
are “cooked” or merely for show. We further find large cross-sectional differences in committee
information processing by issue. Many more witnesses have testified on hearings related to
agriculture and the environment, 17 and 15 on average, respectively. Hearings on these two policy
areas also tend to be more exploratory (78 and 75 percent, respectively) and more focused on
proposed solutions (46 and 55 percent, respectively). Relatively more defense hearings have been
exploratory, 81 percent, than any other policy area, while 61 percent of hearings on public lands
and water issues have been devoted to proposed solutions.
Hearings in other policy areas have focused much more on policy problems and
implementation, and been more positional, over the past four decades. Just 21 percent of hearings
on international affairs have focused on proposed solutions; instead, 45 percent of them have been
devoted to new and emerging problems. Hearings on this topic also involve almost half as many
witnesses (six) as all the hearings. Across all issues, almost 70 percent of the hearings in our
dataset are exploratory. In the cases of commerce and education issues, however, their hearings
are 14 points below the average at 55 percent.
[Table 3 about here.]
16
We now turn to an analysis of important temporal trends in committee information
processing. Committees have always needed to be mindful of the larger partisan and ideological
environments when setting the institution’s agenda (Fenno 1966; Krehbiel 1991; Cox and
McCubbins 2005). If greater partisan competition, control, and warfare have affected information
processing, we should expect three primary trends over the past four decades. In essence, these
trends serve as external validity of our coding procedures. First, over time committees will call
fewer witnesses across fewer panels. Different information sources provide different perspectives
on public policy issues. The more sources committees consult the more information they will
have before crafting solutions. If the committees rely upon fewer panels and fewer witnesses even
while problems get more complex, the quality of information will necessarily decrease.
Second, committees will hold fewer solution-focused hearings as the two parties find
themselves at greater odds over the basic functions of government and what issues they should
address. When a committee holds a hearing about a particular solution it already has determined,
consciously or not, that the issue that solution would address is worthy of congressional attention.
Legislators from parties with very different ideas about what government should and should not
do will instead spend more time trying to justify their arguments by focusing on whether
Congress should become involved in a particular problem (Baumgartner and Jones 2015). If
committees are holding fewer solution-focused (and more problem-focused) hearings, they will
be less able to inform themselves and their colleagues about the solutions that are debated and
voted on.
Third, committees will hold fewer exploratory hearings (those devoted purely to analysis
or that take in multiple viewpoints) over time as they face fewer incentives to become issue
experts for their colleagues. As with the number of witnesses, exploratory hearings on a particular
17
issue provide different perspectives and points of emphasis, and help inform committees and their
members about a subject's substantive (such as what hydraulic fracture drilling or “fracking”
entails on a technical level) and political (which groups support and oppose fracking) dimensions.
If committees are holding more one-sided hearings over time, they necessarily will be restricting
the range of viewpoints from which they hear, and thus be less prepared for any future roadblocks
to action.
A few of the existing explanations for dysfunction also produce different expectations for
when we should expect these changes. If a competitive electoral environment drives changes to
committee information processing, we should see the trends begin in the early to mid-1990s when
Republicans began to make significant gains. If asymmetric partisan warfare plays a role, then we
should expect to see significant differences dependent upon which party is in the majority.
Baumgartner and Jones (2015) provide another hypothesis: that the broader and deeper federal
government that emerged after World War II altered congressional information processing. The
dramatic growth in government, so they argue, serves as a kind of background radiation to which
increasingly competitive and polarized parties have been exposed. A government that addresses a
greater number of issues and is more active on the issues in which it is involved should create
greater demand for oversight as well as a more diverse information search in order to set priorities
in the face of more intense competition for policymaker attention. If this is the case, then we
should see committees hold more implementation-driven hearings beginning in the early 1970s.
Our analyses of the trends lend some preliminary support for the growth-in-government
explanation, but not for the close elections or asymmetric warfare explanations. Committees in
both chambers show a linear downward trend in the number of witnesses and witness panels
called in hearings (see figure 1). House and Senate committees called 14 to 16 witnesses across 6
18
to 7 panels in the 92nd Congress, but less than half as many witnesses and panels in the 111th
Congress. The decline in average witness panels began almost from the outset of our time period,
and the decline in the average number of witnesses heard from followed soon thereafter when
Democrats still held solid electoral and legislative majorities. Interestingly, House committees
heard from more witnesses and panels at the beginning of this time period than did Senate
committees, but the two traded places in the early-to-mid 1990s. While these data do not address
the sources of information from which committees hear, they do provide the first clear indicator
that committee hearings involve fewer information sources today than they did 40 years ago.
[Figure 1 about here.]
We can disaggregate these longitudinal trends by issue with slope coefficients from
regressing a given issue’s information processing measures on a time trend. A positive coefficient
indicates that the relevant indicator is increasing over time for a particular issue, while a negative
coefficient indicates that an indicator is decreasing over time. Comparing the slope coefficients
reveals not only which issues have seen decreases (or increases) over time, but the relative
magnitude of those changes.1
The average number of witnesses at a given hearing has decreased for 16 out of the 19
issues we analyze; only science and technology, international affairs, and environment issues
have seen no statistically significant change in the number of witnesses called per hearing over
the past four decades (see figure 2). The largest decreases in witnesses have been seen in hearings
devoted to social welfare, a little more than one fewer witness with each successive congress.
Education, labor and employment, and agriculture issues display the next-largest decreases. In the
1 We have excluded immigration hearings from this analysis due to the low number of hearings in
our dataset.
social welfare, agriculture, transportation, and environment hearing models, we find relatively
large standard errors for the slope coefficients, which indicate wider variation in the average
number of witnesses called to testify on these issues.
With regards to hearing purposes, committee attention to proposed solutions has waned as
committees have spent more of their time learning about and guiding bureaucratic policymaking
and trying to define which issues the federal government should or should not address. Here, too,
we find fairly linear decreasing trends that begin in the early 1970s (see figure 3). We further find
that both problem- and implementation-focused hearings have increased as solution hearings have
decreased, though they have done so at slightly higher rates in the House than in the Senate.
Among all three purpose types, any difference that may have existed between the two chambers
in the 1970s has all but disappeared. Committees from both chambers devoted about two-thirds of
their hearings to policy solutions 40 years ago. In recent years, hearings on proposed policy
solutions represent just a third of activity in the House and a quarter in the Senate.
More than half of the issues we analyze also saw significant decreases in their attention to
proposed solutions (see figure 4). Put another way, 12 out of 19 topics have seen significant shifts
towards hearings that, first, raise attention to and define a particular policy problem and, second,
towards overseeing implementation of existing solutions. Science and technology hearings exhibit
the largest shift by far, about a four percentage point decrease in solution-focused hearings with
each successive congress, followed by defense, education, agriculture, and government operations
(which includes multi-agency appropriations measures along with matters related to government
employees, tax administration and enforcement, and electoral campaign regulation). Half of these
20
12 issues—health, education, transportation, domestic commerce, science and technology, and
public lands and water—became increasingly structured to the voting dimension most amenable
to partisan and ideological splits since the 1980s (Jochim and Jones 2013).
While our data on hearing witnesses and purposes show very similar trends among House
and Senate committees, a look at hearing stances reveals a striking difference: House committee
stances have become less exploratory over time, while Senate committee stances have become
slightly more exploratory. A linear trend fits these data less well than it does for hearing witnesses
and purposes, but Senate committee stances remain centered around 70 percent exploratory
throughout the time period (see figure 5). House committee stances, by contrast, have decreased
from as high as 77 percent exploratory in the 95th Congress to a low of 63 percent exploratory
percent positional in the 106th Congress and 64 percent exploratory in each congress from 2001
to 2008.2
Finally, just seven of the 19 issue we analyze have seen significant changes in the
percentage of exploratory hearings since 1971 (see figure 6). Of these, science and technology
and defense hearings have become more exploratory over time, and in the former case the change
appear to be quite large, an increase of about two percentage points with each successive
congress. The time trend has positive slope coefficient estimates for four additional issues—
2 The trend of positional hearings within each purpose varies. House hearings of all types have
become less exploratory over time, problem-focused hearings most dramatically so. In the Senate,
problem- and solution-focused hearings have become slightly less exploratory over time, but
implementation hearings have become much more exploratory.
21
are not statistically significant. Five issues show statistically significant decreases in exploratory
hearings; that is, their hearings have become more positional or one-sided over time. Once again
social welfare hearings show the biggest decrease, followed by hearings on foreign trade,
housing, public lands and water, and health.
To summarize these findings, three issues have seen significant decreases in all three of
our indicators: health, social welfare, and public lands and water (see table 3). Over the past 40
years these issues all have seen fewer witnesses called to testify (and thus fewer sources of
information), fewer hearings devoted to learning about proposed solutions, and fewer exploratory,
analytical hearings. In the case of social welfare, this topic saw the largest decrease of any issue in
two of our three measures. Health hearings have been consistently below average in their
attention to proposed solutions throughout this time period, but dropped even lower in the 1990s
and 2000s and fell to just 12 percent solution-focused in 2009-2010. Health hearings were
consistently average or above average in our exploratory measure throughout the 1980s but
similarly became more positional in the late 1990s and early 2000s.
Two additional issues consistently have exhibited no significant change in how
committees process information: international affairs and the environment. Recall from Table 2
that international affairs exhibited the lowest witness average and the lowest percentage of
solution-oriented hearings of all 20 issues. While these patterns have stayed relatively consistent
over time, data from the most recent congresses in our dataset suggest they are declining even
further, with just three witnesses called on average and only eight percent of those hearings being
devoted to proposed solutions in the 2009-2010 congress.
22
Congress has recently enacted some major legislation, including a reauthorization of the
Elementary and Secondary Education Act, a surface transportation bill, and a means for
congressional review of a nuclear agreement with Iran. Yet signs remain that the two parties still
lack policy consensus on an array of issues. Congress narrowly avoided yet another government
shutdown in September 2015 and has continued to rely on a string of temporary, short-term
authorization and funding measures. Deadlines and stopgaps still dictate the rhythms of
congressional attention. Understanding the roots of this dysfunction can help us better develop the
steps needed to turn around this trend.
In this paper we adopt an information processing approach to understanding congressional
dysfunction. Whereas many scholars, journalists, and others see partisan warfare, competitive
elections, and stronger parties as direct contributors to breakdowns in problem solving and
legislative brinkmanship, we believe these factors filter through the committee system. As
committees and their members face fewer incentives to act as legislative workhorses, they have
responded by calling fewer witnesses, spending less time learning about proposed solutions, and
spending less time hearing a range of opinion and analysis on different issues. In turn, our data
provide some preliminary evidence that changes to committee information processing share some
responsibility for the current dysfunction plaguing Congress, particularly the increased reliance
on continuing appropriations resolutions and prevalence of lame duck sessions. Our findings echo
Baumgartner and Jones (2015, 207-8):
The capacity of government to assess problems has leveled off and declined… The process underlying this seems to be a politics of attribute suppression, a self-
conscious attempt to limit serious discussions of aspects of complex problems
through a net of formal and informal restrictions.
23
Understanding the problem better is essential for working toward a solution. We
understand the basic ideological continuum on which policy exists, and our study highlights
another, previously-unexplored continuum, that concerning the information used to inform debate
and how committees process such information. We find it perhaps notable that we do not find
consistent clusters of issues exhibiting the same patterns; changes to committee information
processing do not seem to be fall along familiar lines. Issues concerning the economy,
infrastructure, and the scope of government that typically divide the two parties in Congress are
not always consistent with each other. Health and social welfare undoubtedly are related topics;
their connection to public lands and water issues are less certain, particularly as the latter exhibits
a longitudinal trend different from environment policy. Science and technology hearings have
seen no significant change in the average number of witnesses, a decrease in attention to proposed
solutions, and an increase in exploratory stances over time.
These findings suggest that the type of dysfunction we analyze in this paper is not rooted
to ideological shifts of even gridlock. Rather, it is in the quality of information available to
policymakers, a concept that neither liberal nor conservative and neither majority party or
minority party member should oppose.
While committees and party leaders each act as cue-givers to other members, the
committee system allows different kinds of information to enter the system simultaneously; the
information that party leaders provide, by contrast, typically is used to structure a binary choice
and highlight the dimension most amenable to partisan advantage. Thus even as members of
Congress become more polarized and the institution trends toward dysfunction, committees’
fundamental role of processing and prioritizing information for action remains unchanged. In an
era in which few if any bills are amended and debated at length on the chamber floor, a robust
24
committee process becomes even more vital for learning about policy problems and exploring
effective solutions. Quality information through and from the committee system should render
more effective Congress's ability to solve problems regardless of the solution, partisan or
otherwise, that results. This, in turn, should render American representative government more
effective as well.
25
References
Adler, E. Scott, and John D. Wilkerson. 2012. Congress and the Politics of Problem Solving. New
York: Cambridge University Press.
American Political Science Association (APSA). 1950. “Toward a More Responsible Two-Party
System: A Report of the Committee on Political Parties.” American Political Science
Review 44 (Supplement).
American Political Science Association (APSA). 2013. Negotiating Agreement in Politics: Report
of the Task Force on Negotiating Agreement in Politics. Jane Mansbridge and Cathie Jo
Martin, eds. Washington, D.C.: American Political Science Association.
Bach, Stanley, and Seven S. Smith. 1988. Managing Uncertainty in the House of Representatives:
Adaptation and Innovation in Special Rules. Washington, D.C.: Brookings Institution
Press.
Baumgartner, Frank R., and Bryan D. Jones. 2015. The Politics of Information. Chicago:
University of Chicago Press.
Bipartisan Policy Center. 2014. Governing in a Polarized America: A Bipartisan Blueprint to
Strengthen Our Democracy. Washington, D.C.: Bipartisan Policy Center.
Bradner, Eric. 2014. “2014's sleeper issue: A bank nobody's heard of.” Politico.com. July 13.
http://www.politico.com/story/2014/07/ex-im-bank-splash-campaigns-108833.html (last
accessed July 23, 2014).
Cox, Gary W., and Mathew D. McCubbins. 2005. Setting the Agenda: Responsible Party Gov-
ernment in the U.S. House of Representatives. New York: Cambridge University Press.
Cox, James H. 2004. Reviewing Delegation: An Analysis of the Congressional Reauthorization
Process. Westport, CT: Greenwood Publishing Group.
Representatives. Chicago: University of Chicago Press.
Cyert, Richard M., and James G. March. 1963. A Behavioral Theory of the Firm. Englewood
Cliffs, NJ: Prentice-Hall.
Dennis, Steven T., and Billy House. 2016. “Congress May Rewrite 9/11 Law After Veto
Override.” Bloomberg.com, September 29.
Diermeier, Daniel, and Timothy J. Feddersen. 2000. “Information and Congressional Hearings.”
American Journal of Political Science 44, 51-65.
Dodd, Lawrence C., and Richard L. Schott. 1979. Congress and the Administrative State. Wiley.
Eldridge, David. 2014. “113th Congress Could Yield Fewest Laws in 60 Years.” Rollcall.com.
September 22. http://blogs.rollcall.com/218/113th-congress-legacy-shutdowns-
shoutdowns-gridlock/ (last accessed March 27, 2015).
Everett, Burgess, and Seung Min Kim. 2015. “How abortion politics scuttled a human-trafficking
bill.” Politico.com. March 12. http://www.politico.com/story/2015/03/how-abortion-
politics-scuttled-a-human-trafficking-bill-116042.html?hp=b1_l2 (last accessed March 27).
Fenno, Richard F., Jr. 1966. The Power of the Purse: Appropriations Politics in Congress.
Boston: Little, Brown & Co.
Hall, Thad E. 2004. Authorizing Policy. Columbus: The Ohio State University Press.
Henry, Devin. 2016. “Dems block spending bill over Iran amendment for third time.”
TheHill.com, May 9. http://thehill.com/policy/energy-environment/279284-democrats-
Literatures.” Organization Science 2, 88-115.
Jochim, Ashley E., and Bryan D. Jones. 2013. “Issue Politics in a Polarized Congress.” Political
Research Quarterly 66, 352-69.
Jones, Charles O. 1975. “Somebody Must Be Trusted: An Essay on Leadership of the U.S.
Congress.” In Congress in Change: Evolution and Reform, ed. Norman J. Ornstein. New
York: Praeger.
Kaiser, Robert G. 2013. Act of Congress: How America’s Essential Institution Works, and How It
Doesn’t. New York: Alfred A. Knopf.
Katzmann, Robert A. 1989. “The American Legislative Process as a Signal.” Journal of Public
Policy 9:287-306.
Kingdon, John W. 1973. Congressmen's Voting Decisions. New York: Harper & Row.
Kramnick, Isaac. [1788] 1987. The Federalist Papers. New York: Penguin Books.
Krehbiel, Keith. 1991. Information and Legislative Organization. Ann Arbor: Michigan Press.
Krutz, Glen. 2001. “Tactical Maneuvering on Omnibus Bills in Congress.” American Journal of
Political Science 45:210-23.
Lee, Frances E. 2009. Beyond Ideology: Politics, Principles, and Partisanship in the U.S. Senate.
Chicago: University of Chicago Press.
Lesniewski, Niels, and Bridget Bowman. 2015. “Heading Into Endgame, McConnell, Declares
‘Dysfunction is Over.’” RollCall.com, December 14.
http://www.rollcall.com/wgdb/heading-into-endgame-mcconnell-declares-dysfunction-is-
http://www.rollcall.com/wgdb/democrats-stalling-energy-bill-over-flint-aid/ (last accessed
August 2, 2016)
Mann, Thomas E., and Norman J. Ornstein. 2012. It's Even Worse Than It Looks: How the
American Constitutional System Collided with the New Politics of Extremism. Basic.
Matthews, Donald R., and James A. Stimson. 1975. Yeas and Nays: Normal Decision-Making in
the U.S. House of Representatives. New York: John Wiley & Sons.
Nixon, Ron. 2012. “With Farm Bill Stalled, Consumers May Face Soaring Milk Prices.” New
York Times. December 20. http://www.nytimes.com/2012/12/21/us/milk-prices-could-
double-as-farm-bill-stalls.html?_r=0 (last accessed July 23, 2014).
Ordoñez, Franco. 2016. “Congress has missed its chance to help combat Zika.” Miami Herald,
August 1. http://www.miamiherald.com/latest-news/article93159382.html (last accessed
Panning, William H. 1983. “Information Flow and Party Influence: Theory, Measurement, and
Policy Consequences.” Political Behavior 5:161-89.
Porter, H. Owen. 1974. “Legislative Experts and Outsiders: the Two-Step Flow of
Communication.” Journal of Politics 36:703-730.
Rogers, David. 2012. “House punts agriculture bill.” Politico.com. June 22.
http://www.politico.com/news/stories/0612/77742.html (last accessed July 23, 2014).
Rohde, David, and John Aldrich. 2010. “Consequences of Electoral and Institutional Change: The
Evolution of Conditional Party Government in the U.S. House of Representatives.” In New
Directions in American Political Parties, ed. Jeffrey M. Stonecash. New York: Routledge.
Information: Models of Information Flow.” Legislative Studies Quarterly 10:395-421.
Sherman, Jake, and Anna Palmer. 2015. “House GOP’s appetite for more votes shows limits.”
Politico.com. November 6. http://www.politico.com/story/2015/11/paul-ryan-spending-
bill-215603 (last accessed November 7, 2015).
Simon, Herbert A., and Allen Newell. 1964. “Information Processing in Computer and Man.”
American Scientist 52:281-300.
Sinclair, Barbara. 1983. Majority Leadership in the U.S. House. Baltimore: Johns Hopkins
University Press.
Sinclair, Barbara. 2003. “The Dream Fulfilled? Party Development in Congress, 1950-2000” in
Responsible Partisanship? The Evolution of American Political Parties Since 1950, eds.
John C. Green and Paul S. Herrnson. Lawrence, KS: University of Kansas Press.
Sinclair, Barbara. 2007. Unorthodox Lawmaking: New Legislative Processes in the U.S.
Congress, 3rd ed. Washington, D.C.: CQ Press
Smith, Steven S. 2014. The Senate Syndrome: The Evolution of Procedural Warfare in the
Modern U.S. Senate. Norman: University of Oklahoma Press.
Smith, Steven S., and Christopher J. Deering. 1990. Committees in Congress, 2nd ed. Washington,
D.C.: CQ Press.
Theriault, Sean M. 2013. The Gingrich Senators: The Roots of Partisan Warfare in Congress.
New York: Oxford University Press.
Theriault, Sean M. 2014. “Polarization we can live with. Partisan warfare is the problem.” The
Monkey Cage, washingtonpost.com. http://www.washingtonpost.com/blogs/monkey-
Theriault, Sean M., and Jonathan Lewallen. 2012. “Congressional Parties and the Policy Process.”
In The Parties Respond, 5th ed., eds. Mark D. Brewer and L. Sandy Maisel. Boulder, CO:
Westview Press.
Twight, Charlotte. 1998. “What Congressmen Knew and When They Knew It: Further Evidence
on the Origins of U.S. Broadcasting Regulation.” Public Choice 95:247-76.
Wasson, Erik. 2016. “Congress Punts to Let Trump Decide Current-Year Federal Spending.”
Bloomberg.com, November 17. http://www.bloomberg.com/politics/articles/2016-11-
Whiteman, David. 1995. Communication in Congress. Lawrence: University of Kansas Press.
Chamber
Budget Budget Library
Ethics Environment & Public Works Pepper Commission
Government Operations Ethics
Intelligence Government Affairs
Government Operations 2,248
Banking & Commerce 1,872
32
Table 3. Committee Information Processing by Issue, 1971-2010.
Issue Avg. Witnesses Solution % Exploratory % Avg. Across Issues 11 44 69
Macroeconomics 9 45 72
Health 10 30 65
Agriculture 17 46 78
Labor 12 52 67
Education 12 51 55
Environment 15 51 75
Energy 12 43 75
Immigration 8 28 64
Transportation 11 41 67
Social Welfare 15 55 61
Housing 11 32 63
Defense 9 47 81
Public Lands & Water 11 61 70
Note: cell entries in bold represent above-average values, cell entries in italics
represent below-average values
25
House
0
92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111
Congress
Figure 2. Changes in Average Hearing Witnesses by Issue, 1971-2010.
Note: The data in Figure 1 represent the slope coefficient estimates with standard errors for a
series of OLS regression equations Yi = β0 + β1X + ε, where Y represents the average number of
witnesses called to hearings on a given issue i in congress t and X represents a time trend counter;
Immigration hearings have been excluded from this analysis due to a small number of hearings.
34
Figure 3. Commitee Hearings Devoted to Learning About Proposed Solutions, 1971-2010.
70
60
50
40
30
20
10
0
Congress
Figure 4. Changes in Percentage of Hearings Devoted to Policy Solutions, 1971-2010.
So lu
ti o
n P
er ce
n ta
House Trendline
92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111
Note: The data in Figure 2 represent the slope coefficient estimates with standard errors for a
series of OLS regression equations Yit = β0 + β1X + ε, where Y represents the percentage of
solution-focused hearings on a given issue i in congress t and X represents a time trend counter;
35
0
10
20
30
40
50
60
70
80
90
House Trendline
Senate Trendline
92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111
Congress
Figure 6. Changes in Percentage of Exploratory Hearings, 1971-2010.
Note: Note: The data in Figure 3 represent the slope coefficient estimates with standard errors for
a series of OLS regression equations Yit = β0 + β1X + ε, where Y represents the percentage of
exploratory hearings on a given issue i in congress t and X represents a time trend counter;
36
Table 4. Summary of Changes to Committee Information Processing By Issue.
No. of Witnesses Solution Percentage Exploratory Percentage
Increase None None Defense
Structure Bookmarks
Congressional approval is at an all-time low. Citizens, journalists, current and former members of Congress, and political scientists alike lament the increase in party polarization, the decline of comity and bipartisanship, the resulting fall-off in legislative productivity, and the decline of trust in government among voters (APSA 2013; Bipartisan Policy Center 2014;
dysfunction by beginning with the fact that agenda space in Congress is scarce; too many proposals compete for the institution's limited attention. Managing this competition by privileging reauthorization legislation, Congress, by requiring a review of programs every few years and tying these schedules to program funding changes, ensures that different issues receive attention by imposing precipitous costs on itself for inaction (Adler and Wilkerson 2012; Cox 2004; Hall 2004). Yet the last several years hav
To be clear, the type of breakdown that we examine in this paper is not dependent upon ideology. Neither liberals nor conservatives and neither majority party or minority party members would prefer that Congress engage in lawmaking that “kicks the can down the road.” These measures make governing less efficient, which liberals detest because government problem-solving is impeded and conservatives loathe because tax dollars are squandered. In other words, neither side prefers these types of outcomes, yet the
system—lie at the core of the current congressional dysfunction.
The second explanation for breakdowns in congressional problem solving is stronger parties. In 1950, the American Political Science Association (APSA) issued a report entitled “Toward a More Responsible Two-Party System.” The report was a response to loose linkages between state and national party organizations that made it difficult for whichever party gained
Committee on Political Parties felt at the time that stronger parties would deter elites from misinforming the public and present distinctive choices at the polls. The report made several
In many ways we have the party-driven Congress that APSA wanted (Sinclair 2003); members of the two parties in Congress are voting in patterns that are internally cohesive yet distinct from one another and are doing so at higher rates than any other point in the institution’s history (Rohde and Aldrich 2010). Party leaders have more tools at their disposal to enforce discipline and structure the institution’s agenda (Curry 2015; Theriault and Lewallen 2012). We do not doubt that polarization and partisan wa
Information and analysis are critical to governance. As James Madison writes in arguing against annual elections in the Federalist: “No man can be a competent legislator who does not add to an upright intention and a sound judgment a certain degree of knowledge of the subject on which he
Charles O. Jones (1975) rightly identifies the institutional necessity to gather information and identify and define problems as a means of meeting the American public's policy needs. While Republicans and Democrats in Congress may not always agree on matters of governance, more consensus should exist on the importance of obtaining good information. Without good
problems those solutions are meant to address and potentially frustrate their initial efforts. With good information, the parties can still offer ideologically opposed ideas to present voters with distinct agendas, but they can do so with solutions – either from the left or right – that might actually solve the problems they have identified and in turn lead to more favorable evaluation from constituents.
Consider a human trafficking bill taken up in the Senate in March 2015. Just as the bill was scheduled for debate, Senate Democrats noticed a provision that limited spending on abortion services in other countries; the provision always had been in the bill since its introduction two months prior, but Democrats had not asked whether the bill addressed abortion funding and Republicans were not forthcoming with that information. The anti-trafficking bill finally passed the Senate by a 99-0 vote, but not before
other institutional responsibilities, as it delayed a vote on Loretta Lynch's nomination to be attorney general.
We start our coding of committee hearings by examining the hearing and testimony summaries published by the Congressional Information Service (CIS) as well as the Policy Agendas Project's Congressional Hearings dataset. In addition to the issues they address and the types of witnesses testifying, we code information-gathering in committee hearings along two dimensions. The first dimension we use to describe committee information processing is a hearing's purpose: whether it addresses a problem, policy imple
international events. Implementation-focused hearings ask whether the government's current approach to addressing a particular problem is working or even appropriate. A solution-focused hearing addresses the benefits or costs of a particular proposal; the problem is taken as given. Not every hearing purpose is obvious, especially as the “implementation” code could conceivably describe either a problem or a government solution associated with the federal bureaucracy. The important distinction is whether or n
The second dimension we code is a hearing’s stance. We find that a hearing can take one of two stances: positional or exploratory. Positional hearings typically hear from only one side of the debate. All of the witnesses may praise or, alternately, criticize a program or idea, or the hearing itself may focus only on the positive (or negative) aspects. Exploratory hearings, by contrast, are those in which the committee hears from both sides of a particular debate, or receives testimony that imparts informati
with. Per our coding rules, only one witness needs to have provided a view that differs from other witnesses in order for a hearing stance to qualify as exploratory.
Third, committees will hold fewer exploratory hearings (those devoted purely to analysis or that take in multiple viewpoints) over time as they face fewer incentives to become issue experts for their colleagues. As with the number of witnesses, exploratory hearings on a particular
members about a subject's substantive (such as what hydraulic fracture drilling or “fracking” entails on a technical level) and political (which groups support and oppose fracking) dimensions. If committees are holding more one-sided hearings over time, they necessarily will be restricting the range of viewpoints from which they hear, and thus be less prepared for any future roadblocks to action.
Our analyses of the trends lend some preliminary support for the growth-in-government explanation, but not for the close elections or asymmetric warfare explanations. Committees in both chambers show a linear downward trend in the number of witnesses and witness panels called in hearings (see figure 1). House and Senate committees called 14 to 16 witnesses across 6
Congress. The decline in average witness panels began almost from the outset of our time period, and the decline in the average number of witnesses heard from followed soon thereafter when Democrats still held solid electoral and legislative majorities. Interestingly, House committees heard from more witnesses and panels at the beginning of this time period than did Senate committees, but the two traded places in the early-to-mid 1990s. While these data do not address the sources of information from which c
More than half of the issues we analyze also saw significant decreases in their attention to proposed solutions (see figure 4). Put another way, 12 out of 19 topics have seen significant shifts towards hearings that, first, raise attention to and define a particular policy problem and, second, towards overseeing implementation of existing solutions. Science and technology hearings exhibit the largest shift by far, about a four percentage point decrease in solution-focused hearings with each successive congr
public lands and water—became increasingly structured to the voting dimension most amenable to partisan and ideological splits since the 1980s (Jochim and Jones 2013).
fundamental role of processing and prioritizing information for action remains unchanged. In an era in which few if any bills are amended and debated at length on the chamber floor, a robust

Congressional Dysfunction and the Decline of Problem Solving

Documents

Transcript of Congressional Dysfunction and the Decline of Problem Solving