Representation, Agendas and Institutions
Shaun Bevan ([email protected])
Mannheim Center for European Social Research (MZES), University of Mannheim
Will Jennings ([email protected])
Department of Politics and International Relations, University of Southampton
Acknowledgements: The research is an output of the project, Legislative Policy Agendas in the UK,
funded by the UK Economic and Social Research Council (Award Reference R105938). An earlier
version of this paper won the International Comparative Policy Analysis Forum (ICPA) and APSA
Public Policy Section’s Best Paper in Comparative Policy Award in 2010. We are thankful to
Christian Breunig, Isabelle Engeli, Bob Erikson, Bryan Jones, Peter May, Chris Wlezien and John
Zaller for their thoughtful comments on earlier versions of this paper.
1
Abstract
Dynamic representation can be understood through the transmission of the priorities
of the public onto the policy priorities of government. This differs from representation of
public preferences through shifts in policy. This pattern of dynamic representation in policy
agendas is mediated through institutions due to friction in decision-making and the scarcity of
attention. The paper considers data for the US and UK, from 1951 to 2003, relating to
executive speeches, laws and budgets in combination with data on public opinion about the
‘most important problem’. The results show that responsiveness is greater when institutions
are subject to less friction (executive speeches) and declines as friction against policy change
increases (laws and budgets) and decision-makers become less concerned with agenda-setting
(budgets).
2
1. Policy Agendas and Dynamic Representation
Policy-makers have finite capacity to attend to issues and act upon them (inter alia
Simon 1957; Padgett 1980; Jones 1994; 2001; Jones and Baumgartner 2005a; 2005b; May et
al. 2008; Baumgartner et al. 2009; Green-Pedersen and Mortensen 2009; Walgrave and
Nuytemans 2009; Breunig et al. 2010; Breunig 2011; Adler and Wilkerson 2012). The
relative abundance of information about the state of the world and the costs of information
retrieval and processing require decision-makers to choose which issues to attend to and
which issues to discount. This creates inherent trade-offs in policy-making and necessitates
prioritization between competing policy concerns. The responsiveness of policy agendas to
the public agenda is therefore a basic form of democratic representation.1 In some instances it
may be a necessary condition for policy representation to occur (i.e. there is less likelihood of
policy change on those issues that do not make it onto the agenda). When the attention of
governing institutions is most scarce, the prioritization of issues is most important for patterns
of representation. In institutions where fewer direct trade-offs between issues need to be
made, such as in appropriations and outlays of expenditure, prioritization is less important
and policy-makers can focus on responding to public preferences across a wider portfolio of
issues in parallel.
The representative dynamics of policy agendas are therefore built upon the degree to
which governing institutions translate information signals about the issue priorities of the
public into their own policy priorities and outputs – such as executive decisions, law-making,
budgeting and other government activities. Since the attention of decision-makers is scarce
the responsiveness of policy agendas to public priorities is an important mechanism through
1
Kingdon (1984, p. 3) defines the governmental agenda as ‘...the list of subjects or problems
to which governmental officials, and people outside of government closely associated with
those officials, are paying some serious attention at any given time’.
3
which dynamic representation can occur. Further, in some domains the responsiveness of the
policy agenda is a necessary, but insufficient, condition for actual policy representation. In
itself, the selection of issues for attention on the policy agenda is an important part of the
policy process (Kingdon 1984; Baumgartner and Jones 1993; Jones and Baumgartner 2005a).
This paper builds upon the previous literature on dynamic representation through a
comparative design that examines how attention scarcity and institutional friction affect the
responsiveness of policy agendas to public priorities expressed through the ‘most important
problem’ (MIP) question across three different institutional venues (i.e. executive, legislative
and budgetary agendas) in the US and the UK. Further, it extends research on correspondence
between public priorities and policy agendas (Jones et al. 2009) through analysis of policy
agenda representation over time. This distinguishes it from studies that have examined the
link between public preferences and public policy (e.g. Wlezien 1995; Soroka and Wlezien
2005; 2010). Through estimation of time series error-correction models (ECMs) it finds that
the responsiveness of policy agendas to public priorities decreases as the level of institutional
friction (measured through stochastic process methods) increases and as attention scarcity
decreases. While there is variation between countries in the individual issue categories that
policy agendas are responsive for, there is nevertheless a general pattern of opinionresponsiveness across institutional venues which suggests that the influence of public
priorities is greatest where attention is scarcest and institutional friction is lowest. These
findings highlight the theoretical distinction between responsiveness to public priorities and
public preferences as modes of dynamic representation and suggest that future enquiry might
consider how political institutions give rise to variation in responsiveness to signals about
public priorities compared with shifts in preferences according to the scarceness of attention
and friction against policy change.
Dynamic Representation, Agendas and Public Priorities
4
The responsiveness of public policy to public opinion is a defining characteristic of
democracy (Key 1961; Dahl 1971; Lijphart 1984). This can occur through representation of
the policy preferences or priorities of citizens. There is a large and wide-ranging literature
that demonstrates dynamic representation of public preferences in policy-making at the
macro-level (e.g. Page and Shapiro 1983; Stimson et al. 1995; Wlezien 1995; 1996; Erikson
et al. 2002; Hakhverdian 2010; Bartle et al. 2011). Similar patterns are observed in crossnational comparisons (e.g. Soroka and Wlezien 2005; 2010; Hobolt and Klemmensen 2005;
2008). Such studies tend to emphasize the continuous character of shifts and transitions in the
attitudes of citizens and public policy. The original model of dynamic representation argues
that policy adjusts over time to changes in public preferences both through the mechanism of
elections and through the ‘rational anticipation’ of policy-makers: such that “… when
electoral politicians sense a shift in public preferences, they act directly and effectively to
shift the direction of public policy … [with]… no evidence of delay or hesitation” (Stimson
et al. 1995, p. 560). While preferences are often stable and systematic it is important to
distinguish between policy changes arising due to changes in preferences and those due to
changes in priorities (Jones 1994, pp. 5-10).
Whereas the literature on dynamic representation concerns the mechanisms through
which public preferences are represented in policy decisions and policy outcomes, there is a
growing interest in the responsiveness of policy-making to the issue priorities of the public
(e.g. Cohen 1995; Hobolt and Klemmensen 2005; Jones and Baumgartner 2004; Jennings and
John 2009; Jones et al. 2009; Chaqués Bonafont and Palau 2011; Lindeboom 2012). While
the public hold opinions across a diverse and sometimes competing array of issues and
concerns,2 there is insufficient space on the public agenda to attend to all of them (Carmines
2
This requires the aggregation of judgments across often competing concerns (for example,
see Zaller and Feldman 1992 who argue that at the survey responses of individuals, in
5
and Stimson 1989, p. 4). Some issues tend to be prioritized above others. For example, during
a recession healthcare may remain a concern for the public but the state of the economy will
likely matter more. The same could be true at times of military conflict, as domestic policy
problems drop down the public agenda in response to concerns over national security.
Agendas are therefore a crucial aspect of the making of public policy in shaping the
definition and ranking of the relative importance of issues as well as policy alternatives (e.g.
Schattschneider 1960; Kingdon 1984; Jones and Baumgartner 1993; Jones 1994; Soroka
2002). Agenda-setting describes the process through which ideas and policy solutions make it
onto the decision-making agenda, often leading to policy change. This is an important
mechanism through which actual policy representation can occur.
The transmission of the priorities of the public into the policy agenda of governing
institutions is a basic form of dynamic representation. Policy-makers must selectively assign
their attention across issues. There are prospective risks and rewards at the ballot box for
decision-makers who fail or succeed to recognize and respond to the concerns of the public.
When public concern about an issue increases it sends a signal to policy-makers that the issue
requires attention (and indeed may require policy action). Policy-makers must prioritize their
attention and policy action (e.g. Jones 1994; Jones and Baumgartner 2005a; 2005b) in light of
signals from the public about their ranking of the relative importance of issues. This may, in
turn, lead to representation of priorities in actual policy outputs. Policy-makers may therefore
use information about relative importance of an issue to respond to public opinion. A general
expectation concerning dynamic representation in the attention of governing institutions is
therefore that changes in the prioritization of issues by the public are transmitted into changes
in the policy agenda of governing institutions (H1). This suggests representation may occur
practice, tend to reflect a single underlying, dominant consideration, generated through a
probabilistic scanning of memory, structuring their concerns).
6
through the link between the policy priorities of citizens and the policy agenda pursued by
government.
H1: The issue priorities of the public are represented in the policy agenda of
governing institutions.
Information-Processing, Friction and Representation in Governing Institutions
While government incorporates information about public preferences in its activities
(e.g. Stimson et al. 1995; Wlezien 1996; Soroka and Wlezien 2010; Erikson et al. 2002;
MacDonald and Budge 2005), it faces an abundance of information about the state of the
world but possesses limited time and resources to process it (Simon 1971). It is attention
which is most scarce for policy-makers in this information-rich world (Jones 1994; Jones and
Baumgartner 2005a; 2005b). Executives and legislatures have to juggle a range of concerns
and policies, selecting a portfolio of issues for attention in light of their problem status or
their political platforms, interest groups or the mass public. Some changes in policy are
therefore attributable to changes in priorities while others are due to changes in preferences
(Jones 1994, pp. 5-10). The implicit weighting of information signals about the relative
importance of issues creates inherent trade-offs in decision-making and leads to the
prioritization of certain issues above others. If policy-makers in the higher echelons of
government, such as in executives, legislatures and the judiciary, were able to attend to all
issues in isolation, then their decisions could be based on the processing of information about
preferences alone. However, search and cognitive costs are associated with the retrieval and
processing of information, meaning decision-makers are able to consider a small number of
alternatives (Padgett 1980). Further, think tanks, interest groups, government agencies, expert
communities, the media, political parties and the public all engage in the supply of
information to the policy-making process. Faced with a multitude of competing issues and
demands policy-makers must decide which issues are most urgent and important to them,
7
prioritizing a selection of them for their attention. In conjunction with this, some institutional
features of decision-making exert resistance, i.e. friction, against change (Jones and
Baumgartner 2005a; 2005b).3 Such institutional friction is inbuilt in the design of electoral
systems, the separation or fusion of powers, majority voting rules, formal processes of
budgeting and other features of institutional mechanisms. This friction creates a threshold
that must be exceeded before information signals (i.e. pressure for change) are responded to.
It is attention-shifting and institutional friction, rather than the representation of preferences,
which explains how “…[s]ome aspects of the world are unmonitored, unattended to; [while]
other aspects are incorporated into the decision process beyond their intrinsic merit” (Jones
and Baumgartner 2005b, p. 334). In view of this, it is important to explain how attention
scarcity and friction in governing institutions structure the dynamic representation of public
priorities.
i) Attention-Scarcity:
The finite nature of attention, the abundance of information and the importance of
prioritization in decision-making structure the responsiveness of governing institutions to
public opinion. Governing institutions can communicate their policy priorities through
multiple policy agendas and outputs. Congress, for example, can debate issues, pass bills, set
appropriations, approve appointments, conduct formal oversight and hearings and impeach
3
This idea of institutional friction is consistent with Hobolt and Klemmensen's (2008)
distinction between 'rhetorical' and 'effective' responsiveness, where the use of mediums such
as speeches and manifestos to communicate the policy programs of parties (e.g. manifestos),
executives or legislators is differentiated from the outputs of government such as spending.
There is substantial evidence that such statements of policy intentions are transmitted into
outputs (Hofferbert and Budge 1990; 1992; Bara 2005; Bevan et al. 2011).
8
and remove executive and judicial officers. The issue priorities of the public provide a basis
on which governments can structure their attention. Within some institutional settings such as
legislatures there are practical limits on the number of issues that policy-makers can attend to
at a given time. In a formal statement or speech, such as the State of the Union Addresses, the
speaker must get to the point and emphasize their main message, communicating priorities in
clear terms. The timetables of legislatures, too, are finite in their capacity to debate and pass
legislation and conduct hearings or investigations, requiring implicit choices about issues of
importance. In other settings such as public expenditure, decision-makers are under less
pressure to prioritize their commitments between issues with most re-appropriation decisions
being a question of absolute levels rather than whether or not to spend money on a particular
issue, making it less important for them to respond to the priorities of the public. It makes
sense then, that government exhibits differing rates of responsiveness depending upon the
institutional setting (Jones 1994; Jones et al. 2009), since this reflects variation in the degree
of attention-scarcity.
ii) Institutional Friction:
Political institutions impose decision and transaction costs on collective action and
bargaining: that is, potential blocks or veto points in the legislative and executive branches
exert friction against change (Jones et al. 2003; Jones and Baumgartner 2005a; 2005b). When
change does occur, it must overcome the resistance that has built up. Such institutional costs
determine the extent to which decision-makers can make adjustments to policy in response to
new information, for example gaining a place for their proposals on the legislative timetable
or securing majorities in support of bills.
There is strong evidence that friction produces distributions of change in the policy
agendas and outputs of political institutions that are leptokurtic, characterized by extended
periods of stability and stickiness, a relatively low frequency of moderate adjustments, and a
9
disproportionate number of extreme shifts (see Jones et al. 2003). For example, budgeting is
associated with higher levels of friction due to a combination of cognitive and institutional
costs that increase resistance to change. This determines the capacity of decision-makers to
respond to information signals from the wider political environment. The idea of institutional
friction is that low frequencies of the information signal induce a minimal response, whereas
strong signals produce a disproportionate reaction. For the dynamic representation of political
attention, this suggests that the level of institutional friction must be relatively low for there
to be a smooth and continuous adjustment of the equilibrium relationship between governing
agendas and the priorities of the public. As friction increases there is an accumulation of
equilibrium errors, giving rise to an iterative process of under- and over-reaction (Jennings
and John 2009, p. 841).
There should therefore be institutional variation in the responsiveness of policymakers to the priorities of the public due to differences in the pressure for prioritization in
governing institutions. When attention is most scarce, government must prioritize between
multiple competing issues on the agenda. In government activities where resources and time
are less scarce and there is less direct competition between issues, such as public
expenditures, the relative prioritization of attention is of less consequence. At the same time,
institutional rules and procedures can exert friction against shifts in attention. It therefore
follows that there will be institutional variation in the responsiveness of government to the
issue priorities of the public. This should be observable through the comparison of crossinstitutional and cross-national patterns of representation.
H2: The degree of representation of public priorities in the policy agendas of
governing institutions is higher when decision-making is subject to lower institutional
friction and a greater scarcity of attention.
10
Our model of dynamic representation in the policy agenda of governing institutions can
therefore be summarized as follows:
a) Both the public and policy-makers are engaged in prioritization of information
signals about a range of issues and concerns, selecting some for attention and
disregarding others.
b) Decision-makers in governing institutions have electoral incentives to respond to
the priorities of the public, due to the mechanism of elections for the selection of
representatives and due to the rational anticipation of future electoral outcomes.
Dynamic representation can therefore occur through the continuous adjustment of
policy agendas in response to public priorities (that is, rational anticipation) or
indirect representation through the selection of representatives through elections.
c) While governing institutions provide a means for decision-makers to represent the
concerns of citizens, these institutions are subject to scarcity of attention as well
as friction against policy change. This reduces the capacity of decision-makers to
respond to priorities of the public in certain institutional settings, leading to crossinstitutional and cross-national variation in dynamic representation through policy
agendas.
This model of dynamic representation in the policy agendas of governing institutions is tested
in this paper through the comparison of the responsiveness of agendas to the issue priorities
of the public across a number of institutional venues. It is expected that the issue priorities of
the public will have an effect on the policy content of governing agendas, and that effect will
vary across institutional settings according to scarcity of attention and levels of institutional
friction that create resistance against policy change. The following section describes the data
used to test this model.
11
2. Data
This paper considers the effects of issue priorities of the public on policy agendas
across institutional venues in the US and UK. These countries provide a classic comparative
design in the contrast between governing institutions – between federal-presidentialism in the
US and unitary-parliamentarism in the UK. The design of governing institutions produces
cross-national variation in degrees of democratic responsiveness through its effects on the
clarity of responsibility for policy decisions (Lewis-Beck 1988; Powell and Whitten 1993;
Soroka and Wlezien 2010). At the same time, the internal structure of institutional variation
should be consistent across countries – as responsiveness to the issue priorities of the public
might be expected to differ across institutional settings as a function of institutional friction
and the scarcity of attention. Similarities are therefore expected between countries in the
relative degree of responsiveness in governing agendas that is observed for different types of
institutional venue.
This analysis is based on data on public opinion, executive speeches, law-making and
budgetary expenditure in the US and UK from 1951 to 2003, coded according to the policy
content coding system of the Policy Agendas Project (www.policyagendas.org). This coding
system consists of categories for major topics of public policy, such as macroeconomics,
defense and health. The sixteen categories investigated in this paper are reported in Table 1,
with those topics analyzed in the analysis of budgetary expenditure marked with an asterisk.4
The advantages of this coding framework are twofold: firstly, it is an established method for
coding government and public attention, and secondly it renders the content of governing
agendas comparable across venues and across countries (see www.comparativeagendas.org).
4
Due to the official government categories budgetary expenditure has been classified
according to not all policy topics are covered by the historical record, producing the need for
this particular restriction.
12
[insert Table 1 here]
Public Priorities
The issues priorities of the public are often measured with survey instruments that ask
about the ‘most important problem’ (MIP) facing the nation (see Wlezien 2005; Jennings and
Wlezien 2011). Following previous studies (e.g. Jones 1994; Soroka 2002), we use aggregate
MIP responses to represent the broader public prioritization of particular issues at particular
points over time at the macro-level. Gallup first asked the question in the US in 1935 and in
the UK in 1947. MIP data is not available in the UK after 2001, when Gallup ceased political
polling in the UK. Since 1977, however, Ipsos-MORI has asked a similar question about the
‘most important issue’ (MII) which enables a continuous measure of the issue priorities of the
public.5 The MIP and MII questions have been shown to exhibit a high degree of common
variance and both provide a comparable indication of the issues that are on people’s minds
(Jennings and Wlezien 2011) and, therefore, the degree to which the public prioritizes certain
issues above others. The MIP and MII categories are each recoded to correspond to the Policy
Agendas Project major topic codes.6
5
To measure issue priorities of the public in the UK, data from both Gallup’s MIP and Ipsos-
MORI’s MII is used for these analyses. To accomplish this, the two series have been
combined and averaged for 19 years, from 1982 to 2000, when there is regular overlapping
data. This extends the duration of the data series.
6
Because the Ipsos-MORI MII data used in addition to the Gallup MIP data combines
defense and foreign affairs, our analysis aggregates the Defense (16) and International Affairs
and Foreign Aid (19) topics to ensure that the data is comparable over time and across
countries. This combination generally produced better models and greater responsiveness
than testing these two topics alone. Further, the analysis excludes two topics on which MIP
13
Policy Agendas in Governing Institutions
The governing agendas considered in this analysis are drawn from three institutional
venues in the UK and the US, summarized in Table 2. These provide measures of executive
agendas, legislative outputs and budgetary expenditure for each of the countries at the
national level.
[insert Table 2 here]
Executive Speeches: The State of the Union and the Speech from the Throne
In many political systems the head of state or the head of government delivers an
annual formal statement setting out its policy priorities for the year ahead. These speeches are
forward-looking, communicating general priorities and specific measures that the executive
intends to address in the next year. This substantive function of executive speeches is
reflected in their transmission into policy outcomes (e.g. Bara 2005; Bevan et al. 2011).
The State of the Union Address in the US and the Speech from the Throne7 in the UK
are prominent annual speeches that communicate the governing agenda of the executive. The
policy content of these speeches were divided into quasi-sentences,8 with each quasi-sentence
assigned a single topic code. Because of the timing of each speech (which occurs in January
and MII responses were extremely low in both countries: Banking, Finance, and Domestic
Commerce (15) and Space, Science, Technology and Communications (17).
7
Otherwise known as the King’s or the Queen’s Speech (see Jennings et al. 2011).
8
A quasi-sentence constitutes an expression of a single policy idea or issue (Volkens 2002).
Often this unit of analysis is identifiable from the use of punctuation, though it is possible for
sentences to include multiple references to policy content, in particular sentences which
address a series of major policy issues in a list.
14
in the US and at the start of the parliamentary session in the UK which can occur throughout
the year), the executive agenda is organized by calendar year in the US and by parliamentary
session in the UK. This temporal aggregation is also used for legislative outputs as well as for
public opinion corresponding to the executive and legislative agendas.
Legislative Outputs: Statutes of the US Congress and Acts of the UK Parliament
Law-making provides opportunities both for agenda-setting and the signalling of the
priorities of policy-makers (e.g. Mayhew 1974; Schiller 1995) as well as for the enactment of
substantive policy outputs. Legislative measures are often intended to enact major changes
and can be high profile and observable to the public as they pass through the legislature.
Statutes of the US Congress and Acts of the UK Parliament are the major legislative outputs
that are considered here. Each law is coded with a single topic which indicates the primary
focus of the legislation. The observed time point is the date upon which the bill was signed
into law.9 Commemorative bills are excluded from the US data so it is directly comparable to
the UK data, only containing legislative outputs intended to affect the functional workings of
government.
Budgets: US Federal Budget and UK Government Expenditure
Whereas the executive and legislative agendas are process measures, public budgeting
is a substantive output of government (Wildavsky 1964) affecting the allocation of financial
resources and, in turn, public welfare. Budgeting involves competition over scarce resources,
leading to trade-offs between functional categories. This gives rise to incremental patterns of
9
In the US, the formal process of passing a law is not complete until it has been signed.
While in the UK, the process functionally ends after the vote following the third reading of a
bill. However, all acts in a parliamentary session receive royal assent prior to the start of a
new parliamentary session, which is marked by the Speech from the Throne.
15
decision-making as financial commitments linked to statutes or government agencies become
institutionalized once established, interspersed with occasional large and dramatic shifts as
pressure for change overcomes cognitive and institutional friction (e.g. Padgett 1980; Jones
and Baumgartner 2005a; 2005b; Jones et al. 2003; Baumgartner et al. 2006; 2009; Breeman
et al. 2009; Breunig 2006). The US federal budget and UK central government expenditure
are important from the perspective of the dynamic representation of preferences (e.g. Soroka
and Wlezien 2005; 2010), but are expected to be least responsive to the issue priorities of the
public of all the institutional venues considered here. This is because budgeting enables the
parallel-processing of large numbers of issues and faces fewer trade-offs than either executive
speeches or legislative outputs in the allocation of attention. This might be attenuated because
the MIP measure of public priorities captures the single most important problem to the public
(i.e. it is the plurality winner of important problems). Further, the functional classification of
budgets in both the UK and the US presents a modelling limitation because many issues that
are important to the public and receive attention in executive speeches or are subject to
legislation do not receive funding from national government (or are absorbed into other
functional categories). For this reason, the analyses of responsiveness of budgets to public
priorities are limited to a subset of seven topics from the total of sixteen.
3. Analysis
Friction in Governing Institutions
To test the second hypothesis (H2) concerning the effect of institutions on the
dynamic representation of public priorities it is necessary to first assess the degree of friction
present in each of the policy agendas and to compare processes along the policy cycle within
each country. It is expected that those government processes subject to higher decision costs
also tend to be associated with to higher levels of friction (e.g. Baumgartner et al. 2009). To
ascertain the degree to which decision-making within each of these governing institutions is
16
subject to friction against change we use stochastic process methods (following Breunig and
Jones 2011; Jones and Baumgartner 2005a; 2005b; Jones et al. 2003). This tests the normality
of the distribution of values of the dependent variables for this analysis (i.e. the policy
content of executive speeches, legislation and budgetary expenditure). With the exception of
budgets, the “percentage-percentage” calculation method is used to analyse the overall
distribution of (percentage) change of the difference between (percentage) agenda share in
one year and the next (see Baumgartner et al. 2009, p. 610). Budgetary spending is not treated
as bounded in the same way (i.e. as a percentage) because the numbers involved are much
larger and because the values are reported in real prices, which removes variation due to
inflation. Note that for the executive and legislative agendas, cases in which attention to a
particular topic remains stable at zero are treated as missing to avoid the over-inflation of the
kurtosis scores (which could otherwise lead to the false acceptance of hypotheses) due to the
empirical redundancy of some topic codes under certain circumstances.
The distribution of year-on-year percentage changes is presented in Figure 1 for each
institutional venue in each country, with the kurtosis scores for each variable super-imposed
on each histogram. If the value of the kurtosis statistic is greater than three, the distribution
exhibits positive kurtosis and can be said to be leptokurtic, consistent with the presence of a
high degree of cognitive and/or institutional friction. In addition, kurtosis is tested for via the
Shapiro-Wilk test, which considers whether the sample is drawn from a normal distribution.
The results of the Shapiro-Wilk test are consistent with the kurtosis scores, providing further
validation of the analysis.
[insert Figure 1 here]
Overall, the results are consistent with theoretical expectations regarding the level of
friction in each of the governing institutions: the highest level of kurtosis in both countries is
17
observed for budgetary expenditure, consistent with a large number of studies that show that
public budgeting exhibits incrementalism, interspersed with occasional extreme disturbances
(e.g. Padgett 1980; Jones and Baumgartner 2005a; 2005b; Jones et al. 2003; Baumgartner et
al. 2006; 2009; Breeman et al. 2009; Breunig 2006). This is evidenced in the tall slender peak
and the fat tails of the distributions plotted in Figure 1. Indeed, all of the institutions exhibit
some evidence of disproportionate information-processing in the leptokurtic distributions of
attention change. For the UK, the pattern of institutional differences is clear, with the lowest
kurtosis observed for the executive agenda, the next lowest for the legislative agenda and the
highest for budgets. This is unsurprising in light of the incrementalist tendencies of decisionmakers and the numerous veto points in budgeting processes. In the US, however, the level of
kurtosis is in fact higher for the executive agenda than for legislative outputs. This however
reflects a greater degree of variation in how different presidents use the State of the Union
address. For example, the shortest State of the Union message, delivered by President Nixon
in 1973, contained just 36 policy-related statements, while the longest, presented by President
Carter in 1981, contained 1,336 policy mentions. Part of this variation is no doubt due to the
unique nature of these particular State of the Union messages, which were written instead of
being delivered orally. Indeed, Carter’s speech as outgoing president in particular is an
interesting case. Nevertheless, both these speeches are formally and historically recognized as
State of the Union Addresses and their inclusion is appropriate. However, the ‘friction’
observed in the analysis of the speech may be more representative of the uniqueness of
individual presidential character and speechwriters (Neustadt 1960) rather than institutional
friction. While the unique nature of the State of the Union Address puts the findings in Figure
1 somewhat at odds with expectations concerning the ordering of levels of institutional
friction, the results nevertheless confirm the presence of punctuations – in disproportionate
responses of the political system following the build-up of pressure against change,
18
overcoming the threshold of resistance. It is further expected that these patterns of
institutional friction structure the interaction of the priorities of the public and the policy
content of governing agendas.
Error-Correction Models of Dynamic Representation in Governing Agendas
To test the dynamic representation of the issue priorities of the public in governing
agendas, comparing both across institutional levels and across countries, time series errorcorrection models are now estimated according to policy topic. The responsiveness of policymakers to the issue priorities of the public can be understood in both contemporaneous (Jones
and Baumgartner 2004; Jones et al. 2009) and dynamic (Jennings and John 2009) forms. For
responsiveness to occur public opinion can either be coincident to (at time t) or immediately
precede policy (at time t-1). The use of an error-correction model (ECM) enables diagnosis of
both the short- and long-run effects of the issue priorities of the public on public policy.10 The
error-correction framework is selected in light of past studies which demonstrate that, in both
theory and practice, that agenda-opinion dynamics “… coexist in a long-run equilibrium state
that is subject to short-run corrections” (Jennings and John 2009, p. 838). In other words,
dynamic representation can arise from long-term trends in issue priorities of the public and
from short-run variation and shocks, i.e. events such as the global financial crisis or the swine
flu pandemic. An ECM framework is appropriate when testing for both contemporaneous and
lagged effects. The model can be represented in the form:
10
Note that ordinary least squares estimates are sensitive to departures from normality. Our
analysis of friction in governing institutions suggests that each of the policy agendas exhibit
positive kurtosis and while this does not produce bias in the coefficient estimates it can mean
that the standard errors are no longer efficient (so the p-values are biased downwards, i.e. this
imposes a stricter test upon our test of representation).
19
∆AGENDAt = α0* + α1*AGENDAt-1 + β0*∆OPINIONt + β1*OPINIONt-1 + β2*PARTYt + εt
where short-run changes in legislative outputs relating to a particular issue (∆AGENDAt) are
a function of short-run changes in the public’s prioritization of that same topic (∆OPINIONt),
the long run changes (OPINIONt-1), and where the lagged value of the dependent variable
(AGENDAt-1) measures the speed of re-equilibration (α1*) in response to shocks to the longrun legislative-opinion equilibrium. Consistent both with our theoretical expectations and
other models of dynamic representation (e.g. Wlezien 2004; Jennings and John 2009), this
model includes a variable (PARTYt) to capture the contemporaneous effects of indirect
representation through partisan control of government. This controls for difference in the
governing agendas of political parties, and is coded 1 for the Conservative Party in the UK
case and the Republican Party in the US case and is coded 0 for the Labour Party in the UK
and for the Democratic Party in the US.11
Within the ECM framework, changes in the policy content of governing agendas are
estimated as a function of contemporaneous changes in the issue priorities of the public and
the degree to which these are outside the long-run agenda/opinion equilibrium. This suggests
that if the governing agenda deviates from its long-run equilibrium, as the institution commits
either “too much” or “too little” attention to a particular issue, responsiveness is equal to the
degree of equilibration that restores correspondence between the agenda and public opinion
to its previous status quo.
11
For Statutes in the US and for US budgets the variable was coded according to which party
was the majority party in the House of Congress. Other measures of partisan control in the
US, such as Senate majority party and a variable measuring House, Senate and presidential
control were also tested and produce the same general inferences of the models using the
House control.
20
Representation, Agendas and Institutions
Pooled Analyses
To first assess the general pattern of dynamic representation in each of the governing
institutions, a time series cross-sectional ECM is estimated for the seven policy topics that are
common to each institution. This pooled model specification has the advantage of measuring
the degree to which the policy agenda of each governing institution is representative, overall,
of public priorities.12 The results that are reported in Table 3 show rather limited evidence of
dynamic representation: the long-run (lagged) effect of the issue priorities of the public is
positive and significant, at the 95 per cent confidence level, for the executive agenda in the
US and for the legislative outputs in the US and the UK, while the short-run effect of public
priorities is significant for legislative outputs in the UK. Further, there is no evidence of
representation in the link between public priorities and changes in budgetary expenditures in
either the US or the UK. In addition, the negative and significant error-correction parameter
(AGENDAt-1) for the executive and legislative agendas indicates that shocks to the long-run
equilibrium of the policy agenda are corrected over time. The absence of a similar effect for
budgetary expenditure (indeed, the parameter is positive and significant for the UK) suggests
that errors tend to accumulate. This is consistent with the high level of institutional friction in
budgeting compared to the other venues, with the build-up of errors contributing to
punctuations in policy change. The absence of a significant link between public priorities and
the executive agenda in the UK, in contrast, might be attributable to the lack of a general
12
Using a time series cross-sectional ECM framework, the model of dynamic representation
takes the general form: ∆AGENDAit = α0* + α1*AGENDAit-1 + β0*∆OPINIONit +
β1*OPINIONit-1 + β2*PARTYit.
21
pattern of representation, rather than suggesting that the policy agenda is unresponsive on all
topics.
[insert Table 3 about here]
While the use of a time series cross-sectional framework enables comparison of the
structure of representation across governing institutions and across countries, it is possible
that unresponsiveness on some issues might cancel out responsiveness on others, obscuring
important features of the underlying categories, i.e. the policy agenda might be responsive for
a number of the most salient topics (such as the economy and defense) but this would not
necessarily be reflected in the pooled analysis. Indeed, variation in rates of responsiveness for
individual issues is to be expected due to differences in the intrinsic importance of certain
issues to citizens (Page and Shapiro 1983; Jones 1994; see Burstein 2003 for a review) and
because government agendas possess policy ‘tools’ or instruments that are optimised to solve
certain types of problem (Hood 1983). Therefore, an issue-level analysis is needed to fully
test the dynamic interrelationship between public priorities and policy agendas. This analysis
therefore next turns to assess the dynamic representation of public priorities on specific issues
across each agenda.
Individual Topic Analyses
To assess issue-specific patterns of dynamic representation in each of the governing
institutions, seventy-eight13 individual ECMs are next estimated. Table 4 summarizes the
findings on the short-run and long-run effects of the issue priorities of the public on the
policy content each of the governing agendas (executive speeches, legislative outputs and
13
This figure includes sixteen models for the executive and the legislative agenda and seven
models for the budgetary agenda in each of the two countries (16 + 16 + 7) * 2 = 78.
22
budgetary expenditure), in each of the countries, by presenting the coefficient estimates for
each topic (the full results are reported in the Online Appendix, Tables A1 to A6). The topics
are reported in the first column and the responsiveness coefficients for each of the governing
agendas are presented in turn across the columns of Table 4, enabling comparison across the
two countries and across institutions. For each topic, the direction, size and significance of
the responsiveness of governing agendas is measured with the short- and long-run effects of
public issue priorities. The final three rows of Table 4 further indicate the general pattern of
dynamic representation, summarizing the total number of topics for which the issue priorities
of the public have a positive short-run, long-run or either effect on the policy content of each
institutional agenda.
[insert Table 4 about here]
These results show that the policy agendas of governing institutions in the US and the
UK exhibit a substantial degree of responsiveness to the issue priorities of the public over the
period between 1951 and 2003 (H1). Specifically, statistically significant short- or long-run
effects of the issue priorities of the public on governing agendas are observed in four or more
of the sixteen topics for the executive agenda and legislative outputs. In the US, the executive
agenda is responsive to public priorities for seven out of the sixteen topics (health,
agriculture, education, environment, energy, law and order and social welfare) while in the
UK it is also responsive for six (macroeconomic issues, health, education, environment, law
and order and housing). In the US, legislative outputs are responsive for five of the sixteen
topics (macroeconomic issues, environment, energy, social welfare and defence and foreign
affairs) and in the UK for four (macroeconomic issues, environment, public lands and
territorial issues and defence and foreign affairs). There are no significant effects, in contrast
to these findings, for budgetary expenditure in any of the seven topics tested here. This is
23
consistent with the theoretical expectation that a decision-making venue like budgeting which
lacks the same pressure for serial-processing and the prioritization of issues is less likely to
be responsive to information signals about public priorities. Further, in a few cases across the
executive, legislative and budgetary agendas the coefficient estimates are negative and
significant suggesting that change in the policy agenda leads public concern about that issue.
For example, a short-run increase in US spending on education precedes a decrease in the
public’s prioritization of the same issue. It is therefore possible for governing agendas to
shape the issue priorities of the public. (This is consistent with Wlezien’s thermostatic model
of public preferences, as well as with Page and Shapiro’s findings, since an increase in the
attention of government to an issue leads the public to downgrade its concern about that
issue.) Overall, these results provide strong evidence of a link between the issue priorities of
the public and the policy agenda of governing institutions in both countries (H1).
Of all governing institutions, the issue priorities of the public are found to have the
strongest effect, on average, on the executive agenda in terms of the frequency of statistical
significance of short- and long-run effects. The results do not enable direct comparison of the
level of responsiveness either between countries or institutions because the measure of the
policy agenda of each governing institution uses a different unit of analysis (i.e. speeches,
legislation and expenditure) and the mean level varies between countries. For example, in
both countries the executive agenda is measured by units of text (i.e. quasi-sentences) within
speeches while legislative outputs are measured by individual statutes. Further, the volume of
statutes passed in the US far exceeds the volume of acts passed in the UK.14
14
Although a proportionalized modeling strategy might seem a solution to this problem,
because these agendas are substantively different across institutions in the unit of analysis,
this would only mask the underlying issue.
24
The results reported at the bottom of Table 4 indicate that the issue priorities of the
public have a positive and significant effect at the 95 per cent confidence level on the policy
content of executive speeches in seven out of the sixteen topics in both the US and the UK.
The next strongest link between public priorities and governing agendas is found for
legislative outputs. The issue priorities of the public have a positive and significant effect on
the legislative agenda in five topics in the US and four topics in the UK. Lastly, there is no
positive and significant effect of the issue priorities of the public on budgetary expenditure in
either the US or the UK for any of the seven topics tested. These findings provide evidence
consistent with the link between public priorities and governing agendas (H1), subject to
variation in the short- and long-run, across institutions and across countries. Further, the
higher levels of dynamic representation observed in each of the countries for executive
speeches followed by legislation, in comparison to budgetary spending, indicate variation
between institutional settings, consistent with theoretical expectations (H2) and the pattern of
friction reported earlier. The responsiveness of governing agendas to the issue priorities of
the public is highest in institutional settings where attention is most scarce (executive
speeches), which requires that decision-makers prioritize between issues, and is lowest in the
institutional venue subject to the highest degree of friction (budgetary expenditure).
While there is variation at the topic level, there are many similarities between the US
and the UK in the responsiveness of governing institutions to the issue priorities of the public.
For example, there is no opinion-responsiveness in the executive agenda or budgetary
spending for defence and foreign affairs in either the US or the UK while there is evidence of
responsiveness in legislative outputs in both countries in the short- and the long-run (with the
long-run effects in the UK being significant at the 90 per cent confidence level). For law and
order on the other hand, the executive agenda is responsive to public priorities in the long-run
in both the US and the UK (as well as being responsive in the short-run in the US), while
25
public priorities do not have a significant effect on legislative outputs for the same topic in
either country. There is nevertheless variation in responsiveness across institutions between
the US and the UK that is likely driven by differences in the level of importance of issues to
citizens (Page and Shapiro 1983; Jones 1994; also see Burstein 2003) and due to the ability of
particular institutional agendas to better address certain issues (Hood 1983). For example, the
executive agenda is responsive to public opinion for macroeconomic issues in the long-run in
the UK, but not in the US. Legislative outputs are responsive, however, in the short-run in the
US and in the long-run in the UK (the short-run effect is significant at the 90 per cent
confidence level in the UK). Institutional variation therefore structures the overall pattern of
representation (H2), where governing agendas in which the attention of decision-makers is
most scarce or subject to lower levels of friction tend to be more responsive, but also leads to
variation in dynamic representation across issues.
The non-responsiveness of budgetary expenditure is an interesting finding that further
confirms previous evidence that spending is not responsive to public concern about the “most
important problem”, in contrast to relative preferences (Wlezien 2005). This is perhaps
because whereas budgets have directional implications, and the public can prefer either more
or less spending in a particular policy domain, changes in the issue priorities of the public do
not signal the desired direction of change. For example, the issue of healthcare might be
highly salient to the public either because the government is spending too much or too little
on it. The issue priorities of the public are, however, a useful information signal for valence
issues; i.e. those issues on which there is broad consensus over ends, such as lower crime or
economic growth (Stokes 1963).
4. Conclusion
The sorting of issues for attention by political institutions is an important mechanism
through which dynamic representation occurs. The transmission of the priorities of the public
26
into the priorities of government is mediated through institutions – due to institutional friction
in decision-making or the scarcity of attention. This comparative analysis provides extensive
evidence of the effect of the issue priorities of the public on the policy agenda of governing
institutions in the US and the UK over more than half a century. There is wide-ranging
responsiveness of the executive and legislative agenda to public priorities in both the US and
UK across a range of issues in the short- or the long-run. There is no similar opinionresponsiveness with respect to budgetary expenditures. The pattern of dynamic representation
across governing institutions is consistent within countries – supporting the theoretical
expectation of variation across institutional settings due to attention scarcity and institutional
friction. The degree of responsiveness is greater in policy agendas when attention is most
scarce (i.e. speeches) and declines as decision-makers’ attention becomes less scarce, and less
concerned with agenda-setting and the signalling of policy priorities, and is subject to higher
levels of cognitive and institutional friction (i.e. expenditure).
There are similarities in the pattern of dynamic representation of public priorities
observed within countries – as executive speeches are the most responsive institutional venue
and legislative outputs the next most responsive venue, with no responsiveness observed for
budgeting. In contrast, budgetary expenditure is not found to be responsive to the issue
priorities of the public in the same way.15 This finding differs from studies that demonstrate
the effect of public preferences on budgetary spending (e.g. Ostrom and Marra 1986; Wlezien
1995; Soroka and Wlezien 2010). Budgets have distributive implications and changes in the
issue priorities of the public do not signal the desired direction of change, unlike preferences.
Nevertheless, it might be expected that an increase in the public’s prioritization of a particular
15
Note that our inferences regarding budgets are limited to the seven policy topics on which
data is available.
27
issue might be associated with a corresponding preference for increased spending. The results
here suggest that this is not the case.
This paper has provided comparative analyses of the dynamic representation of issue
priorities of the public in multiple policy agendas across sixteen policy topics in the US and
the UK. It reveals cross-national similarities in the pattern of responsiveness to priorities.
Systematic differences in responsiveness between institutional venues are consistent with the
institutional scarcity of attention and with cognitive and institutional friction in decisionmaking. Such variation occurs because government attends to a multitude of issues across a
number of different institutional venues and must prioritize information concerning the issue
priorities of the public. Further, the ability and willingness of policy-makers to respond to
public priorities varies according to institutional venue, due to the scarcity of attention and
the capacity of decision-makers to simultaneously handle multiple issues competing for space
on the agenda.
Processes of dynamic representation can be understood through transmission of the
priorities of the public onto the policy priorities of decision-makers. Theories of opinionresponsiveness should recognize that policy-makers prioritize their attention to issues in light
of signals from the public about their relative importance. Variation in patterns of dynamic
representation between policy agendas suggests that the institutional scarcity of attention and
friction in decision-making structures the degree of responsiveness to the issue priorities of
the public. Further investigation is required to determine whether these patterns of dynamic
representation in policy agendas are replicated elsewhere.
28
References
Adler, E. Scott, and John D. Wilkerson. (2012). Congress and the Politics of Problem
Solving. Cambridge: Cambridge University Press.
Bara, Judith. (2005). ‘A Question of Trust: Implementing Party Manifestos.’ Parliamentary
Affairs 58(3): 585-599.
Bartle, John, Sebastian Dellepiane-Avellaneda, and James A. Stimson. (2011). ‘The Moving
Centre: Preferences for Government Activity in Britain, 1950-2005.’ British Journal of
Political Science 41(2): 259-285.
Baumgartner, Frank R., Christian Breunig, Christoffer Green-Pedersen, Bryan D. Jones, Peter
B. Mortensen, Michiel Neytemans and Stefaan Walgrave. (2009). ‘Punctuated Equilibrium
in Comparative Perspective.’ American Journal of Political Science 53(3): 602-619.
Baumgartner, Frank R., Martial Foucault, and Abel François. (2006). ‘Punctuated
Equilibrium in French Budgeting Processes.’ Journal of European Public Policy 13(7):
1082-1099.
Bevan, Shaun, Peter John and Will Jennings. (2011) ‘Keeping Party Programmes on Track:
The Transmission of the Policy Agendas of the Speech from the Throne to Acts of the UK
Parliament’, European Political Science Review 3(3): 395-417.
Breeman, Gerard, David Lowery, Caelesta Poppelaars, Sandra L. Resodihardjo, Arco
Timmermans and Jouke de Vries. (2009). ‘Political Attention in a Coalition System:
Analyzing Queen’s Speeches in the Netherlands 1945-2007.’ Acta Politica 44(1): 1-27.
Breunig, Christian. (2006). ‘The More Things Change, the More Things Stay the Same: A
Comparative Analysis of Budget Punctuations.’ Journal of European Public Policy 13(7):
1069-1085.
Breunig, Christian. (2011). ‘Reduction, Stasis, and Expansion of Budgets in Advanced
Democracies.’ Comparative Political Studies 44(8): 1060-1088.
29
Breunig, Christian, and Bryan D. Jones. (2011). ‘Stochastic Process Methods with an
Application to Budgetary Data.’ Political Analysis 19(1): 103-117.
Breunig, Christian, Chris Koski and Peter B. Mortensen. (2010). ‘Stability and Punctuations
in Public Spending: A Comparative Study of Budget Functions.’ Journal of Public
Administration Research and Theory 20(3): 703-722.
Burstein, Paul. (2003). ‘The Impact of Public Opinion on Public Policy: A Review and an
Agenda.’ Political Research Quarterly 56(1): 29-40.
Chaqués Bonafont, Laura, and Anna M. Palau. (2011). ‘Assessing the Responsiveness of
Spanish Policymakers to the Priorities of their Citizens.’ West European Politics 34(4):
706-730.
Cohen, Jeffrey E. (1995). ‘Presidential Rhetoric and the Public Agenda.’ American Journal of
Political Science 39(1): 87-107.
Erikson, Robert S., Michael B. MacKuen and James A. Stimson. (2002). The Macro Polity.
Cambridge: Cambridge University Press.
Hakhverdian, Armen. (2010). ‘Political Representation and its Mechanisms: A Dynamic
Left–Right Approach for the United Kingdom, 1976–2006.’ British Journal of Political
Science 40(4): 835-856.
Hobolt, Sara B., and Robert Klemmensen. (2005). ‘Responsive Government? Public Opinion
and Government Policy Preferences in Britain and Denmark.’ Political Studies 53(2): 379402.
Hobolt, Sara B., and Robert Klemmensen. (2008). ‘Government Responsiveness and Political
Competition in Comparative Perspective.’ Comparative Political Studies 41(3): 309-337.
Hood, Christopher. (1983). The Tools of Government. London: Macmillan.
30
Jennings, Will, and Peter John. (2009). ‘The Dynamics of Political Attention: Public Opinion
and the Queen’s Speech in the United Kingdom.’ American Journal of Political Science
53(4): 838-854.
Jennings, Will, Shaun Bevan, Arco Timmermans, Laura Chaques, Gerard Breeman, Sylvain
Brouard, Christoffer Green-Pedersen, Peter John, Peter B. Mortensen and Anna Palau.
(2011). ‘Effects of the Core Functions of Government on the Diversity of Executive
Agendas.’ Comparative Political Studies 44(8): 1001-1030.
Jennings, Will, Shaun Bevan and Peter John. (2011). ‘The Agenda of British Government:
the Speech from the Throne, 1911-2008.’ Political Studies 59(1): 74-98.
Jennings, Will, and Christopher Wlezien. (2011). ‘Distinguishing between Most Important
Problems and Issues?’ Public Opinion Quarterly 75(3): 545–555..
Jones, Bryan. (1994). Reconceiving Decision-making in Democratic Politics: Attention,
Choice, and Public Policy. Chicago: University of Chicago Press.
Jones, Bryan. (2001). Politics and the Architecture of Choice. Chicago: University of
Chicago Press.
Jones, Bryan D., and Frank R. Baumgartner. (2004). ‘Representation and Agenda Setting.’
Policy Studies Journal 32(1): 1-24.
Jones, Bryan D., and Frank R. Baumgartner. (2005a). The Politics of Attention: How
Government Prioritizes Problems. Chicago: University of Chicago Press.
Jones, Bryan D., and Frank R. Baumgartner. (2005b). ‘A Model of Choice for Public Policy.’
Journal of Public Administration Research and Theory 15(3): 325-351.
Jones, Bryan D., Heather Larsen-Price, and John Wilkerson. (2009b). ‘Representation and
American Governing Institutions.’ Journal of Politics 71: 277-290.
Jones, Bryan D., Tracy Sulkin, and Heather Larsen. (2003). ‘Policy Punctuations in American
Political Institutions.’ American Political Science Review 97(1): 151-169.
31
Key, V.O., Jr. (1961). Public Opinion and American Democracy. New York: Knopf.
Lewis-Beck, Michael. (1988). Economics and Elections: The Major Western Democracies.
Ann Arbor: University of Michigan Press.
Lindeboom, Gert-Jan. (2012). ‘Public priorities in government's hands: Corresponding policy
agendas in the Netherlands?’ Acta Politica doi: 10.1057/ap.2012.14.
McDonald, Michael D., and Ian Budge. (2005). Elections, Parties, Democracy. Oxford:
Oxford University Press.
May, Peter J., Samuel Workman and Bryan D. Jones. (2008). ‘Organizing Attention:
Responses of the Bureaucracy to Agenda Disruption.’ Journal of Public Administration
Research and Theory 18 (4): 517-541.
Mayhew, David. (1974). Congress: The Electoral Connection. New Haven: Yale University
Press.
Neustadt, Richard. (1960). Presidential Power: The Politics of Leadership. New York: John
Wiley & Sons.
Ostrom Jr., Charles W. and Robin F. Marra. (1986). ‘U.S. Defense Spending and the Soviet
Estimate.’ American Political Science Review 80(3): 819-842.
Page, Benjamin I. and Robert Y. Shapiro. (1983). ‘Effects of public opinion on policy.’
American Political Science Review 77: 175-190.
Powell, G. Bingham, and Guy D. Whitten. (1993). ‘A Cross-National Analysis of Economic
Voting: taking account of the political context.’ American Journal of Political Science 37:
391-414.
Schattschneider, Elmer E. (1960). The semi-sovereign people: A realist’s view of democracy
in America. New York, NY: Holt, Rinehart & Winston.
Schiller, Wendy. (1995). ‘Senators as Political Entrepreneurs: Using Bill Sponsorship to
Shape Legislative Agendas.’ American Journal of Political Science, 39(1): 186-203.
32
Simon, Herbert A. (1971). ‘Designing Organizations for an Information-Rich World.’ In
Computers, Communication, and the Public Interest. ed. Martin Greenberger. Baltimore,
MD: Johns Hopkins Press.
Soroka, Stuart N. (2002). Agenda-Setting Dynamics in Canada. University of British
Columbia Press, Vancouver.
Soroka, Stuart and Christopher Wlezien. (2005). ‘Opinion-Policy Dynamics: public
preferences and public expenditure in the United Kingdom.’ British Journal of Political
Science, 35: 665-689.
Soroka, Stuart and Christopher Wlezien. (2010). Politics, Public Opinion, and Policy.
Cambridge: Cambridge University Press.
Stimson, James A., MacKuen, Michael B. and Erikson, Robert S. (1995). ‘Dynamic
Representation.’ American Political Science Review 89: 543-565.
Stokes, Donald. (1963). ‘Spatial Models and Party Competition.’ American Political Science
Review 57(2): 368–377.
Walgrave, Stefaan, and Michiel Nuytemans. (2009). ‘Friction and Party Manifesto Change in
25 countries (1945-1998).’ American Journal of Political Science 53 (1): 190-206.
Wildavsky, Aaron. (1964). The Politics of the Budgetary Process. Boston: Little Brown.
Wlezien, Christopher. (1995). ‘The Public as Thermostat: dynamics of preferences for
spending.’ American Journal of Political Science, 39: 981–1000.
Wlezien, Christopher. (1996). ‘Dynamics of Representation: The case of U.S. spending on
defense.’ British Journal of Political Science 26: 81-103.
Wlezien, Christopher. (2004). ‘Patterns of Representation: Dynamics of Public Preferences
and Policy.’ The Journal of Politics 66(1): 1-24.
Wlezien, Christopher. (2005). ‘On the Salience of Political Issues: The problem with ‘most
important problem’.’ Electoral Studies 24(4): 555-79.
33
Figure 1. Change Distributions in Executive, Legislative and Budgeting Agendas in the US and the UK, 1951 – 2003
US Executive
US Legislation
Mean: 60.57
SD: 295.57
Kurtosis: 35.70
SW Test: 12.91***
150
100
50
Mean: 19.87
SD: 97.36
Kurtosis: 18.19
SW Test: 12.10***
150
100
50
0
300
500
700
900
Mean: 6.13
SD: 59.79
Kurtosis: 87.67
SW Test: 13.64***
150
100
50
0
-100 0 100
0
-100 0 100
300
500
700
900
-100 0 100
300
500
700
Percentage change
Percentage change
Percentage change
UK Executive
UK Legislation
UK Budget
Mean: 11.06
SD: 94.57
Kurtosis: 8.87
SW Test: 10.50***
150
100
50
Mean: 16.55
SD: 116.13
Kurtosis: 27.33
SW Test: 11.58***
150
100
50
0
300
500
700
Percentage change
900
Mean: 1.74
SD: 23.60
Kurtosis: 59.15
SW Test: 12.11***
150
100
50
0
-100 0 100
900
200
Frequency
200
Frequency
200
Frequency
200
Frequency
200
Frequency
Frequency
200
US Budget
0
-100 0 100
300
500
700
Percentage change
900
-100 0 100
300
500
700
900
Percentage change
34
Table 1. Policy Agendas Project Major Topic Codes
Topic
1
2
3
4
5
6
7
8
10
12
13
14
18
20
21
16/19
Abbreviation
Economy
Civil
Health
Agriculture
Labor
Education
Environment
Energy
Transport
Law
Social
Housing
Trade
Gov't
Lands
Foreign
Name
Macroeconomics*
Civil Rights, Minority Issues, and Civil Liberties
Health*
Agriculture
Labor, Employment, and Immigration
Education*
Environment
Energy
Transportation
Law, Crime, and Family Issues*
Social Welfare*
Community Development, Planning and Housing Issues*
Foreign Trade
Government Operations
Public Lands and Water Management
Defense, International Affairs and Foreign Aid*
Note: * Indicates a Major Topic that is tested for budgetary expenditure in this paper.
Policy Agendas Topic Codebook, see www.policyagendas.org
35
Table 2. Government Agendas by Country and Venue
United States
United Kingdom
Executive
State of the Union
Speech from the Throne
Legislative
Statutes of the US Congress
Acts of the UK Parliament
Budgetary
The US Federal Budget
UK Government Expenditure
36
Table 3. Time Series Cross-Sectional Model of Public Priorities and Government
Agendas
∆Agendat
Executive
US
∆ Opiniont
Opiniont-1
Partyt
Constant
R2
UK
US
0.233
-0.053
0.077
(0.321)
(0.033)
(0.087)
0.890***
(0.171)
Agendat-1
Legislation
-0.864***
0.020
(0.015)
-0.518***
0.228***
UK
0.072**
258.254
39.629
(202.967)
(31.735)
96.024
-14.862
(105.212)
(14.928)
(0.027)
0.048***
-0.622***
-0.994***
(0.053)
(0.050)
-2.186
-0.468
-6.013***
(3.472)
(0.331)
(1.287)
3.414**
US
(0.013)
(0.047)
14.375***
UK
(0.049)
(0.053)
7.390***
Budget
0.013
0.020**
(0.010)
(0.007)
0.275
6064.818†
-336.452
(0.276)
(3140.560)
(324.371)
803.301
739.139†
(1971.571)
(389.697)
3.556***
(3.307)
(0.446)
(0.887)
(0.333)
0.438
0.267
0.306
0.498
0.031
0.034
Note * p ≤ .05, ** p ≤ .01, *** p ≤ .001, † p ≤ 0.10; N=364 (7 Topics * 52 Years); Start=1951,
End=2003
37
Table 4. Summary of the Effects of Public Priorities on Government Agendas
1 - Economy
2 - Civil
3 - Health
4 - Agriculture
5 - Labor
6 - Education
7 - Environment
8 - Energy
10 - Transport
12 - Law
13 - Social
14 - Housing
18 - Trade
20 - Gov't
21 - Lands
16/19 - Foreign
Total
Responsive
Topics
Short
Long
Short
Long
Short
Long
Short
Long
Short
Long
Short
Long
Short
Long
Short
Long
Short
Long
Short
Long
Short
Long
Short
Long
Short
Long
Short
Long
Short
Long
Short
Long
Executive
US
UK
-0.064
0.056
0.269
0.083***
-0.156
-0.333*
-0.140
-0.399*
0.515
-0.030
0.972*
0.111***
4.435*
-0.090
2.274
-0.169
-1.001
0.040
1.826
0.059†
6.476** -0.031
4.014*** 0.283*
3.407** 0.376*
0.249
0.527***
0.628
0.020
2.759*** 0.236
14.760
0.500†
47.080
0.121
1.312** 0.032
1.235*** 0.462***
1.860†
0.064
1.321*
0.100
3.028
-0.174
1.808
0.161*
-1.080
0.070
0.600
0.123
-0.617
0.359
0.247
0.134
-34.261
0.123
15.206
0.019
-1.191
-0.367**
0.904
-0.181**
Legislation
US
UK
0.129*
0.080†
-0.032
0.055*
-0.119†
0.088
-0.101†
0.038
-0.558*
0.072
-0.042
0.031
1.929†
-0.206
0.740
-0.322
0.253
0.007
-0.247
0.055†
0.487
0.110
0.191
0.035
1.846†
-0.396†
2.293**
0.490*
0.215
0.058
0.446*
-0.006
14.725
0.259
36.067
0.517
0.085
-0.178
-0.047
0.113
0.160
-0.113
0.411*
-0.196**
1.321
-0.132
1.257
0.138
-3.594*
0.030
-5.615**
0.097
1.370
-0.502
-2.049†
-0.695
-383.446
0.073
-269.913
0.445**
0.446*
0.122*
0.348*
0.045†
Short
Long
Either
4
5
7
2
4
5
1
6
6
1
3
4
Budget
US
0.441
-0.272
UK
76.661
-20.561
72.673
-72.474
-12.042
62.198
-1675.423*
-421.279
49.732
146.878
-5.905
-7.519
1151.909
855.517
-6772.070
-2370.936
-82.274
-155.718**
-106.974
-246.707
-216.407
62.027
403.245
555.18 1
12.608
-25.406
0
0
0
0
0
0
Note * p ≤ .05, ** p ≤ .01, *** p ≤ .001, † p ≤ 0.10
38
ONLINE APPENDIX
Table A1. State of the Union ECMs by Major Topic, 1951-2003
1
2
3
4
5
6
7
8
∆OPINIONt
-0.064
(0.281)
-0.156
(0.256)
0.515
(0.448)
4.435*
(1.916)
-1.001
(1.496)
6.476**
(1.77)
3.407**
(1.195)
0.628
(0.63)
OPINIONt-1
0.269
(0.169)
-0.140
(0.209)
0.972*
(0.437)
2.274
(2.319)
1.826
(2.017)
4.014***
(0.98)
0.249
(1.022)
2.759***
(0.617)
AGENDAt-1
-0.799***
-0.991***
-0.852***
-0.884***
-0.810***
-0.991***
-1.072***
-0.952***
(0.139)
(0.146)
(0.142)
(0.146)
(0.138)
(0.154)
(0.136)
(0.129)
PARTYt
4.772
(6.149)
1.140
(3.951)
-8.397
(5.122)
1.814
(3.127)
-4.277
(2.752)
-0.956
(3.363)
-0.339
(2.219)
5.658
(3.995)
Constant
20.532*
(7.734)
10.309*
(4.122)
12.642**
(4.658)
3.424
(2.449)
9.160**
(3.355)
5.182†
(3.052)
5.591**
(1.798)
-1.776
(3.24)
Adj. R2
Bgodfrey
0.376
0.456
0.404
0.445
0.408
0.459
0.575
0.511
1.038
1.008
1.931
0.093
3.415†
3.697†
1.136
3.496†
10
12
13
14
18
20
21
16/19
∆OPINIONt
14.760
(35.44)
1.312**
(0.452)
1.860†
(1.043)
3.028
(6.215)
-1.080
(1.751)
-0.617
(1.403)
-34.261
(52.008)
-1.191
(0.823)
OPINIONt-1
47.080
(49.621)
1.235***
(0.325)
1.321*
(0.618)
1.808
(7.851)
0.600
(1.897)
0.247
(1.088)
15.206
(67.079)
0.904
(0.62)
AGENDAt-1
-1.030***
(0.145)
-0.967***
(0.146)
-0.870***
(0.144)
-1.006***
(0.146)
-0.779***
(0.142)
-0.907***
(0.146)
-0.988***
(0.145)
-0.829***
(0.135)
PARTYt
3.089
(3.145)
-2.368
(4.529)
1.966
(3.583)
-0.154
(2.212)
-0.810
(1.825)
0.779
(5.982)
1.477
(1.839)
6.618
(21.96)
Constant
0.737
(2.548)
3.427
(4.479)
3.990
(3.716)
4.068*
(1.816)
5.272**
(1.817)
15.914*
(6.468)
2.108
(1.497)
47.569†
(24.063)
Adj. R2
Bgodfrey
0.481
0.498
0.419
0.463
0.356
0.411
0.468
0.442
1.980
1.781
1.064
0.873
0.136
1.219
5.943*
0.655
Note * p ≤ .05, ** p ≤ .01, *** p ≤ .001, † p ≤ 0.10, N=52
39
Table A2. Speech from the Throne ECMs by Major Topic, 1951-2003
1
2
3
4
5
6
7
8
∆OPINIONt
0.056
(0.037)
-0.333*
(0.130)
-0.030
(0.041)
-0.090
(0.104)
0.040
(0.046)
-0.031
(0.204)
0.376*
(0.152)
0.020
(0.167)
OPINIONt-1
0.083***
(0.021)
-0.399*
(0.176)
0.111***
(0.025)
-0.169
(0.149)
0.059†
(0.034)
0.283*
(0.133)
0.527***
(0.154)
0.236
(0.234)
AGENDAt-1
-0.840***
(0.145)
-1.108***
(0.135)
-1.078***
(0.12)
-0.561***
(0.131)
-0.761***
(0.140)
-0.919***
(0.14)
-0.906***
(0.144)
-0.852***
(0.138)
PARTYt
-2.089*
(0.836)
-2.716***
(0.653)
-1.338***
(0.347)
0.507
(0.457)
-0.697
(0.555)
-0.192
(0.761)
-0.383
(0.31)
-0.578
(0.427)
Constant
3.925***
(1.163)
3.626***
(0.625)
2.056***
(0.372)
0.925*
(0.447)
2.447***
(0.663)
2.501**
(0.91)
1.003***
(0.297)
1.295**
(0.4)
Adj. R2
0.377
0.605
0.630
0.221
0.380
0.439
0.433
0.430
Bgodfrey
0.069
2.416
0.407
0.818
0.141
1.468
0.051
0.388
10
12
13
14
18
20
21
16/19
∆OPINIONt
0.500†
(0.272)
0.032
(0.223)
0.064
(0.089)
-0.174
(0.123)
0.070
(0.231)
0.359
(0.478)
0.123
(0.179)
-0.367**
(0.12)
OPINIONt-1
0.121
(0.22)
0.462***
(0.119)
0.100
(0.084)
0.161*
(0.079)
0.123
(0.285)
0.134
(0.45)
0.019
(0.124)
-0.181**
(0.068)
AGENDAt-1
-0.743***
(0.143)
-0.862***
(0.133)
-0.952***
(0.15)
-0.806***
(0.136)
-0.443***
(0.124)
-0.745***
(0.136)
-0.625***
(0.146)
-0.654***
(0.133)
PARTYt
-0.800
(0.528)
-1.830*
(0.739)
-1.466*
(0.586)
-0.261
(0.595)
0.166
(0.567)
-3.060**
(1.005)
0.591
(0.821)
3.242*
(1.583)
Constant
2.229***
(0.619)
4.180***
(0.894)
2.643***
(0.698)
1.405*
(0.66)
1.028†
(0.596)
6.167***
(1.26)
4.653***
(1.211)
15.434***
(3.52)
Adj. R2
0.407
0.436
0.422
0.442
0.161
0.346
0.247
0.359
Bgodfrey
0.821
3.791†
0.074
1.265
4.955*
5.226*
0.596
1.573
Note * p ≤ .05, ** p ≤ .01, *** p ≤ .001, † p ≤ 0.10, N=52
40
Table A3. Statutes of the US Congress ECMs by Major Topic, 1951-2003
1
2
3
4
5
6
7
8
∆OPINIONt
0.129*
(0.06)
-0.119†
(0.062)
-0.558*
(0.249)
1.929†
(1.148)
0.253
(0.345)
0.487
(0.816)
1.846†
(0.978)
0.215
(0.179)
OPINIONt-1
-0.032
(0.037)
-0.101†
(0.052)
-0.042
(0.268)
0.740
(1.358)
-0.247
(0.492)
0.191
(0.547)
2.293**
(0.843)
0.446*
(0.168)
AGENDAt-1
-0.856***
-1.318***
-0.196
-0.974***
-1.333***
-0.948***
-1.425***
-1.271***
(0.138)
(0.14)
(0.133)
(0.151)
(0.134)
(0.142)
(0.135)
(0.14)
PARTYt
-5.296**
(1.949)
0.659
(1.232)
-2.792
(3.593)
-9.346***
(2.469)
-3.548***
(0.925)
-5.931
(3.663)
-4.812*
(2.304)
-0.996
(1.435)
Constant
7.567***
(1.767)
6.984***
(1.065)
3.345†
(1.908)
12.872***
(2.059)
7.515***
(0.861)
8.522***
(1.523)
12.169***
(1.565)
7.365***
(1.041)
Adj. R2
Bgodfrey
0.445
0.627
0.222
0.466
0.663
0.446
0.682
0.606
1.228
3.866*
7.406**
1.509
4.016*
2.343
10.509***
3.800†
10
12
13
14
18
20
21
16/19
∆OPINIONt
14.725
(32.966)
0.085
(0.207)
0.160
(0.256)
1.321
(1.553)
-3.594*
(1.449)
1.370
(1.48)
-383.446
(324.636)
0.446*
(0.186)
OPINIONt-1
36.067
(47.364)
-0.047
(0.199)
0.411*
(0.172)
1.257
(1.999)
-5.615**
(1.749)
-2.049†
(1.171)
-269.913
(416.094)
0.348*
(0.165)
AGENDAt-1
-0.746***
(0.141)
-1.219***
(0.14)
-1.241***
(0.151)
-1.086***
(0.14)
-1.093***
(0.147)
-1.076***
(0.146)
-1.066***
(0.144)
-0.769***
(0.145)
PARTYt
-11.196*
(4.3)
-2.438
(4.343)
-4.669**
(1.419)
-3.462***
(0.807)
-4.590*
(2.021)
1.353
(8.105)
-26.276†
(15.008)
-22.812**
(7.512)
Constant
15.307***
(3.46)
17.153***
(2.442)
5.131***
(0.847)
5.070***
(0.734)
11.704***
(1.822)
78.287***
(11.446)
78.418***
(12.314)
32.993***
(7.31)
Adj. R2
Bgodfrey
0.328
0.587
0.580
0.527
0.508
0.571
0.515
0.345
14.843***
2.307
1.148
0.333
1.410
2.743
7.233**
3.559†
Note * p ≤ .05, ** p ≤ .01, *** p ≤ .001, † p ≤ 0.10, N=52
41
Table A4. Acts of the UK Parliament ECMs by Major Topic, 1951-2003
1
2
3
4
5
6
7
8
∆OPINIONt
0.080†
(0.041)
0.088
(0.077)
0.072
(0.05)
-0.206
(0.151)
0.007
(0.041)
0.110
(0.14)
-0.396†
(0.219)
0.058
(0.139)
OPINIONt-1
0.055*
(0.021)
0.038
(0.102)
0.031
(0.028)
-0.322
(0.216)
0.055†
(0.030)
0.035
(0.09)
0.490*
(0.188)
-0.006
(0.19)
AGENDAt-1
-0.876***
-1.077***
-1.002***
-0.960***
-1.094***
-1.168***
-1.203***
-1.215***
(0.144)
(0.159)
(0.144)
(0.145)
(0.144)
(0.143)
(0.126)
(0.146)
PARTYt
-0.897
(0.844)
-0.736†
(0.394)
0.345
(0.416)
0.793
(0.657)
0.019
(0.498)
0.176
(0.523)
0.045
(0.434)
0.527
(0.358)
Constant
4.914***
(1.226)
1.831***
(0.394)
2.213***
(0.504)
2.542***
(0.692)
1.712**
(0.557)
1.926**
(0.605)
1.892***
(0.412)
1.909***
(0.372)
Adj. R2
Bgodfrey
0.406
0.470
0.481
0.441
0.517
0.553
0.648
0.588
2.183
0.360
0.101
1.132
0.981
0.708
1.309
0.151
10
12
13
14
18
20
21
16/19
∆OPINIONt
0.259
(0.482)
-0.178
(0.404)
-0.113
(0.074)
-0.132
(0.15)
0.030
(0.12)
-0.502
(0.563)
0.073
(0.177)
0.122*
(0.05)
OPINIONt-1
0.517
(0.396)
0.113
(0.181)
-0.196**
(0.073)
0.138
(0.096)
0.097
(0.148)
-0.695
(0.531)
0.445**
(0.137)
0.045†
(0.024)
AGENDAt-1
-0.872***
(0.145)
-0.975***
(0.144)
-1.118***
(0.143)
-0.934***
(0.142)
-0.617***
(0.14)
-1.182***
(0.144)
-0.870***
(0.134)
-1.336***
(0.137)
PARTYt
1.793†
(0.949)
1.755
(1.291)
-0.243
(0.444)
0.570
(0.723)
-0.111
(0.302)
-0.126
(1.046)
1.958*
(0.856)
-0.233
(0.622)
Constant
2.444*
(0.997)
7.374***
(1.803)
2.865***
(0.566)
2.307**
(0.843)
0.489†
(0.283)
7.797***
(1.305)
1.076
(0.815)
4.263***
(0.738)
Adj. R2
Bgodfrey
0.415
0.466
0.536
0.464
0.253
0.554
0.438
0.651
0.068
1.274
0.118
0.342
0.041
3.402†
1.122
0.013
Note * p ≤ .05, ** p ≤ .01, *** p ≤ .001, † p ≤ 0.10, N=52
42
Table A5. US Federal Budget ECMs by Major Topic, 1951-2003
1
3
6
12
13
14
16/19
∆OPINIONt
0.441
(1.013)
72.673
(197.173)
-1675.423*
(641.526)
-5.905
(27.024)
1151.909
(898.006)
-6772.070
(22334.3)
403.245
(479.816)
OPINIONt-1
-0.272
(0.648)
-72.474
(236.916)
-421.279
(438.149)
-7.519
(25.762)
855.517
(812.166)
-2370.936
(28777.44)
555.181
(383.886)
AGENDAt-1
-0.210*
0.067***
-0.033
0.046*
0.025*
-0.287**
-0.057
(0.097)
(0.014)
(0.044)
(0.024)
(0.012)
(0.100)
(0.078)
PARTYt
86.409*
(40.049)
-7106.637
(4866.701)
6499.114*
(2921.977)
696.608
(718.447)
-3978.508
(5809.023)
-3138.519
(10426.77)
31913.240†
(16866.25)
Constant
53.544*
(23.8)
1860.910
(1459.816)
1927.041
(1167.794)
167.307
(249.261)
5677.128*
(2764.024)
13580.090†
(7002.519)
9509.406
(38286.48)
Adj. R2
Bgodfrey
0.079
0.589
0.135
0.270
0.326
0.078
0.040
0.002
0.941
0.012
1.643
2.533
5.412*
0.483
Note * p ≤ .05, ** p ≤ .01, *** p ≤ .001, † p ≤ 0.10, N=52
43
Table A6. UK Government Expenditure ECMs by Major Topic, 1951-2003
1
3
6
12
13
14
16/19
∆OPINIONt
76.661
(61.078)
-12.042
(70.173)
49.732
(137.384)
-82.274
(51.654)
-106.974
(268.968)
-216.407
(136.771)
12.608
(35.874)
OPINIONt-1
-20.561
(28.593)
62.198
(63.558)
146.878
(92.504)
-155.718**
(52.625)
-246.707
(276.053)
62.027
(97.299)
-25.406
(18.33)
AGENDAt-1
-0.094
0.019
0.020
0.118***
0.017
-0.089
-0.175*
(0.076)
(0.028)
(0.019)
(0.027)
(0.017)
(0.066)
(0.076)
PARTYt
-659.534
(1386.867)
-295.495
(546.515)
-494.412
(519.929)
-66.166
(138.258)
-1.154
(1657.102)
-810.323
(657.275)
594.086
(499.731)
Constant
3370.918†
(1863.273)
484.794
(770.895)
375.286
(771.689)
-203.253
(198.318)
2748.573
(2600.71)
1419.903
(1271.659)
5065.639*
(2221.256)
Adj. R2
Bgodfrey
0.023
0.107
0.129
0.353
-0.016
0.100
0.035
0.085
4.079*
5.807*
0.817
4.456*
0.870
1.988
Note * p ≤ .05, ** p ≤ .01, *** p ≤ .001, † p ≤ 0.10, N=52
44