Valence and Government Priorities: How issue ownership
and issue salience shape US and U.K. policy agendas
Jane Green
University of Manchester
[email protected]
Will Jennings
University of Southampton
[email protected]
DRAFT: PLEASE DO NOT CITE
Paper prepared for the Elections, Public Opinion and Parties conference,
University of Oxford, 7-9 September 2012.
1
Valence and Government Priorities: How issue ownership
and issue salience shape US and U.K. policy agendas
Abstract
Theories of party electoral competition expect that political parties prime voters to cast a
ballot on a party’s ‘owned’ issues. This paper applies these theories to the policy agendas
of parties in government. We theorise that governments attend to policy issues – in
executive speeches and in laws – according to their public rankings for competence.
Furthermore, we theorise that such effects should be mediated by the salience of issues
and by a party’s electoral popularity. Governments will focus on their owned issues when
they are not responding to important public concerns, and when their electoral popularity
is waning. The theory is tested using data on the policy agendas in State of the Union
Addresses and Statutes of Congress in the U.S. (1945-2007) and Speeches from the Throne
and Acts of Parliament in the U.K. (1945-2011). The results support the theoretical
expectations, and also reveal the limitations of generalising across all issue domains. The
implications are important for evaluating how party incentives, as well as issue salience
incentives, motivate governing policy agendas, and when those incentives matter.
The data on US Policy Agendas were originally collected by Frank R. Baumgartner and Bryan D.
Jones, with the support of National Science Foundation grant numbers SBR 9320922 and 0111611,
distributed through the Department of Government at the University of Texas at Austin. Neither NSF
nor the original collectors of the data bear any responsibility for the analysis reported here. The data
on U.K. Policy Agendas were collected by Peter John, Will Jennings and Shaun Bevan with the support
of ESRC research award RES-062-23-0872. The collection of data on party competence is supported
by ESRC small grant award ES/J001678/1. We thank Shaun Bevan for advice on model specification,
and Frank Baumgartner for valuable comments on an earlier version of this paper.
2
There is an extensive literature predicting the issue agendas of political parties and
candidates during election campaigns. A party or candidate is expected to emphasise issues
on which they are considered more competent or which they ‘own’, and avoid issues on
which they are considered less competent - on which another party therefore has a relative
advantage or ownership. These saliency and selective emphasis theories (Budge and Farlie
1977; 1983; Budge et al. 1987; Budge 1993), or issue ownership theories (Petrocik 1996;
Petrocik et al. 2003) offer predictions about how parties use their policy reputations, and
the salience of issues, to maximize their votes during election campaigns. Parties seek to
prime voters to cast a ballot on the issues that benefit that party (Iyengar and Kinder 1987;
Iyengar and Simon 2002; Vavreck 2009). However, these theories have so far been confined
to the intense campaign periods preceding elections. They have also been confined to the
policy statements, manifestos and speeches of political candidates and parties, but not to
the policy attention of political parties in government. However, if these campaign
strategies and emphases matter, they should also matter insofar as they influence the
legislative outcomes of elected government. In this paper we argue that parties govern, in
part, according to issue competence-based incentives. The paper fills an important gap via a
theory of issue competence and issue salience effects on the policy agendas of
governments. As Mortensen et al. (2011: 2) state, “Prioritizing problems and setting policy
goals is a crucial aspect of the work of a government…understanding the causes behind
these agendas deserves considerably more attention than it has received until now”.
We propose a party competition theory to explain how political parties and
presidents use their period of time in government in ways related to their incentives
displayed during election campaigns. Our theoretical explanation for this focus relies on the
electoral incentives highlighted in issue ownership theory (Petrocik 1996; Petrocik et al.
2003); a party will use a period of government to highlight its issues of strength, given the
attention of mass media to governing party activity, and its likely effect on a party’s
electoral fortunes. We also expect that a party will focus on its areas of strengths due to its
ideological or partisan commitment to those issues, the likely success in gaining majorities
on such issues, and the tendency of parties to govern on issues they have emphasised in
campaigns – since parties emphasise issues in campaigns on which they have a relative
advantage. Our theory expects these likelihoods to also be a function of the salience of
different policy issues. The public salience of issues may influence the policy agendas of
3
governing parties in two ways; via a direct effect, since we know that governments are more
likely to attend to issues of public concern (Baumgartner and Jones 2004; Jones and
Baumgartner 2005; Hobolt and Klemmensen 2005; 2008; Jennings and John 2009; Jones et
al. 2009; Bevan and Jennings 2010), but also via the attenuating effect of competence as a
function of salience, and vice versa. We expect governments to attend to their owned issues
to the extent that they are not otherwise responding to policy problems and issues of high
public prioritisation. Furthermore, our theory expects issue ownership effects to be
attenuated by the popularity of the governing party. An ownership strategy is more likely
when a party needs to rely on its issue-based strengths (Green 2011), when it has a
narrower electoral constituency, lower political capital, and when it relies on a smaller
legislative majority more likely to support the government on issues the party owns.
Governments fall back onto issue ownership incentives the weaker their electoral mandate.
The evidence presented in the paper provides support for these theoretical
expectations, drawing on analyses of U.S. and U.K. policy agendas between 1945-2007 and
1945-2011, respectively. Our analyses suggest that issue-based incentives exert a significant
effect upon governing policy agendas. We also reveal the conditions under which these
effects are more or less likely; via issue salience and governing party popularity. However,
the analyses in the paper also reveal, via a two-step analytic procedure moving from pooled
issue models to analyses specific to nine comparable policy domains, that none of these
explanations, including for the widely accepted main effect of public importance of public
policy problems (see Baumgartner and Jones 2004; Jones and Baumgartner 2005; Hobolt
and Klemmensen 2005; 2008; Jennings and John 2009; Jones et al. 2009; Bevan and
Jennings 2010), can always be generalised for any one-issue or across issues, irrespective of
political and country context. These insights urge caution in generalising findings from
single-issue case studies, or for extrapolating across issues from general effects.
The findings offer important implications for our understanding of legislative and
executive policy outcomes, and for the relationship of government behaviour to the
electoral strategies of parties. To the extent that parties or candidates adopt issue
ownership considerations in devising their campaigns, our findings suggest that election
campaigns may provide important cues to voters about the likely policy agendas of
governments. Campaigns may therefore serve an additional role in informing voters of the
likely focus of parties while in government. Furthermore, we reveal that these predictive
4
implications may be weaker for governments with strong public popularity. Governments
tend to fall back onto an issue ownership basis of attentiveness the lower their popular
appeal. Given the cyclical nature of governing party popularity and well-known ‘costs of
governing’ dynamics for parties in government (Stimson 1976; Rose and Mackie 1983;
Paldam 1986; Sanders 2005), issue ownership incentives may be particularly relevant when
governments are least popular, often following a period already in office. Finally, we
propose that while governments have incentives to attend to their ‘own’ issues, they are
less likely to respond to issue ownership considerations when there are important issues of
public concern. This is an encouraging finding from a normative point of view. Because issue
salience is closely tied to the degree to which policy areas pose significant problems to the
administration, these findings suggests that governments de-prioritise an issue ownership
agenda in government when responding to public problems and concerns.
Party Competition Incentives and Election Campaigns
Extensive scholarship has focused on the issue agendas of political parties and candidates in
election campaigns. Whereas parties may shift their ideological positions in response to
electoral losses and gains over the long-term (Adams 2001; Adams et al. 2005, Somer-Topcu
2009), parties have fewer strategic options at their disposal come the shorter election
campaign. The saliency or selective emphasis theory of Budge and Farlie (1977; 1983; Budge
et al. 1987; Budge 1993) and the issue ownership theory of Petrocik (1996; Petrocik et al.
2003) argue that parties seek to win votes by emphasizing the issues on which they have
their greatest advantage, on which they have a reputation for competence, or ownership,
and by seeking to neutralize issues on which their opponent has an advantage, downplaying
those issues in campaigns. Political parties vote maximise when the issues on which they are
evaluated most positively are also salient with voters (Budge and Farlie 1983). These
theories predict that parties seek to emphasise different issues in elections, rather than
engage in direct confrontation, and the expectation finds parallels in Riker’s (1993)
dominance/dispersion principle: the purpose of campaign agenda shaping is not to engage
other parties in debate or dialogue but to increase the salience of issues over which the
party or candidate is perceived to be credible.
These theories have been challenged by scholars identifying instances in which
parties and candidates confront the same issues (Simon 2002; Sigelman and Buell 2004;
5
Damore 2004; 2005; Kaplan et al. 2006; Sides 2006). Parties may confront issues as a means
to highlight policy differences, to highlight policy or governance failures in other parties, to
respond and defend attacks on their policies, or competence, and because parties do not
have complete control over campaign issue agendas. Damore (2004; 2005) investigated a
series of other factors influencing issue trespassing or convergence and found that a
candidate’s competitive standing, their partisanship, the importance of the issue for the
electorate, and the campaign message tone were all associated with candidates’ decisions
to address issues owned by their opponent’s party. Given the importance of salience to
issue ownership theories, following the expected multiplicative or conditional effect of issue
salience and issue ownership upon vote choices (Bélanger and Meguid 2008; Green and
Hobolt 2008), it seems likely that issue importance is a particularly important explanation of
the tendency for parties to dialogue on some issues. Nevertheless, even among scholars
who report convergence and dialogue, candidates and parties still exhibit a preference to
focus on their owned issues – to the extent that the political context permits this (Spiliotes
and Vavreck 2002; Damore 2005; Sides 2006; Green-Pedersen and Mortensen 2010).
It is somewhat surprising that these theories have mostly been confined to
explaining the issue agendas of parties in elections (for notable exceptions, see Bräuninger
and Debus2009; Green-Pedersen and Mortensen 2009; Cummins 2010). The Comparative
Manifestos Project, on which issue saliency theory is based (Budge et al. 1987; Budge 1993;
2001), naturally turns attention to election-focused effects. The understanding of party
incentives of any kind is tied to an assumption of vote maximization incentives in elections.
However, parties have other means of seeking re-election and vote-maximization beyond
their emphases prior to and during the election campaign itself. A party or a president also
has four or five years in government during which voters evaluate their performance,
competence and delivery, the agenda of the government, and during which a party and/or a
president can determine the policy agenda and emphasis of a legislative period. Indeed,
Petrocik (1996: 826) defined issue ownership as follows: “the ability to resolve a problem of
concern to voters. It is a reputation for policy and program interests, produced by a history
of attention, initiative and innovation toward these problems, which leads voters to believe
that one of the parties (and its candidates) is more sincere and committed to doing
something about them.” This definition implies that parties acquire issue ownership via
their attention to policy issues, and there is no clearer arena than government but to display
6
this attentiveness, both in terms of public awareness and the scope for delivery. A growing
literature considers how presidential agendas influence Congress, the public or media and
are also shaped by them (Edwards and Wood 1999), and how presidents shape public
opinion (Cohen 1995), and respond to it (Cohen 1997), but the literature in this area is
specific to presidents, and not to parties, nor to presidential policy attentiveness as a
function of issue incentives (see also Hill 1998; Edwards and Wood 1999; Eshbaugh-Soha
and Peake 2004; Yates and Whitford 2005).
A Party Competition Theory of Government Agendas
The question of government attentiveness is highly important; both in theoretical and in
empirical terms. If parties and governments have incentives to win elections, and the
scholarship in this area is immense, we should care about the election winning strategies
because those election results have substantively important and distinctive outcomes.
Government policy agendas provide an important outcome variable through which to
understand the legislative outcomes of elections.
To the extent that scholarship has sought to explain the policy agendas of
governments, answers point to the exogenous policy environment and to the emergence of
policy areas as problems and salient public issues (Baumgartner and Jones 2004; Jones and
Baumgartner 2005; Hobolt and Klemmensen 2005; 2008; Jennings and John 2009; Jones et
al. 2009; Bevan and Jennings 2010). Governments also respond to the emergence of new
social issues and problems (Baumgartner and Jones 1993; Jones and Baumgartner 2005;
Kingdon 1984). There is also a body of research pointing to an absence of any partisan
influence of government agendas of any kind (Rose 1980; 1993; Rose and Davies 1994;
Baumgartner et al. 2009; Mortensen et al. 2011), whereas other studies have found
significant differences between the policy agendas of different parties in government, with
findings of partisan effects on macroeconomic outcomes (e.g. Hibbs 1977), public spending
(e.g. Castles and McKinlay 1972; Castles 1982) and attention to issues (Jennings et al.
2011a). We need to better understand the nature of party incentives, which may link the
party-in-government to partisan differences in policy agendas and in legislative outcomes.
We contend that governments have incentives to attend to their ‘owned’ policy
issues, but that this attentiveness should be conditioned by the importance of salient policy
7
concerns. These within-party differences may obscure between-party differences when
scholars seek to establish partisan effects on government agendas.1
The first reason to support the expectation of an ownership effect on government
policy agendas arises from the literature on issue ownership itself (Petrocik 1996; Petrocik
et al. 2003). According to Petrocik (1996: 826), ‘ownership’ relates to “a reputation…which
leads voters to believe that one of the parties is more sincere and committed to doing
something about [the issue]”. Following this definition, parties gain ownership by
committing to issues and ‘doing something about them’. If a party has a reputation for
ownership of an issue via its long-term association with that area of policy, it is likely that
party will maintain a commitment to that issue in government. Such a strategy will only
prove effective if ‘doing something about a policy issue’ does not result in a negative
consequence for the government, causing an embarrassment or a loss in issue ownership.
Given the reputation risks of governing, we would expect parties to focus on issues on which
they already have a reputation for competence, rather than take risks on issues they do not
own. Since the fundamental assumption of ownership is that two parties cannot ‘own’ the
same issue, we must expect parties to focus on issues selectively in order to capitalise on a
reputation for commitment and competence: we expect selective emphasis of issues in
government, just as we observe a degree of selective emphasis in party campaigns.
The second reason to support our expectations comes from scholarly insights that
parties represent their electoral strategies when they enter periods of government,
according to median mandate theory (Budge and Hofferbert 1990; Hofferbert and Budge
1992; McDonald et al. 2004; McDonald and Budge 2005). This theory is derived relating the
content of party manifestos to the policy outcomes of parties in government. Those party
manifesto contents are a function of the relationship between owned issues and the
selective emphasis of parties towards issues they consider they benefit from, and away from
the issues that benefit their opponents. If a party gives disproportionate emphasis to a
policy issue in its manifesto, and it does so because it believes an issue will benefit it, we
1
It might alternatively be argued that parties will use a period in government to build a reputation
for competence on an issue; to gain ownership of an issue, especially when an issue is salient in the
electorate. However, we believe the arguments in the opposite direction are more compelling.
8
should see a corresponding disproportionate emphasis in government, at least if median
mandate theory is supported.
Third, we expect parties to ‘own’ issues because they are particularly committed to
them, but commitment should derive from ideological and/or partisan preoccupations and
from policy-seeking party incentives. While scholars typically agree that parties exhibit voteseeking strategies in elections, we also expect that parties gain office for policy-seeking
objectives (Aldrich 1985; Strom 1990; Kitschelt 1989; 1994). Thus, while some scholars
argue that there is little evidence for party-specific agendas in the attention of governments
to issues (Rose 1980; 1993; Rose and Davies 1994; Baumgartner et al. 2009; Mortensen et
al. 2011) closer focus on issues by ownership may result in partisan priority differences
which are masked by across-issue aggregation. Parties may focus on their ‘best’ issues,
simply because those best issues are also those issues on which they are most committed;
and on which they will therefore focus attention, when possible (when not driven off the
agenda by salient problem policy issues).
These expectations lead to the following hypothesis.
H1: Governing parties attend to an issue more when they have a reputation for issue
ownership on that issue.
These tendencies for selective government emphasis and attention should be
moderated by policy salience. In government, policy-makers must prioritise between a
range of competing concerns (Jones 2001; Jones and Baumgartner 2005). While parties in
government might own a number of issues, the scarcity of attention often requires them to
deal with them in rank order of importance, such that reputation considerations will be
processed through the ordering of transitive preferences between issues A, B, C … k.
Scarceness of attention requires decision-makers to prioritise between policy alternatives: a
party cannot govern for reputation reasons or for partisan or policy-seeking reasons if it is
responding to the salient policy issues of the day. Changes in issue salience, often due to
exogenous shifts in problem status (e.g. Hibbs 1979; Baumgartner and Jones 1993; Jennings
and Wlezien 2011), might therefore trump considerations of the pre-existing reputation of
the governing party for competence on an issue. Indeed, such a trade-off between focusing
on issues of public concern and focusing on a party’s owned issues may account for some of
9
the evidence on issue trespassing and convergence in the campaign literature on ownership.
Alternatively, it might be argued that the effects of issue ownership would be
greater the salience of the issue; parties have greater incentives to govern on issues they
own and on issues of public concern, just as issue ownership theories suggest that parties
seek to raise the salience of their owned issues, in order to prime voters to cast a ballot on
these issues. We should therefore expect to find a mediating relationship between issue
salience and issue ownership effects on governing party agendas, remaining open to either
attenuating or enhancing issue ownership effects.
H2: The effects of Issue ownership evaluations on governing policy agendas will be
mediated by the salience of the policy issue.
We also contend that issue ownership incentives should be moderated by party
popularity. Research on the issue ownership campaign strategies of parties or candidates
suggest that elites may attend primarily to public concerns, gaining electoral support by
being in touch with those issues of concern to the public, or they instead focus on the issues
they own (Ansolabehere and Iyengar 1994). Green (2011) suggests that an issue ownership
strategy is particularly likely for parties with relatively weak electoral support. This
prediction arises because owned issues tend to those on which relatively unpopular parties
retain a potential advantage in terms of public support, whereas popular parties gain from a
wide number of issues on which they are positively rated (a greater number than their
traditionally associated issues), and are freer, therefore, to respond both to public concerns
and indeed to any other explanations for a focus on issues. Wagner and Meyer (2012), in a
cross-national analysis of 26 countries, also find a correspondence between the electoral
size of a party and its tendency to attend to its owned issues. We therefore expect parties to
attend more to an ownership strategy, the weaker that party’s electoral mandate.
H3: The effects of Issue ownership evaluations on governing policy agendas will be
attenuated by the electoral popularity of the governing party.
Data and Measures
10
We use two country cases upon which to test our hypotheses; the U.S. and U.K. We further
examine two within-country institutional contexts; the executive and legislative agenda, in
the form of the U.S. State of the Union address and the U.K. Speech from the Throne (the
Queen’s Speech), and the outputs of each country’s legislature, via Statutes of U.S. Congress
and Acts of the U.K. Parliament. The State of the Union Address in the U.S. and the ‘Speech
from the Throne’ in the U.K. (known as the Queen’s Speech during our data points) are the
prominent annual speeches that communicate the policy programme of the executive. The
use of two countries and two contexts allows us to explore whether the strength of support
for the hypotheses should be different across these institutional and country contexts.
Given the electoral and campaign basis of issue ownership theories, it might be plausible to
suggest that issue ownership effects will be greatest in executive speeches, since these
events are forums for priming of the media and public agenda, as well as for legislative
programming. However, research finds some consistency between stated executive agendas
and legislative action, both in the U.S. where there is evidence of limited presidential
influence over Congress (e.g. Edwards and Wood 1999; Edwards and Barrett 2000) and the
U.K. where the executive has a high rate of transmission between its programme and
statutes (Bara 2005; Bevan et al. 2011). We might alternatively contend that executive
speeches will be similar to legislative outcomes, particularly in terms of the possible
attenuating effect of issue salience on issue competence effects. We can further consider
whether our hypotheses are likely to be supported similarly in the two country contexts.
While the two cases are chosen for their similarity in terms of majoritarian party
government, we might contend that issue ownership effects will be greatest in the U.S., in
which two parties consistently compete for office, and in which presidential issue handling is
a more powerful signal of competence. Alternatively, the strength of explanatory power of
issue ownership effects on party agendas and vote choices in Britain (Clarke et al. 2004;
2009; Green and Hobolt 2008; Johns et al. 2009; Green and Jennings 2012a; 2012b), many
other European countries (Budge and Farlie 1983; Bellucci 2006; Blomqvist and Pedersen
2004), and in Canada (Nadeau et al. 2001; Bélanger and Meguid 2009) may suggest that
incentives for government attention to owned issues may be generalisable across the two
countries. Our use of two country cases permits some exploration of these questions.
Moreover, the analysis of executive and legislative outputs, in the U.S. and in the U.K.
11
provides a means of examining the robustness of the hypothesised findings and a way of
exploring the generalisability to similar country systems.
The analysis is based upon data on public opinion, executive agendas and legislative
outputs in the U.S. from 1945 to 2007 and in the U.K. from 1945 to 2011, coded according
to a modified version of the policy content coding system of the Policy Agendas Project
(www.policyagendas.org). A number of categories are combined in order to match the
Policy Agendas Projects topics with our own dataset on public evaluations of the
competence of political parties, required in order for there to be sufficient poll observations
on a given topic. The advantage of this approach is that it enables comparison across
governing institutions, across countries and over time. The issue categorizations are
summarized in Table 1.
Table 1 about here
The analyses are based on more than 5,000 survey questions asked about party competence
across the two countries. These questions measure public evaluations of party competence
to handle particular issues or policy problems. In the U.K. the wording of these questions
tend to focus on the party that is ‘best able to handle’ defined issues or problems, while in
the U.S. the question wording often asks which party the public ‘trust to do a better job of
handling’ a given issue (further examples of questions are provided in the appendix). In the
U.K., the survey data consist of 2,600 administrations of 162 different questions by five
polling organizations, while in the U.S. there are 2,433 administrations of 257 different
questions by sixteen polling organizations. Most of the questions require respondents to
choose between specified political parties: in the U.K. the options tend to be Labour, the
Conservatives or the Liberals, with the option of a non-response (‘none’ or ‘don’t know’). In
the U.S., the choice is typically between the Republican or the Democratic Party. Because
we are interested in governing parties, we drop responses for third parties and nonresponses, and recalculate these issue competence figures as the percentage of the twoparty share. This enables comparison of relative competence evaluations across topics and
over time, since the concern of our analysis is in whether one party holds a reputational
advantage compared to another on a given topic.
The competence data are based on questions concerning handling and trust. Such
data sources have been used extensively as measures of competence. However, we assume
12
a degree of partisan bias in respondents’ answers to these questions, and therefore control
for the governing party’s share of vote intention in each model.
We use three operationalisations of party competence in our initial analysis,
enabling exploration of effects according to rival operationalisations of the governing party’s
advantage on competence. The issue ownership literature encompasses different ways of
measuring issue ownership, and so these measures serve to examine our hypotheses across
three alternative constructions of issue competence evaluations and their effects.
1.
Mean competence in the previous electoral cycle (level): To test for the effect of the
level of party competence on a given topic, we calculate the governing party’s mean score
across available competence evaluations in the previous electoral cycle (which takes a value
in the range between 0% and 100%). This should be higher (lower) when a governing party
has a competence advantage (disadvantage) on an issue. This competence score is then
used as a predictor of the level of attention in the next electoral cycle. For example, if the
governing party’s rating for better handling the economy was 65% in the previous electoral
cycle, this 65% is the value assigned to indicate its level of competence on the issue.
2.
Between-party rank of competence in the previous electoral cycle (majority): To test
for the effect of the competence of a party on a given topic relative to its opponents, we
calculate the plurality winner of the above mean competence evaluations on each topic for
each electoral cycle. This variable is coded dichotomously; equal to 1 if a party holds greater
than 50% of the two-party share and being equal to 0 if a party holds less than 50% of the
two-party share. This measure denotes a short-term issue competence advantage on a
given topic, aggregated over topics (longer-term ownership would need to be measured
through the relative competence evaluations over a longer period).
3.
Within-party rank of competence in the previous electoral cycle (ownership): To test
for the effect of the competence of a party on a given topic relative to its competence on
other issues, we calculate the rank of mean competence evaluations for each topic for each
electoral cycle. This variable is coded between 10-k and 9, where k is the total number of
topics on which competence evaluations are available during a given time period (and can
take a value up to 9). For example, if competence scores are available for a party across all
nine issues(A=50%, B=45%, C=70%, D=75%, E=55%, F=55%, G=40%, H=70%, I=40%) their
rank order will indicate the relative advantage of party competence between all issues (9=D,
8=C/H, 6=E/F, 4=A, 3=B, 2=G/I). The competence rank is therefore highest for issue D, next
13
highest for issues C and H, and so on. The scale is constructed such that ‘9’ refers to the topranked issue and ‘1’ refers to the lowest-possible ranked, meaning that the estimated
coefficient for the effect of competence rank should be positive. Because of the
discontinuous nature of survey data on party competence, there is missing data for some
categories for certain periods. Over the full time period, our data consists of an average of
5.5 issue categories in the US and 6.6 categories in the U.K. (as shown in Tables A2 and A4),
meaning that such missing categories account for around a third of the data. In the reported
analyses missing categories are not given a ranking (note that the highest-scoring issue is
assigned a score of 9 regardless of whether there are missing categories or not, because it is
the highest observed competence ranking). Further robustness checks are conducted to
ensure the results hold for periods when there is less missing data or when missing data is
assigned the median rank, or the minimum rank, rather than treated as missing.2
To allow for the lagged effect of competence evaluations on governing party
agendas, and to avoid endogeneity where governing party agendas shape competence
evaluations in a given time period, we take all available observations for each of the nine
topics for the previous election cycle. For example, party competence ratings for the
duration of the first Bush administration between 2001 and 2004 are taken for the
Republican Party from the 1997 to 2000 period, while competence ratings for the Blair
government between 2001 and 2005 are taken for the Labour Party from the 1997 to 2001
period.
We use aggregate data on public responses to the survey question about the ‘most
important problem’ (“MIP”) to measure the salience of particular issues to the public at
particular points in time.3 This data is available for the period between 1947 and 2007 in the
2
We replicated the pooled time series cross-sectional and the issue-specific models for the period
between 1974 and 2011 in the US and the U.K. This period provides a far more comprehensive range
of survey data (the data consists of an average of 6.9 in the US and 8.3 in the U.K.). These analyses
did not alter the substantive inferences drawn, and in some instances led to a better model fit for
issue-specific models. The level of significance was reduced in a small number of cases. All
robustness checks are discussed in a later section of the paper.
3
In the U.K., data for Gallup’s “most important problem” question is not available after 2001, so we
use Ipsos-MORI data for the question about the “most important issue”, which enables construction
of a continuous measure of issue salience. The MIP and MII questions have been shown to exhibit a
14
U.S. and between 1945 and 2010 in the U.K. Responses are standardised for each survey to
total 100% and then averaged across the calendar year where multiple surveys are available
(for further details see Feeley et al. 2006; Jennings and Wlezien 2011).
Popular support for the governing party is measured using data on vote intentions
for each country by calendar year. In the U.S. we use support for the president’s party in the
generic congressional ballot which is asked in a number of forms by numerous polling
houses about how respondents intend to vote in their district in the upcoming congressional
elections.4 The data used here consists of 1,997 polls, dating back to 1942, from Wlezien
and Erikson (2002; Bafumi et al. 2010), supplemented with data from the Gallup ‘Brain’
database and trend data on generic ballot reported at www.pollingreport.com. In the U.K.,
we use a comprehensive dataset of available national surveys on vote intention for the
period between 1945 and 2008. We have 1,095 polls where respondents were asked about
which party they would vote ‘if the election were held tomorrow’ (from the pollsters Gallup
and Ipsos-MORI, who offer the longest periods of continuous polling on vote intention in the
U.K.).
The dependent variable is the proportion of the policy agenda that is assigned to a
topic in a given institutional venue in a given year. This can take a value between 0% and
100%, making it directly comparable to data on party competence and issue salience. The
policy content of the State of the Union and the Speech from the Throne are divided into
‘quasi-sentences’; expressions of a single policy idea or issue. In practice these are often
identifiable from use of punctuation, with each quasi-sentence assigned a single topic code.
Symbolic or non-policy content is dropped for the purposes of analysis. The unit of analysis
is the proportion of each speech allocated to a particular topic (economy, health, labor,
education, environment, law, social, foreign or other). Statutes of the U.S. Congress and
Acts of the U.K. Parliament provide cases on legislative outputs. Each piece of legislation is
high degree of common variance, providing a comparable indication of the issues that are on
people’s minds (Jennings and Wlezien 2011). To accomplish this, the two series are combined and
averaged for the period between 1980 and 2000 when there is regular overlapping data.
4
Gallup’s original formulation of the generic ballot question asked respondents: “... If the elections
for Congress were being held today, which party’s candidate would you vote for in your
congressional district -- the Democratic Party’s candidate or the Republican Party’s candidate?”
15
coded with a single topic, indicating the primary focus of the legislation. The observed time
point is the date upon which the bill was signed into law. Commemorative bills are excluded
from the U.S. data in order for these data to be directly comparable to U.K. Acts of
Parliament, limited to legislative outputs intended to affect the functional workings of
government. The unit of analysis is the proportion of legislative outputs that relate to a
particular topic.
Method of Analysis
A time series cross-sectional AR(1) model is estimated for all nine topics (i.e. panels)
common to each of the executive and legislative agendas, in order to assess general
patterns of effects of party competence on governing party agendas. This pooled model
specification has the advantage of estimating the degree to which the government’s agenda
is a function of the governing party’s reputation for competence overall, across issues, and
the conditional relationships with salience and popularity.5 The model takes the form:
AGENDAit
= α* 0
+ α*1COMPETENCEit-c
+ β*1MIPit
+ β*2MIPit*COMPETENCEit-c
+ β*3POPULARITYit
+ β*4POPULARITYit*COMPETENCEit-c
(1)
Where AGENDAit refers to the proportion of the policy agenda assigned to a given issues, α *0
represents the intercept, COMPETENCEit-c refers to party competence in the previous
electoral cycle (hence t-c, not t-1), MIPit refers to the percentage of respondents assigning
5
Prior to modelling, the dependent variable (i.e. the executive and legislative agendas in the US and
in the U.K.) was tested for stationarity using the Augmented Dickey-Fuller test and in all instances
rejected the presence of unit root at the 95 per cent confidence level (the data is therefore modelled
in level form rather than in first differences).
16
each issue as the most important in the current period, t, and POPULARITYit refers to the
share of popular support gained by the governing party, also at time period t.
The model is estimated with panel-corrected standard errors (Beck and Katz 1995),
which controls for panel heteroscedasticity and contemporaneous correlations of the
errors. The model is fitted with the Prais-Winsten method to test for serial autocorrelation
(μit), with the rho estimated separately for each panel as the first-order autoregressive
process: μit = ρμit-1 + εit. This allows the rate of persistence to vary across units (i.e. panels),
consistent with the expectation that there will be greater stability in attention to some
issues compared to others (see Jennings et al. 2010).
In subsequent analysis, we model each panel (i.e. issue) separately, since it is
plausible to us that substantively important effects might be cancelled out through pooling
the data. Governments may exhibit the expected issue emphasis on some issues but not
others, and effects may not all be in the same direction. The individual topic models take
the same form as above and are again fitted with the Prais-Winsten method.
Results
We first present the pooled models for competence effects in the U.S. for the State of the
Union address (i.e. the executive and legislative agenda) and for Statutes of U.S. Congress
(i.e. legislative outputs). Table 2(a) presents the results for party competence level,
ownership and competence rank, and the results of the interactions of each competence
measure with salience and with governing party popularity.
Table 2(a) about here
Examining the first three rows of Table 2(a) reveals that the rank measure of party
competence helps account for the policy agendas in the State of the Union address and in
Statutes of Congress, but this is not true for a measure based on the average level of
competence, nor whether the governing party has a competence majority on each issue.
These patterns persisted whether or not popularity (and its interaction with competence)
was included in the model, suggesting that these null findings do not arise because of
potential endogeneity of competence measures resulting from partisan bias. The results
suggest that U.S. governments do not attend to issues because they have a higher level of
competence on a given issue, or a lead over a rival party. They attend to issues when those
issues are their ‘best’ issues relative to other issues. We consider this measure to best
17
represent the concept of issue ownership. The coefficient on the rank measure is positive
and significant in both models, providing support for Hypothesis 1; namely, that
governments will attend to issues they own. They do not appear to use a period of
government to build a reputation for competence on their lower ranked issues.
The results reported in Table 2(a) also provide some support for Hypothesis 2, which
predicted that the effects of issue ownership would be mediated by the salience of an issue.
The coefficients for the MIP*Party competence (issue ownership) interaction are significant
and negative for the policy agenda of the State of the Union address, although this is not the
case for U.S. Statutes of Congress. The main effect for issue salience (MIP) suggests that the
policy agenda of the State of the Union address is highly responsive to issues of public
concern. The absence of a significant issue ownership*MIP interaction coefficient for
Statutes of Congress suggests that the President’s party attends to its owned issues and to
salient issues, but the effect of one is not mediated by the other.
The coefficient for the competence*MIP interaction in the State of the Union
address is represented in Figure 1(a), which plots the marginal effects of issue salience over
values of issue ownership ranking, which are scored from 1 to 9.
Figure 1(a) about here
The slope in Figure 1(a) indicates a drop in effect size from 1.00 to 0.27, equivalent to
approximately one third in change across the range on the y axis, with all values and
confidence intervals falling above zero. When parties are ranked highest on policy issues in
terms of competence, the effect of salience is significantly reduced.
Moving to the sixth and seventh row of Table 2(a), the results provide some support
for Hypothesis 3, which predicted that issue ownership effects would be greatest when
parties are less popular. The interaction effect is significant and negative, indicating an
attenuating effect of issue ownership with party popularity, in the policy agenda of Statutes
of Congress but not for the State of the Union. The coefficient for the significant interaction
for the policy agenda in Statutes of Congress is represented by Figure 1(b), which
demonstrates that issue ownership effects decline across values of popularity.
Figure 1(b) about here
The slope displayed in Figure 1(b) reveals a significant decline in the marginal effect of issue
ownership, from a value of 2.0 to -1.2. This is a substantial and significant drop in effect size
as party popularity increases, crossing zero at around 50%. We confine the range of the X
18
axis to the minimum and maximum values of the popularity variable in these cases. These
results suggest that issue ownership effects are attenuated by governing party popularity in
the policy agendas of Statutes of Congress, whereas issue salience effects are attenuated by
issue ownership in the State of the Union address.
The models for the U.K. Queen’s Speech and Acts of Parliament are presented in
Table 2(b).
Table 2(b) about here
These models reveal an interesting difference in comparison to those for the U.S. Whereas
issue salience had a significant effect on the agendas of the State of the Union address and
Statutes of Congress (in Table 2(a)), Table 2(b) reveals no significant relationship between
issue salience and government policy agendas in the British context, as we might have
expected given theories of government attention (see Jones and Baumgartner 2004; Jones
et al. 2009; Bevan and Jennings 2010).
The results in Table 2(b) support the use of the issue ownership rank measure of
party competence, revealing that where U.K. governments attend to issues they own, they
do so according to the within-party ranking of issues, not the competence level or issue
advantage on competence. The effect is not significant within the 95% confidence level for
the agenda of the Speech from the Throne (p = 0.18), but strongly significant in Acts of
Parliament. This finding provides partial support to Hypothesis 2, which predicted that
parties in government would attend to issues they own. Table 2(b) further reveals no
significant interactions between issue ownership and issue salience, plausibly due to the
absence of explanatory power on the MIP measure in the U.K., but row seven of Table 2(b)
offers a parallel finding with the U.S. with respect to the conditional relationship between
governing party popularity and the effect of issue ownership. The interaction effect
between popularity and ownership is negative in direction and significant, providing
additional support to Hypothesis 3, which predicted that parties would attend to an issue
ownership strategy, mediated by their electoral strength. It is interesting to note that this
interaction is significant in the policy outputs of U.K. governments, just as it was significant
(and negative) in the policy outputs of U.S. administrations, over the same time period and
using comparable data and measures.
The interaction effect for the issue ownership*popularity effect on U.K. Acts of
Parliament is represented in Figure 1(c).
19
Figure 1(c) about here
Figure 1(c) illustrates that the effect of issue ownership declines as governing parties gain in
popularity. Here we see that the effect of issue ownership is only significant for governing
parties with low vote intention shares, less than 35% and below. Such values of vote
intention occur quite frequently in British politics, due to the presence of the third party and
minor parties.
The results from the pooled models therefore offer support to the hypotheses. They
suggest that governing parties attend more to their owned issues, where ownership is
reflected by the within-party rank of individual policy categories, and that this effect is
conditioned by the salience of issues and by the popularity of the incumbent’s party. Issue
salience has a greater effect for issues not owned by a party, and issue ownership has a
greater effect for parties less popular with a weaker electoral mandate. The hypotheses are
supported in both countries, but not equally across institutional contexts.
These general trends are useful up to a point. They give us an indication of the
support for the hypothesised directions of effect, and are important and interesting in this
regard. However, they do not provide insight into the substantive application of these issue
ownership effects, such as whether they apply across all issues. They may also mask effects
in different directions or significant effects in support of the hypotheses across institutional
context. Scholarship in the field of agenda-setting has often been characterised by the
examination of single domains or cases (Baumgartner et al. 2006). We know, for example,
that there are by-issue differences in patterns of policy-opinion responsiveness (e.g. Page
and Shapiro 1983) and presidential agenda-setting (e.g. Edwards and Wood 1999). However,
these single-issue case studies do not allow us to understand examples of issues for which
we may find hypothesised effects, and issues where we may not, and extrapolating from
single-issue studies is also something to be avoided. We therefore combine our pooled
models where we identify general effects with more in-depth analysis across a range of
available issues; examining our hypotheses across nine policy domains in both countries.
Issue-by-Issue Analyses
We proceed to estimate the equivalent model as used above, with Prais-Winsten regression
to control for serial autocorrelation, but for each individual issue category. The executive
and legislative agenda series are found to be stationary using the Augmented-Dickey Fuller
20
test, so can be used in level form without differencing. The Prais-Winsten transformation is
known to improve model efficiency for small sample sizes, although can still underestimate
standard errors (Park and Mitchell 1980).
Following the results displayed in Tables 2(a) and 2(b), we adopt the rank measure of
issue ownership in all analyses. We first present the results for the State of the Union
address, then for Statutes of Congress, followed by the U.K. Queen’s Speech, and then Acts
of the U.K. Parliament.
Table 3(a) about here
Table 3(a) reveals the limitations of applying any theory of governing party agendas across
all policy domains. Analysing the agendas of the State of the Union address, we find that
issue ownership is only associated with emphasis of one issue, the economy. Issue
ownership effects are mediated by issue salience, as we found in the results using the
pooled models, but the effect is only found in four policy areas; the economy, health,
foreign affairs and the category ‘other’. The result is weakly significant in the opposite
direction for the issue of labour relations.
Moving to the interaction effect for issue ownership*popularity, we find a weakly
positive coefficient on the issue of foreign affairs, rather than a negative coefficient. Recall
that no effect was identified using the pooled models of the agendas of the State of the
Union address, but this result suggests that Presidents will take issue ownership
considerations into account the greater their party’s electoral popularity. The absence of
significant interaction effects for popularity and issue ownership is consistent with the null
finding reported earlier in Table 2(a).
The results for the agenda of U.S. Statutes of Congress are reported in Table 3(b)
Table 3(b) about here
These results further underline the importance of examining governing agendas by issue,
and also the importance of comparing results across institutional context. Whereas the
effect of the issue ownership and salience interaction was positive for Labour issues in the
State of the Union address, the effect is negative (as hypothesised) for Statutes of Congress
on this policy topic. This difference suggests an imperfect correspondence between the
State of the Union address and in legislative outputs, suggesting that while issue ownership
may be more important in the State of the Union address the greater the salience of the
issue, the effect of issue ownership is weaker in explaining Congressional Statutes the higher
21
the salience of issues relating to Labour. Figure 2(a) displays the interaction for labour issues
derived from Table 3(b).
Figure 2(a) about here
Issue salience considerations have a negative effect on the policy agenda of Congress on
issues of labour, the higher ranking a party’s ownership of the issues.
Returning to the examination of issue-specific results in Table 3(b), the results
support the hypotheses in the majority of significant cases, but most results are not
significant. Issue ownership helps explain an incumbent party’s focus on the issues of labour
and social affairs, but not on the economy (unlike in the State of the Union address), health,
education, environmental issues, crime, foreign affairs and the category ‘other’. There is a
significant negative interaction between issue ownership and issue salience for the issues of
labour, crime and foreign affairs, but the relationship is weakly positive for education. The
interaction between issue ownership and incumbent party popularity is negative, as
predicted, but significant only on the issues of Labour and of social affairs. For the purpose
of interpretation and illustration, Figure 2(b) reveals how issue ownership exerts a positive
effect on the policy agenda in Statutes of Congress on social issues when party popularity is
low, but a negative effect when party popularity is high.
Figure 2(b) about here
These results are revealing. While support is given to the theoretical hypothesis for the
majority of significant cases, we find exceptions to the expected directions in a small
number of issues, and null findings over a large number of policy issues. Note that even the
main effect of issue salience is not significant in a majority of cases. These insights into U.S.
politics should provide caution in generalising agenda theories across policy domains.
We next turn to the analysis of U.K. legislative agendas. Table 3(c) presents the issueby-issue models for the agendas of the Queen’s Speech.
Table 3(c) about here
It is noticeable in Table 3(c) that there are few significant effects displayed in the agendas of
the Queen’s Speech, in relation to the findings reported for both cases for the U.S.
legislative agendas. This helps to explain the lack of significant findings across all main
effects and interactions in the pooled model reported in Table 2(b). Not only are the issue
ownership main effects and interactions not significant, on the whole, but issue salience
only explains the policy agenda for the issues of the economy and social affairs. Given that
22
issue salience explanations are the most prominently argued mechanisms in this literature,
these results are revealing and important. However, whereas the interaction effect between
MIP and ownership was not significant across issues, there is evidence of a possible
relationship, as predicted, on the issue of social affairs, significant at the 90% confidence
level. The interaction effect between issue ownership and governing party popularity is
significant on the issue of foreign affairs. This effect was significant in the pooled model, but
Table 3(c) reveals the danger in generalising this finding across issues. Not only is there no
effect on any other policy category, but there is evidence of a small effect in the opposite
direction on policies associated with labour. Note that this ‘labour’ category provides reason
to deduce that agenda incentives may be somewhat different and isolated on this issue,
since there was also an unexpected finding (for a positive interaction between MIP and
issue ownership) for this issue category in the agendas of the State of the Union address
(see Table 3(c)).
Finally, the analysis proceeds to examine the hypotheses for within-party rankings
on issues in Acts of the U.K. Parliament.
Table 3(d) about here
The analyses presented in Table 3(d) provide greater support for the hypotheses in terms of
significant results for the main effect of ownership, and the interactions between ownership
and party popularity, but these findings can by no means be generalised across all issues.
The overall pattern is still one of issue-specific results, and it should be clear that there are
no obvious generalisations that can be made within issue area. The results in Table 3(d) also
reveal that there is no support for Hypotheses 2, which predicted that issue ownership
effects would be mediated by issue salience, except where one effect is found in the
opposite direction for the economy, due to the negative effect of issue salience for
economic issues. These null findings are consistent with the null findings reported in the
pooled model for U.K. legislative outputs, reported in Table 2(b). There is much stronger
support for the third hypothesis; namely, that issue ownership effects will be mediated by
party popularity, and this relationship is demonstrated by the plotting of marginal effects in
Figure 2(c), for the issue of crime/law and order.
Figure 2(c) about here
23
The slope in Figure 2(c) reveals that issue ownership effects have no significant effect on this
issue for low levels of party popularity, but have a negative effect on the policy agenda of
the U.K. Parliament on crime, the greater the party’s electoral standing.
We conclude that there is support for our hypotheses and expectations, but with
important and revealing exceptions. Overall, there is evidence of competence-based
attention to policies in government agendas across issues and across political systems, in
particular with regard to the rank order of party competence evaluations. These effects
appear to be conditional upon the salience of issues, such that issue salience effects weaken
on a party’s owned issues, or issue ownership effects weaken the greater the salience of
issues. Governments’ incentives to attend to salient or owned issues appear to exist as a
trade-off. Either issue salience incentives influence government agendas, or issue ownership
incentives have greater weight, but it is not the case that an issue’s salience enhances the
effect of an issue ownership incentive. Furthermore, we find support for our expectation
that issue ownership strategies are more influential upon governing party agendas for
parties with a weaker electoral mandate, and note that these effects appear to be more
influential in legislative outputs rather than in executive speeches, and especially so in the
U.K. It was also notable that issue salience explanations appear to have more relevance to
executive and legislative policy agendas in the U.S., and less so in the U.K.
However, on further exploration of the hypotheses for individual issues, we conclude
that these theories – and those in existing scholarship concerning issue salience, for
example – are not generalisable across the issue domain. The largest majority of significant
cases continue to support our hypotheses, but null findings on many issues confirms the
importance of analysing effects in the aggregate and across issue, and also the limitations of
generalisation. While we do not find unequivocal asymmetries across our four cases, which
is encouraging with respect to the validity and applicability of the theory advanced, and
while we find support for our hypotheses across various issue domains, there is a clear need
for caution in applying government attention theories across any issue domain. In particular,
the issues of labour and the economy may warrant special attention as possible unique
cases, whereby the effect of issue ownership of these issues is sometimes greater the higher
the salience of the issue (for the U.S. State of the Union and U.K. Acts of Parliament), or the
greater a government’s popularity (U.K. Queen’s Speech). There was also some evidence of
a stronger issue ownership effect for government popularity on the issue of foreign affairs
24
(U.S. State of the Union). These anomalies might suggest that these executive issues provide
governments with different incentives, such that they address them the higher their
salience or the higher a government’s electoral mandate, but this explanation would not
apply to an individual finding for a multiplicative issue ownership and salience effect for
education (U.S. Statutes of Congress), although just significant at the 90% level.
Overall, the results reveal that while presidents, prime ministers and their parties
attend to issues in ways consistent with party competence and competition-based
incentives, the salience of issues to the public remain important and can require governing
parties to shift attention away from their preferred issues. Presidents, prime ministers and
parties are also more likely to respond to issue ownership incentives, the weaker their
electoral support in the country. While these explanations account for the policy agendas of
U.S. and U.K. governments in the vast majority of significant cases, the absence of effects in
individual issue-topics suggests that these expectations cannot be consistently applied.
Robustness of Results
We re-estimated the pooled and issue specific models to take into account a number of
potential confounding factors, which may potentially influence the validity of the results.
We re-estimated all models controlling for the party of the incumbent.6 Although we
have no theoretical reason to expect Republicans or Democrats to be more inclined to focus
on an ownership or salience strategy as a result of issue salience or party popularity (the
mediating hypotheses), it is plausible that some issue ownership and salience effects may be
accounted for by stable party-based priorities over time. Controlling for party does not alter
the significance or direction of the reported main effects or interactions, but we find some
substantively interesting party effects in addition those reported in the results. We found no
party effect whatsoever on the policy agendas of the State of the Union address. There were
two party effects in the agendas of U.S. Statutes; Democrats attend more to the
environment issue and to the ‘other’ category than Republicans. Moving to the U.K., we
found that Labour attend more to health and social issues in the policy agendas of the
Queen’s Speech, and with respect to Acts of Parliament, the Conservatives attend more to
6
A dummy variable for party was entered where 1 = Republican or Conservative and 0 = Democrat
or the Labour party.
25
economy and to the ‘other’ category, and Labour attend more to foreign affairs. As noted,
controlling for party resulted in no change to our substantive conclusions.
We next took a series of steps to ensure that our coding of missing data in our rank
issue ownership variable could not be responsible for the results. Whereas we treated
missing data as missing in the analyses reported above, a decision made in order to offer a
conservative test of the hypotheses, we re-estimated all the models using one of two
different treatments of missing data. In the first we reassigned missing values to the median
value on the 1-9 scale, and in the second we re-assigned missing values to the lowest rank
value, 1. The coding of missing data made no difference to the reported results. Also, in
order to be more certain of the robustness of the results, we re-estimated models using
data from 1974 onwards, since survey data on party competence tended to be far less
frequent prior to this. The number of statistically significant results was slightly lower, as
would be expected given the smaller sample size of the data, but the direction of results
remained consistent, and support for each hypothesis remained similarly robust.
Finally, we re-estimated the models using a variable for party popularity based on
the proportion of party seats, rather than the proportion of the incumbent party’s vote
intention. This may be especially relevant in the U.S. case, where the proportion of the
incumbent party’s Congressional seat share will be an important determinant of legislative
initiative and the ability of the incumbent party to achieve legislative outcomes. The vote
and seat share measures are strongly aligned in the U.K. (correlated at 0.82***) but not in
the U.S. (0.39***). However, it is also plausible that a model based on seat shares may be
substantially important in the U.K., since if parties fall back onto an ownership strategy
when they are less popular, this may arise because a party’s owned issues are partisan
priorities – which the party backbench will be more inclined to support in voting lobbies.
Such an explanation may also hold in the U.S. Analysis of the results suggests that the
measure of popularity does not make a difference to the results, although the results for the
conditional relationship between issue ownership and popularity are slightly stronger in the
issue specific models for Acts of the U.K. Parliament than the model reported in Table 3(d),
above. The issue-specific results are reported in a reviewer appendix Table RA1(a-d). Note
that the pooled model results were completely unchanged, and these models are therefore
omitted from the reviewer appendix. Additionally, we modelled the results using vote
intention while controlling for seat shares in additional robustness checks. While some of
26
the main effects for popularity (vote intention share) dropped out of significance, the main
effects for ownership and salience, and the interaction effects for ownership*salience and
ownership*popularity all continued to support the conclusions reported in the paper.
Conclusions
Issue ownership theories (Budge and Farlie 1977; 1983; Budge et al. 1987; Budge 1993;
Petrocik 1996; Petrocik et al. 2003) have been applied extensively to our understanding of
the election campaign and advertising strategies of parties and candidates. This paper
argued that if parties and candidates have incentives to prime voters to cast a ballot on their
owned issues in elections, they should also be more inclined prioritise these issues while in
government. Ownership denotes greater commitment to partisan policy priorities over
time, and parties can demonstrate such long-term commitment while in office. However,
since government is thought to be about responding to policy problems and public
concerns, we theorised that the policy attention of government would be governed by issue
ownership considerations mediated by the importance of issues. We also theorised that
issue ownership considerations would be conditional on the popularity of the incumbent
government’s party. A party with a large electoral mandate can govern on any issue, but a
party with a weaker electoral mandate will likely fall back onto issue ownership
considerations in delivering its partisan legislative agenda in office.
The analysis of executive legislative priorities in U.S. State of the Union speeches
(1945-2007) and the U.K. Speech from the Throne (or the Queen’s Speech), and of
legislative outcomes in U.S. Statutes of Congress (1945-2007) and Acts of the U.K.
Parliament (1945-2011) gave support to our theoretical expectations.
In order to specify the substantive limitations and extent of the support for our
hypotheses, we further examined the hypotheses breaking down the analyses across nine
policy issue areas. In the largest majority of significant cases, our hypotheses were
supported, but it is important to note that no mechanism modelled in the paper – neither
for issue salience, issue ownership or the mediating effects of issue salience or party
popularity – were able to explain the policy agendas of U.S. and U.K. governments across all
issues and cases. This is not particularly surprising to us, but it is an important observation in
a literature which tends to take individual issue case studies and extrapolate their
implications. For many issues, null findings were reported across each case. In a handful of
27
issues and cases, parties may employ issue ownership considerations more when issues are
salient and when the party is popular, but these results were confined to a small number of
executive issues, such as the economy, labour issues, and foreign affairs. These examples
serve to underline the relevance of issue ownership applications to governing party
attention to the policy domain.
We conclude that issue ownership is an important concept to apply to the analysis of
executive agendas and legislative outputs, but it should not be applied in general. It is
encouraging that we found support for our hypotheses across countries, and in executive
agendas (in speeches) and in actual legislative outcomes. However, we cannot conclude that
the effects we report in the paper can be generalised across one policy issue for any party in
any particular context. We also cannot be sure whether issue ownership evaluations reflect
the partisan priorities of governments, since ownership should be associated with long-term
policy prioritisation, or whether issue ownership evaluations cause parties to focus on the
issues on which they are positively rated. Our robustness checks for party control would
suggest the latter, but the answer is also likely to be circular. Although our analyses span six
decades of legislative agendas in both countries, further research is needed to understand
when – and on what issues – governments will respond to the incentives we reveal in this
paper. It seems plausible to us that it is the salience of issues that makes these incentives
more or less important, and this may be a fruitful avenue for future research. So too may
the diversity of the issue agenda be relevant to understanding the ability of governments to
attend to their owned issues, or to attend to their owned issues as a function of salience of
governing party popularity. Certain ‘core issues’ of government tend to crowd out
peripheral/conditional issues when the policy agenda becomes less diverse (Jennings et al.
2011b; also see Adler and Wilkerson n.d.). These questions may usefully be the subject of
further research.
Notwithstanding remaining questions, there are three important implications of the
analyses we report in this paper, both for electorates and for governments.
The first implication is that the issue agendas of parties in elections provide a useful
cue to voters of the likely focus of that party in government. If we assume that parties and
candidates employ issue ownership strategies in their campaigns, at least to some extent,
our results reveal that rather than simply prime votes to cast a ballot on the issues
beneficial to a given party, the campaign agendas of political parties provide an
28
informational source on a party’s intentions and behaviours in government. This
relationship will not be perfect, but issue ownership provides a very helpful cue. Campaigns
may therefore provide an additional role to the electorate, informing voters of the likely
actions and outcomes of an incoming government, and aiding a vote choice, accordingly.
The second implication is that the cycle of government popularity may tell us a great
deal about the policy agendas of governments. It is often the case that parties and
presidents enter government with a bounce in terms of popularity, followed by gradual
decline. These so-called cycles of government popularity, and costs of governing effects, are
well documented in the literature (Stimson 1976; Rose and Mackie 1983; Paldam 1986;
Sanders 2005). The results we report for the relationship between incumbent party
popularity and the effects of issue ownership on government policy agendas may suggest
that a bias towards a party’s own issues may be greatest towards the end of a government
cycle. Thus, while the issue ownership strategies of parties and candidates in campaigns
may provide a useful cue to voters regarding the policy priorities of future governments,
they are less likely to do so towards the beginning of an administration’s period in office.
This offers a potentially important insight into the possible cyclical nature of government
policy attentiveness, and the importance of context, and future research may explore the
length of time in office as a conditioning factor on the effects of issue ownership.
The final implication is that although the issues of greatest concern to the public do
not exert a strong direct effect upon the attentiveness of governments to policy issues – in
the U.K. especially but also in the U.S. – it is normatively desirable and encouraging that
issue salience considerations appear to outweigh issue ownership considerations when
issues are high on the public agenda, or when a party is less positively ranked on a given
issue domain. Although governments may wish to attend to their ‘own’ issues, they are less
inclined to do so when there are important issues of public concern, and more inclined to do
so when their ranking is less positive.
Through the combination of insights of theories of attention-driven choice in
government (e.g. Jones 2001; Jones and Baumgartner 2005) with theories of the role of
selective emphasis, issue ownership and competence (e.g. Budge and Farlie 1977; 1983;
Budge et al. 1987; Budge 1993; Petrocik 1996; Petrocik et al. 2003), there is potential for
better understanding of when governments attend to their party-based competence
strengths and when other issues are prioritised above them. The extension of election29
focused theories of selective emphasis and issue ownership to the policy agendas of
government therefore raise exciting prospects.
30
Bibliography
Adams, James. (2001). Party Competition and Responsible Party Government: A Theory of
Spatial Competition Based upon Insights from Behavioral Voting Research. Ann Arbor:
University of Michigan Press.
Adams, James, Samuel Merrill, and Bernard Grofman. (2005). A Unified Theory of Party
Competition: A Cross-National Analysis Integrating Spatial and Behavioral Factors.
Cambridge: Cambridge University Press.
Adler, E. Scott, and John D. Wilkerson. (2012). Congress and the Politics of Problem Solving.
Cambridge: Cambridge University Press.
Adler, E. Scott, and John D. Wilkerson. (n.d.). ‘Governing Priorities and Agenda Setting in the
U.S. House of Representatives.’ Unpublished manuscript.
Aldrich, John Herbert. (1995). Why Parties? The Origin and Transformation of Political
Parties in America, American politics and political economy. Chicago: University of
Chicago Press.
Ansolabehere, Stephen, and Shanto Iyengar. (1994). ‘Riding the Wave and Claiming Ownership over
Issues: The Joint Effects of Advertising and News Coverage in Campaigns.’ Public Opinion
Quarterly 58: 335-357.
Bara, Judith. (2005). ‘A Question of Trust: Implementing Party Manifestos.’ Parliamentary
Affairs 58(3): 585-599.
Baumgartner, Frank R., Christoffer Green-Pedersen and Bryan D. Jones. (2006).
‘Comparative Studies of Policy Agendas.’ Journal of European Public Policy 13 (7): 95974.
Baumgartner, Frank R., Emiliano Grossman, and Sylvain Brouard. (2009). ‘Agenda-setting
Dynamics in France: Revisiting the ‘Partisan Hypothesis’.’ French Politics 7(2): 75–95.
Beck, Nathaniel and Jonathan N. Katz. (1995). ‘What To Do (and Not To Do) with Time-Series
Cross-Section Data.’ American Political Science Review 89: 634-47.
Bélanger, Éric, and Bonnie Meguid. (2008). ‘Issue salience, Issue Ownership and Issue Based
Vote Choice: Evidence from Canada.’ Electoral Studies 27(3): 477-491.
Bellucci, Paolo. (2006). ‘Tracing the cognitive and affective roots of ‘party competence’: Italy
and Britain, 2001.’ Electoral Studies 25(3): 548-569.
Bevan, Shaun, and Will Jennings. (2010). ‘Opinion-Responsiveness of Governing Agendas in
the US and U.K.: Institutional Filtering of Issue Priorities of the Public’, Annual Meeting of
the American Political Science Association conference, Washington D.C., September 3,
2010.
31
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.
Blais, Andre, Donald Blake, and Stephane Dion. (1993). ‘Do Parties Make a Difference?
Parties and the Size of Government in Liberal Democracies.’ American Journal of Political
Science 37(1): 40-62.
Blomqvist, Paula and Christoffer Green-Pedersen (2004). ‘Defeat at Home? Issue-Ownership
and Social Democratic Support in Scandinavia.’ Government and Opposition 39(4): 587613.
Bräuninger, Thomas, and Marc Debus. (2009). ‘Legislative agenda-setting in parliamentary
democracies.’ European Journal of Political Research 48(6): 804-839.
Budge, Ian. (1993). ‘Issues, Dimensions, and Agenda Change in Post-war Democracies: Longterm Trends in Party Election Programs and Newspaper Reports in Twenty-three
Democracies.’ In W.H. Riker (ed.) Agenda Formation. Ann Arbor: The University of
Michigan Press.
Budge, Ian. (2001). Mapping Policy Preferences: Estimates for Parties, Electors, and
Governments, 1945-1998. Oxford: Oxford University Press.
Budge, Ian, and Dennis Farlie. (1977). Voting and Party Competition: A Theoretical Critique
and Synthesis Applied to Surveys from Ten Democracies. London: Wiley.
Budge, Ian, and Dennis Farlie. (1983). Explaining and Predicting Elections: Issue Effects and
Party Strategies in Twenty-Three Democracies. London: Allen & Unwin.
Budge, Ian, Derek Hearl, and David Robertson. (1987). Ideology, Strategy and Party Change:
Spatial Analyses of Post-War Election Programmes in 19 Democracies. Cambridge:
Cambridge University Press.
Budge, Ian, and Richard I. Hofferbert. (1990). ‘Mandates and Policy Outputs: U.S. Party
Platforms and Federal Expenditures.’ American Political Science Review 84(1): 111-132.
Castles, Frank G. (ed.). (1982). The Impact of Parties Politics and Policies in Democratic
Capitalist States. London: Sage.
Castles, Frank G., and Robert D. Mckinlay. (1979). ‘Does Politics Matter. An Analysis of the
Public Welfare Commitment in Advanced Democratic States.’ European Journal of
Political Research 7(2):169-186.
Clarke, Harold D., David Sanders, Marianne C. Stewart, and Paul F. Whiteley. (2004). Political
Choice in Britain. Oxford: Oxford University Press.
32
Clarke, Harold D., David Sanders, Marianne C. Stewart, and Paul F. Whiteley. (2009).
Performance Politics and the British Voter. Cambridge: Cambridge University Press.
Cohen, Jeffrey E. (1997). Presidential Responsiveness and Public Policy-making. Ann Arbor:
University of Michigan Press.
Cummins, Jeff. (2010). ‘The Partisan Considerations of the President’s Agenda.’ Polity 42(3):
398-422.
Damore, David F. (2004). The Dynamics of Issue Ownership in Presidential Campaigns.
Political Research Quarterly, Vol. 57(3): 391 – 397.
Damore, David F. (2005). ‘Issue Convergence in Presidential Campaigns.’ Political Behavior
27(1): 71-97.
Edwards, George C. III, and Andrew Barrett. (2000). ‘Presidential Agenda-Setting in
Congress.’ In Jon R. Bond and Richard Fleischer (eds.) Polarized Politics: Congress and the
President in a Partisan Era. Washington, DC: CQ Press.
Edwards III, George C., and B. Dan Wood. (1999). ‘Who Influences Whom? The President,
Congress and the Media.’ American Political Science Review 93(2): 327-44.
Eshbaugh-Soha, Matthew, and Jeffrey Peake. (2004). ‘Presidential Influence Over the
Systemic Agenda.’ Congress and the Presidency 31(2): 161-81.
Feeley, T. Jens, Bryan D. Jones and Heather Larsen. (2001). Public Agendas: Annualized Most
Important Problem Polling Data, 1939-2001. Seattle, WA: University of Washington
(Portions of this data originally compiled by the Gallup Organization) [Computer file].
Green, Jane. (2011). ‘A Test of Core Vote Theories: The British Conservatives, 1997-2005.’
British Journal of Political Science 41(4): 667-688.
Green, Jane, and Sara B. Hobolt. (2008). ‘Owning the Issue Agenda: Explaining Party
Strategies in British General Election Campaigns.’ Electoral Studies 27(3): 460-476.
Green, Jane, and Will Jennings. (2012a). ‘Valence as Macro-Competence: An Analysis of
Mood in Party Competence Evaluations in the U.K.’ British Journal of Political Science
42(2): 311-343.
Green, Jane, and Will Jennings. (2012b). ‘The Dynamics of Issue Competence and Vote
Choice for Parties in and Out of Power.’ European Journal of Political Research 51(4):
469-503.
Green-Pedersen, Christoffer, and Peter B. Mortensen. (2010). ‘Who sets the agenda and
who responds to it in the Danish parliament? A new model of issue competition and
agenda-setting.’ European Journal of Political Research 49(2): 257-281.
33
Grossback, Lawrence J., David A.M. Peterson, and James A. Stimson. (2006). Mandate
Politics. Cambridge: Cambridge University Press.
Hibbs, Douglas A., Jr. (1977). ‘Political Parties and Macroeconomic Policy.’ American Political
Science Review 71(4): 1467-1487.
Hibbs, Douglas A. (1979). ‘The Mass Public and Macro-Economic Policy: The Dynamics of
Public Opinion towards Unemployment and Inflation.’ American Journal of Political
Science 2(3): 705-31.
Hill, Kim Q. (1998). ‘The Policy Agendas of the President and the Mass Public: A Research
Validation and Extension.’ American Journal of Political Science 42(4): 1328-34.
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.
Hofferbert, Richard I., and Ian Budge. (1992). ‘The Party Mandate and the Westminster
Model: Party Programmes and Government Spending in Britain, 1948-1985.’ British
Journal of Political Science 22(2): 151-182.
Iyengar, Shanto, and Donald R. Kinder. (1987). News that Matters: Television and American
Opinion, American Politics and Political Economy. Chicago: University of Chicago Press.
Iyengar, Shanto, and Adam F. Simon (2000). ‘New Perspectives and Evidence on Political
Communication and Campaign Effects.’ Annual Review of Psychology 51(1): 149-169.
Kitschelt, Herbert. (1989). The Logics of Party Formation: Ecological Politics in Belgium and
West Germany. Ithaca: Cornell University Press.
Kitschelt, Herbert. (1994). The Transformation of European Social Democracy, Cambridge
studies in comparative politics. Cambridge: Cambridge University Press.
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 and Peter John. (2011a). ‘The Agenda of British Government:
the Speech from the Throne, 1911-2008.’ Political Studies 59(1): 74-98.
Jennings,Will, Shaun Bevan, Arco Timmermans, Laura Chaques, Gerard Breeman, Sylvain
Brouard, Christoffer Green-Pedersen, Peter John, Peter B. Mortensen and Anna Palau.
(2011b). ‘Effects of the Core Functions of Government on the Diversity of Executive
Agendas.’ Comparative Political Studies 44(8): 1001-1030.
34
Jennings, Will, and Christopher Wlezien. (2011). ‘Distinguishing between Most Important
Problems and Issues?’ Public Opinion Quarterly 75(3): 545–555.
Johns, Robert, James Mitchell, David Denver and Charles Pattie. (2009). ‘Valence Politics in
Scotland: Towards an Explanation of the 2007 Election.’ Political Studies 57(1): 207-233.
Jones, Bryan D. (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. (2005). The Politics of Attention: How
Government Prioritizes Problems. Chicago: University of Chicago Press.
Jones, Bryan D., Heather Larsen-Price, and John Wilkerson. (2009). ‘Representation and
American Governing Institutions.’ Journal of Politics 71(1): 277-290.
Kaplan, Noah, David K. Park, and Travis D. Ridout. (2006). ‘Dialogue in American Political
Campaigns: An examination of issue engagement in candidate television advertising.’
American Journal of Political Science 50(3):724-736.
Kingdon, John W. (1984). Agendas, Alternatives, and Public Policies. New York: Harper
Collins.
McDonald, Michael D., Sylvia M. Mendes and Ian Budge. (2004). ‘What Are Elections For?
Conferring the Median Mandate.’ British Journal of Political Science 34(1): 1-26.
McDonald, Michael D., and Ian Budge. (2005). Elections, Parties, Democracy: Conferring the
Median Mandate. Oxford: Oxford University Press.
Mortensen, Peter B.,Christoffer Green-Pedersen, Gerard Breeman, Will Jennings, Peter
John, Arco Timmermans, Laura Chaques and Anna Palau. (2011). ‘Comparing
Government Agendas: Executive Speeches in the Netherlands, United Kingdom and
Denmark.’ Comparative Political Studies 44(8): 973-1000.
Nadeau, Richard, André Blais, Elisabeth Gidengil, and Neil Nevitte. (2001). ‘Perceptions of
party competence in the 1997 election.’ In Hugh G. Thorburn and Alan Whitehorn (eds.)
Party Politics in Canada, eighth ed. Toronto: Prentice-Hall, pp. 413-430.
Paldam, Martin. (1986). ‘The distribution of election results and the two explanations of the
cost of ruling.’ European Journal of Political Economy 2(1): 5–24.
Page, Benjamin I. and Robert Y. Shapiro. (1983). ‘Effects of public opinion on policy.’
American Political Science Review 77(1): 175-190.
35
Park, Rolla E., and Bridger R. Mitchell. (1980). ‘Estimating the Autocorrelated Error Model
with Trended Data.’ Journal of Econometrics 13(2): 185–201.
Petrocik, John R., William L. Benoit and Glenn J. Hansen. (2003). ‘Issue Ownership and
Presidential Campaigning, 1952-2000.’ Political Science Quarterly 118(4): 599-626.
Petrocik, John R. (1996). ‘Issue Ownership in Presidential Elections, with a 1980 Case Study.’
American Journal of Political Science 40(3): 825-50.
Rose, Richard. (1980). Do Parties Make a Difference? London: Macmillan.
Rose, Richard. (1983). ‘Still the era of party government.’ Parliamentary Affairs 36(3): 282299.
Rose, Richard, and Philip L. Davies. (1994). Inheritance in Public Policy: Change Without
Choice in Britain. New Haven, CT: Yale University Press.
Rose, Richard, and Thomas T. Mackie. (1983). ‘Incumbency in Government: Asset or
Liability?’ in Hans Daalder and Peter Mair (eds.) Western European Party Systems:
Continuity and Change. Beverley Hills, Calif.: Sage.
Sanders, David. (2005). ‘The political economy of UK party support, 1997-2004: Forecasts for
the 2005 general election.’ Journal of Elections, Public Opinion & Parties 15(1): 47-71.
Sigelman, Lee, and Emmett H. Buell. (2004). ‘Avoidance or Engagement? Issue Convergence
in U.S. Presidential Campaigns, 1960-2000.’ American Journal of Political Science 48(4):
650-661.
Simon, Adam F. (2002). The Winning Message: Candidate Behavior, Campaign Discourse,
and Democracy. Cambridge: Cambridge University Press.
Somer-Topcu,Zeynep. (2009). ‘Timely Decisions: The Effects of Past National Elections on
Party Policy Change.’ Journal of Politics 71(1): 238-248.
Spiliotes, Constantine J., and Lynn Vavreck. (2002). ‘Campaign advertising: partisan
convergence and divergence?’ Journal of Politics 64(1): 249–261.
Stimson, James A. (1976). ‘Public Support for American Presidents: A Cyclical Model.’ Public
Opinion Quarterly 40(1): 1-21.
Strom, Kaare. (1990). ‘A Behavioural Theory of Competitive Political Parties.’ American
Journal of Political Science 34(2):565 - 598.
Vavreck, Lynn. (2009). The Message Matters: The Economy and Presidential Campaigns.
Princeton: Princeton University Press.
36
Wagner, Markus and Thomas Meyer. (2012). ‘Explaining Parties’ Salience Strategies: the
moderating role of party size.’ Unpublished manuscript.
Wlezien, Christopher, Will Jennings, Stephen Fisher, Robert Ford and Mark Pickup. (2012).
‘Polls and the Vote in Britain.’ Forthcoming in Political Studies.
Yates, Jeff, and Andrew Whitford. (2005). ‘Institutional Foundations of the President’s Issue
Agenda.’ Political Research Quarterly 58(4): 577-85
37
Figure 1(a): Marginal Effects of Issue Salience for values for Issue Ownership in the
Policy Agendas of the US State of the Union Address
State of the Union - Effect of MIP
2.00
1.75
Marginal Effect
1.50
1.25
1.00
0.75
0.50
0.25
0.00
1
2
3
4
5
6
Party Competence
7
8
9
Figure 1(b): Marginal Effects of Issue Ownership for values for Popularity in the
Policy Agendas of US Statutes of Congress
Statutes
Competence
Statutesof
ofUS
USCongress
Congress -- Effect
Effect of
of Party
Party Competence
5.0
10.0
4.0
9.0
3.0
8.0
Marginal Effect
Marginal Effect
2.0
7.0
1.0
6.0
0.0
5.0
-1.0
4.0
-2.0
3.0
2.0
-3.0
1.0
-4.0
0.0
-5.0
300
535
10 40 15
4520
25
3055
50
Popularity
Popularity
35 60 40
6545
50
70
38
Figure 1(c): Marginal Effects of Issue Ownership for values for Popularity in the
Policy Agendas of British Statutes of Parliament
Acts of UK Parliament - Effect of Party Competence
5.0
Marginal Effect
2.5
0.0
-2.5
-5.0
30
35
40
45
50
Popularity
55
60
65
70
39
Figure 2(a): Marginal Effects of Issue Salience for values for Issue Ownership in the
Policy Agendas of US Statutes of Congress on Labour issues
Statutes of US Congress - Effect of MIP
3
2
1
0
-1
-2
-3
1
2
3
4
5
6
Party Competence
7
8
9
Figure 2(b): Marginal Effects of Issue Ownership for values for Popularity in the
Policy Agendas of the US State of the Union Address on Social issues
Statutes of US Congress - Effect of Party Competence
20.0
15.0
Marginal Effect
10.0
5.0
0.0
-5.0
-10.0
-15.0
-20.0
30
35
40
45
50
Popularity
55
60
65
70
40
Figure 2(c): Marginal Effects of Issue Ownership for values for Popularity in the
Policy Agendas of U.K. Statutes of Parliament on the issue of Crime
Acts of UK Parliament - Effect of Party Competence
5.0
Marginal Effect
2.5
0.0
-2.5
-5.0
30
35
40
45
50
Popularity
55
60
65
70
41
Table 1. Policy Agendas Project Major Topic Codes
Topic
Abbreviation
Policy Agendas Major Topic Codes
Issue Competence Categories
1
Economy
Macroeconomics
Economy
3
Health
Health
Health
5
Labor
Labor, Employment, and Immigration
Labor, Employment, and Immigration
6
Education
Education
Education
7
Environment
Environment
Environment
12
Law
Law, Crime, and Family Issues
Law, Crime, and Family Issues
13
14
16
19
Social
Foreign
2
Social Welfare
Community Development, Planning and Housing Issues
Defense
International Affairs and Foreign Aid
4
Civil Rights, Minority Issues, and Civil Liberties
Agriculture
8
Energy
10
Other
Transportation
18
Foreign Trade
20
Government Operations
21
Public Lands and Water Management
Welfare, Pensions & Housing
Defense, Foreign Affairs & Terrorism
Other (includes transport, agriculture,
privacy women, general performance,
and morality issues).
Policy Agendas Topic Codebook, see www.policyagendas.org
42
Table 2(a). Time Series Cross-Sectional AR(1) Model of Party Competence and Government Agendas US 1945 – 2007
AGENDAit
Executive
Legislative Outputs
(State of the Union)
(Statutes of US Congress)
Model 1
Model 2
Model 3
Model 4
Model 1
Model 2
Model 3
Model 4
-0.031
(0.063)
-
-
-
0.021
(0.029)
-
-
-
Party Competenceit (majority)
-
-0.500
(2.207)
-
-
-
0.739
(0.697)
-
-
Party Competenceit (ownership)
-
-
1.197*
(0.536)
2.044
(5.331)
-
-
0.347*
(0.171)
4.387*
(1.799)
MIPit
0.474+
(0.243)
0.402**
(0.091)
1.065**
(0.234)
1.073**
(0.246)
0.047
(0.059)
0.066**
(0.026)
0.147**
(0.055)
0.081
(0.059)
MIPit * Party Competenceit
-0.000
(0.004)
0.079
(0.108)
-0.082**
(0.030)
-0.083**
(0.032)
0.000
(0.001)
0.003
(0.023)
-0.009
(0.007)
-0.005
(0.007)
Popularityt
-
-
-
0.106
(0.852)
-
-
0.639*
(0.277)
Party Competencet * Popularityt
-
-
-
-0.016
(0.107)
-
-
-0.081*
(0.035)
Constant
9.350*
(3.768)
7.949**
(1.868)
-0.461
(3.590)
-5.806
(42.224)
10.548***
(1.591)
11.279***
(0.912)
8.831***
(1.305)
-22.736
(14.133)
R-squared
0.166
0.153
0.219
0.224
0.410
0.425
0.461
0.282
265
265
265
265
281
281
281
281
9
9
9
9
9
9
9
9
Party Competenceit (level)
N
Panels
+ p ≤ 0.10, * p ≤ .05, ** p ≤ .01
43
Table 2(b). Time Series Cross-Sectional AR(1) Model of Party Competence and Government Agendas U.K.1945 – 2011
AGENDAit
Executive
Legislative Outputs
(Speech from the Throne)
(Acts of U.K. Parliament)
Model 1
Model 2
Model 3
Model 4
Model 1
Model 2
Model 3
Model 4
0.015
(0.039)
-
-
-
0.036
(0.060)
-
-
-
Party Competenceit (majority)
-
-0.035
(0.840)
-
-
-
0.353
(1.122)
-
-
Party Competenceit (ownership)
-
-
0.232‡
(0.172)
0.491
(0.850)
-
-
-0.027
(0.285)
3.106**
(0.892)
MIPit
-0.061
(0.136)
0.035
(0.043)
0.075
(0.060)
0.054
(0.071)
0.193
(0.234)
0.039
(0.063)
-0.129
(0.110)
0.118
(0.106)
MIPit * Party Competenceit
0.002
(0.002)
0.023
(0.030)
-0.004
(0.009)
-0.002
(0.011)
-0.003
(0.004)
0.013
(0.069)
-0.014
(0.018)
-0.013
(0.017)
Popularityt
-
-
-
0.019
(0.106)
-
-
-
0.401**
(0.115)
Party Competencet * Popularityt
-
-
-
-0.004
(0.020)
-
-
-
-0.079**
(0.022)
Constant
6.947**
(2.340)
7.857**
(1.231)
6.239***
(1.360)
5.561
(4.485)
7.674*
(3.390)
9.160***
(0.056)
9.775***
(1.770)
-6.108
(4.779)
R-squared
0.049
0.057
0.038
0.043
0.199
0.180
0.195
0.229
394
394
394
376
376
376
376
376
9
9
9
9
9
9
9
9
Party Competenceit (level)
N
Panels
+ p ≤ 0.10, * p ≤ .05, ** p ≤ .01
‡ p=0.18
44
Table 3(a). Prais-Winsten AR(1) Model of Party Competence and Government Agendas US 1945 – 2007
AGENDAt
Executive
(State of the Union)
Economy
Health
Labour
Education
Environ
Crime
Social
Foreign
Other
Issue Ownershipt
27.879+
(14.293)
13.745
(8.684)
4.675
(6.644)
3.565
(16.172)
-0.337
(5.601)
49.186
(30.045)
-30.042
(36.211)
-33.350
(20.760)
-6.958
(26.390)
MIPt
1.160**
(0.418)
6.542**
(1.709)
-6.942
(4.521)
2.299
(2.425)
-1.140
(2.546)
-0.079
(0.512)
3.174
(2.009)
1.244*
(0.481)
2.620+
(1.288)
MIPt * Issue Ownershipt
-0.131*
(0.051)
-0.659**
(0.207)
1.269+
(0.735)
-0.106
(0.346)
0.302
(0.466)
0.016
(0.131)
-0.327
(0.247)
-0.121+
(0.063)
-0.306+
(0.156)
Popularityt
3.680+
(2.196)
2.249
(1.333)
0.998
(0.924)
1.025
(1.946)
0.276
(0.755)
8.446+
(4.001)
-5.392+
(2.413)
-5.291
(3.226)
-2.131
(3.674)
Issue Ownershipt * Popularityt
-0.455
(0.275)
-0.215
(0.162)
-0.126
(0.115)
-0.058
(0.313)
0.003
(0.120)
-0.965
(0.585)
0.635
(0.685)
0.704+
(0.407)
0.262
(0.482)
-209.800+
(113.654)
-137.914+
(71.646)
-35.088
(51.285)
-55.544
(99.591)
-11.256
(34.936)
-422.944+
(213.437)
260.051+
(132.541)
282.834+
(167.265)
72.309
(200.059)
Rho
0.088
-0.483
-0.027
-0.272
-0.148
-0.682
-0.189
0.253
-0.019
Adjusted R-squared
0.178
0.863
0.250
0.620
0.229
0.619
0.579
0.203
0.050
Durbin-Watson d-statistic
(transformed)
1.945
1.761
1.782
2.119
1.987
1.632
1.933
1.909
1.840
59
17
33
17
17
13
13
59
37
Constant
N
+p ≤ 0.10, * p ≤ .05, ** p ≤ .01
45
Table 3(b). Prais-Winsten AR(1) Model of Party Competence and Government Agendas US 1945 – 2007
AGENDAt
Legislative Outputs
(Statutes of US Congress)
Economy
Health
Labor
Education
Environ
Crime
Social
Foreign
Other
Issue Ownershipt
-1.653
(2.141)
-5.502
(3.405)
4.022*
(1.919)
-6.878
(4.066)
0.731
(1.919)
10.708
(10.405)
23.304*
(8.672)
6.126
(4.120)
10.941
(13.336)
MIPt
0.074
(0.069)
0.201
(0.671)
1.285+
(0.714)
-1.052
(0.623)
-0.822
(1.391)
0.347+
(0.183)
0.312
(0.560)
0.316**
(0.101)
-0.570
(0.548)
MIPt * Issue Ownershipt
-0.007
(0.008)
-0.013
(0.082)
-0.294*
(0.131)
0.162+
(0.089)
0.106
(0.262)
-0.101+
(0.049)
-0.003
(0.072)
-0.028*
(0.013)
0.072
(0.065)
Popularityt
-0.307
(0.329)
-0.713
(0.447)
0.685*
(0.280)
-0.804
(0.455)
-0.044
(0.216)
2.147
(1.482)
0.943+
(0.461)
0.885
(0.639)
1.821
(1.889)
Issue Ownershipt * Popularityt
0.039
(0.041)
0.103
(0.067)
-0.076*
(0.035)
0.115
(0.079)
-0.018
(0.040)
-0.173
(0.201)
-0.438*
(0.165)
-0.109
(0.081)
-0.244
(0.243)
Constant
15.098
(16.874)
40.301+
(22.490)
-32.011*
(15.010)
49.400+
(23.075)
7.480
(10.212)
-112.264
(78.823)
-52.541+
(25.475)
-37.020
(33.091)
-15.136
(102.520)
Rho
0.203
0.268
-0.254
-0.120
-0.661
-0.464
-0.494
-0.084
0.645
Adjusted R-squared
0.010
0.423
0.259
0.510
0.519
0.374
0.565
0.321
0.710
Durbin-Watson d-statistic
(transformed)
1.958
1.603
2.052
1.898
1.273
2.077
2.052
1.916
1.801
60
19
35
19
19
15
15
60
39
N
+p ≤ 0.10, * p ≤ .05, ** p ≤ .01,
46
Table 3(c). Prais-Winsten AR(1) Model of Party Competence and Government Agendas U.K.
AGENDAt
Executive
(Speech from the Throne)
Economy
Health
Labour
Education
Environ
Crime
Social
Foreign
Other
Issue Ownershipt (rank)
0.942
(2.227)
0.590
(0.416)
-1.793
(1.237)
1.142
(1.722)
-2.682
(2.570)
-0.897
(1.916)
2.960*
(1.409)
6.119**
(1.656)
4.725
(3.574)
MIPt
0.176**
(0.064)
0.091
(0.075)
0.037
(0.139)
1.280
(0.764)
1.197
(1.011)
0.180
(0.289)
1.479*
(0.547)
0.067
(0.215)
0.468
(0.625)
MIPt * Issue Ownershipt
-0.015
(0.013)
0.011
(0.010)
0.009
(0.022)
-0.156
(0.133)
-0.162
(0.185)
0.068
(0.072)
-0.142+
(0.071)
-0.042
(0.029)
-0.050
(0.082)
Popularityt
0.151
(0.263)
0.001
(0.071)
-0.168
(0.156)
0.336
(0.270)
-0.377
(0.330)
-0.006
(0.236)
0.226
(0.229)
0.198
(0.251)
0.282
(0.508)
Issue Ownershipt * Popularityt
0.002
(0.052)
-0.009
(0.011)
0.045+
(0.025)
-0.024
(0.050)
0.086
(0.063)
0.007
(0.044)
-0.050
(0.038)
-0.103*
(0.041)
-0.061
(0.089)
Constant
-5.523
(11.441)
0.363
(2.553)
10.086
(7.878)
-10.820
(9.650)
13.009
(13.292)
9.349
(11.060)
-11.733
(8.166)
10.914
(11.436)
7.670
(19.679)
Rho
-0.034
-0.103
-0.036
-0.124
-0.059
0.147
0.334
0.197
0.305
Adjusted R-squared
0.462
0.565
0.214
0.290
0.234
0.093
0.315
0.540
0.189
Durbin-Watson d-statistic
(transformed)
1.918
1.697
1.942
2.004
1.940
1.983
2.020
1.843
2.010
49
41
45
43
22
34
45
49
48
N
+p ≤ 0.10, * p ≤ .05, ** p ≤ .01
47
Table 3 (d). Prais-Winsten AR(1) Model of Party Competence and Government Agendas U.K.
AGENDAt
Legislative Outputs
(Acts of U.K. Parliament)
Economy
Health
Labor
Education
Environ
Crime
Social
Foreign
Other
-2.962
(3.618)
-0.284
(0.463)
0.539
(1.282)
2.677*
(1.069)
6.761
(5.944)
4.122*
(1.806)
2.251*
(0.887)
1.203
(0.771)
12.177**
(4.462)
MIPt
-0.306**
(0.100)
0.101
(0.083)
0.204
(0.143)
1.027*
(0.494)
1.974
(2.110)
0.072
(0.247)
0.421
(0.407)
0.122
(0.110)
0.231
(0.748)
MIPt * Issue Ownershipt
0.068**
(0.021)
0.009
(0.012)
-0.028
(0.023)
-0.119
(0.087)
-0.339
(0.388)
0.088
(0.065)
-0.047
(0.054)
-0.009
(0.015)
0.010
(0.097)
Popularityt
0.041
(0.431)
-0.075
(0.078)
0.087
(0.162)
0.448*
(0.170)
0.801
(0.747)
0.500*
(0.213)
0.288+
(0.156)
0.255*
(0.122)
1.444*
(0.637)
Issue Ownershipt * Popularityt
-0.003
(0.086)
0.003
(0.012)
-0.001
(0.026)
-0.073*
(0.031)
-0.166
(0.145)
-0.115**
(0.042)
-0.061*
(0.025)
-0.027
(0.020)
-0.301*
(0.113)
Constant
25.263
(18.638)
7.388*
(2.826)
-2.697
(8.218)
-14.118*
(5.984)
-28.974
(30.495)
-6.327
(9.955)
-2.508
(5.321)
-5.844
(5.651)
-19.108
(24.243)
Rho
-0.253
-0.420
-0.324
-0.553
-0.307
-0.314
-0.251
-0.347
0.158
Adjusted R-squared
0.155
0.433
-0.007
0.212
0.012
0.427
0.226
0.016
0.124
Durbin-Watson d-statistic
(transformed)
1.983
2.044
2.068
1.908
2.020
1.950
1.890
1.614
1.920
49
41
45
43
22
34
45
49
48
Issue Ownershipt(rank)
N
+p ≤ 0.10, * p ≤ .05, ** p ≤ .01,
48
Appendix –U.K. Data
Table A1.Categories of survey questions on issue competence, U.K.
Topic
Obs.
%
842
32.38
3
Questions
Economy, inflation, prices, interest rates,
unemployment, exchange rates, taxation.
Health, NHS
206
7.92
5
Strikes, labor relations, trade unions, employment
270
10.38
6
7
12
1314
Education, schools
Environment, climate change
Law and order, crime
Welfare, benefits, pensions, poverty, housing
Defense, international/foreign affairs, nuclear arms,
Europe, terrorism
Immigration, agriculture, transport
202
79
214
195
7.77
3.04
8.23
7.50
330
12.69
262
2,600
10.08
100.00
1
1619
99
Total
Table A2.Number of survey questions on issue competence, U.K.
Election Cycle
1939 – 1945
1
0
3
0
5
0
6
0
7
0
12
0
1314
1
1619
0
99
0
Total
1
1950 – 1951
0
0
0
0
0
0
0
0
1
1
1951 – 1955
1955 – 1959
2
11
0
1
0
4
0
0
0
0
0
1
2
7
2
8
2
4
8
36
1959 – 1964
10
1
2
0
0
0
1
3
4
21
1964 – 1966
18
2
8
2
0
3
6
10
7
56
1966 – 1970
1970 – 1974
11
23
0
6
17
12
0
6
0
0
0
0
4
0
5
13
1
1
38
61
1974 – 1979
60
8
48
8
0
14
19
17
34
208
1979 – 1983
46
6
37
9
0
18
16
24
31
187
1983 – 1987
1987 – 1992
75
145
39
40
37
15
38
41
5
21
39
36
5
48
74
74
0
46
312
466
1992 – 1997
134
29
26
26
25
29
45
55
66
435
1997 – 2001
74
11
10
10
10
11
17
20
33
196
2001 – 2005
40
9
7
8
4
10
14
17
19
128
2005 – 2010
118
28
21
28
14
27
10
8
11
265
2010 – 2015
75
26
26
26
0
26
0
0
2
181
49
Appendix – US Data
Table A3.Categories of survey questions on issue competence, US
Topic
Questions
Obs.
%
1: Economy
Economy, inflation, prices, interest rates,
unemployment, exchange rates, taxation.
703
28.89
3: Health
Health, medicare
184
7.56
5: Labor
Labor relations, unions, employment
127
5.22
6: Education
Education, schools
97
3.99
7: Environment
Environment, pollution
62
2.55
12: Law & Order
Law and order, crime
48
1.97
1314: Social & Housing
Welfare, benefits, pensions, housing
111
4.56
1619: Defense& foreign
affairs
Defense, international/foreign affairs, nuclear
arms, terrorism
495
20.35
99: Other
Agriculture, transport
606
24.91
2,433
100.00
Total
Table A4.Number of survey questions on issue competence, US
Election Cycle
1944 – 1948
1948 – 1952
1952 – 1956
1956 – 1960
1960 – 1964
1964 – 1968
1968 – 1972
1974 – 1976
1976 – 1980
1980 – 1984
1984 – 1988
1988 – 1992
1992 – 1996
1996 – 2000
2000 – 2004
2004 – 2008
2008 – 2012
1
24
12
13
16
14
17
12
4
10
14
13
74
86
54
70
134
136
3
1
0
0
0
0
0
0
0
0
0
0
11
33
30
21
42
46
5
12
4
0
0
0
0
0
1
1
3
0
5
5
1
6
49
40
6
1
0
0
0
0
0
0
0
0
0
0
12
12
18
16
22
16
7
0
0
0
0
0
0
0
0
0
1
0
10
9
11
13
8
10
12
0
0
0
0
0
0
0
0
0
0
0
8
25
11
1
3
0
1314
0
0
0
0
0
0
0
0
0
0
0
9
13
19
25
30
15
1619
9
7
11
15
15
15
10
3
6
15
11
32
28
24
68
142
84
99
22
11
8
24
17
22
14
13
13
28
42
57
79
50
40
110
56
Total
69
34
32
55
46
54
36
21
30
61
66
218
290
218
260
540
403
50
Sample Questions
Q. If Britain were in economic difficulties, which party do you think could handle the
problem best – the Conservatives or Labour?
Q. I am going to read out a list of problems facing the country. Could you tell me for
each of them which political party you personally think would handle the problem
best? Pensions
Q. Who do you trust to do a better job of handling the economy: the Democrats or
the Republicans?
Q. When it comes to … Ensuring a strong national defense ... which party do you
think would do a better job--the Democratic Party, the Republican Party, or both
about the same? If you think that neither would do a good job, please just say so.
51
REVIEWER APPENDIX I. Percentage of Seats in the House of Representatives / Parliament
Table RA1(a). Prais-Winsten AR(1) Model of Party Competence and Government Agendas US 1945 – 2007
AGENDAt
Executive
(State of the Union)
Economy
Health
Labour
Education
Environ
Crime
Social
Foreign
Other
Issue Ownershipt (rank)
12.592*
(5.073)
-14.419
(14.338)
-7.533
(5.318)
7.939
(14.820)
7.222
(6.538)
22.909
(54.320)
32.749+
(15.965)
-9.562
(7.715)
9.480
(13.061)
MIPt
0.990*
(0.436)
6.657*
(2.582)
-10.278**
(3.706)
2.833
(2.984)
1.533
(2.576)
-0.326
(0.830)
3.922
(2.800)
1.204*
(0.476)
2.492+
(1.283)
MIPt * Issue Ownershipt
-0.110*
(0.053)
-0.655+
(0.301)
1.902**
(0.597)
-0.218
(0.446)
-0.117
(0.549)
-0.061
(0.313)
-0.230
(0.483)
-0.117+
(0.062)
-0.285+
(0.153)
Seatst
1.290
(0.796)
-2.886
(2.553)
-0.774
(0.697)
0.590
(1.554)
0.803
(0.782)
0.722
(2.043)
6.550**
(1.839)
-1.434
(1.156)
0.020
(1.612)
Issue Ownershipt * Seatst
-0.171+
(0.100)
0.345
(0.299)
0.090
(0.091)
-0.120
(0.260)
-0.132
(0.118)
-0.465
(0.984)
-0.647*
(0.256)
0.244
(0.153)
-0.045
(0.217)
Constant
-82.127*
(39.868)
119.971
(120.856)
61.141
(39.200)
-39.995
(88.310)
-43.180
(42.425)
-12.633
(114.223)
-339.152*
(104.259)
81.662
(59.275)
-41.388
(98.837)
Rho
0.056
-0.444
-0.023
-0.190
-0.227
-0.513
0.317
0.197
-0.037
Adjusted R-squared
0.183
0.828
0.269
0.533
0.237
0.221
0.623
0.227
0.078
Durbin-Watson d-statistic
(transformed)
1.959
1.758
1.873
1.895
2.110
1.614
1.808
1.922
1.830
59
17
33
17
17
13
13
59
37
N
+p ≤ 0.10, * p ≤ .05, ** p ≤ .01
52
Table RA1(b). Prais-Winsten AR(1) Model of Party Competence and Government Agendas US 1945 – 2007
AGENDAt
Legislative Outputs
(Statutes of US Congress)
Economy
Health
Labour
Education
Environ
Crime
Social
Foreign
Other
Issue Ownershipt (rank)
-0.245
(0.878)
-4.918+
(2.565)
-2.062
(1.857)
1.076
(4.097)
0.427
(3.883)
-10.574**
(2.117)
5.394
(3.826)
1.721
(1.711)
-1.502
(5.769)
MIPt
0.066
(0.071)
-0.338
(0.629)
-0.359
(0.612)
-0.856
(0.793)
-0.834
(2.126)
-0.082
(0.099)
-0.117
(0.646)
0.282**
(0.104)
-0.581
(0.536)
MIPt * Issue Ownershipt
-0.006
(0.009)
0.054
(0.078)
0.006
(0.121)
0.125
(0.119)
0.010
(0.437)
0.008
(0.020)
-0.035
(0.127)
-0.024+
(0.014)
0.072
(0.064)
Seatst
-0.068
(0.136)
-0.780*
(0.348)
-0.229
(0.247)
0.266
(0.420)
0.071
(0.440)
-0.646**
(0.123)
0.778+
(0.369)
0.210
(0.261)
-0.063
(0.689)
Issue Ownershipt * Seatst
0.011
(0.017)
0.078
(0.045)
0.034
(0.032)
-0.040
(0.072)
-0.011
(0.068)
0.198**
(0.040)
-0.096
(0.059)
-0.024
(0.034)
-0.006
(0.095)
Constant
2.936
(6.857)
49.670*
(20.029)
17.002
(13.737)
-4.788
(23.556)
1.890
(24.724)
41.168**
(7.172)
-36.901
(21.533)
-1.760
(13.234)
82.535+
(41.795)
Rho
0.172
0.482
-0.137
0.150
-0.588
-0.658
-0.355
-0.067
0.547
Adjusted R-squared
0.026
0.456
0.042
0.223
0.394
0.781
0.395
0.300
0.683
Durbin-Watson d-statistic
(transformed)
1.916
1.898
1.896
1.477
1.215
2.092
2.083
1.936
1.734
60
19
35
19
19
15
15
60
39
N
+p ≤ 0.10, * p ≤ .05, ** p ≤ .01,
53
Table RA1(c). Prais-WinstenAR(1) Model of Party Competence and Government Agendas U.K.
AGENDAt
Executive
(Speech from the Throne)
Economy
Health
Labour
Education
Environ
Crime
Social
Foreign
Other
Issue Ownershipt (rank)
-5.451
(8.070)
0.664
(0.956)
1.898
(2.482)
4.242
(3.259)
0.975
(8.237)
5.478+
(3.060)
-3.012
(3.744)
-0.948
(6.239)
-5.614
(11.115)
MIPt
0.099
(0.095)
0.073
(0.068)
0.096
(0.152)
0.059
(0.680)
1.597
(1.190)
0.410+
(0.220)
1.557*
(0.597)
-0.011
(0.275)
0.143
(0.663)
MIPt * Issue Ownershipt
-0.003
(0.020)
0.019+
(0.011)
0.002
(0.025)
0.054
(0.111)
-0.250
(0.221)
0.015
(0.055)
-0.141+
(0.077)
-0.018
(0.036)
-0.018
(0.089)
Seatst
-0.507
(0.601)
-0.078
(0.094)
0.127
(0.255)
0.633
(0.413)
0.057
(0.772)
0.935**
(0.308)
-0.260
(0.367)
-0.018
(0.652)
-0.932
(1.294)
Issue Ownershipt * Seatst
0.112
(0.143)
-0.009
(0.018)
-0.034
(0.043)
-0.083
(0.063)
0.002
(0.147)
-0.099+
(0.051)
0.065
(0.066)
0.044
(0.113)
0.141
(0.215)
Constant
31.181
(34.651)
4.821
(5.212)
-4.387
(15.169)
-29.509
(21.042)
-6.780
(43.074)
-47.289*
(18.757)
12.895
(20.449)
23.012
(36.763)
73.103
(67.616)
Rho
0.076
-0.217
0.148
-0.134
-0.051
-0.117
0.548
0.643
0.408
Adjusted R-squared
0.328
0.629
0.070
0.188
0.138
0.421
0.260
0.024
0.156
Durbin-Watson d-statistic
(transformed)
1.904
1.931
2.024
2.029
2.003
1.902
2.181
1.927
1.955
51
43
47
45
24
36
47
51
50
N
+p ≤ 0.10, * p ≤ .05, ** p ≤ .01
54
Table RA1(d). Prais-Winsten AR(1) Model of Party Competence and Government Agendas U.K.
AGENDAt
Legislative Outputs
(Acts of U.K. Parliament)
Economy
Health
Labour
Education
Environ
Crime
Social
Foreign
Other
Issue Ownershipt (rank)
-28.284+
(14.923)
-1.829
(1.192)
6.556**
(2.230)
7.613*
(2.825)
15.512
(16.804)
7.953+
(3.977)
6.453**
(2.366)
2.086
(2.009)
31.417*
(14.341)
MIPt
-0.546**
(0.160)
0.141
(0.084)
0.271*
(0.121)
0.477
(0.411)
2.795
(2.503)
0.117
(0.276)
0.843*
(0.411)
0.043
(0.136)
0.292
(0.789)
MIPt * Issue Ownershipt
0.120**
(0.035)
-0.011
(0.014)
-0.036+
(0.019)
-0.043
(0.069)
-0.509
(0.475)
0.103
(0.069)
-0.123*
(0.054)
0.001
(0.017)
-0.012
(0.104)
Seatst
-1.662
(1.053)
-0.047
(0.119)
0.604**
(0.220)
0.740*
(0.323)
1.316
(1.533)
0.804+
(0.402)
0.529*
(0.239)
0.285
(0.217)
3.674*
(1.713)
Issue Ownershipt * Seatst
0.446+
(0.263)
0.031
(0.022)
-0.109**
(0.038)
-0.137*
(0.053)
-0.268
(0.293)
-0.144*
(0.066)
-0.110*
(0.041)
-0.040
(0.036)
-0.595*
(0.277)
122.119+
(60.680)
7.193
(6.631)
-33.570*
(13.323)
-38.123*
(16.981)
-72.422
(87.446)
-33.095
(24.579)
-23.204+
(13.741)
-9.808
(11.665)
-155.927+
(90.051)
Rho
-0.310
-0.372
-0.360
-0.534
-0.251
-0.239
-0.236
-0.325
0.281
Adjusted R-squared
0.239
0.431
0.099
0.209
-0.050
0.332
0.229
-0.042
0.098
Durbin-Watson d-statistic
(transformed)
2.050
1.974
2.133
1.901
2.063
1.868
1.876
1.621
1.973
49
41
45
43
22
34
45
49
48
Constant
N
+p ≤ 0.10, * p ≤ .05, ** p ≤ .01
55