12 Jan 2007 20:17
V I E W
P1: OKZ
ANRV312-PL10-13.tex
XMLPublishSM (2004/02/24)
AR REVIEWS IN ADVANCE10.1146/annurev.polisci.10.050806.155344
(Some corrections may occur before final publication online and in print)
A
N
I N
C E
S
R
E
AR
D V A
Annu. Rev. Polit. Sci. 2007. 10:271–96
doi: 10.1146/annurev.polisci.10.050806.155344
c 2007 by Annual Reviews. All rights reserved
Copyright !
First published online as a Review in Advance on January 22, 2007
THE END OF ECONOMIC VOTING? Contingency
Dilemmas and the Limits of Democratic
Accountability
Annu. Rev. Polit. Sci. 2007.10. Downloaded from arjournals.annualreviews.org
by CORNELL UNIVERSITY on 04/30/07. For personal use only.
Christopher J. Anderson
Department of Government, Cornell University, Ithaca, New York 14853;
email: [email protected]
Key Words elections, representation, political economy, democratic theory,
political behavior
■ Abstract The predominant normative justification for research on economic voting has been its essential role in shaping democratic accountability. A systematic examination of this literature reveals, however, that economic voting is highly contingent
on two critical moderating factors: voters themselves and the political context in which
they make judgments. The trend toward a better and more realistic understanding of
economic voting produced by almost four decades of empirical research has created
what I label “contingency dilemmas” for the field’s normative foundations because
economic voting does not function as envisioned by advocates of democratic accountability. This essay reviews these empirical findings and critically examines how they
affect the economic voting paradigm. It argues that, when viewed from a normative
perspective, contingent accountability is clearly problematic, and it calls for a reconsideration of the normative underpinnings of the economic voting paradigm in light of
the current state of knowledge.
INTRODUCTION
Over the years, the notion that voters judge democratic governments by how well
they manage the economy has taken on the ring of an incontrovertible social
scientific fact analogous to knowing that economic development and democracy
go together or that plurality electoral rules are associated with fewer political
parties in a political system. What is more, the belief that governments are judged
by their economic record has made its way into the repertoire of pundits and
has spilled over into popular views of voter behavior. Perhaps its most famous
articulation came during the 1992 U.S. presidential campaign, when Bill Clinton’s
chief strategist James Carville put up a sign in campaign headquarters that read:
“It’s the economy, stupid!”
The trouble with these claims—as Al Gore and John Major can attest—is that
they are only intermittently borne out by the facts. Empirical findings accumulated
1094-2939/07/0615-0271$20.00
271
12 Jan 2007 20:17
Annu. Rev. Polit. Sci. 2007.10. Downloaded from arjournals.annualreviews.org
by CORNELL UNIVERSITY on 04/30/07. For personal use only.
272
AR
P1: OKZ
ANRV312-PL10-13.tex
XMLPublishSM (2004/02/24)
AR REVIEWS IN ADVANCE10.1146/annurev.polisci.10.050806.155344
ANDERSON
in recent years strongly suggest that the influence of the economy on government
popularity and election outcomes is far from inevitable. Moreover, these findings
reveal that such a link, when it exists, is typically contingent for both institutional
and psychological reasons. The consequences of this improved and more realistic understanding of economic voting are not widely appreciated. In particular,
these newest findings of conditional economic effects on government support have
had the unintended consequence that the reward-punishment thesis—which has
guided most of the work in the area of economics and elections and which makes
economic voting such an appealing explanation for how elections come out—has
been systematically challenged.
Below, I discuss the literature on the relationship between the economy and
government support in contemporary democracies with an eye toward what I call
“contingency dilemmas.” I argue that recent findings of substantial variability in
the effect of the economy on voter behavior have fundamentally reshaped how
we think about economic voting and pointed to theoretical shortcomings in the
economic voting paradigm. More importantly, the highly contingent nature of
the economy-vote link seems to pose a fundamental dilemma for research on
economic voting. It calls into question the explanatory and predictive power of the
economic voting approach, and it threatens to undercut the normative importance
of economic voting research as a field of inquiry.
On its face, the most recent wave of economic voting scholarship could be
taken to imply that, perhaps, too much of the impact of the economy on government support or election outcomes is too contingent to justify the appeal of
economic voting research on normative or empirical grounds. Although such a
reading of the findings would be premature, they do point to issues that merit
separation: what the findings show empirically and what the findings mean for
our theoretical understanding of economic voting as a mechanism of accountability. Empirically speaking, impediments to economic voting merit attention for
improving the predictive accuracy of economic voting models. Viewed from a normative perspective, however, contingent accountability is more complex. Below, I
argue that the primary threat to democratic accountability comes from institutional
barriers to economic voting rather than barriers rooted in how voters collect and
interpret information. I conclude by suggesting ways of alleviating this normative
contingency dilemma and reorienting research on economic voting to sustain it
successfully as an area of empirical and theoretical political science scholarship
that also has something to say about the real world of politics.
ECONOMIC VOTING AND THE DISCOVERY
OF CONTINGENT EFFECTS
Although early empirical efforts to study the relationship between the business
cycle and election outcomes exist (Tibbits 1931), research on how the economy
influences voters’ decisions began in earnest only in the 1950s with the advent
12 Jan 2007 20:17
AR
P1: OKZ
ANRV312-PL10-13.tex
XMLPublishSM (2004/02/24)
AR REVIEWS IN ADVANCE10.1146/annurev.polisci.10.050806.155344
Annu. Rev. Polit. Sci. 2007.10. Downloaded from arjournals.annualreviews.org
by CORNELL UNIVERSITY on 04/30/07. For personal use only.
THE END OF ECONOMIC VOTING?
273
of mass surveys in the United States and other developed democracies (Berelson
et al. 1954, p. 185; Campbell et al. 1960, ch. 14). By the time the postwar economic
boom turned to bust in the late 1960s and early 1970s, the study of the relationship
between the economy and voter behavior took off. It was helped along by several
seminal works, published in rapid succession, by Key (1966), Goodhart & Bhansali
(1970), Mueller (1970), Kramer (1971), Frey & Garbers (1971), Stigler (1973),
and Tufte (1975). By the end of the twentieth century, the flow of scholarly papers
on the topic had “changed from a trickle to a torrent of over 300 articles and
books on economics and elections” (Lewis-Beck & Stegmaier 2000, p. 183) and
covered virtually every democracy for which data on economics and elections were
available.1
Scholars commonly argued that the importance of economic voting derives from
its consistency with normative theorists’ postulates of democratic accountability:
“after all, nothing is more fundamental to popular control than the idea that citizens
hold government officials accountable for their collective actions” (Kuklinski &
West 1981, p. 437; see also Kramer 1971, Nannestad & Paldam 1994, Rudolph
2003). As Fiorina (1981, p. 11) puts it, “Given political actors who fervently
desire to retain their positions and who carefully anticipate public reaction to
their records as a means to that end, a retrospective voting electorate will enforce
electoral accountability.” The reward-punishment hypothesis lies at the heart of
such an understanding of accountability (see Downs 1957, Key 1966, Nannestad
& Paldam 1994, Lewis-Beck & Paldam 2000). Lewis-Beck & Stegmaier (2000,
p. 183) articulate that hypothesis simply: “The citizen votes for the government if
the economy is doing all right; otherwise, the vote is against.”
In addition to asserting the normative importance of economic voting, researchers have routinely motivated their investigations by claiming that it is an
empirical fact: “When you think economics, think elections; when you think elections, think economics” (Tufte 1978, p. 65). Similarly, Lewis-Beck & Stegmaier’s
(2000, p. 211) review of the economic voting literature concluded, “Economics
and elections form a tight weave. . . . For all democratic nations that have received
a reasonable amount of study, plausible economic indicators, objective or subjective, can be shown to account for much of the variance in government support.”
Or, as the British Labour Prime Minister Harold Wilson famously declared, “All
political history shows that the standing of a Government and its ability to hold
the confidence of the electorate at a General Election depend on the success of its
economic policy” (Watt 1968, p. 15).
These assertions have had many echoes over the years and have become part
of scholarly and popular lore. Yet, as several decades of research have revealed,
1
To keep this topic manageable, I focus on democratic elections in the developed countries,
and I focus on the domestic political economy (unlike Hellwig 2001, for example). For
contributions to the economic voting paradigm in the developing world, see, e.g., Pacek &
Radcliff (1995a), Remmer (1993), and the essays in Stokes (2001). For a review, see Tucker
(2002).
12 Jan 2007 20:17
274
AR
P1: OKZ
ANRV312-PL10-13.tex
XMLPublishSM (2004/02/24)
AR REVIEWS IN ADVANCE10.1146/annurev.polisci.10.050806.155344
ANDERSON
things are not quite so simple. In fact, the question of how exactly the economy relates to how voters view governments and what they do at the ballot box
has been the subject of intense inquiry and debate. The effects of the economy
on election outcomes or voter behavior have been difficult to pin down, and
this difficulty influences how we think about economic voting as an intellectual
enterprise.
Annu. Rev. Polit. Sci. 2007.10. Downloaded from arjournals.annualreviews.org
by CORNELL UNIVERSITY on 04/30/07. For personal use only.
Contingent Effects as an Empirical Problem
One explanation for why these effects are difficult to pin down has been methodological. Although scholars have long recognized the variability in the relationship
between economic conditions and government support, they traditionally attributed
it to inadequacies in operationalizing and measuring the state of the economy or
estimating its effects on the proper indicators of voter behavior and election outcomes. Over the years, scholars have followed distinct approaches to studying
the economy and voter behavior. They focused either on the behavior of individuals with the help of cross-sectional survey studies (microlevel approach) or
on the behavior of aggregates—usually national electorates (the macrolevel approach). Scholars have used widely varying indicators of both the economy and
the dependent variable of interest—vote choice, approval, or election outcome—
and have employed a variety of frequently quite disparate methods of statistical analysis. And whereas some researchers relied on objective indicators of a
country’s economic situation, others found it more sensible to employ measurements of public perceptions of economic conditions that—when available—may
differ from how the economy is performing in reality, a point I will return to
below. In addition, scholars debated the proper use of control variables (such as
political events, issues, and temporal dynamics) and model specification in economic voting models (see the review in Nannestad & Paldam 1994). It is perhaps
not surprising that scholars have disagreed over whether individuals react to national economic performance (sociotropic voting) or their own economic situation (egocentric voting); whether citizens evaluate actual or potential economic
performance retrospectively or prospectively; and whether they react to growth,
unemployment, or inflation, for example (Lewis-Beck & Stegmaier 2000). Suffice it to say that different researchers analyzing slightly different time periods
and/or different countries—and using different indicators of economic performance, different measures of these indicators, and different estimation methods—
frequently disagreed about the magnitude of the economic effects, and even about
whether the economy is an important predictor of government support in the first
place.
Painting with a broad brush, economic voting research during the 1970s and
1980s can be characterized as focused on finding just the right way to model
the economy-voter relationship. Scholars debated technical issues such as model
specification, measurement, and econometrics, rather than the fundamental theoretical notions—such as the reward-punishment thesis—that originally guided the
12 Jan 2007 20:17
AR
P1: OKZ
ANRV312-PL10-13.tex
XMLPublishSM (2004/02/24)
AR REVIEWS IN ADVANCE10.1146/annurev.polisci.10.050806.155344
THE END OF ECONOMIC VOTING?
275
generation of such modeling efforts. Looking back on more than 20 years of quite
rigorous research, it appeared that social scientists by and large had come to believe
that a relationship existed between economics and political attitudes/behavior and
that the “true” relationship between economic conditions and public support could
be demonstrated if we only looked hard enough. And looking hard enough meant
using the right data and modeling and estimating the economy–mass opinion relationship in the “right” way rather than questioning the underlying assumptions
that guided economic voting models.
Annu. Rev. Polit. Sci. 2007.10. Downloaded from arjournals.annualreviews.org
by CORNELL UNIVERSITY on 04/30/07. For personal use only.
Contingent Effects as a Theoretical Problem
A turning point for research on economic voting came in 1991 with the publication
of a volume edited by Norpoth, Lewis-Beck, and Lafay. In it, Paldam reported the
results of an exhaustive, broadbased, cross-national study of economic voting and
warned that the evidence for economic effects on government popularity or vote
choice is seldom conclusive or straightforward:
First X presents an impressive study of the Vote or Popularity function for
country Z, with a nice theory and—most important—very fine econometric
fits: a high R2 , very significant t-ratios, and, in addition, some new econometric trick like the ζ ζ-test from the latest issues of Esoterica. Everybody is
impressed, until a few years later Y demonstrates that, by one little change,
X’s result collapses. (Paldam 1991, p. 10)
Paldam called on scholars to attack the sources of this problem. Following the
exposure of what Paldam ingeniously labeled the “instability” dilemma (see also
Nannestad & Paldam 1994), the fundamental question of economic voting research
quickly changed from “How can we best identify a relationship we know exists?”
to “How can we best make sense of a relationship we know to be unstable?” (See
also Lewis-Beck & Paldam 2000, p. 119.)
To those who approached the question as a technical matter, the instability
dilemma wasn’t much of a dilemma. After all, in the imperfect world of social science research, the use of different model specifications and different time periods
(in time-series research) is bound to lead to at least somewhat different estimates
of the relationship between, say, government popularity and the rate of economic
growth. Thus, it would be difficult to obtain stable estimates of its strength because of measurement and estimation problems even if the relationship between
economic conditions and government support were truly stable. Viewed in this
way, the instability of the relationship thus poses mostly empirical problems because it points to difficulties with modeling the underlying process that are inherent
in the methodologies of the social science enterprise.
While some continued to treat the instability dilemma as a technical matter,
researchers approaching the issue of economic voting from the perspective of
democratic accountability were troubled by inconsistent economic effects for
altogether different reasons. If true, these effects suggested that governments
12 Jan 2007 20:17
Annu. Rev. Polit. Sci. 2007.10. Downloaded from arjournals.annualreviews.org
by CORNELL UNIVERSITY on 04/30/07. For personal use only.
276
AR
P1: OKZ
ANRV312-PL10-13.tex
XMLPublishSM (2004/02/24)
AR REVIEWS IN ADVANCE10.1146/annurev.polisci.10.050806.155344
ANDERSON
were not consistently punished for bad economic times or rewarded for good
performance—in short, that democratic accountability was the exception rather
than the rule. Viewed from this perspective, inconsistencies in the relationship
between economic performance and voter behavior undermine the normative
foundation of the economic voting paradigm. Ultimately, this meant a turn away
from the simple reward-punishment model to a more complicated model positing that we should not expect the relationship between economic performance
and voter behavior to be stable over time for all incumbents in all countries. In
the ensuing years, this shift has produced a wealth of new insights, all predicated on the idea that economic effects are likely to be contingent, showing that
the influence of the economy varies systematically and depends on a variety of
constraints.
ECONOMIC VOTING AND DEMOCRATIC
ACCOUNTABILITY
The logic of the reward-punishment argument, which students of economics and
elections seldom spell out explicitly, is more multifaceted than is frequently acknowledged. More importantly, it affects how we view the consequences of the
instability dilemma for our understanding of democratic accountability. To begin with, the reward-punishment idea is based on the social contract tradition and
rooted in the idea that representative government is the only practical way to govern
nation-states democratically. To writers like Mill and the authors of the Federalist
Papers, for example, Greek-style direct democracy was impractical or perhaps even
foolish for modern nation-states because their size, scope, and complexity make it
virtually impossible for a meaningful number of people to participate directly in
the day-to-day administration of the state.
In addition, these writers worried about the inexperience and instability of the
general electorate and about balancing the need for professionalism and expertise in the administration of government with the need for public accountability.
According to this vision of democracy, the limits imposed by representative government on citizens’ ability to directly influence the operation of government were
justified in order to ensure good governance, as Mill (1955, pp. 229–30) made
clear in his distinction “between controlling the business of government and actually doing it.” Thus, representative democratic institutions tame the impulses
and predispositions of flawed or uninformed electorates while ensuring effective
governance.
The idea of representative government and the temporary suspension of citizen
control as the only sensible and legitimate form of democratic rule for nation-states
became even more compelling with the advent of empirical findings on political
behavior in the 1940s and 1950s, which showed that most citizens most of the
time do not live up to the Aristotelian ideal of a competent citizen (Converse 1964,
12 Jan 2007 20:17
AR
P1: OKZ
ANRV312-PL10-13.tex
XMLPublishSM (2004/02/24)
AR REVIEWS IN ADVANCE10.1146/annurev.polisci.10.050806.155344
Annu. Rev. Polit. Sci. 2007.10. Downloaded from arjournals.annualreviews.org
by CORNELL UNIVERSITY on 04/30/07. For personal use only.
THE END OF ECONOMIC VOTING?
277
Dalton 2002). For a variety of reasons, citizens have neither detailed information
about government activities nor the necessary motivation and capacity to make
use of such information even when it is available (see Bartels 1996, Hardin 2002,
Zaller 1992).
This led many modern-day students of democracy to equate representative
democracy with accountability as a matter of course (cf. Riker 1965, Powell 2000):
“To support the Ins when things are going well; to support the Outs when they
seem to be going badly, this . . . is the essence of popular government” (Walter
Lippmann quoted in Powell 2000, p. 10). Similarly, “Democracy means only that
the people have the opportunity of accepting or refusing the men who are to rule
them” (Schumpeter 1942, p. 269). The point was echoed in Schmitter & Karl’s
(1991, p. 76) oft-cited definition of democracy: “Modern political democracy is
a system of governance in which rulers are held accountable for their actions in
the public realm by citizens acting indirectly through the competition and cooperation of their elected representatives.” And Dahl (1971, p. 3) reasoned in
Polyarchy, “at a minimum, it seems to me, democratic theory is concerned with
processes by which ordinary citizens exert a relatively high degree of control
over leaders.” Using slightly different language, rational choice theorists have
construed democratic accountability as a so-called principal-agent relationship,
where voters are the principals who delegate authority via elections to agents
who will act on their behalf. Viewed in this way, elections are designed to grant
power to citizens, but limited power, and provide incentives for politicians to
act responsibly (Barro 1973, Ferejohn 1986, Przeworski 1999; but see Fearon
1999).
In light of these theoretical and empirical insights into the underpinnings of citizen competence and institutional design, the reward-punishment thesis motivating
research on economic voting is compelling. Given citizens’ limited willingness and
capacity to process complex information about politics, reward and punishment
should most easily be detectable with regard to the performance of the economy—
after all, the economy is perhaps the most perennially talked-about issue during
election campaigns in democracies (Anderson 1995). What is more, judging economic performance is more straightforward for average citizens than judging other
areas of government performance, such as biotechnology or air traffic control. Because high levels of transparency and monitoring of elected representatives are
potentially more attainable in this area of government action, elected representatives are also more likely to behave responsibly and seek to maximize the general
welfare (Kiewiet 2000, Ferejohn 1999).
In summary, democratic elections are expected to ensure accountability via economic voting because they are held intermittently but regularly, and because they
allow citizens to express broad and blunt approval or disapproval of the government’s record in an area of policy they can relate to and understand. Accountability
rooted in economic voting thus constitutes a minimalist but nevertheless legitimate
vision of democracy as a form of government that allows ill-informed electorates
12 Jan 2007 20:17
278
AR
P1: OKZ
ANRV312-PL10-13.tex
XMLPublishSM (2004/02/24)
AR REVIEWS IN ADVANCE10.1146/annurev.polisci.10.050806.155344
ANDERSON
to exert a circumscribed measure of control through their ability to “throw the
rascals out.”2
Annu. Rev. Polit. Sci. 2007.10. Downloaded from arjournals.annualreviews.org
by CORNELL UNIVERSITY on 04/30/07. For personal use only.
INDIVIDUAL AND INSTITUTIONAL LIMITS
OF ACCOUNTABILITY
In the context of economic voting, the foundations of the reward-punishment
model as sketched above mean that accountability can be compromised either
by individuals or by institutions. The principals/voters may fail to impose negative
sanctions on the agents/representatives either because of the characteristics of
voters (the individual constraints on economic voting) or because it is difficult for
the principals to obtain the necessary information about the outcomes of agents’
activities (the institutional constraints on economic voting). This model leads to
a conception of accountability that is conditional: It implies that both the design
of democratic institutions and the behavior of individuals can blunt the impact
of the economy on election outcomes or voter behavior. Consistent with this,
the empirical literature on contingent economic effects can be separated into (a)
research that probes how individuals’ characteristics affect the connection between
the economy and voters’ decisions, and (b) research that examines how institutions
affect the impact of the economy on electoral outcomes.
The First Contingency Dilemma: Individual-Level Constraints
on Economic Voting
Most of the work on how individual differences affect the connection between the
objective economy and citizens’ support for the government can be categorized
(very broadly and perhaps inadequately) under two headings: (a) informational and
cognitive limits and (b) the impact of values and predispositions on the formation
of economic perceptions and evaluations as well as attributions of responsibility.
Both streams of research examine the conditions under which actual economic
conditions are translated into voter behavior most truthfully via citizens’ subjective
perceptions of economic conditions.
2
This view of accountability is not uncontested. In fact, the concern over individuals and
institutions as the sources as well as cures of democracy’s imperfections is central to longstanding debates among political theorists. In contrast to Lippmann’s skepticism regarding
individuals’ ability to play their assigned role as good citizens (see also Hardin 2002,
for example), Dewey, and others (e.g., Smiley 1999, Waldron 1998) are more optimistic.
And with regard to institutions as the source of democracy’s limitations, the literature on
democratic deliberation is premised on the notion that officials can be and ought to be
held accountable to the reasoned views of citizens (Ferejohn 1999, Warren 1996) despite
the weakness and opaqueness of democratic institutions identified by Schumpeter, Riker,
Przeworski (1999) and others.
12 Jan 2007 20:17
AR
P1: OKZ
ANRV312-PL10-13.tex
XMLPublishSM (2004/02/24)
AR REVIEWS IN ADVANCE10.1146/annurev.polisci.10.050806.155344
Annu. Rev. Polit. Sci. 2007.10. Downloaded from arjournals.annualreviews.org
by CORNELL UNIVERSITY on 04/30/07. For personal use only.
THE END OF ECONOMIC VOTING?
279
As it turns out, the translation of “objective” economic conditions into voter
motivations to reward and punish is anything but straightforward and requires
several steps (Anderson & O’Connor 2000). First of all, it assumes that there is
such a thing as the objective economy. Starting from this premise, the translation
process requires that this objective economy is perceived by voters and perceived
somewhat accurately. These perceptions are presumed to translate into negative,
positive, or neutral evaluations of the economy (e.g., is it better, worse, or no
different from before?). The next step in the chain of necessary events is that these
evaluations translate into a vote for or against the government. This is possible only
if voters attribute responsibility to the incumbents for the economic situation they
have formed an evaluation of. All of these conditions are rarely, if ever, met, and a
considerable body of research has investigated the ways in which each link of the
chain can be broken. The fragility of the chain is likely to compromise democratic
accountability (but see Gomez & Wilson 2006 for a contrasting argument).
To begin with, accurate, objective information about the economy is frequently
difficult to come by for most citizens (Lohmann 1999). In fact, the assumption
that there is some objective state of the economy that citizens can perceive and
evaluate is in itself problematic; although there may be economic facts, their interpretation is frequently contested (Keech 1995). Without a consensus on what
economic facts mean, accountability is difficult if not impossible to judge. As
research on the effect of mass media on economic perceptions has pointed out,
most people learn about the national economy indirectly via mass media. Mass
media tend to overreport negative economic conditions (Goidel & Langley 1995),
and this tends to heavily condition voters’ economic perceptions (Hetherington
1996). Moreover, reports about the economy help shape, in particular, people’s
perceptions of national economic conditions and their relative importance in economic voting models relative to personal economic conditions (Mutz 1992, 1994;
Nadeau et al. 1999; Shah et al. 1999; for a contrasting view, see Haller & Norpoth
1997). In fact, a recent study shows that economic evaluations derive more from
the media’s presentation of economic developments than from objective changes
in the real economy (Sanders & Gavin 2004; see also Nadeau et al. 1999).
It should not be surprising that all this makes it difficult for voters to form
an accurate perception of actual economic conditions in the country. However,
even without those biases and limits, voters would have incentives to be ignorant
regarding their nation’s economy; information about the economy is costly (Aidt
2000, Downs 1957). Even if we assume that voters are motivated to seek out
accurate information about the condition of the macroeconomy, a number of factors
still hamper the collection and decoding of this information. For one, citizens’
cognitive limits reduce the accuracy with which objective information is likely to
be coded (Krause 1997, Krause & Granato 1998; however, also see Sanders 2000).
Moreover, citizens do not learn about different aspects of the macroeconomy in the
same way or at the same pace (Nannestad et al. 2003, Weatherford 1983); lag times
in citizens’ learning about economic trends differ significantly for inflation and
unemployment, for example (Conover et al. 1986), and significant proportions
12 Jan 2007 20:17
Annu. Rev. Polit. Sci. 2007.10. Downloaded from arjournals.annualreviews.org
by CORNELL UNIVERSITY on 04/30/07. For personal use only.
280
AR
P1: OKZ
ANRV312-PL10-13.tex
XMLPublishSM (2004/02/24)
AR REVIEWS IN ADVANCE10.1146/annurev.polisci.10.050806.155344
ANDERSON
of voters are predominantly static and myopic in their evaluations of economic
conditions (Paldam & Nannestad 2000). Differently situated voters are exposed
to systematically different economic experiences that help determine how they
perceive the state of the economy (Holbrook & Garand 1996, Lohmann 1999,
Weatherford 1983, Welch & Hibbing 1992).
Even if citizens had no motivations to be oblivious, had no cognitive limits on
their ability to make sense of the information, and had uniform access to unbiased
economic information, their own biases and values would interfere with the relationship between actual economic conditions and economic evaluations. That is,
citizens are likely to systematically misjudge the state of the economy even when
it is presented to them on a silver platter. Citizens take shortcuts to make sense of
their environment, and they prefer to avoid inconsistencies in their behaviors and
attitudes. Thus, partisans are likely to form evaluations of the state of the economy
to be consistent with their previously held beliefs (Anderson et al. 2004, Borre
1997, Evans & Andersen 2006, Pattie & Johnston 2001, Wlezien et al. 1997).
Moreover, voters’ affective reactions to personal and national economic conditions help shape economic evaluations (Conover & Feldman 1986), and people
have been found to rely on heuristics such as the well-being of similarly situated
individuals to form evaluations of the economy (Mutz & Mondak 1997).
For years, all these individual differences were assumed to matter little for
understanding the efficacy of the electorate’s collective choice. It had long been
assumed that a significant portion of individual-level differences are simply the
result of “noise” or random variation associated with survey data (e.g., Converse
1990, Page & Shapiro 1992, Sanders 2000, Wittman 1989). As a result, aggregation of individual responses was believed to cancel out the random variation,
thereby leaving only the underlying meaningful (or rational) component of public
opinion (Kramer 1983). This argument presumes, however, that individual errors
in measures of public opinion are truly random. As Bartels (1996) notes, however,
if these individual errors are systematic, aggregation will not produce unbiased
measures of public sentiment. Rather, these aggregate measures will include systematic variation with factors unrelated to objective economic performance, such
as partisanship or personal experiences. As recent research suggests, the various sources of individual “error terms” in national economic evaluations are not
random (Duch et al. 2000). As a result, aggregate deviations of individual-level
economic evaluations from objective conditions are not idiosyncratic but reflect
the systematic effects of respondent characteristics. This, in turn, compromises the
connection between actual economic conditions and voters’ evaluations of those
conditions.
The failure of individual variation in economic evaluations to cancel out in the
aggregate has important implications. If voters systematically perceive and then
evaluate the state of the economy to be different from what it actually is, accountability is undermined. But even if they were accurate—and it should be clear by
now that this is a big “if”—voters’ evaluations will not translate into a vote for
or against the government unless citizens attribute responsibility for economic
12 Jan 2007 20:17
AR
P1: OKZ
ANRV312-PL10-13.tex
XMLPublishSM (2004/02/24)
AR REVIEWS IN ADVANCE10.1146/annurev.polisci.10.050806.155344
Annu. Rev. Polit. Sci. 2007.10. Downloaded from arjournals.annualreviews.org
by CORNELL UNIVERSITY on 04/30/07. For personal use only.
THE END OF ECONOMIC VOTING?
281
conditions to the incumbent government. That is, accurate perceptions and evaluations are necessary but not sufficient for economic voting to take place; they
require the attribution of responsibility to do their job. As it turns out, attribution
is tricky business, and citizens frequently make the wrong attribution (Fischhoff
1976, Nisbett and Ross 1980). They attribute problem-solving competency and responsibility for good performance to the party they support and blame parties they
do not like for inferior economic performance (Rudolph 2003; see also Peffley et al.
1987, Peffley & Williams 1985; however, see Basinger & Lavine 2005). Comparing
people’s assignments of responsibility for economic performance across different
institutional actors in the United States, recent findings strongly suggest that economic values, knowledge of divided government, and partisanship leads voters to
credit those institutional actors they already favor when times are good and blame
actors they already dislike when times are bad (see Norpoth 2001).
Taken as a whole, this growing body of research shows that that the relationship
between the state of the economy and voter behavior, at the level of individuals, is
highly contingent and varies systematically because of biases in sources of information and differences in citizens’ motivations and cognitive abilities. Moreover, and
in contrast to what has been commonly assumed, it is likely that these differences
do not cancel out in the aggregate. Thus, people’s biased economic perceptions
systematically undermine the truthful translation of actual economic conditions
into voters’ economic evaluations and eventually their voting behavior. At a minimum, this undermines the assumption that significant portions of the electorate
judge a government’s record based on what occurs in actuality. But as importantly,
even if voters have biased judgments of economic reality, we do not know whether
and under what conditions these judgments are sufficient motivations for citizens
to vote against incumbents.
The Second Contingency Dilemma: Institutional
Limits to Economic Voting
Aside from individual limits on economic voting that compromise democratic accountability, voters may fail to impose negative sanctions on their representatives
because structural features of polities can hinder voters’ access to necessary information about representatives’ activities. This is where institutions come in. Even
if voters’ economic perceptions and evaluations were accurate, and even if they
sought to attribute responsibility to the proper authorities, they still might not be
able to do so. Yet again the accountability chain of economic voting is broken.
Institutions allow representatives to escape attention and shift blame. Examples
include the frequent practice of coalition government in the continental European
countries, divided government in the United States, and cohabitation in France.
But even when representatives do not try to hide information about their activities
or seek to shift blame, the complexity of political institutions makes it difficult for
voters to figure out who among the possibly honest officeholders or parties is to
blame for a bad economy. This insight is hardly new, as Hamilton’s quote from
12 Jan 2007 20:17
282
AR
P1: OKZ
ANRV312-PL10-13.tex
XMLPublishSM (2004/02/24)
AR REVIEWS IN ADVANCE10.1146/annurev.polisci.10.050806.155344
ANDERSON
Annu. Rev. Polit. Sci. 2007.10. Downloaded from arjournals.annualreviews.org
by CORNELL UNIVERSITY on 04/30/07. For personal use only.
Federalist No.70 regarding the proposal of a plural or cabinet executive makes
clear:
But one of the weightiest objections to a plurality in the Executive . . . is,
that it tends to conceal faults and destroy responsibility. . . . It often becomes
impossible, amid mutual accusations, to determine on whom the blame or the
punishment of a pernicious measure, or series of pernicious measures, ought
really to fall. It is shifted from one to another with so much dexterity, and under
such plausible appearances, that the public opinion is left in suspense about the
real author. The circumstances which may have led to any national miscarriage
or misfortune are sometimes so complicated that, where there are a number
of actors who may have had different degrees and kinds of agency, though
we may clearly see upon the whole that there has been mismanagement, yet
it may be impracticable to pronounce to whose account the evil which may
have been incurred is truly chargeable.
Following this line of thought, political scientists in recent years have begun
to systematically tackle the issue of how differences in political institutions impose limits on the relationship between the economy and government support
(Anderson 1995, Lewis-Beck 1988, Lewis-Beck & Mitchell 1993, Norpoth 2001,
Paldam 1991, Powell & Whitten 1993, Rudolph 2003, Stein 1990). This research
has sought to resolve the instability dilemma by examining how a country’s political context moderates the relationship between the economy and citizen support.
Investigating the question of institutional design and the role of political contexts
in the assignment of credit and blame to incumbent governments is a direct result
of the quest to make sense of Paldam’s instability dilemma. This approach assumes
that the inconsistent economic effects are the actual effects, and that the challenge
lies in accounting for these differences by understanding how political context
structures the assignment of credit and blame.
The basic idea that political contexts matter and may help explain differences in
economic effects is not new (e.g., Eulau & Lewis-Beck 1985, Lewis-Beck 1988),
but explicit theorizing about and systematic tests of the idea that political contexts
can be measured and incorporated into economic voting models are. Eschewing any
reference to individual constraints on economic voting, this research usually starts
with the assumption that voters seek to reward or punish incumbents for economic
performance and are variably helped or hindered by the political context in which
they attempt to do so. This context is viewed as a constraint that varies either
cross-nationally (cross-institutionally) or across time within countries. Although
the studies in this new wave of economic voting models have slightly different
ways of formally incorporating politics, they are all centered on the notion of
responsibility. Responsibility differs from accountability: Whereas the assignment
of responsibility to political actors for the state of the economy is a necessary
condition for accountability to exist, the reverse does not hold.
This research on responsibility has argued that different institutional setups
provide greater or lesser “clarity of responsibility” and thus make it easier or
12 Jan 2007 20:17
AR
P1: OKZ
ANRV312-PL10-13.tex
XMLPublishSM (2004/02/24)
AR REVIEWS IN ADVANCE10.1146/annurev.polisci.10.050806.155344
Annu. Rev. Polit. Sci. 2007.10. Downloaded from arjournals.annualreviews.org
by CORNELL UNIVERSITY on 04/30/07. For personal use only.
THE END OF ECONOMIC VOTING?
283
harder for voters to assign credit and blame (Anderson 1995, Powell 2000, Powell
& Whitten 1993). For example, actual policy-making responsibility frequently
is shared by competing political actors through mechanisms such as coalition
government, or simply obscured because of multiple levels of decision making
and political control. As a result, the impact of the economy on voter behavior is
variably strong, muted, or nonexistent because institutional contexts differ with
regard to the ease with which voters can identify who is in charge of government
policy and the extent to which political actors have actual influence on economic
outcomes (see, however, Chappell & Veiga 2000, Dorussen & Taylor 2001).
Because this approach is explicitly comparative in theory and practice, it is very
different from the initial, seminal, studies of government popularity and economic
conditions, which focused on countries with relatively well-defined incumbents
and oppositions, (approximate) two-party systems, and single-actor (executive)
incumbents, such as the United States and Great Britain (Goodhart & Bhansali
1970, Kramer 1971). In retrospect, it is clear that the field’s initial focus on the
Anglo-American systems had a significant impact on the formulation of hypotheses for subsequent work; a simple reward-punishment mechanism can be expected
to function most smoothly in two-party, single-party-incumbent executive systems, which the United States and Great Britain approximate. Yet the definition of
incumbency often varies considerably across countries, and this variation matters.
In research on presidential systems such as France and the United States, for example, it is usually presumed that the president is blamed or rewarded by the mass
public (see, e.g., Lewis-Beck 1980, Lafay 1985, Kernell 1978, MacKuen 1983,
Monroe 1984, Norpoth 1984, Chappell & Keech 1985), whereas in parliamentary
systems, the focus is typically on the party or coalition constituting the government (Kirchgässner 1974, 1991; Frey & Schneider 1980; Paldam & Schneider
1980; Dorussen & Taylor 2001; van der Brug et al. 2007).
This new comparative literature was pushed along by Powell & Whitten (1993),
who were among the first to model economic effects as being moderated by how
much a country’s formal institutional structure allows citizens to establish who
is responsible for economic management. Powell & Whitten’s operationalization
and measurement of the clarity-of-responsibility concept are encompassing. They
sort political systems into those where responsibility is clear and those where it is
not, based on factors such as one-party versus multiparty rule, whether there is bicameral opposition, decision-making powers for opposition parties in parliament,
and party cohesion. Their seminal analysis linking objective economic conditions
and aggregate election outcomes shows that economic effects are stronger in countries with clearer responsibility (see also Lewis-Beck 1988, pp.108–10; Bengtsson
2004; Palmer & Whitten 2003; but see also Royed et al. 2000 for a dissenting
view).
As originally conceptualized, Powell & Whitten’s clarity-of-responsibility index measures political context as varying across countries in predictable ways,
but it does not account for changes that may occur in the countries’ political contexts over time. A complementary view of how political context may moderate the
12 Jan 2007 20:17
Annu. Rev. Polit. Sci. 2007.10. Downloaded from arjournals.annualreviews.org
by CORNELL UNIVERSITY on 04/30/07. For personal use only.
284
AR
P1: OKZ
ANRV312-PL10-13.tex
XMLPublishSM (2004/02/24)
AR REVIEWS IN ADVANCE10.1146/annurev.polisci.10.050806.155344
ANDERSON
relationship between economy and government support suggests that clarity varies
not only across countries (because of differences in formal decision-making procedures that are relatively stable) but also within and across countries over time. For
example, democracies shift the power to govern and enact policy at more or less
regular intervals by way of the electoral process. And every election takes place
in a different political context that offers different choices to changing electorates
(Bengtsson 2004). Elections also shuffle the cards of government and—depending
on the political system—they install new actors, change the partisan composition
of governing coalitions, or confirm political parties and executives in office with an
expanded (reduced) or large (small) mandate. Thus, even when formal institutions
do not change, the extent to which voters are able to assign responsibility to political actors changes from one election to the next because of the political dynamics
created by electoral systems, party systems, the process of government formation,
and the like (Anderson 1995, 2000; Bengtsson 2004; Nadeau et al. 2002).
Unfortunately, few of the published studies that investigate the clarityof-responsibility thesis explicitly, with the help of indicators that score countries on
different dimensions of clarity, are directly comparable. They are based on different
data sources, ranging from monthly aggregate polling results to individual-level
surveys to national election results. They also differ with regard to objective indicators of economic performance (e.g., level of unemployment, change in unemployment, deviation in unemployment rates from international average performance),
or they eschew objective indicators altogether and rely exclusively on economic
evaluations. Moreover, they define “clarity” in different ways. Whereas some (e.g.,
Powell & Whitten 1993, Royed et al. 2000, Whitten & Palmer 1999) rely on more
encompassing indicators of clarity, others focus on coalition government, cabinet
strength, and the like (e.g., Anderson 1995, Dorussen & Taylor 2001), and one
study even combines several of these (Nadeau et al. 2002). In addition, several
studies use complicated interaction terms, interacting economic performance with
government partisanship as well as with clarity of responsibility. Finally, and perhaps most importantly, many of the published studies do not furnish descriptive
statistics, making it difficult to know how much the dependent variable is in fact
moved by the independent variables.
Although they focus on different cross-national or temporal properties of the
political context, these arguments all assume that voters will seek to throw incumbents out of office if the economy performs poorly. However, this can work only if
there are other credible actors who can benefit from such behavior. Thus, a critical
condition has recently been added to this debate: Clarity of responsibility really
matters only when voters perceive viable alternatives to the incumbents (Sanders
& Carey 2002). For example, in the 1980s, the British Tories went largely unchallenged over the course of several elections, regardless of the country’s economic
fortunes, because the opposition Labour Party was not perceived to constitute a
viable alternative government. In contrast, the Tories lost power in the early 1990s
despite a good economy in part because Labour had been able to resurrect itself
as a credible alternative under the leadership of Tony Blair.
12 Jan 2007 20:17
AR
P1: OKZ
ANRV312-PL10-13.tex
XMLPublishSM (2004/02/24)
AR REVIEWS IN ADVANCE10.1146/annurev.polisci.10.050806.155344
Annu. Rev. Polit. Sci. 2007.10. Downloaded from arjournals.annualreviews.org
by CORNELL UNIVERSITY on 04/30/07. For personal use only.
THE END OF ECONOMIC VOTING?
285
Unsurprisingly, the extent to which voters have or perceive such alternative
choices varies considerably both across countries and over time because of a
country’s structural features and political dynamics. There is, for example, considerable variation in the structure of party systems across contemporary democracies,
and not all systems provide clearcut or easily predictable alternatives to incumbent governments. During most of the post–World War II period, Sweden and
Japan, for example, saw very little alternation in government, irrespective of economic conditions, in part because of the dominance of one party. The nature of
the party system should influence the effects of macroeconomic performance on
governing party support because it should be more difficult for voters to identify
a clear alternative to the incumbent government if the party system is fragmented
(i.e., if there is a large effective number of parties), volatile, or dominated by one
party (Anderson 2000, Paldam 1991). In such countries, there tends to be greater
uncertainty about the likely shape of an alternative future government that will
form after the election has been held, and this diminishes the likelihood that voters will turn out the incumbent government even when economic conditions are
bad. This, in turn, exacerbates any potential moral-hazard problem inherent in the
voter-government relationship (cf. Fearon 1999).
Thus, in addition to the clarity of responsibility, the effects of economic conditions also should be moderated by the clarity of available alternatives, or what
Lewis-Beck (1988) has called “incumbent alternatives for dissent.” A more clearly
defined set of viable alternatives to the incumbent government should lead citizens
to more readily express content or discontent with the ruling party or parties
(Bengtsson 2004). In summary, when political conditions and institutional structures do not change, what matters are the available alternatives. Citizens will desert
the governing party only when they have somewhere to go to express their discontent (Anderson 2000).3
Although the approaches to understanding contingent effects from an institutional or contextual perspective differ, they all view the truthful translation of
economic evaluations into an economic vote as contingent—either because of the
difficulty of assigning responsibility to the right actor, institution, or party, or because of a lack of choices that would enable citizens to use their vote to mete out
rewards and punishments (for a slightly different approach, see Duch & Stevenson
2005). Parallel to the individual-level findings on economic voting, this body of
3
Thus, the basic economic voting model has an incumbent running with (or away from) the
economy, depending on whether economic performance was good or bad. Implicit in many
of the models, and in many of the conceptual elaborations, is the assumption that voters
sanction bad economic performance in order to provide incentives for a successor to strive
for good performance. Yet, given that the quality of challengers is likely to be an endogenous
consequence of incumbent performance, the quality of challengers will determine whether
the electorate is in a position to produce these incentives. And as the literature on government
termination suggests, in some countries the availability of alternatives is clearly endogenous
to economic conditions (Palmer & Whitten 2000).
12 Jan 2007 20:17
286
AR
P1: OKZ
ANRV312-PL10-13.tex
XMLPublishSM (2004/02/24)
AR REVIEWS IN ADVANCE10.1146/annurev.polisci.10.050806.155344
ANDERSON
research demonstrates that that the relationship between the state of the economy
and voter behavior is highly conditional at the cross-national (or cross-institutional
level) as well. Even if none of the individual-level biases I have enumerated existed, democratic institutions and political contexts frequently serve to weaken
the impact of economic conditions on voting behavior and election outcomes in
systematic ways. And this, too, undercuts the assumption that the economy drives
elections and government support.
Annu. Rev. Polit. Sci. 2007.10. Downloaded from arjournals.annualreviews.org
by CORNELL UNIVERSITY on 04/30/07. For personal use only.
THE END OF ECONOMIC VOTING, OR BACK TO BASICS?
Viewed from the perspective of democratic accountability, these new findings
about the nature of the link between the economy and government support are quite
coherent. What are we to make of them? Cumulatively, the evidence available to
date strongly suggests that the economic voting effect—defined by a bad economy
leading to lower government support or loss of office—is intermittent, highly
contingent, and substantively small (McDonald & Budge 2005). In fact, it may well
be that statistically significant coefficients for economic variables systematically
overstate the true effect of the economy because incumbents, on average, lose about
2.3% of voter support in elections independent of their performance in office
(McDonald & Budge 2005). When seen in conjunction with the magnitude of
the lower-than-expected performance of the governing party in elections, negative
economic effects may have little to do with accountability per se but simply capture
the long-run dynamics of electoral processes (McDonald & Budge 2005). In short,
economic effects on election outcomes do not qualify as a “robust fact” about
elections (cf. Bengtsson 2004; see also Cheibub & Przeworski 1999, van der Brug
et al. 2007).
When we add to this the ongoing challenges to the rationality paradigm by
psychology and behavioral economics, it is not unreasonable to ask whether the
research agenda on economics and elections has run its course. If the current state
of understanding is not to spell the end of economic voting as a field of inquiry,
researchers must recognize the need for stronger and more explicit theoretical
foundations (see also Hibbs 2006). As the study of economic voting continues
seemingly unabated (for innovative extensions, see, e.g., Anderson 2006, Hellwig
2001, Hellwig & Samuels 2007, Pacek & Radcliff 1995b, Samuels 2004), has
started to branch out into the world of transitioning democracies and the Third
World (cf. the contributions in Stokes 2001 and the review in Tucker 2002), and
shows few signs of slowing down, a normative correction seems to be in order—
namely, to ask students of economic voting to be conscious of theories of accountability and representation and to explicate how their findings square with those
theories. One starting point would be to focus much more explicitly on questions of
democratic accountability and citizenship and to confront the institutional and individual contingency dilemmas portrayed here in light of these questions. One way
of doing so would be to address the different contingency dilemmas directly via
12 Jan 2007 20:17
AR
P1: OKZ
ANRV312-PL10-13.tex
XMLPublishSM (2004/02/24)
AR REVIEWS IN ADVANCE10.1146/annurev.polisci.10.050806.155344
Annu. Rev. Polit. Sci. 2007.10. Downloaded from arjournals.annualreviews.org
by CORNELL UNIVERSITY on 04/30/07. For personal use only.
THE END OF ECONOMIC VOTING?
287
research designs that ask whether the right conditions can be created to maximize
the potential for accountability.
A second recommendation is to develop a better understanding of the normative implications of the specific findings that both individual voters and political
institutions play a role in blunting the edge of bad economic times. Voters are
predisposed not to acquire much information about the economy, and they are
motivated to interpret this information in ways consistent with previously held
beliefs or their own voting behavior. Given that people use different standards,
have different perceptions, and form different evaluations of the economy, accurate perceptions of economic conditions are possible only seldom and only for a
small subset of citizens. As well, democratic institutional designs and contextual
conditions commonly allow governments to avoid blame and make it difficult to
determine who is responsible for economic conditions, even for voters who are
well informed and motivated to mete out the proper sanction.
These findings have a number of implications. Empirically, the consistent inconsistency of economic effects explains why some authors have concluded that
economic voting is retrospective and sociotropic in nature while others argue that
it is prospective and egocentric in nature. Depending on the country and electorate, as well as the institutional and informational environment, it is plausible
that economic voting could be all of those things for some people in some places
at some points in time. Moreover, contingent economic effects help explain why
governments are retained in office despite a poor economic record (such as the
British Conservatives in 1992) or kicked out of office during good economic times
(such as the British Conservatives in 1997).
None of this is liable to improve people’s faith in the predictive capacity of the
economic voting paradigm, as the answer to the questions of whether and when
the economy matters seems to be a resounding “it depends.” But aside from their
purely empirical significance, these questions affect how we view the effectiveness
of economic voting as a mechanism of democratic accountability. According to
Manin et al. (1999, p. 40), “Governments are ‘accountable’ if voters can discern
whether governments are acting in their interest and sanction them appropriately,
so that those incumbents who act in the best interest of citizens win reelection and
those who do not lose them.” I would add that, as an empirical matter, accountability
happens when voters can and want to discern whether governments are acting in
their interest. These factors converge only intermittently.
However, in the long run, a lack of accountability rooted in institutions is
more problematic than one based on individual differences. One reason is practical: Institutions are hard to change, and governments have an incentive to avoid
transparency. Another reason is normative: In a democracy, citizens are free to
misperceive what qualifies as objective reality to political economists and misjudge the government as a result. That is, even when citizens’ misperceptions do
not cancel out in the aggregate, biased evaluations of the real economy are less
troubling normatively because accountability as it has been construed does not
require citizens to be all-knowing, nor does it see a role for political economists to
12 Jan 2007 20:17
Annu. Rev. Polit. Sci. 2007.10. Downloaded from arjournals.annualreviews.org
by CORNELL UNIVERSITY on 04/30/07. For personal use only.
288
AR
P1: OKZ
ANRV312-PL10-13.tex
XMLPublishSM (2004/02/24)
AR REVIEWS IN ADVANCE10.1146/annurev.polisci.10.050806.155344
ANDERSON
instruct people what standards and indicators of performance they should consider.
As Manin (1997) notes, voters can decide whether to reelect the incumbent on any
basis they want, including qualifying for the World Cup, and they can change
their minds between the beginning and the end of a term. In this way voters are
sovereign.
In contrast, institutions (as well as the political conditions they produce) whose
design is usually beyond the control of citizens are to blame when voters are prevented from discriminating good performance from bad and thus from sanctioning
governments appropriately, whereas voters only have themselves to blame when
economic conditions do not translate accurately into economic perceptions. The
latter can be viewed at least partly as a choice, but the former cannot, unless voters
had a hand in creating these institutional and contextual conditions. Thus, a normative contingency dilemma exists primarily because of institutions and contexts
that thwart voters, not because of voters themselves.
How can this dilemma be resolved practically? Although the accurate translation of objective economic conditions into people’s perceptions could be facilitated
by more and better information about the country’s economic state, the complete
answer to this question is far from obvious. To be sure, accountability cannot induce representation when voters have incomplete information (Manin et al. 1999,
p. 44). Thus, Manin et al. (1999, p. 50) have suggested that “we need institutions
that would provide citizens with independent information about the government:
‘accountability agencies,’ in the terms of an Australian Commission on Government Reform.” But more, better, and unbiased information cannot improve citizens’
motivations or cognitive abilities to process complex information, and it does little
to invalidate political biases that might undermine accurate perceptions. Short of
modifying the ways in which political and economic information is delivered, one
way out of this dilemma might be to welcome the well-documented decline in parties and partisan attachments across western democracies (Dalton & Wattenberg
2002), which should “free” citizens from their own biases and thus allow them to
focus on the real performance of the government. Yet this too has the potential to
be problematic, since politics is more easily understood when simple structuring
principles like partisanship or ideology help citizens arrange the world in their
own minds. And, as Hardin (2000, p. 46) points out, when there is little hope of
parties’ aligning as coherently as in the neat division between economically conservative and economically liberal, “the accountability of elected officials becomes
increasingly murky.”
This leaves institutional fixes to allow greater accountability through economic
voting in potential and reality. This insight is well known to institutional designers. The most obvious starting point would be the construction of institutions
that facilitate voters’ assignment of credit and blame, and subsequent reward and
punishment when warranted (given whatever standard of evaluation they choose
to apply). Moreover, these institutions should be able to produce viable governing alternatives—a set of clear and feasible choices—so that voters can replace
the incumbent. Whether such institutions should be imposed from the outside
12 Jan 2007 20:17
AR
P1: OKZ
ANRV312-PL10-13.tex
XMLPublishSM (2004/02/24)
AR REVIEWS IN ADVANCE10.1146/annurev.polisci.10.050806.155344
Annu. Rev. Polit. Sci. 2007.10. Downloaded from arjournals.annualreviews.org
by CORNELL UNIVERSITY on 04/30/07. For personal use only.
THE END OF ECONOMIC VOTING?
289
(if possible) or whether they should evolve endogenously—perhaps through competition, which creates incentives for agents (governments) to offer principals
(voters) tools to control and monitor officials (Ferejohn 1999, p. 133)—is beyond
the scope of this paper but an important question of institutional design. Yet, this
seemingly simple fix runs into the old problem that it is difficult, if not impossible,
to state a priori how much popular control is sufficient for ensuring accountability
while simultaneously guaranteeing effective government and stability. Some institutional designers, like Madison and his successors, are troubled by the notion of
too much popular control and have therefore installed political systems that resolve
the tradeoff between popular control and effectiveness in favor of the latter. But, of
course, these institutional devices are likely to be exactly those that make accountability more difficult because attribution of responsibility is hampered. A related
issue is the design of economic policy institutions. Many countries rely on unaccountable institutions such as central banks to make decisions (see, e.g., Stiglitz
1998), further complicating the question of who ought to be held accountable for
economic performance.
Perhaps this issue can be resolved empirically by asking whether systems with
more transparent institutions also produce the expected benefit of inducing governors to behave responsibly (Ferejohn 1999, Kiewiet 2000). There is some evidence
that this is indeed the case. For example, recent studies have found that political
systems with greater transparency have higher levels of public good provision and
systematically lower levels of public debt (Alt & Lassen 2006, Lassen 2003; but
also see Persson et al. 2000), and they allow for the more general assignment of
credit and blame for other policy areas, such as fiscal policy (Lowry et al. 1998).
BEYOND (SIMPLY) ACCOUNTABILITY?
All the while, it is worth asking whether the assumption that voters should reward
and punish politicians for economic performance is worth sustaining in the first
place, and, if so, whether accountability should be the prime normative standard to
guide scholarship in the area of economic voting. Regarding the former question,
two arguments can be made against the accountability model. First, the complete or
“real” economic voting model as it has been characterized in the empirical literature
discussed above simply requires too much of citizens to be of much use. To properly
judge the government’s record, citizens ideally should be well informed, unbiased
consumers of accurate and plentiful information in an environment of clear and
transparent political decision making. Such conditions are unlikely to be met in
reality save for a tiny number of citizens.
The second argument against the accountability model can be phrased thus:
Is it useful to expect that politicians can and do affect economic performance, or
that they seek to improve the country’s well-being writ large? After all, given that
the impact of governments on economic performance tends to be contingent as
well, and given that political parties are generally interested in benefiting their
12 Jan 2007 20:17
Annu. Rev. Polit. Sci. 2007.10. Downloaded from arjournals.annualreviews.org
by CORNELL UNIVERSITY on 04/30/07. For personal use only.
290
AR
P1: OKZ
ANRV312-PL10-13.tex
XMLPublishSM (2004/02/24)
AR REVIEWS IN ADVANCE10.1146/annurev.polisci.10.050806.155344
ANDERSON
own supporters rather than maximizing the general welfare (Hibbs 1977, Lange
& Garrett 1985), it is unclear whether accountability is the yardstick students of
economic voting should employ. But even if we stick with accountability as a
standard for democratic performance, it probably should not be the only or even
most prominent standard for assessing the quality of democracy. After all, the
issue of representation is much broader and potentially much more interesting
than the simple reward-punishment model. Thus, instead of focusing too narrowly
on accountability concerns, perhaps the broader notion of responsiveness, which
focuses on the various mechanisms of translating citizen preferences into policy
outcomes, offers an alternative (Powell 2004). Here, parties play a crucial role,
as do individual preferences and institutional conditions that subvert or facilitate
conditions of responsiveness via political parties (McDonald & Budge 2005; see
also van der Brug et al. 2007). This moves the economic voting paradigm away
from the government-opposition dichotomy, and it allows for both prospective and
retrospective judgments.
Even if aspects of the reward-punishment model of economic voting are here
to stay, future research needs to embed empirical studies in the larger literatures
on democratic citizenship and authority. It also ought to focus on institutions as
moderators of the economy–government support link, especially if it is concerned
with notions of democracy and accountability (see, e.g., Samuels 2004, Hellwig
& Samuels 2007). Aside from homing in on critical aspects of the contingency
dilemma explicated here, this would allow those engaged in economic voting
research to connect their findings with research programs in related fields that
examine political business cycles, principal-agent models, comparative political
economy, and accountability more generally (see Alt 2002 for a review of this
literature).
In the end, not all is doom and gloom. Research into the economic determinants
of government support constitutes one of the few areas of political science where
students of American and comparative politics have united to successfully push
the boundaries of what we know (see, e.g., Tucker 2006). Moreover, it is one area
of political science scholarship that has much to add to the received conventional
wisdom popularly propagated by political commentators about the connection
between the fortunes of democratic governments the state of the economy. Yet,
the contingency dilemmas outlined here should serve as a warning. Economic
voting research is more likely to remain viable as a field of inquiry if it can
be recast as a general set of important relationships that can illuminate various
aspects of democratic governance—including the role of the media, elite behavior
(blame avoidance), the study of institutions, and voter behavior—but also as an
explicit concern with normative aspects of democratic politics. Economic voting
research will thrive if it is integrated with broader and related research agendas
on democracy, elections, and representation. The field and those who peruse its
findings will benefit from a self-conscious reconsideration of the role of democratic
theory as well as from making connections with findings and theorizing in other,
related corners of the discipline.
12 Jan 2007 20:17
AR
P1: OKZ
ANRV312-PL10-13.tex
XMLPublishSM (2004/02/24)
AR REVIEWS IN ADVANCE10.1146/annurev.polisci.10.050806.155344
THE END OF ECONOMIC VOTING?
291
ACKNOWLEDGMENTS
I thank Rachel Cremona for research assistance. I am grateful to Elizabeth Cohen,
Michael D. McDonald, David Rueda, and the editors for their generous and
thoughtful comments. Thanks also to the Department of Politics and International
Relations and Nuffield College at the University of Oxford for their hospitality
while I worked on this paper.
Annu. Rev. Polit. Sci. 2007.10. Downloaded from arjournals.annualreviews.org
by CORNELL UNIVERSITY on 04/30/07. For personal use only.
The Annual Review of Political Science is online at
http://polisci.annualreviews.org
LITERATURE CITED
Aidt TS. 2000. Economic voting and information. Elec. Stud. 19:349–62
Alt JE. 2002. Comparative political economy:
credibility, accountability, and institutions.
In Political Science: The State of the Discipline, ed. I Katznelson, HV Milner, pp.147–
71. New York: WW Norton
Alt JE, Lassen DD. 2006. Fiscal transparency,
political parties, and debt in OECD countries.
Eur. Econ. Rev. 50:1403–39
Anderson CD. 2006. Economic voting and multilevel governance: a comparative individuallevel analysis. Am. J. Polit. Sci. 50:449–
63
Anderson CJ. 1995. Blaming the Government:
Citizens and the Economy in Five European
Democracies. Armonk, NY: ME Sharpe
Anderson CJ. 2000. Economic voting and political context: a comparative perspective. Elec.
Stud. 19:151–70
Anderson CJ, O’Connor KM. 2000. System
change, learning, and public opinion about
the economy. Br. J. Polit. Sci. 30:147–72
Anderson CJ, Mendes S, Tverdova YV.
2004. Endogenous economic voting: evidence from the 1997 British election. Elec.
Stud. 23:683–708
Barro R. 1973. The control of politicians:
an economic model. Public Choice 14:19–
42
Bartels LM. 1996. Uninformed votes: information effects in presidential elections. Am. J.
Polit. Sci. 40:194–230
Basinger SJ, Lavine H. 2005. Ambivalence,
information, and electoral choice. Am. Polit.
Sci. Rev. 99:169–84
Bengtsson A. 2004. Economic voting: the effect
of political context, volatility and turnout on
voters’ assignment of responsibility. Eur. J.
Polit. Res. 43:749–67
Berelson B, Lazarsfeld P, McPhee W. 1954.
Voting: A Study of Opinion Formation in
a Presidential Campaign. Chicago: Univ.
Chicago Press
Borre O. 1997. Economic voting in Danish electoral surveys 1987–1994. Scand. Polit. Stud.
20:347–65
Campbell A, Converse PE, Miller W, Stokes S.
1960. The American Voter. New York: Wiley
Chappell HW, Goncalves Veiga L. 2000. Economics and elections in western Europe:
1960–1997. Elec. Stud. 19:183–97
Chappell HW, Keech WR. 1985. A new
view of political accountability for economic
performance. Am. Polit. Sci. Rev. 79:10–
27
Cheibub JA, Przeworski A. 1999. Democracy,
elections, and accountability for economic
outcomes. See Przeworski et al. 1999, pp.
222–50
Conover P, Feldman S. 1986. Emotional reactions to the economy: I’m mad as hell and I’m
not going to take it anymore. Am. J. Polit. Sci.
30:50–78
Conover P, Feldman S, Knight K. 1986. Judging inflation and unemployment: the origins
of retrospective evaluations. J. Polit. 48:565–
88
12 Jan 2007 20:17
Annu. Rev. Polit. Sci. 2007.10. Downloaded from arjournals.annualreviews.org
by CORNELL UNIVERSITY on 04/30/07. For personal use only.
292
AR
P1: OKZ
ANRV312-PL10-13.tex
XMLPublishSM (2004/02/24)
AR REVIEWS IN ADVANCE10.1146/annurev.polisci.10.050806.155344
ANDERSON
Converse PE. 1964. The nature of belief systems in mass publics. In Ideology and Discontent, ed. D Apter, pp. 206–61. New York:
Free
Converse PE. 1990. Popular representation and
the distribution of information. In Information and Democratic Processes, ed. JA
Ferejohn, JH Kuklinski, pp. 369–88. Urbana:
Univ. Ill. Press
Dahl RA. 1971. Polyarchy: Participation and
Opposition. New Haven, CT: Yale Univ.
Press
Dalton RJ. 2002. Citizen Politics: Public Opinion & Political Parties in Advanced Industrial Democracies. Chatham, NJ: Chatham
House
Dalton RJ, Wattenberg MP. 2002. Parties Without Partisans: Political Change in Advanced
Industrial Democracies. New York: Oxford
Univ. Press
Dorussen H, Taylor M. 2001. The political context of issue-priority voting: coalitions and
economic voting in the Netherlands, 1970–
1999. Elec. Stud. 20:399–426
Downs A. 1957. An Economic Theory of
Democracy. New York: Harper & Row
Duch RM, Stevenson RT. 2005. Context and the
economic vote: a multilevel analysis. Polit.
Anal. 13:387–409
Duch RM, Palmer HD, Anderson CJ. 2000.
Heterogeneity in perceptions of national economic conditions. Am. J. Polit. Sci. 44:635–
52
Eulau H, Lewis-Beck MS, eds. 1985. Economic Conditions and Electoral Outcomes in
the United States and Western Europe. New
York: Agathon
Evans G, Andersen R. 2006. The political conditioning of economic perceptions. J. Polit.
68:194–207
Fearon JD. 1999. Electoral accountability and
the control of politicians: selecting good
types versus sanctioning poor performance.
See Przeworski et al. 1999, pp. 55–97
Ferejohn J. 1986. Incumbent performance
and electoral control. Public Choice 50:5–
26
Ferejohn J. 1999. Accountability and authority:
towards a model of political accountability.
See Przeworski et al. 1999, pp. 131–54
Fiorina M. 1981. Retrospective Voting in American National Elections. New Haven, CT:
Yale Univ. Press
Fischhoff B. 1976. Attribution theory and judgment under uncertainty. In New Directions in
Attribution Research. Vol. 1, ed. JH Harvey,
WJ Ickes, RF Kidd, pp. 421–52. New York:
Wiley
Frey B, Garbers H. 1971. Der Einfluss
wirtschaftlicher Variabler auf die Popularität
der Regierung: Eine empirische Analyse.
Jahrb. Natl. ökon. Stat. 186:281–95
Frey B, Schneider F. 1980. Popularity functions: the case of the U.S. and West Germany.
In Models of Political Economy, ed. P
Whiteley, pp.47–84. Beverly Hills, CA: Sage
Goidel RK, Langley RE. 1995. Media coverage of the economy and aggregate economic
evaluations: uncovering evidence of indirect
media effects. Polit. Res. Q. 48:313–28
Gomez BT, Wilson JM. 2006. Cognitive heterogeneity and economic voting: a comparative
analysis of four democratic electorates. Am.
J. Polit. Sci. 50:127–45
Goodhart CAE, Bhansali RJ. 1970. Political
economy. Polit. Stud. 18:43–106
Haller HB, Norpoth H. 1997. Reality bites:
news exposure and economic opinion. Public
Opin. Q. 61:555–75
Hardin R. 2000. The public trust. In Disaffected
Democracies: What’s Troubling the Trilateral Democracies, ed. SJ Pharr, RD Putnam,
pp. 31–51. Princeton, NJ: Princeton Univ.
Press
Hardin R. 2002. Street-level epistemology
and democratic participation. J. Polit. Phil.
10:212–29
Hellwig TT. 2001. Interdependence, government constraints and economic voting. J.
Polit. 63:1141–62
Hellwig TT, Samuels D. 2007. Electoral accountability and the variety of democratic
regimes. Br. J. Polit. Sci. 37: In press
Hetherington MJ. 1996. The media’s role in
forming voters’ national economic evaluations in 1992. Am. J. Polit. Sci. 40:372–95
12 Jan 2007 20:17
AR
P1: OKZ
ANRV312-PL10-13.tex
XMLPublishSM (2004/02/24)
AR REVIEWS IN ADVANCE10.1146/annurev.polisci.10.050806.155344
Annu. Rev. Polit. Sci. 2007.10. Downloaded from arjournals.annualreviews.org
by CORNELL UNIVERSITY on 04/30/07. For personal use only.
THE END OF ECONOMIC VOTING?
Hibbs Douglas A. 2006. Voting and the macroeconomy. Oxford Handbook of Political
Economy, ed. BR Weingast, D Wittman,
pp. 565–86. New York: Oxford Univ. Press
Hibbs D. 1977. Political parties and macroeconomic policy. Am. Polit. Sci. Rev. 71:1467–
87
Holbrook T, Garand JC. 1996. Homo economus? Economic information and economic
voting. Polit. Res. Q. 49:351–75
Keech WR. 1995. Economic Politics: The Costs
of Democracy. New York: Cambridge Univ.
Press
Kernell S. 1978. Explaining presidential popularity. Am. Polit. Sci. Rev. 72:506–22
Key VO. 1966. The Responsible Electorate.
New York: Vintage Books
Kiewiet DR. 2000. Economic retrospective voting and incentives for policymaking. Elec.
Stud. 19:427–44
Kirchgässner G. 1974. Ökonometrische Untersuchungen des Einflusses der Wirtschaftslage auf die Popularität der Parteien.
Schweiz. Z. Volkswirtsch. Stat. 100:409–45
Kirchgässner G. 1991. Economic conditions
and the popularity of West German parties: before and after the 1982 government
change. In Economics and Politics: The Calculus of Support, ed. H Norpoth, MS LewisBeck, JD Lafay, pp.103–22. Ann Arbor:
Univ. Mich. Press
Kramer G. 1971. Short-term fluctuations in U.S.
voting behavior, 1896–1964. Am. Polit. Sci.
Rev. 65:131–43
Kramer G. 1983. The ecological fallacy revisited: aggregate- versus individual-level findings on economics and elections, and sociotropic voting. Am. Polit. Sci. Rev. 77:92–
111
Krause GA. 1997. Voters, information heterogeneity and the dynamics of aggregate
economic expectations. Am. J. Polit. Sci.
41:1170–200
Krause GA, Granato J. 1998. Fooling some of
the public some of the time? A test for weak
rationality with heterogeneous information
levels. Public Opin. Q. 62:135–51
Kuklinski JH, West DM. 1981. Economic ex-
293
pectations and voting behavior in United
States House and Senate elections. Am. Polit.
Sci. Rev. 75:436–47
Lafay JD. 1985. Important political change and
the stability of the popularity function: before
and after the French general election of 1981.
In Economic Conditions and Electoral Outcomes, ed. H Eulau, MS Lewis-Beck, pp.78–
97. New York: Agathon
Lange P, Garrett G. 1985. The politics of growth
strategic interaction and economic performance in the advanced industrial democracies, 1974–1980. J. Polit. 47:792–827
Lassen DD. 2003. Political accountability and
the size of government: theory and crosscountry evidence. EPRU Work. Pap. Ser. 00–
20, Econ. Policy Res. Unit (EPRU), Univ.
Copenhagen
Lewis-Beck MS. 1980. Economic conditions
and executive popularity: the French experience. Am. J. Polit. Sci. 24:306–23
Lewis-Beck MS. 1988. Economics and Elections: The Major Western Democracies. Ann
Arbor: Univ. Mich. Press
Lewis-Beck MS, Mitchell G. 1993. Bringing
politics back in: the French lesson for popularity functions. Presented at Ann. Meet.
Midwest Polit. Sci. Assoc., Chicago
Lewis-Beck MD, Paldam M. 2000. Economic
voting: an introduction. Elec. Stud. 19:113–
22
Lewis-Beck MS, Stegmaier M. 2000. Economic determinants of electoral outcomes.
Annu. Rev. Polit. Sci. 3:183–219
Lohmann S. 1999. What price accountability?
The Lucas Island model and the politics of
monetary policy. Am. J. Polit. Sci. 43:396–
430
Lowry RC, Alt JE, Ferree KE. 1998. Fiscal policy outcomes and electoral accountability in
American states. Am. Polit. Sci. Rev. 92:759–
74
MacKuen MB. 1983. Political drama, economic
conditions, and the dynamics of presidential
popularity. Am. J. Polit. Sci. 28:165–92
Manin B. 1997. The Principles of Representative Government. New York: Cambridge
Univ. Press
12 Jan 2007 20:17
Annu. Rev. Polit. Sci. 2007.10. Downloaded from arjournals.annualreviews.org
by CORNELL UNIVERSITY on 04/30/07. For personal use only.
294
AR
P1: OKZ
ANRV312-PL10-13.tex
XMLPublishSM (2004/02/24)
AR REVIEWS IN ADVANCE10.1146/annurev.polisci.10.050806.155344
ANDERSON
Manin B, Przeworski A, Stokes SC. 1999.
Elections and representation. See Przeworski
et al. 1999, pp. 29–54
McDonald MD, Budge I. 2005. Elections, Parties, Democracy: Conferring the Median
Mandate. New York: Oxford Univ. Press
Mill JS. 1955. On Liberty. Chicago: Henry
Regnery
Monroe K. 1984. Presidential Popularity and
the Economy. New York: Praeger
Mueller JE. 1970. Presidential popularity from
Truman to Johnson. Am. Polit. Sci. Rev.
64:18–34
Mutz D. 1992. Mass media and the depoliticization of personal experience. Am. J. Polit. Sci.
36:483–508
Mutz D. 1994. Contextualizing personal experience: the role of mass media. J. Polit. 56:689–
714
Mutz D, Mondak J. 1997. Dimensions of sociotropic behavior: group-based judgements
of fairness and well-being. Am. J. Polit. Sci.
41:284–308
Nadeau R, Niemi R, Fan DP, Amato T. 1999.
Elite economic forecasts, economic news,
mass economic judgments, and presidential
approval. J. Polit. 61:109–35
Nadeau R, Niemi R, Yoshinaka A. 2002. A
cross-national analysis of economic voting:
taking account of the political context across
time and nations. Elec. Stud. 21:403–23
Nannestad P, Paldam M. 1994. The vpfunction: a survey of the literature on vote
and popularity functions after 25 years. Public Choice 79:213–45
Nannestad P, Paldam M, Rosholm M. 2003.
System change and economic evaluations:
a study of immigrants and natives in Israel.
Elec. Stud. 22:485–501
Nisbett RE, Ross L. 1980. Human Inference:
Strategies and Shortcomings of Social Judgment. Englewood Cliffs, NJ: Prentice-Hall
Norpoth H. 1984. Economics, politics, and the
cycle of presidential popularity. Polit. Behav.
6:253–73
Norpoth H. 2001. Divided government and economic voting. J. Polit. 63:414–35
Pacek A, Radcliff B. 1995a. The political econ-
omy of competitive elections in the developing world. Am. J. Polit. Sci. 39:745–59
Pacek A, Radcliff B. 1995b. Economic voting
and the welfare state: a cross-national analysis. J. Polit. 57:44–61
Page B, Shapiro R. 1992. The Rational Public.
Chicago: Univ. Chicago Press
Paldam M. 1991. How robust is the vote function? In Economics and Politics: The Calculus of Support, ed. H Norpoth, MS LewisBeck, JD Lafay, pp. 9–32. Ann Arbor: Univ.
Mich. Press
Paldam M, Nannestad P. 2000. What do voters know about the economy? A study of
Danish data, 1990–1993. Elec. Stud. 19:363–
92
Paldam M, Schneider F. 1980. The macroeconomic aspects of government and opposition popularity in Denmark 1957–1978.
Natl. Økon. Tidsskr. 118:149–70
Palmer HD, Whitten GD. 2000. Government
competence, economic performance and endogenous election dates. Elec. Stud. 19:413–
26
Palmer HD, Whitten GD. 2003. Questionable
analyses with no theoretical innovation: a response to Royed, Leyden and Borrelli. Br. J.
Polit. Sci. 33:139–49
Pattie C, Johnston R. 2001. Routes to party
choice: ideology, economic evaluations and
voting at the 1997 British general election.
Eur. J. Polit. Res. 39:373–89
Peffley M, Williams JT. 1985. Attributing presidential responsibility for national economic
problems. Am. Polit. Q. 13:393–425
Peffley M, Feldman S, Sigelman L. 1987. Economic conditions and party competence: process or belief revision. J. Polit. 49:100–21
Persson T, Roland G, Tabellini G. 2000. Comparative politics and public finance. J. Polit.
Econ. 108:1121–61
Powell GB. 2000. Elections as Instruments
of Democracy. New Haven, CT: Yale Univ.
Press
Powell GB. 2004. The chain of responsiveness.
J. Democr. 15:91–105
Powell GB, Whitten GD. 1993. A crossnational analysis of economic voting: taking
12 Jan 2007 20:17
AR
P1: OKZ
ANRV312-PL10-13.tex
XMLPublishSM (2004/02/24)
AR REVIEWS IN ADVANCE10.1146/annurev.polisci.10.050806.155344
Annu. Rev. Polit. Sci. 2007.10. Downloaded from arjournals.annualreviews.org
by CORNELL UNIVERSITY on 04/30/07. For personal use only.
THE END OF ECONOMIC VOTING?
account of the political context. Am. J. Polit.
Sci. 37:391–414
Przeworski A. 1999. A minimalist conception
of democracy: a defense. In Democracy’s
Value, ed. I Shapiro, C Hacker-Cordon, pp.
23–55. New York: Cambridge Univ. Press
Przeworski A, Manin B, Stokes SC. 1999.
Democracy, Accountability, and Representation. New York: Cambridge Univ. Press
Remmer K. 1993. The political economy of
elections in Latin America, 1980–1991. Am.
Polit. Sci. Rev. 87:393–407
Riker WH. 1965. Democracy in the United
States. New York: Macmillan. 2nd ed.
Royed TJ, Leyden KM, Borrelli SA. 2000.
Is ‘clarity of responsibility’ important for
economic voting? Revisiting Powell and
Whitten’s hypothesis. Br. J. Polit. Sci.
30:669–85
Rudolph TJ. 2003. Institutional context and
the assignment of political responsibility. J.
Polit. 65:190–215
Samuels D. 2004. Presidentialism and accountability for the economy in comparative perspective. Am. Polit. Sci. Rev. 98:425–36
Sanders D. 2000. The real economy and the
perceived economy in popularity functions:
How much do voters need to know? A
study of British data, 1974–1997. Elec. Stud.
19:275–94
Sanders D, Carey S. 2002. Temporal variations
in economic voting: a comparative crossnational analysis. In Economic Voting, ed. H
Dorussen, M Michaell Taylor, pp. 200–32.
London: Routledge
Sanders D, Gavin N. 2004. Television news,
economic perceptions and political preferences in Britain, 1997–2001. J. Polit.
66:1245–66
Schmitter P, Karl T. 1991. What democracy
is . . . and is not. J. Democr. 2:75–89
Schumpeter JA. 1942. Capitalism, Socialism
and Democracy. New York: Harper
Shah V, Watts MD, Domke D, Fan DP, Fibison
M. 1999. News coverage, economic cues and
the public’s presidential preferences, 1984–
96. J. Polit. 61:914–43
Smiley M. 1999. Democratic citizenship: a
295
question of competence? In Citizen Competence and Democratic Institutions, ed. S
Elkin, K Soltan, pp. 371–84. Univ. Park:
Penn. State Univ. Press
Stein RM. 1990. Economic voting for governor
and U.S. senator: the electoral consequences
of federalism. J. Polit. 52:29–53
Stigler GJ. 1973. General economic conditions and national elections. Am. Econ. Rev.
63:1960–67
Stiglitz J. 1998. Central banking in a democratic
society. De Econ. 146:199–226
Stokes SC, ed. 2001. Public Support for Market Reforms in New Democracies. New York:
Cambridge Univ. Press
Tibbits C. 1931. Majority votes and the business
cycle. Am. J. Soc. 36:596–606
Tucker JA. 2002. The first decade of postcommunist elections and voting: What have we
studied, and how have we studied it? Annu.
Rev. Polit. Sci. 5:271–304
Tucker JA. 2006. Regional Economic Voting: Russia, Poland, Hungary, Slovakia, and
the Czech Republic, 1990–1999. New York:
Cambridge Univ. Press
Tufte ER. 1975. Determinants of the outcomes of midterm congressional elections.
Am. Polit. Sci. Rev. 69:812–26
Tufte ER. 1978. Political Control of the Economy. Princeton, NJ: Princeton Univ. Press
van der Brug W, van der Eijk C, Franklin
MN. 2007. The Economy and the Vote: Understanding Voter Preferences and Election
Outcomes in Fifteen Countries. New York:
Cambridge Univ. Press. In press
Waldron J. 1998. Participation: the right of
rights. Proc. Aristot. Soc. 98:307–37
Warren ME. 1996. Deliberative democracy and
authority. Am. Polit. Sci. Rev. 90:46–60
Watt D. 1968. Labour’s hard road to recovery.
Fin. Times March 8:15
Weatherford MS. 1983. Economic voting and
the “symbolic politics” argument: a reinterpretation and synthesis. Am. Polit. Sci. Rev.
77:158–74
Welch S, Hibbing J. 1992. Financial conditions,
gender and voting in American national elections. J. Polit. 54:197–213
12 Jan 2007 20:17
296
AR
P1: OKZ
ANRV312-PL10-13.tex
XMLPublishSM (2004/02/24)
AR REVIEWS IN ADVANCE10.1146/annurev.polisci.10.050806.155344
ANDERSON
Annu. Rev. Polit. Sci. 2007.10. Downloaded from arjournals.annualreviews.org
by CORNELL UNIVERSITY on 04/30/07. For personal use only.
Whitten GD, Palmer HD. 1999. Cross-national
analyses of economic voting. Elec. Stud.
18:49–67
Wittman DA. 1989. Why democracies produce
efficient results. J. Polit. Econ. 97:1395–
1424
Wlezien C, Franklin MN, Twiggs D. 1997. Economic perceptions and vote choice: disentangling the endogeneity. Polit. Behav. 19:7–17
Zaller JR. 1992. The Nature and Origins of
Mass Opinion. New York: Cambridge Univ.
Press