The primary argument among PC proponents is that voters’ partisanship acts as a filter for their economic perceptions. Voters who are ill-informed about political and economic issues tend to respond more to cues provided to them by political parties than more informed voters (Keech, 1995; Gerber and Huber, 2009). Political parties, in turn, help shape the perceptions of voters who are unable or unwilling to sift through and process the political and economic information available (Zaller, 1992; Nelson and Kinder, 1996). Voters welcome information from their political parties that share their values and belief systems.
EC scholars, in contrast, argue that economic forces condition voters’ economic impressions (MacKuen et al., 1992). When voters observe the economy, they form assessments that reflect current conditions (Nadeau and Lewis-Beck, 2001; Lewis-Beck, 2006; Lewis-Beck et al., 2008). Their personal well-being and how they perceive the overall performance of the economy then define their perception of the economy (Kinder and Kiewet, 1978).
A critical review of the literature illuminates some of the limitations of the PC and EC arguments on the economic voting process. Namely, the literature on the conditioning of voters’ economic perceptions does not offer a comprehensive argument on how economic forces inform voters’ decisions. PC scholars say that voters make choices at the polls primarily on partisanship, which in turn influences their economic evaluations. This point of view ignores outside forces that influence voters, suggesting that political parties can simply prime them on to how perceive the economy and lead them to particular voting decisions. The limit of this rationale is that it presumes that voters have already made up their minds going into an election, because their partisanship has determined this, and that partisan cues determine the level of consideration voters give to these arguments when making their voting decisions (Just et al., 1996). A more persuasive argument is that voters make their considerations about the economy the moment that changes in economic conditions are significant enough to resonate with them (Zaller, 1992).
Further, PC and EC scholars do not fully explain the process by which electoral outcomes shape voters’ economic perceptions. PC scholars contend that if the opposition party wins, then voters are likely to form positive economic assessments. In this case causality is defined by winning the election, which, in effect, changes partisan supporters’ view of the economy. In addition, an implicit claim in EC scholars’ research is that voters make selections based on the economy to protect their own economic interests. The limitation of this claim is that electoral forces can at times moderate voters’ economic evaluations. Voters may make a political decision based not only on their economic interests, but also on their political preferences. What ensues from this process is that electoral outcomes cause voters’ economic assessments to change. Partisan supporters of the party that has just lost the election develop more negative economic perceptions. From their vantage point, the major opposition party that is now in power will have policies that will likely make the economy worse.
Defining economic-minded partisans
What I call the ‘economic-minded partisan model’ is based on how voters’ economic perceptions and partisanship in combination condition their voting behavior. In this model, partisanship influences voters to make decisions reflecting their main political beliefs and values. Economic perceptions simultaneously influence voters as they assess the incumbent government’s past competence in handling the economy and evaluate its prospects for managing it in the future. This model reflects the reality that partisanship and economic perceptions are sometimes difficult to separate.
Economic-minded partisans engage in limited economic voting. Voters’ partisanship dampens their economic perceptions in the voting process. Economic-minded partisans make their voting decisions when changes in the economy are distinct enough for them to take notice. Voters then decide if they will reward or punish the incumbent party based on these changes. At the same time, voters’ partisanship moderates their evaluations of the economy as a means to cope with its uncertainty. Partisanship, therefore, not only mitigates whether individuals approve or disapprove of the incumbent government’s management of the economy, but also determines if voters will choose a party that differs from their existing partisan preferences. As a result, the presumption that partisan voters uniformly engage in economic voting is unlikely. The level of support for a political party tends to be a strong and reliable factor in explaining voting behavior.
This model does not assume a high level of political sophistication among voters. Rather, it assumes the following: First, individuals pay varying levels of attention to politics. Second, voters’ reactions to issues and events are limited by their knowledge of particular phenomena. Third, voters’ perceptions of the economy are not constant, but reflect immediate reactions that result from changes in the political and economic environments. Fourth, voters use existing partisan attitudes to process new economic information, as well as to assist them in considering previous information on the economy when voting (Fiske and Linville, 1980; Conover et al., 1986). Fifth, voters’ economic perceptions are more likely to be based on sociotropic assessments – defined as voters’ evaluation of the broader economy – of the economy that can either be retrospective or prospective (Kinder and Kiewiet, 1978). Lastly, voters’ perceptions of the economy and their partisan intensity are based upon the information that is most salient to them (Zaller, 1992).
The intent of the model is to demonstrate how the economic vote works when it is conditioned by partisanship and economic perceptions. In addition, the economic-minded partisan model argues that non-partisan voters, like partisan voters, are susceptible to partisan conditioning and economic conditioning when voting. The theory postulates that non-partisans, when compared to partisans, are more sensitive to differences between the parties in their evaluations of how well each party has managed economic and political conditions during an election cycle. Since non-partisan voters are not bound by pre-existing attachments to a political party when voting, they actively look for differences between the parties and their candidates in competence in managing economic and political conditions (Popkin, 1992). In addition non-partisans, like partisans, are receptive to partisan and economic conditioning because they make their voting decisions based upon some combination of the following factors: issue preferences, candidate evaluations, knowledge of politics, ideology, assessment of the economy, and information provided to them during the campaign (Basinger and Levine, 2005).
Analyzing the 2010 British general election
Table 10.1 summarizes the two-stage probit estimates from the 2010 UK general election taken during the pre-election period – June 2008 through March 2010 – and during the actual campaign – from 12 April 2010 to polling day on 6 May 2010. Overall, the model meets the critical threshold, and all of its coefficients are statistically significant.
Of particular interest to this research is the economic-minded partisan variable. A one-unit change in this variable, other things being equal, yields a –0.17 drop in the likelihood of voting for the incumbent party during the campaign. During the pre-election period, a one-unit change on this variable resulted in a –0.37 drop in the probability of voting for Labour, other things being equal.5
The results also uncover the impact economic voting had on the campaign. To start, the effect size of the model’s coefficients diminish in the campaign period, compared to voters’ prior preferences in the pre-election period. More specifically, when comparing the pre-campaign with the campaign period, there was a drop in the direct effects of voters’ partisanship and voters’ positive economic perceptions by roughly –22 percent and –46 percent respectively. This trend suggests that as UK voters began to make up their minds on the party they would support, the campaign’s effects may not have helped either party gain traction. Rather, the campaign period provided voters with enough information to guide their voting decisions and gave them an opportunity to determine how intensely they would punish th
e Labour Party, either by voting for another party, or by voting at all, due to an overall pessimism among voters as a result of the economic downturn (Gelman and King, 1992).
Table 10.1 Two-stage probit estimates 2010 UK general election (cross-sectional data)
Source: British Election Survey (2010).
The economic-minded partisan variable indicates that not all voters uniformly engaged in economic voting. This is due to both the conditioning process of voters’ partisanship and voters’ economic perceptions in explaining voting behavior. To illustrate the conditioning process between these two variables, Figure 10.1 shows average changes in voting for the Labour Party based on voters’ partisanship. It compares changes in economic perceptions between those who felt the economy had gotten better or stayed the same and those who felt the economy had gotten worse.
Voters who were most likely to engage in economic voting in the election were leaning partisans who supported Labour and the Conservative parties. For instance, during the campaign there was a drop of 12 percentage points in the probability of voting for Labour, based on changes from positive to negative economic perceptions. Prior to the campaign, there was a decrease of 26 percentage points in voting for Labour among leaning Labour supporters. This type of behavior also occurred among leaning partisans who supported the Conservatives. During the campaign, the change from positive to negative economic perceptions resulted in a nine percentage point drop and during the pre-election period, an 18-point decrease, in voting for the Labour party. As one moves toward each end of the scale, changes in voting for the incumbent party based on the economic vote decline. This finding suggests stronger partisans did not engage in economic voting, compared to nominally partisan voters. Both weak and strong partisans who supported Labour were not likely to change their initial decisions to vote for their parties’ candidates as a result of their economic perceptions; the same was true among strong and weak Conservative voters.
Figure 10.1 Average difference in probability of voting for the Labour Party based on economic vote (economy has ‘stayed the same or better’ from economy has gotten ‘worse’)
Source: British Election Survey (2010).
Campaign effects appeared to cause a drop in swings among individual British voters. Forced to make a choice between the two parties’ platforms the closer they got to election day, voters were not as likely to solely cast an economic vote. This may have been a problem for the Conservative Party; despite voting swings in its favor reflecting the economic vote, the size of these swings decreased when the official campaign was underway. This may have been a result of many Labour and Liberal Democrat voters finding it difficult to vote for the Conservative Party because their partisanship would not let them do so.
Analyzing the 2010 US congressional vote
Table 10.2 presents estimates from the 2010 US congressional vote model in February and September 2010. The data provide a point of comparison that shows voters’ behavior before major campaigning began and during the middle of the campaign. This approach allows the analysis to evaluate the degree to which campaign effects changed voting behavior. It controls for voters’ partisanship, their economic perceptions, and the interaction between these two variables.
The two-stage probit results indicate that all of the variables included in the model exceed its critical threshold and that all of its variables are significant. Prior to the campaign, the variable that specifies economic-minded partisans yields a –0.58 decrease in the probability of voting for the Democratic Party. In September 2010, the effect size of this variable increased to –0.32 units, other things being equal.6
As was the case in the UK, there was also a decline in the effect size of the model’s parameters during the campaign (September 2010) when compared to the period before the campaign (February 2010). The findings indicate a significant drop of –11 percent in the effect size of voters’ partisanship and a –46 percent drop in voters’ positive economic perceptions. The impact for the latter parameter was on par with the UK results. As was the case in the British election, the American results point to the asymmetric impact of the economic downturn on a drop in overall intensity levels among incumbent party supporters the closer to election day. Voters were, therefore, more likely to use the campaign as a means to reinforce their negative assessments of the economy and political environment in general, thus causing them to punish the Democratic Party, rather than embrace the Republican Party as the more desirable outcome. This may have resulted in an overall decline in the likelihood of turnout for the Democratic Party among its own supporters on election day. However, this possibility was not directly tested in this model. At the same time, these factors may have contributed to an overall boost in the propensity of individuals who were motivated to punish the incumbent party for poor economic conditions to turn out to vote.
Table 10.2 Two-stage probit estimates 2010 US congressional elections
Source: Pew Research Center for the People and the Press (2010).
Last, Figure 10.2 plots changes in the probability of voting for the incumbent party, based on differences between voters’ negative and positive economic perceptions by voters’ partisan level. The US findings indicate a similar distribution among voters as in the UK. Namely, less partisan voters were more likely to engage in economic voting than strong partisans. These individuals appear to be swing voters who left the Democratic Party and voted for the Republican Party during the 2010 election. Among these less partisan voters, a change from positive to negative economic perceptions resulted in a 32 percent drop in the probability of voting for the party in February 2010 and a 38 percent decrease by September 2010. By comparison, Democratic voters had a 10 percent drop in February 2010 and a 17 percent decrease in voting for the Democratic Party in September 2010. The differences between February 2010 and September 2010 indicated that over the course of the campaign, voters were more likely to engage in economic voting. At first blush, these effects appear to have hurt the Democratic Party, diminishing the impact of campaigning efforts undertaken by President Obama. Clearly, these results suggest that American voters engaged in retrospective voting, treating the 2010 midterm election as a referendum on the President rather than a choice over the two party’s platforms.
Figure 10.2 Average difference in probability of voting for the Democratic Party based on the economic vote (economy has ‘stayed the same or better’ from economy has gotten ‘worse’)
Source: Pew Research Center for the People and the Press (2010).
The politics of aggrieved acquiescence
The results indicate that in the last election both British and American voters engaged in economic voting. Yet the data also reveal that not all voters did so equally. Strong partisan voters proved less likely to be economic voters, compared to leaning and weak partisans, as well as independent voters. This heterogeneity among voters’ voting decisions leads us to ask: Did voters truly vote based on the economy, or did voters use the economy to reinforce their pre-existing partisan bias?
This research suggests an answer to this question. Voters used their partisanship as the baseline for updating their political assessments. Thus, individuals were likely to make judgments about the economy the moment that changes in economic conditions were significant enough to resonate with them. In the case of the 2010 elections, partisanship among major opposition party supporters served as the conduit by which voters punished the Labour and Democratic parties.
The results also imply that the electoral strategies employed by the Obama administration and Brown government did not persuade voters that the elections would be a choice over which parties’ economic platforms would best steer each country out of the current economic downturn. A more plausible explanation was that campaigning by the incumbent president and prime minister reminded voters that they wanted to punish their party for the poor economy. Voters who were less partisan, non-partisan, or strong partisans who supported the major opposition party tended to reject the claims
made by both the Obama administration and Brown government about the state of the economy, their ability to manage it, and how their party’s policies would lead to economic growth in the near future. Yet less partisan and non-partisan voters who engaged in rejecting the incumbent party in each country were not likely to wholeheartedly embrace the alternative parties’ platforms. Frustration with the status quo and acceptance of the economic policy choices offered by the Republican and Conservative parties are not necessarily part of the same decision-making process among voters. Thus, the 2010 elections in the UK and US can be understood as a feeling of aggrieved acquiescence among voters. When presented with two choices, individuals tend to be reluctant to accept the alternative choice when this decision is based primarily on their dissatisfaction with the other option. Voters’ emotional reaction to the recent economic downturn triggered an immediate political response to punish the incumbent party. This process is likely to continue in the near term if the economy does not improve, thus, threatening the stability of the recently elected incumbent parties in upcoming elections.7
This outcome may be part of the reason that Britain’s Conservative Party has struggled to secure a majority of parliamentary seats in the House of Commons and that the Republican Party has not won control of both chambers in Congress. Even though British voters saw the economy, as was also the case among American voters, as the most important issue during the campaign, and voters sought to punish the incumbent Labour Party for it, not all voters were ready to swing to the Tories in their voting decisions.
Two popular arguments have emerged as to why this happened in the UK. First, voters were acting strategically in their voting decisions. Therefore weak and strong Labour partisan supporters were left with supporting their party or the Liberal Democrats but not the Conservatives. Second, systemic problems caused this outcome; the first past the post electoral system limited the ability of large swings in voting behavior. Though the national mood was pessimistic, and voters were not happy with Labour’s economic performance, people across all districts didn’t share these feelings equally. Conservatives were unable to swing enough Labour districts to their side because many voters in these districts either stayed with their party or went with the Liberal Democrats as a safe ideological alternative to the Tories.
The Legacy of the Crash Page 26