Sabato's Crystal Ball

Health Care as an Issue in the Midterm Election

Alan I. Abramowitz, Senior Columnist May 6th, 2010

Since the passage of the Democratic health care reform bill in March, congressional Republicans have been promising to turn the 2010 midterm election into a referendum on the new law. Party leaders including House Minority Leader John Boehner and Senate Minority Leader Mitch McConnell have argued that campaigning for repeal of health care reform will help Republicans pick up enough additional seats to win back control of the House and Senate. On the other side, Democratic leaders and strategists have been urging their party’s candidates to vigorously defend the new law and to use the threat of repeal to energize the Democratic base. All of this begs the question, however, of whether health care reform has the potential to influence enough voters to affect the results of the House and Senate elections.

In this article I examine the potential influence of health care reform as an issue in the 2010 midterm election. We know that the most important influence on voter decision-making in congressional elections is party identification. Over 90% of voters identify with or lean toward a party and in recent years the vast majority of party identifiers, including the vast majority of independents who lean towards a party, have cast their ballots for their own party’s House and Senate candidates.

In order to influence voter decision-making, an issue must cause some partisans to defect from their party’s candidate because of disagreement with their party’s position on that issue. Therefore the influence of an issue depends on the proportion of partisans on each side who disagree with their own party’s position. The potential of an issue to influence the outcome of an election is greatest when the proportion of cross-pressured partisans is much larger in one party than in the other party.

I will use data from a recent Gallup Poll of the U.S. voting age population to test the potential impact of health care reform on voter decision-making in the 2010 midterm elections. The survey was conducted from March 26-28 and involved interviews over landlines and cellular telephones with 1033 voting age adults including 925 registered voters. Respondents were asked about their preference for a generic Democrat vs. a generic Republican for the House of Representatives as well as their opinions about the recently passed health care bill.

The first thing that stands out when one examines the results of this poll is the powerful influence of party identification on vote choice. Among all registered voters, 48% favored a generic Republican, 46% favored a generic Democrat, and 6% were undecided. However, the results in Table 1 show that well over 90% of party identifiers and leaning independents supported their own party’s candidate. There was almost no difference in this regard between identifiers and leaners. In fact, Republican leaners were more loyal to their party than regular Republicans: 92% of Republican identifiers and 97% of Republican leaners favored a generic Republican while 95% of Democratic identifiers and 90% of Democratic leaners favored a generic Democrat.


Table 1. Generic Ballot Preference by Party Identification



Source: Gallup Poll, March 26-28, 2010


When registered voters were asked about the effect of the health care law on their congressional vote, they divided fairly evenly: 40% said they would be more likely to vote for a candidate who supported the law, 46% said they would be more likely to vote for a candidate who opposed the law, and 13% said it would have no effect on their vote.

The data in Table 2 show that these opinions were generally consistent with voting intentions. 75% of Democratic voters said they would be more likely to vote for a candidate who favored the law while only 8% said they would be more likely to vote for a candidate who opposed the law; 84% of Republican voters said they would be more likely to vote for a candidate who opposed the law while only 9% said they would be more likely to vote for a candidate who favored the law. Among undecided voters, 28% said they would be more likely to vote for a candidate who favored the law, 34% said they would be more likely to vote for a candidate who opposed the law, and 38% said it would have no effect on their vote.


Table 2. Position on Health Care by Generic Ballot



Source: Gallup Poll, March 26-28, 2010


These results indicate that only a small minority of voters are cross-pressured on the issue of health care reform and that the numbers of cross-pressured Democrats and Republicans are about equal. Moreover, among undecided voters, there is a fairly even split between those saying they would be more likely to vote for a candidate who supports the law and those saying they would be more likely to vote for a candidate who opposes the law. Based on these results, there appears to be little potential for this issue to produce a shift in voter preferences. The main effect of health care as an issue would probably be to reinforce voters’ partisan preferences.

There is one other way that a debate about health care reform during the campaign could affect the congressional elections, however. It could increase interest in the election and therefore voter turnout. So far this year most polls have found Republicans to be more enthusiastic about voting than Democrats and that was also the case in the Gallup survey. 71% of those supporting a generic Republican indicated that they were more enthusiastic than usual about voting this year compared with only 56% of those supporting a generic Democrat.

Among both Democrats and Republicans, those who agreed with their party on health care were more enthusiastic about voting than those who disagreed with their party but Republicans who strongly opposed reform were a good deal more enthusiastic (78% were more enthusiastic than usual) than Democrats who strongly favored reform (63% were more enthusiastic than usual).
It is possible that an intense debate over health care reform in the months between now and Election Day could increase enthusiasm for voting among those with strong preferences on this issue but there appears to be more room for increased enthusiasm among pro-reform Democrats than among anti-reform Republicans.