IDEOLOGY, PARTISANSHIP, AND COGNITIVE SHORTCUTS
To move beyond such simplistic candidate selection methods, some American voters turn to ideology. A small portion of the electorate evaluates candidates through indirect sources of information about their policy views. For example, a sophisticated voter might know that the Center for Tax Responsibility shares his or her views on public spending. The voter can then evaluate an incumbent based upon the rating that the center gives to that representative's voting record. Though the voter still does not know the incumbent's voting record, the voter may infer with confldence that a high rating of the incumbent by the center signals that the representative acted in sync with the voter's own policy preferences. Such inferences are quite logical, and an ideologically self-aware voter familiar with a political organization's rating system has as good a chance as any of making sensible candidate evaluations. When such voters have adequate information and a viable opponent to consider, their behavior comes close to the ideal presented in chapter 2. If a representative's constituents have a clear ideological bent and obtain solid information about all candidates, these voters have the potential for successful electoral rejection of incumbents who stray too far from their ideologically deflned interests. That threat, in turn, might win such a constituency sound representation in government.
Such a combination of factors is rare, however. Even when there are alternative candidates to consider, limited political sophistication, low
Samuel Popkin, a polling consultant and public opinion scholar, has examined these short cuts carefully in relation to presidential campaigns. One of the most commonly used metaphors for such thinking is the Drunkard's Search: when a drunken man stumbles outside of a bar at midnight and drops his keys, he drags himself over to the streetlight to look for them; the keys probably fell elsewhere, but he can only see the pavement that's under the light. Thus, rather than trying to piece together all the information they have obtained, voters make judgments based on the one or two pieces of information in which they have some confldence. Popkin stresses the importance of personal information in a presidential race, but the metaphor works equally well for other salient facts that a voter might glean from an election—or even from the ballot itself.[26]
Many political observers argue that one particular fact—a candidate's party—is enough information to make an informed choice about which candidate will best represent a voter's interests. Though relatively few voters have strong ideological orientations, as many as three-quarters of the electorate identify with either the Democratic or the Republican Party. For example, in the 1998 House elections, 73 percent of voters described themselves as either Democrats or Republicans. Eightynine percent of Democratic voters backed Democrats, and 91 percent of Republican voters backed Republicans. Nonaligned voters, by contrast, split their vote almost evenly between the two parties. Statistics for House elections from 1980 to 1996 show a consistent pattern of from 77 to 92 percent of partisans voting for their party's candidate.[27] Numerous other studies have found that political partisanship is the most powerful predictor of voting choices in a wide range of partisan general elections in the United States. For example, voter partisanship predicts roughly 50 percent of all voting in the presidential election.[28] Highly sophisticated partisans sometimes even use party-identiflcation cues in nonpartisan elections by reading party "voting guides" or "candidate slates."
Nonetheless, both the prevalence and value of partisan voting cues are overestimated. After having chosen to vote Republican, for example, voters are more likely to identify themselves as Republican in response
In any case, there is considerable evidence that the power of the partisan cue is declining. After studying the past forty years of party politics in the United States, Martin Wattenberg concluded, "Once central and guiding forces in American electoral behavior, the parties are currently perceived with almost complete indifference by a large proportion of the population." Consistent with this view, surveys conducted in 1994, 1996, and 1997 have found that between two-thirds and threequarters of Americans report that they regularly vote for candidates of parties other than their own in at least one race per election.[30]
Table 1 shows, however, that over the past quarter-century, strong partisans of the two major parties have never made up more than a third of the voting-age population. The table also shows that the percentage of partisans in the general population has changed little during that same period. Some observers have argued that despite the scarcity of self-identifled party loyalists, independents are actually far more rare. In this view, "pure independents" make up only 10–15 percent of the electorate, because most independents actually "lean" toward one of the two major parties. Self-reported voting data from 1962 to 1990 show that these "leaners" were no less partisan than "weak" Democrats and Republicans. When viewed from another standpoint, however, it is just as noteworthy that over one-quarter of "weak" partisans vote against their party in U.S. House elections. Whether this makes the weak partisans independent or the leaning independents partisan is a deflnitional question. The fact remains that only a fraction of the electorate consistently vote for just one party.[31]
One example of the decline of political party identiflcation is the case of Maine, which some political observers view as a bellwether state. In the 1998 election, the Democratic Party campaigned vigorously in the gubernatorial election, but not with an eye toward recapturing the governor's seat. Late polls showed the Democratic candidate, Tom Connolly, with just 6 percent of the vote, well behind the Republican candidate and the independent incumbent, Governor Angus King Jr. As the
| Strong Democrat (%) | Strong Republican (%) | Total Strong Partisans (%) | |
|---|---|---|---|
| SOURCE: U.S. Bureau of the Census, Statistical Abstract of the United States, 1997, table 461, based on surveys conducted by the Center for Political Studies, University of Michigan, Ann Arbor. | |||
| 1972 | 15 | 10 | 25 |
| 1980 | 18 | 9 | 27 |
| 1984 | 17 | 12 | 29 |
| 1986 | 18 | 11 | 29 |
| 1988 | 18 | 14 | 32 |
| 1990 | 20 | 10 | 30 |
| 1992 | 18 | 11 | 29 |
| 1994 | 15 | 16 | 31 |
Maine's successful independent governor has inspired candidates and voters in other states to abandon the major parties. The most surprising example of this influence was the successful 1998 campaign of the wrestler-turned-governor Jesse "The Body" Ventura in the historically liberal state of Minnesota. In a three-way open-seat race against nondescript Republican and Democratic challengers, Ventura persuaded just over a third of Minnesotans to support his Reform Party candidacy. The Minnesota electorate that supported Ventura did not do so because of his political party per se. Rather, voters largely ignored his Reform Party affiliation and embraced him because his personal history, campaign style, and rhetoric were so different from those of traditional Democrats and Republicans.[33]
Not only is support for the two major parties declining, but partisanship is also most common among those voters who are the most knowledgeable and ideological. Strong Democrats are the most liberal voters, while strong Republicans are the most conservative, and this relationship is even stronger for those voters with the most political
Making matters worse, those less sophisticated voters who rely upon partisan cues are more likely to have incidental party affiliations. For most Americans, political party membership is not a "fundamental belief … but rather an inherited trait."[36] When a person votes for Democrats because his or her parents voted for Democrats, that voting pattern represents the person's actual concerns only to the extent that child and parent share similar values, circumstances, and knowledge. Historical party realignments, such as the African-American shift toward the Democratic Party during the Civil Rights era, show that people can change party membership in response to changing party platforms or shifts in voters' own attitudes. Nonetheless, party identity remains stable for most Americans regardless of changing personal and political circumstances.[37] Party membership influences other attitudes and candidate evaluations far more often than these beliefs influence membership.[38]
Even if a large proportion of the electorate do rely on partisan cues when making voting choices, this dependence is often counterproductive. Recalling the general model of representative democracy in chapter 2, one of the primary purposes of elections is to give an electorate the opportunity to reject an incumbent who has failed to represent its interests. When voters base reelection decisions entirely on the candidates' party membership, they necessarily overlook other information about the incumbent and challenger. Though publicly available, an official's voting record often goes unnoticed. Thus, a congressional district populated by liberal Democrats might consistently reelect a conservative Democrat over a liberal Republican because the former shares the majority's party identity. More generally, voters' reliance on partisan cues permits representatives to act against their constituents' interests. If a single representative consistently deviates from the party line even on roll-call votes, the party may sabotage that particular incumbent's reelection bid, but if the party as a whole violates its membership's
The use of "open primaries" worsens the situation by permitting voters to cross party lines in the primary election. In states such as California, primary voters select one candidate from among the entire pool of candidates in each race, and candidates advance to the general election if they win more votes than any other in the same party. Voters can support Democrats in some primary races and Republicans in others, regardless of their own party affiliation. Californians embraced this electoral reform as offering "more choice" in the primaries, but the system can easily be abused to select general election candidates out of sync with their own party. The candidacy of an unopposed Democratic incumbent might, for example, encourage Democratic voters to support the most incompetent or unpopular Republican in the primary. With Republicans divided among a group of candidates, this Democratic voting bloc can thus bring victory in the primary to the worst Republican choice. Alternatively, Democratic activists in a strong Republican district can use this system to support a liberal Republican in the primary, leaving Republican voters with nothing but liberals in the general election.[40] When voters use a candidate's party as a shortcut to evaluation, they risk serious misjudgment, and the open primary makes the risk even greater.
A more serious problem with partisan cueing lies in the limited range of parties Americans can choose from today. In most races in the United States, voters must choose either a Democrat or a Republican. In a satire of this situation, the animated television program The Simpsons imagined what would happen if the Democratic and Republican candidates in the 1996 presidential election were actually Kang and Kodos, tentacled space reptiles seeking to conquer Earth. On the eve of election day, Homer Simpson unmasks the aliens before a large crowd assembled on the steps of the nation's capitol. Unperturbed, one of the aliens explains, "It makes no difference which one of us you vote for. Either way, your planet is doomed. Doomed!" In reply to this boast, a bold spectator declares, "Well, I believe I'll vote for a third-party candidate." Kang and Kodos writhe and laugh maniacally. "Go ahead," they retort.
"Throw your vote away!" In the closing scene, legions of human slaves toil to build a giant ray gun, while an alien administrator bellows, "All hail President Kang!"
Though the Democratic and Republican parties often see eye to eye on fundamental economic class issues, they differ in many important respects. Party allegiance is a meaningful choice between the ideological left-to-center and center-to-right, and the party unity displayed in rollcall votes in Congress has increased in recent decades.[41] If party identity is becoming more ideologically grounded, as some election scholars claim, this trend would result in more meaningful partisan voting.[42]
Nonetheless, reliance upon just two parties makes a citizen's choices extremely narrow. Beyond the factions within parties, there are many ideological identities that cross party lines. The Libertarian Party, the Reform Party, and other independent and multifaceted parties provide other choices, but the reality of winner-take-all elections is that those parties have few opportunities to elect their candidates to public office. Voters can register with and vote for a third party to play the role of spoiler, as the Green Party has done in New Mexico, but taken as a whole, third parties play only a marginal role in U.S. politics.[43] In sum, because two choices obscure the real diversity of meaningful viewpoints among which one might otherwise choose, parties only crudely represent Americans' concerns.
The ultimate failing of partisan cue reliance, however, is an inescapable feature of conventional American elections. Simply put, partisan cues are useless in nonpartisan elections and party primaries. When choosing among finalists in a nonpartisan general election, less sophisticated voters usually remain unaware of the candidates' party loyalties, or the candidates may belong to the same party. Innumerable political observers have declared the party voting cue to be the connection between public preferences and political representation, yet no such connection exists in numerous local and statewide nonpartisan races for judgeships, executive offices, councils, and boards. Though nonpartisan systems are often designed to make local government "less political" and avoid graft and corruption, the net result is the removal of the one cue that most voters rely on to distinguish among competing candidates. For this very reason, many political observers oppose the very concept of nonpartisan elections, let alone changing some local elections from partisan to nonpartisan.[44]
For the same reason, it turns out that every election suffers from dependence on partisan cues. When two Democrats face off against each
One might counter that political parties manage their own primaries and indirectly ensure that the most representative party member goes on to the general election. The problem with this argument is that it is false. That may have been the case at one time, but incumbents are now stronger than the local parties. Renomination and reelection depend upon the decisions of the incumbent, over whom the party has precious little control.[46]"Few congressional candidates flnd opposition from the local party leaders to be a signiflcant handicap; neither is their support very helpful," Gary Jacobson observes.[47] In other words, party primaries are often critical decisions for voters, yet overreliance on partisan cues makes these choices extremely difficult.
In sum, relatively few voters use their ideology as a guide when making voting decisions. Many more voters use partisan cues as a guide to voting. By doing so, they hope to overcome the limitations of their own underdeveloped interests. If a voter decides that the Democratic or Republican Party adequately represents his or her interests, candidate party membership can then be used to guide otherwise complex voting decisions. The use of these cues may be in decline, but partisan voting has limited value, in any case. Strong partisan loyalty can actually obscure poor representation by diverting attention from the actual views of candidates and the performance of incumbents. The limited twoparty system in the United States also makes it difficult to represent diverse public viewpoints in partisan voting. When the diversity of viewpoints on current issues is condensed into a binary choice, many ideas and concerns get lost in the process. Finally, partisan cues are useless in nonpartisan and primary elections. Since election to nearly every office requires winning either a party primary or a nonpartisan election, reliance on party membership as a voting guide is inadequate.