Are American Voters Racist?
Friday, May 17, 2013
Racial prejudice plays a very small role in American politics, but a highly disciplined sense of self-interest on the part of one group may play a very large role in the way the federal government functions.
A question that has frequently been raised in presidential election campaigns is: are American voters racist? The way this question is framed almost always centers on a concern that some Americans might be unwilling to vote for a black candidate to be their president. Prior to the last two campaigns, the question was strictly hypothetical and many observers were convinced that the answer was affirmative, that Americans simply would not vote for such a candidate.
The last two elections have demonstrated clearly that Americans are not prejudiced in this way — a very useful event, since the charge can never be taken seriously again.
But what about another interpretation of this question: do black voters vote for black candidates on a racial basis? To answer this question requires a study of long-term voting patterns, something that is not only interesting but timely, given the recent census report indicating that in 2012, the percentage of blacks who voted surpassed that of whites for the first time, a significant milestone in U.S. electoral history.
The easiest and broadest database from which to see Americans’ voting patterns is the well-documented voting in presidential elections. For this discussion, ten elections were studied, those of 1976 through 2012, using demographic data available on the Roper Center website.
Table 1 below shows voting percentages for three groups: whites, blacks, and Hispanics. Asians and others are not included in the data because the results for these groups have been inconsistently reported during the elections sampled.
In each case, the data shown is the percent differential between votes cast for Democrats and votes cast for Republicans. A positive number represents more votes for Democrats; a negative number denotes more votes for Republicans. Third-party candidate percentages are not included.
There is no discernible difference in voting patterns across any of the three groups of voters based on the race of the candidates.
Whites do show a tendency to strongly support Republicans. In none of the elections analyzed did a Democrat receive a majority of the votes cast by whites, even in the two elections where whites cast large numbers of votes for a third-party candidate who was widely regarded to be more fiscally conservative than the Republican candidate. The white voter percent differentials in the two elections involving a black candidate are indistinguishable from the eight preceding elections.
Blacks show a strong tendency to support Democrats. In none of the elections analyzed did the Republican receive a majority of the votes cast by blacks. In the two elections where a black candidate was running, blacks provided slightly higher vote differentials (the average differential during elections from 1976 through 2004 was 75 percent, with a standard deviation of 5 percent). While the differentials in the two most recent elections are higher than any previous elections, the basic voting pattern is similar.
Hispanics show the most variation in voting differentials. The average positive differential in favor of Democrats is 36 percent. But the standard deviation is also high, at 15 percent, and George W. Bush received only a 9 percent unfavorable differential from Hispanics in 2004.
Percentages tell only part of the story in any political contest, however. The actual number of votes is the most important statistic to any candidate for office. The following table converts the percentage numbers provided in the first table into millions of votes. Once again, positive numbers favor Democrats and negative numbers denote more Republican votes. All vote differentials are shown in millions of votes, with one decimal place.
While the majority of white votes in favor of Republicans has been relatively stable in recent years, after reaching a high point in the 1984 election that has never been approached again, the result is different for blacks: the majorities provided to Democrats by blacks have risen substantially over the course of the ten elections. Furthermore, in four out of six elections where Democratic presidential candidates received more popular votes, the vote majority provided to the Democrat by blacks exceeded the total votes by which the Democrats won the popular vote. Only in 1992 and 1996 did the Democrat receive overall majorities from non-black voters, and in each of these cases, the results are skewed by the large number of third-party votes cast by white voters.
Hispanics display a third pattern. While the percentage of Hispanic votes is always strongly Democratic in each of the ten elections, the actual differential in votes provided has been small in relation to the majorities provided by blacks. And the strong performance by George W. Bush in 2004 indicates that Hispanics are less certain to vote Democratic than are blacks.
Interpreting Voting Patterns
This numerical analysis cannot determine why blacks so strongly favor Democrats or why whites consistently favor Republicans. Raw numbers do not speak to motivations. However, some general observations can be made about the role of economic differences in voting patterns.
Democrats are generally regarded as more in favor than Republicans of programs that provide government support to lower-income groups. The percentage of blacks with lower incomes tends to be higher than for whites. So from an economic perspective, it would make sense for blacks to consistently vote Democratic in large numbers, as they do.
In none of the elections analyzed did a Republican receive a majority of the votes cast by blacks.
Democrats typically favor higher taxes than do Republicans. To the extent that whites tend to have higher incomes, they tend to pay higher taxes. So from an economic perspective, it would make sense for whites to consistently vote Republican in large numbers, as they do — although to a lesser degree than blacks favor Democrats.
New Hispanic immigrants tend to have income levels close to those of blacks, so it would make economic sense for these Hispanics to vote Democratic. Some research indicates that after several generations, many Hispanics experience income levels more similar to whites. Based solely on this factor, the Hispanic vote in the future may shift from a strongly Democratic vote differential to a smaller vote differential, if multiple generation Hispanics grow in numbers relative to first generation Hispanics.
The consistently strong support of blacks for Democratic presidential candidates is well-known to political campaign managers. The inconsistency of Hispanic voting patterns and the somewhat lower correlation between the economic interests and voting patterns of white voters send a less powerful message to campaign managers and to politicians seeking re-election.
All in all, the data does not support the contention that racial prejudice plays a large role in American politics. But a consistent and large difference in the voting pattern of blacks, whites, and Hispanics may play a meaningful role in influencing the policies of politicians favored by these groups.
Kenneth Gould is a retired banker, living in Lake Bluff, IL.
FURTHER READING: Gould also writes “Sound Money vs. Stable Money” and “The High Cost of College: An Economic Explanation.” Jonah Goldberg contributes “The GOP and Racism, Yet Again” and “2 Cheers for Rebranding.” James Pethokoukis explains “For GOP, Immigration Reform and Sweet Talk Won’t Be Enough.”
Image by Dianna Ingram / Bergman Group