The day after the 2004 presidential election, columnist Dick Morris proclaimed that George W. Bush owed his victory to Hispanics—a startling conclusion that Morris based on reports that 45 percent of Hispanics had voted for the president and that their vote constituted a whopping 12 percent of the electorate. Given that 1996 Republican nominee Bob Dole had won just 21 percent of the Hispanic vote, the more than doubling of support over eight years was sufficient, in Morris’s eyes, to appear decisive.

The only problem with this thesis was that few of its facts were correct. Exit polls, recall, were wildly inaccurate that year. Subsequent academic analysis of the polls concluded that only about 37 percent of Hispanics had pulled the lever for Bush—about the same percentage as had voted for Ronald Reagan in 1984, and hardly a historic shift. Moreover, a Census Bureau analysis estimated that Hispanics accounted for 6 to 8 percent of the 2004 electorate, not 12 percent. Nevertheless, the idea that Hispanics had given unprecedented support to a Republican presidential candidate became a compelling media story line, especially during the immigration debates of 2007, when commentators wondered whether a Hispanic backlash against the GOP’s tough position on illegal immigration would sink Republicans at the polls.

Then, in the wake of Barack Obama’s 2008 election victory, commentators dredged up the old, discredited 2004 numbers and used them as evidence that Hispanics had shifted impressively toward the Democratic Party and had even been instrumental in the outcome. USA Today argued that Obama had “reversed” gains made among Hispanics by Bush, who, the paper erroneously said, had carried 44 percent of the Hispanic vote in 2004. The paper also noted that Hispanics were the country’s fastest-growing ethnic group—a misleading statistic in this context, because Hispanics don’t vote at anywhere near the rate that their numbers would suggest. But USA Today wasn’t alone. The New York Times claimed that Obama had garnered 13 percentage points more of the Hispanic vote than John Kerry did four years ago (he did only a few points better) and implied, again, that Bush had won somewhere in the mid–40 percent range of the Hispanic vote.

The numbers actually show something quite different: Hispanics shifted away from Republican candidate John McCain at virtually the same rate that the entire electorate did. McCain earned 5 percentage points fewer of the total vote than Bush did in 2004, and similarly got 4 to 5 percentage points fewer than Bush’s share of Hispanic voters. This was in line with how other constituencies voted, too: Obama peeled off several percentage points of support from McCain among vastly dissimilar voting blocs, from Christian evangelicals to women voters, from those making $200,000 or more in annual income to voters lacking a high school education. None of this is surprising. According to a Pew Hispanic Center poll, the issues that mattered most to Hispanic voters were education, the cost of living, jobs, and health care—at the top of the list for almost every voting group.

Pew also found no sharp increase in Hispanics’ voting rates in 2008, estimating that they made up about 8 percent of the electorate, roughly the same percentage as in 2006 and just slightly more than in 2004. What some pundits have referred to as the Latino voting tsunami is more like ripples in a stream.

What the 2008 election really demonstrates is that Hispanic voting patterns remain pretty much where they’ve been for the past 30 years. The most successful Republican candidates—Reagan in the 1980s and Bush in 2004—garner somewhere in the high 30 percent range, and the Hispanic vote deviates from that figure depending on the quality of the Republican candidate.

Why has the GOP been unable to make more headway among Hispanics? One answer has to do with income. As political scientist Andrew Gelman notes in his new book, Red State, Blue State, Rich State, Poor State, lower-income voters continue to vote disproportionately Democratic, despite a popular notion among pundits that many of them have shifted to the GOP for cultural reasons. That fact suggests that Hispanics—nearly half of whom live in households whose earnings fall in the country’s bottom two income quintiles—would naturally trend Democratic. And in fact, in the McCain-Obama contest, 83 percent of Hispanic voters with annual incomes of $15,000 or less voted for Obama, as did 71 percent of those earning between $15,000 and $30,000. By contrast, 51 percent of those with household incomes between $150,000 and $200,000 voted for McCain.

McCain hardly ran a campaign that appealed to voters worried about the country’s economic woes. Those voters, especially in lower-income households, went strongly for Obama. That, more than anything else, explains much of the “shift” in Hispanic voting last November.


City Journal is a publication of the Manhattan Institute for Policy Research (MI), a leading free-market think tank. Are you interested in supporting the magazine? As a 501(c)(3) nonprofit, donations in support of MI and City Journal are fully tax-deductible as provided by law (EIN #13-2912529).

Further Reading

Up Next