When President Bush's immigration reform bill collapsed this summer, largely because of objections from his own party, open-borders advocates warned that the GOP would pay a harsh political price for killing the bill. Latino support had been crucial in electing Bush, the argument went, and Latino voters represented a rising electoral tide that Republicans were ignoring at their peril.
But such commentary is based on an inaccurate picture of the Latino voting public that emerged after the 2004 election and persists today. Just days after the election, for instance, Dick Morris, a former pollster and advisor to President Clinton, declared that Latinos had elected Bush; they represented 12% of the electorate, Morris reasoned, and 45% of them had pulled the levers for the president, enough to be decisive.
The Latino vote for Bush was far from decisive, however, and it may be years before it plays a pivotal role in a national election. Latinos may represent about 14% of the U.S. population, but they constituted just 6% of the 2004 electorate -- 7.5 million voters out of 125 million. According to Census Bureau data, only 34% of the nation's adult Latino population registered to vote in 2004, and 28% voted. By contrast, 67% of the country's adult white, non-Latino population and 56% of its adult black population voted in 2004. Black voters outnumbered Latino voters nearly 2 to 1 in 2004.
Exit polls taken during 2004 also indicate Latino support for Bush may have been exaggerated. In different polls, Bush's share of the Latino vote ranged from a high of 44% to a low of 33%. Yet subsequent academic studies have estimated Bush's actual level of Latino support at the lower end, somewhere between 35% and 37%. Seen in this context, the "swing" of voters from Bob Dole, who garnered 21% of the Latino vote in 1996, to George W. Bush was hardly historic. In 1984, Ronald Reagan captured 37% of the Latino vote -- a performance at least equal to Bush's.
This suggests that the key to winning Latino votes may be running good candidates, not pandering. Latino voters themselves seem to agree. A 2004 Washington Post poll found that immigration was the least important issue among Latino voters, with only 3.5% placing it at the top of their concerns.
The decline in Latino support was not a unique phenomenon for Republicans; from 2004 to 2006, the GOP lost support among virtually all constituencies, including union members (down 10 percentage points) and even white evangelicals (down 8 percentage points). In many places, the falloff was larger among core Republican voters than among Latinos. In California, for instance, 2006 GOP Senate candidate Richard Mountjoy's share of the Latino vote was 10 percentage points below Bush's 2004 share -- while his share of white male voters was a whopping 12 percentage points below Bush's comparable showing two years earlier.
Many commentators make another mistake by assuming that a liberal immigration policy will attract voters for a candidate but not cost him any votes. In fact, polls suggest that Bush's immigration policy might be more of a detriment at the ballot box in some places. In Arizona, a state with a large Latino population that some commentators have suggested could wind up in the Democratic column in future elections, 78% of the electorate told pollsters in 2004 that immigration should either be maintained at current levels or decreased.
One can only imagine how those voters would have reacted in the 2004 election if the president had introduced an immigration reform bill that offered a path to citizenship for about 12 million illegals along with vastly expanded legal immigration slots.
Given what the voting numbers show us, it's unlikely that Latinos will become an important voting bloc in most places as soon as many predict. And by the time that they do, Latino citizens might find that an immigration policy based on enforcing borders and increasing the number of better-skilled immigrants, which many Republicans advocate, actually benefits them. Recent economic studies show that the country's current levels of immigration are hurting immigrants who are already here -- and hurting native-born Latinos more than most U.S. residents. A saner immigration flow is likely to boost the average wages of our current Latino population and free up resources, like housing, in Latino communities.
But much of the commentary on Latino voting power tends to ignore such issues, focusing instead on Latino voters' supposed anger at Republicans and comparing it to black voters' desertion of the party after key Republicans opposed civil rights legislation in the 1960s. But the analogy hardly stands up. American blacks were striving to obtain rights guaranteed in the Constitution but denied to them. By contrast, the current immigration debate is not about denying immigrants anything; it's about dealing with those now here illegally and those yet to come.
Steven Malanga is the coauthor of the forthcoming "The Immigration Solution" and senior editor of City Journal, from whose forthcoming issue this is adapted.