At first glance, the case against Republican support for immigration restrictions on purely political grounds seems compelling, and a comprehensive report in the New York Times soon after the 1996 election seemed to buttress it.¹ In California, the Hispanic population numbered about 30 percent of the state in 1996, and in Orange County, long a Republican stronghold, it represented about 25 percent. Hispanic voters in California constituted “nearly 15 percent” of the state’s voting population in 1996, up by some 40 percent since 1992, while in Texas Hispanics were about 12 percent of the electorate, and 10 percent in Arizona.

Nationally, the Hispanic portion of the voting population was about 5 percent of the total electorate. In 1996, actual Hispanic turnout was considerably larger than it had been in pervious elections (in which Hispanics were about 80 percent as likely to register to vote as other Americans). In Florida, the Hispanic part of the vote increased 10 percent over 1992, while in Texas it increased nearly 60 percent, and in California 40 percent. In the 1996 election, President Bill Clinton won 71 percent of the Hispanic vote in California, 15 points ahead of his showing in 1992, and defeated Dole in the state by 51 percent to 38 percent. “The Republicans once could compete for our vote,” Antonio Gonzalez of the Southwest Voter Research Institute in Texas told the Times, “but now it’s clear that they’re facing a powerful and united bloc of opposition from us.”²

The California statistics were especially worrisome to Republicans, since the party cannot hope to win presidential elections without that state — until 2000, it had gone Republican in every election Republicans have won since 1948 — and they cannot usually expect to win California without carrying Orange County by at least a two-thirds majority. Orange County itself saw Hispanic voter registration increase by a massive percentage between 1992 and 1996, and the county’s Hispanic voters supported the Democrats by more than 3 to 1 in the latter year.³ By the year 2025, the Hispanic part of the state’s population will be 43 percent of the total, and 30 percent of the voting-age population was projected to be Hispanic by the year 2000.

But other states with sizable Hispanic populations showed similar trends. In Arizona, which prior to 1996 had voted Republican in every presidential election year since 1948. Hispanics helped Clinton win in 1996 by three percentage points. In Florida, where most Hispanics are anti-communist Cubans who have traditionally supported the Republican presidential ticket, Clinton in 1996 won the state by 48 percent to Dole’s 42 percent, assisted by what the Times called “a heavy Hispanic vote — helped by a heavy switch to the Democratic column by usually conservative Cuban-Americans.”⁴

The Hispanic support for the Democratic ticket in 1996 was due to several factors, according to the Times, among them “a sharp increase in the number of Hispanic immigrants becoming citizens and a push by the Democrats to get them registered and get them to the polls” (a push that may have involved the deliberate and illegal naturalization of immigrants by the Clinton White House to enhance its Hispanic support⁵), but it also supposedly was due to Hispanic fear of Republican support for immigration restrictions. As the Times reported,

The big presence also resulted, the experts say, from deep concern among the Hispanic population that Republicans are intent on cutting Hispanic immigration and reducing civil rights protections and welfare benefits for immigrants already in the country, including those who arrived legally. That concern ran especially deep in California, where a ballot initiative, Proposition 187, proposed immigration strictures so severe that it drew attention well beyond California’s borders.⁶

Yet despite the statistics and the apparent plausibility of the argument that Republican support for Prop 187 and similar immigration restriction measures alienated the Hispanic vote, a deeper examination of election figures challenges this interpretation. In the first place, Hispanic voters have traditionally been more Democratic than Republican, win the exception of Cuban emigres in Florida. It was not Prop 187 or recent Republican support for immigration restriction that turned Hispanics into Democrats.

From 1972 through 1988, Republican presidential candidates won an average of less than 32 percent of the Hispanic vote, while their Democratic rivals won an average of nearly 66 percent — more than twice the Republican share. Even Bill Clinton’s 72-percent share of the national Hispanic vote in 1996 was smaller than Jimmy Carter’s 76 percent in 1976; Gerald Ford in 1976 won only 24 percent of the Hispanic vote; and in 1992 — well before Proposition 187 — President George H. W. Bush won only 25 percent nationally.⁷ Strong Republican candidates like Richard Nixon and Ronald Reagan could win 30 to 35 percent of the Hispanic vote nationally, but weaker candidates such as Gerald Ford in 1976 and George H.W. Bush in 1992 were able to win only smaller shares — well before Proposition 187. In short, there was no massive swing of Hispanics from Republicans to Democrats in the wake of Prop 187; Hispanics have largely always voted Democratic, as their congressional representative shows.

Secondly, it remains to be proved that increased Hispanic support for the Democrats in 1996 was due to Hispanic support for immigration and fear of Republican opposition to it. In 1996, there was, in fact, little concerted opposition to immigration, and only in the last days of the campaign did the Dole-Kemp ticket campaign against illegal immigration in California. Although the GOP platform of that year contained language calling for immigration restrictions, the first action Bob Dole took after securing the GOP nomination in 1996 was to announce his rejection of the platform; his running mate, Jack Kemp, was well known for his long support for large-scale immigration and his opposition to Prop 187 in 1994. While Dole had earlier expressed support for limiting legal immigration, he had little record of support for such restriction.

Nor did there seem to be any close correlation between support for immigration restrictions and electoral defeat in 1996 among Republican congressional candidates. Rep. Bob Dornan lost his re-election bid in 1996 in part because of the large Hispanic vote in his district (and perhaps because of voter fraud by illegal immigrants), but not because he opposed immigration. Indeed, Dornan had been a vocal supporter of legal immigration, regardless of the political consequences. As for the strong immigration restrictionists among the Republican congressman, neither California’s Elton Gallegly nor Texas’ Lamar Smith seemed to have any problem winning re-election, while the strong support for immigration that then-Texas Governor George W. Bush expressed did not deter two-thirds of the state’s Hispanics from registering Democratic and a majority of them voting for his Democratic opponent.⁸

Moreover, it was probable that Hispanic voters, many of whom tend to be low-income, supported the Democrats in 1996 not because of supposed Republican opposition to immigration but because of Republican support for welfare reform, reducing eligibility for food stamps, and similar budget-cutting policies that threatened to reduce or limit benefits and government protections for any low-income group. Nor was it by any means a valid assumption that all (or even most) Hispanics are in favor of more immigration or that their position on immigration determines how they vote. Proposition 187 won the support of some 37 percent of Hispanics in California, and in 1992 the Latino National Political Survey found that more Mexican-Americans and Puerto Ricans (75 and 79 percent, respectively) favored reducing immigration than did non-Hispanic whites (74 percent). In 1996, a Roper Poll showed that some 83 percent of Americans favored reducing immigration, with 52 percent of Hispanics expressing support for limiting legal immigration to 300,000 a year (as opposed to what was then 1 million legal immigrants per year).⁹ Immigration authority Roy Beck reported in the same year,

The Latino National Political Survey by Rodolfo de la Garza of the University of Texas discovered that 75 percent of Mexican-American citizens, for example, said there are too many immigrants. That compared to 74 percent of non-Hispanic white American citizens who said so.¹⁰

There is no reason why Republicans supporting immigration restrictions cannot effectively appeal to many (perhaps a majority of) Hispanics and win their support, despite (or perhaps even because of) the Republican position on immigration. Nor is there any decisive reason to think that, even among pro-immigration Hispanics, a candidate’s position on immigration is what determines how the Hispanic voter will cast his ballot. As Cuban-born Republican consultant Alex Castellanos told the New York Times in 1997, speaking of the Hispanic electorate

I do think it’s a constituency that we should have a lot of affinity with . . . . It’s an absolutist culture, very black and white, with strong moral values. I think that’s Republican culture. It’s a strong Catholic culture, one of the few groups where the abortion issue is not fuzzy . . . . It is also, and this is something that’s hard to talk about, but it’s more of a daddy-bear, father-figure culture, and Republicans have been very successful with crime and other patriarchal issues.¹¹

Appeals to conservative values involving family issues, crime, and the economy ought to attract low-income as well as middle-class Hispanics at least as much as supporting increased immigration.

Although supporters of immigration argue that opposing in alienates Hispanics, there is little evidence for this. Indeed, the evidence tends to support the arguments of those who oppose mass immigration for a variety of reasons and who argue that Hispanics, as a typically low-income, low-skill, non-white minority group, will naturally tend to vote Democratic. Thus, immigration critic Peter Brimelow writes in his 1995 Alien Nation, “The post-1965 immigrants are overwhelmingly visible minorities. And these are precisely the groups that the Republican party has had the most difficulty recruiting . . . . The numbers are indisputable: Current immigration policy is inexorably reinforcing Jesse Jackson’s Rainbow Coalition.”¹²

Two years after his book appeared, Brimelow (with co-author Edwin S. Rubenstein) published an article in National Review (of which each was an editor) arguing that the Hispanic immigration tide was permanently reducing the Republican Party to a minority status. Their argument was based on the premise that “Nine-tenths of the immigrant influx is from groups with significant — sometimes overwhelming — Democratic propensities. After thirty years, their numbers are reaching critical mass. And there is no end in sight.”¹³

Considering the Republicans’ 53 percent of the vote in the 1988 presidential race, Brimelow and Rubenstein argued that this was “the Republican high-water mark” — comparable to the average Republican national vote in every election since 1968, and the same percentage that the Republicans received in the congressional elections of 1994. If the Republicans could maintain their 1988 level of support among each ethnic group, they would find that winning percentage successively shrinking them into defeat in every future election after 2008. Assuming each ethnic group continues to vote as it did in 1988, the Republicans would receive less and less of the increasing numbers of immigrant (Asian and Hispanic) voters.

In 2000, Brimelow and Rubenstein projected, the Republicans would win only 50.7 percent of the national vote (in fact, George W. Bush won 47.9 percent); in 2004, 50.3 percent; in 2008, 49.9 percent; and so on until 2052, when they would win only 45.3 percent of the national vote. As the authors acknowledged, their projections made several assumptions that might turn out to be false — that the Asian and Hispanic vote increases, that the Republicans do not increase their share of Hispanic votes and they do gain Asian votes, and that the Republicans do not gain more white electoral support — though they defended each such assumption as reasonable. So far, with only one presidential election taking place since they published their projections, the actual results have been worse for the Republicans than the authors foresaw.

1. B. Drummond Ayres, Jr., “The Expanding Hispanic Vote Shakes Republican Strongholds,” New York Times, November 10, 1996.
2. Ibid.
3. Maria Elena Fernandez, “Two Ways of Looking at Latino Vote,” Los Angeles Times, December 28, 1999; see also Ayres, New York Times, and Todd S. Purdum, “California G.O.P. faces a crisis as Hispanic voters turn away,” New York Times, December 9, 1997.
4. Ayres, New York Times.
5. Ibid.; for citizenship fraud, see William Branigan, “INS accused of giving in to politics,” Washington Post, March 4, 1997; Ruth Larson, “INS faces Hill grilling over scandals, snafus,” Washington Times, March 4, 1997.
6. Ayres, New York Times.
7. All exit poll data from the 1992, 1996, and 2000 presidential elections cited in this monograph are taken from Voter News Service exit polling data as reported in the New York Times on November 10, 1996, p. 28, and November 12, 2000, p. 24. Exit poll data for the 1998 congressional elections come from VNS data published in the New York Times, November 9, 1998, p. A20.
8. Michael Barone and Grant Ujifusa, Almanac of American Politics, 2000 (Washington: National Journal, 1999), p. 1509.
9. Press Release, Negative Population Growth, February 19, 1996.
10. Roy Beck, The Case Against Immigration: The Moral, Economic, Social, and Environmental Reasons for Reducing U.S. Immigration Back to Traditional Levels (New York: W.W. Norton, 1996), pp. 63-64.
11. Purdum, New York Times, December 9, 1997.
12. Peter Brimelow, Alien Nation: Common Sense about America’s Immigration Disaster (New York: Harper Perennial, 1996), p. 196.
13. Peter Brimelow and Edwin S. Rubenstein, “Electing a New People,” National Review, June 16, 1997, p. 32.