The Hispanic Giant -- Numbers, Yes -- Power? Not Yet

By David Simcox
Volume 11, Number 3 (Spring 2001)
Issue theme: "George W. Bush, last Republican president? And does it matter?"

Was the Republican National Committee whistling past the graveyard with its January announcement that a recently completed survey showed the Republican Party has a future with the nation's Hispanics? Party spinmeisters claimed the survey showed that 38 percent of all Hispanics had voted for Bush in November and 50 percent of all respondents said they would vote for him in 2004 'if he showed concern for Hispanics during his first administration.'1 There is no arguing that Bush lost decisively to Gore among Latinos. Only when contrasted to his paltry nine percent of the African-American vote does Bush's 38 percent national showing among Hispanic voters seem almost impressive. The unforgiving facts are that Bush lost the Hispanic vote in all but one heavily Hispanic state - Florida - and the best he could get there was a virtual tie. His margin of defeat among Hispanic voters in California, the premier Hispanic state where fourteen percent of all voters are Latino, was 68 percent to 29 percent. In New York state, Bush made his worst showing, with 80 percent of Hispanics voting for Gore. In Arizona, New Mexico and New Jersey, he was limited to about a third or less of the Hispanic vote. Bush's performance in Texas was his best outside of Florida There he won 43 percent of Hispanic voters.

Combining these weak showings with the prospect that the number of Hispanic voters in states rich in electoral votes such as California and Texas will swell to near majority status within two decades stokes fear that the GOP may have won its last presidential election. Such a gloomy prospect, however, overlooks some of the imponderables of the demography, ideology, and voting eligibility and voting propensities of the Hispanic population now and in the future.

Latinos Generally, But Not Unalterably, Support the Democrats

Worth noting at the outset is that the Hispanic preference for the Democrats has long been the rule - with occasional exceptions when hugely popular Republicans, such as Ronald Reagan capture their allegiance. Latinos' heavy preference for Democrats surged in the 1960s and 1970s with the rise of the Chicano movement. John Kennedy's defeat of Nixon in 1960 drew in part from successful Democratic mobilization.2 Plainly, preference for the Democrats was a fact well before Proposition 187 and other immigration, affirmative action, and language issues commencing in 1994, rather than a product of them .

Table 2. Hispanic Registration and Voting in Presidential Elections (in millions)

Year 1980 1984 1988 1992 1996 2000

Registered 2.98 3.79 4.57 5.14 6.57 7.6*

Voted 2.45 3.09 3.71 4.24 4.93 7.35**

% of US Vote 2.8% 3.3% 4.0% 4.7% 5.1% 7.0%**

Sources Federal Elections Commission, Census and Velazquez Institute

* Estimate of Velazquez Institute ** Based on Voter News Service Exit Polls Bush's 38 percent in 2000 was a far greater share of Hispanic votes than any Republican presidential candidate has won since Ronald Reagan. He surpassed Republican candidate Bob Dole's Hispanic vote in 1996 by 80 percent. In 1992, before Proposition 187, Bush's father polled only 27 percent among California Hispanics to Bill Clinton's 51 percent. In 1990 in California, Gubernatorial candidate Pete Wilson was considered a virtual victor for attracting 45 percent of the Hispanic vote. Latino loyalty to the Democrats is strong and often trumps even ethnic identity. The increasing number of Hispanic Republican candidates for house seats in Latino majority districts have generally not polled well against Hispanic democrats. In Los Angeles and New York, however, Republican Mayors Richard Riordan and Rudy Giuliani won the Hispanic vote. Latinos' adhesion to the Democrats, though formidable, does not approach the monolithic loyalty of African-Americans to that party.

Stunning Increase in Hispanic Voters in 2000 Where Did It Come From?

Heightening the damage to the GOP in the 2000 elections was the far larger than expected turnout of Hispanic voters. Hispanic think tanks such as the Tomas Rivera Foundation, the Velazquez Institute, and the Southwest Voter Registration Education Project had projected a total Latino turnout of 5.5 million to 5.7 million, based on an estimated total Hispanic registration of 7.6 million.3

But if exit polls are to be believed, Hispanics actually cast some 7.3 million votes - 28 percent more than even the optimists had predicted. Clearly, the Latino advocacies' estimate of 7.6 million registered voters was way too low for 2000. Assuming a) the Hispanic percentage of registered voters actually going to the polls was 80 percent (their historical average for presidential elections since 1984) and b) that Hispanic voters were in fact generally required to establish their registration in order to vote, the number of Hispanic registered voters would be about 9.2 million, not 7.6 million.

The explosive growth in the known Hispanic population makes us wonder what the Census Bureau has been doing and thinking since 1990? Astoundingly, Census' experts released a report in early March 2001 based on the 'Current Population Survey' a year earlier estimating the Hispanic population at 32.5 million, some 2.8 million less than its decennial count would show a few days later.4 And indications are that even the lofty new figure of 35.3 million may be a significant undercount. There has long been an element of denial in Census estimates of net immigration, perhaps in part out of fear of alarming the public.

Now Census has just told us that the Hispanic population has actually grown by 57 percent, from 22.4 million to at least 35.3 million (even then, some undercounting is likely) between 1990 and 2000. Census' ineffectual monitoring of Latinos' and the nation's population growth raise serious credibility 'Persistent high immigration will also preserve many of the social liabilities of Hispanics that work against political participation. The Latino, on average, is younger, poorer, less educated, and more linguistically isolated than his non-Hispanic white and African-American counterparts.'problems about existing projections of growth to 2050 and confounds best efforts to gauge future Hispanic voting strength.

The surge in Hispanic turnout may well have bureaucratic as well as demographic roots. The eight years of the Clinton administration saw several projects intended to open the vote to masses of those loosely tied to the community or those who simply could not handle the paperwork. The Motor-Voter Act of 1993 made registration easy and painless for even the most marginalized, while minimizing documentary proofs of citizenship. Eased registration by mail had similar effects. The Clinton administration's hurry-up naturalization of 1.1 million people in the year before the election brought still more Hispanic immigrants, ready or not, into the electorate.

Strength Commensurate with Numbers? Not Yet

But does the Hispanics' raw population growth translate into corresponding voting power in the nation and among heavily Hispanic states in the near future? The estimated national turnout in 2000 of 7.3 million voters means that only 20.6 percent of the Hispanic population voted. Contrast that with the proportion of those voting from other ethnic groups. African-American voters were 31.5 percent of their population, 50 percent higher than Hispanics. Asians, though nearly two-thirds foreign born, almost matched Hispanics in the portion of their population going to the polls (19.8 percent to 20.6 percent). Non-Hispanic whites voting were 43.4 percent of their total population. In California, non-Hispanic whites who make up a little less than half the population constitute three-quarters of the electorate. It will be years before that share of the vote shrinks below half. The large and recently arrived foreign-born share of their population and the high percentage of population below voting age are two critical demographic factors that are likely to keep Hispanic turnout below that of non-Hispanic whites and blacks for some time. Almost 36 percent of the Hispanic population is below the voting age of 18, compared to only 23.5 percent of non-Hispanic whites.

In 2000, 39 percent of Hispanics were foreign-born. In 1997, only 24 percent of Hispanic immigrants had gained naturalization. Among the 43 percent of all immigrants entering since 1990, only 6.7 percent have been naturalized. The naturalization rate among Mexican immigrants in 1997 was a mere 15 percent, creating a special challenge to voter registration drives in such heavily Mexican states as California and Texas.

The 39 percent foreign-born figure is likely to be revised upward as Census 2000 data now indicates much higher immigration than projected. While time will remedy some of these adverse demographic factors, in the near-term Hispanics will remain underrepresented at the ballot box. A quick reprise of the status of the 35.3 million Hispanic population shows that some 12 million are now underage for voting, and of the 23.3 million remaining some 6.7 million are not citizens. Hispanic America is thus left with a core electorate of only 16.6 million. A refusal to accept this fact may well heat up new pressures to rubber stamp naturalizations, further ease the documentary proof for registration, or change the election rules entirely to allow non-citizens, even illegal aliens, to vote.

Census population projections imply that Hispanics' under-representation at the polls will taper off only slowly. High immigration will ensure a foreign-born component of 30 percent at least until 2020 - probably more when the projections are adjusted to the reality of higher immigration. By mid-century the projected Hispanic population will top 100 'These differences, along with polls showing that Hispanics feel low identification with other Hispanics outside their national origin groups, raise skepticism about the future viability of a ‘Hispanic voting bloc.'' million, but 20 percent of it will be immigrants.5 Persistent high immigration will also preserve many of the social liabilities of Hispanics that work against political participation. The Latino, on average, is younger, poorer, less educated, and more linguistically isolated than his non-Hispanic white and African-American counterparts. As an official of the Southwest Voter Registration Education Project (SVREP) explained it, '[The Latino vote] is not a working middle-class electorate; it's a working poor electorate. And you have to go door to door to reach people.' This sort of campaigning is expensive and difficult, according to another SVREP official, and lags current campaign practices emphasizing expensive media and telephone contact with the most likely voters. Unless Hispanics step up personal contacts and canvassing, he fears, 'Their voice will remain muted.'6

Will There Be a ‘Hispanic Bloc'?

The Democratic party has not won the entire ideological battle for the Latino vote. Latinos are liberal on social welfare and immigration issues, but conservative on some critical social issues. More than half of Latinos are pro-life and support the death penalty. They voted against California's proposition to legalize marijuana and they tend to share Republican views on issues such as school vouchers and gay rights.7 Yet there are significant differences in public opinion on these and other issues among Hispanic subgroups. Cuban-Americans strongly support abortion rights, for example. The sense of racial otherness that undergirds the political cohesion and direction of African-Americans is not mirrored among Hispanics, notwithstanding frequent invocations of 'La Raza' among Hispanic leaders. According to the census, a little less than half of all Hispanics identify themselves as white. These differences, along with polls showing that Hispanics feel low identification with other Hispanics outside their national origin groups, raise skepticism about the future viability of a 'Hispanic voting bloc.'

There is nothing in Latino political culture that predisposes them to support liberal, welfare-state parties. Mexicans in Mexico, given a chance with free elections, overwhelmingly voted in a more conservative, less statist, business-minded administration. Puerto Rico's counterpart of the U.S. Republican party is a powerful political competitor and regularly wins elections on that island. Contemporary politics in Latin America is characterized by democratic, conservative, free-market governments. A sign of the times was that Latin America's archetypal populist, pro-labor, welfare state party, the Justicialistas (Peronists), recreated themselves as a more conservative, free enterprise party and regained power in the 1990s.

No Allegiances Are Permanent in the U.S. Electorate

The American electorate - even its segments driven by ethnicity, race and religion - has a short attention span. It can be swept by spasms of aversion toward a particular party or movement, rarely is the rejection permanent. The electorate grows and changes with prosperity or poverty, secularization, education, social mobility, intermarriage, and the appeal of newer and more relevant needs and grievances. And the political parties, seeking to be microcosms of the electorate, change their messages, self-definitions, target groups, interests, and what goes in the United States for their 'ideologies,' in an unceasing whirlwind of opportunistic adaptation.

A post-Civil War observer of American politics might have concluded that the adherence of millions of newly enfranchised slaves to the ascendant Republican party would have left the Democratic party a chronic also-ran for national power. That same observer returned to our era would be astonished by the near unanimity of Black Americans' preference for the Democratic party, and the massive flight of once atavistic southern Democrats to the Republican party. Would not such a time-traveling political observer from the late nineteenth century also note that the majority of the descendants of the massive waves of immigrants of that period are now themselves restrictionists?

None of this is remarkable. Political parties in the U.S. may have at one time been bodies of believers united by a common faith and militancy. But now their prime roles are as electoral apparatuses, animated more by power than program to organize this vast and varied population for regular rituals of selection, nomination, and promotion of candidates in their electoral tests. Political competition has become a market process. Governance is a commodity and the electors are consumers. Consumer preferences are shaped by advertising, tested and retested by polls, and cemented by pork and patronage. Ideology can become a matter of spin in this merchandising. As an important New York Tammany boss once commented after an upsurge in Socialist Party votes, 'They don't have to vote socialist. If they want socialism, Tammy will give it to them.' This role of modern parties is hardly inspirational. But it is hard to imagine how the democratic process as we now know it would work without it. True, American history shows that major political parties do succumb. Where are the Federalists and the Whigs now? And third parties come and go in dizzying succession. But the unceasing churning of the electorate and the proven market adaptability of both major parties make it unlikely that even the explosive growth of the overwhelmingly Democratic Hispanic minority could make the GOP a perennial also-ran in presidential politics.

Mass Immigration and the Likely

Growth of Rival Constituencies

The United States' continuing indifference to mass immigration is itself a factor that could well reduce the top-heavy primacy of the Hispanics among America's ethnic and immigrant groups. The mass immigration that has fed the Hispanic minority's growth may well nourish competing ethnic groups. Hispanics now loom large in the population of the United States, but their source countries are but a modest portion of world population. Asian immigration to the United States is now only in its infancy. The Census Bureau, with its characteristic disingenuousness, predicts that the population share of Asians in America, with their higher propensity to vote, will approach forty million by 2050. Expect these projections to be even more understated than the Census Bureau's myopic reading of U.S. population growth between 1990 and 2000.

China and India alone now comprise 2.2 billion people, 36 percent of the world's population. Both are undergoing rapid urbanizing that will sharply boost immigration expectations. Both have developed elaborate migration networks to and within the United States and sizable anchor populations. Within two decades combined legal, illegal, and nominally 'temporary' immigration - now about a third of the flow - could exceed current Latino immigration, now 44 percent of legal immigration and over half of illegal. There is nothing in the Asian immigration experience so far that suggests that the new masses of immigrants from China and India will sign on en masse with the Democrats.

So, are the Republicans done for? Clearly, they will be here for a while and will continue to win elections, though expect to see the party offering considerably more Hispanic candidates. A tougher question is this In the identity group politics likely to prevail in mid-century in an America of 400 million very different peoples, will the two-party system for aggregating diverse interests survive, or will it give way to an India-style congeries of regional, ethnic and religious parties?


1 Gannett News Service, 18 January 2001.

2 Hero, Rodney, 'Latino Participation, Partisanship and

Office Holding.' PS Political Science & Politics,

September 2000.

3 Velazquez Institute and 21 March 2001; and Tomas

Rivera Policy Institute, 14 August 1999.

4 Bureau of the Census, CB01-41, 6 March 2001.

5 Bureau of the Census Projection, Middle Series. NP-T4-E.

6 'Absentee Ballot, Apathy, and Noncitizenship.' Hispanic, 30 September 1998.

7 Hero, cited above.

About the author

David Simcox is chair of the Policy Board of the Washington-based Center for Immigration Studies, and director of Migration Demographics, a research organization.