This article originally appeared in Miller-McCune magazine and is reprinted with permission.
By Norman Nie
Norman H Nie is professor emeritus at the University of Chicago and Stanford University.
A comprehensive look at voter behavior and demographics reveals a momentous prospect: a Hispanic electorate that turns out to vote en masse, allies itself strongly with one political party and changes America’s political balance for decades.
Common wisdom notwithstanding, neither Democrats nor Republicans have yet sealed the Hispanic deal.
The rapid growth in the US Hispanic population over the last 40 years—both in terms of raw numbers and percentage of the population—is probably the most important emergent force in American politics today. The evidence is around us: In 2008, each party conducted an entire presidential primary debate in Spanish. In 2009, the first Hispanic judge, Sonia Sotomayor, was appointed to the US Supreme Court. And in 2010, for the first time ever in a single election, three Hispanic candidates won top statewide offices: Republican Susana Martinez won in New Mexico to become the nation’s first Hispanic woman elected governor; Republican Brian Sandoval became Nevada’s first Hispanic governor; and Republican Marco Rubio was elected to represent Florida in the US Senate.
Despite these notable top-of-the-ticket wins by Hispanic Republicans in 2010, most political observers continue to assume there is significant and stable support for Democrats among Hispanics, similar to the support that African Americans have shown in recent decades. Indeed, new Hispanic voters have entered the electorate more often as Democrats than as Republicans in recent elections.
But the current degree of Hispanic attachment to the Democratic Party is by no means a future certainty.
We examined Hispanic turnout and partisanship in midterm elections—that is, the elections where turnout isn’t significantly affected by charismatic presidential candidates—from 1978 to 2006 by collecting and matching data from several sources. First, we combined more than 150 academic and high-quality commercial public opinion surveys, each of which employed a nationally representative sample of at least 1,000 respondents. In addition to this combined dataset, we added voter turnout data from the US Census Bureau’s Current Population Survey, creating what we believe is the most comprehensive demographic and voting behavior dataset ever compiled for midterm elections. The data allowed us to examine both overall trends and individual behavior. (Individual level data is not yet publicly available for the 2010 midterm election, but aggregate results are considered here.)
On the surface, the information we gathered supports some common political wisdom: A vast buildup of potential Hispanic voters, primarily composed of immigrant citizens and young US-born Hispanics, has generally tended to favor Democrats, but turned out to vote at far lower levels than whites or African-Americans during the last three decades.
Our dataset, however, shows that the common wisdom misses a potentially momentous prospect.
Once we accounted for demographic differences known to affect turnout, we found that Hispanics actually vote at rates very similar to those of whites and blacks. In other words, much of the explanation for the low turnout rates for Hispanics is not related to being Hispanic, but to Hispanics being younger and having less education on average than whites or blacks.
Similarly, we found that, in terms of both party identification and the strength of that partisanship, the differences between whites and Hispanics disappear when individual-level characteristics such as age and education are taken into account. In short, the data suggest that Hispanics have not been genuinely incorporated into the party system or made anything like a deep commitment to either party.
But that incorporation and commitment may come soon.