Discussion: (16 comments)
Comments are closed.
A public policy blog from AEI
The changing ethnic makeup of the electorate spells doom for the GOP unless it can figure out a way to appeal to minorities, right? Everybody knows that.
I thought I knew it too, and so, as part of the book I’m working on, I set out to document this universally acknowledged truth with numbers. I hereby concede defeat. The GOP is facing a demographic headwind, but for the next few election cycles it is a zephyr rather than a gale.
It seems impossible that the headwind is not already a gale, or even a hurricane. After all, we know from the national census that non-Latino whites (hereafter just whites) fell to 64% of the population in 2010, while Latinos continued their skyrocketing rise, now constituting 16% of the population, overtaking African Americans as the nation’s largest minority. We know from the National Election Pool exit polls, the one used by all the major news organizations, that Democrats captured large majorities of Latinos (averaging 64% of the vote), blacks (92%) and Asians (62%) in the four presidential elections from 2000 through 2012. The Census Bureau’s projections tell us that America’s minorities will continue to increase as a proportion of the population, with whites becoming a minority of all Americans in the early 2040s.
And yet, when these numbers are plugged into the standard arithmetic for predicting voting outcomes, the expected increase in the Democratic vote in 2016 is not five, six, or seven percentage points. Nor even one or two percentage points. The demographic changes I just described may be expected to produce an increase in the Democratic presidential vote of just three-tenths of one percentage point.
How is that possible? Because I neglected to mention one other set of numbers that goes into that arithmetic, also produced by the Census Bureau in periodic special surveys for the November Current Population survey: Voter turnout. In the presidential elections from 2000 through 2008 (the 2012 figures aren’t yet available), the percentage of Americans eighteen years and older who actually voted averaged 57%. But those percentages varied widely by ethnic group. Among whites, the average turnout was 64%. Among blacks, 57%. Among Latinos and Asians, just 29%.
That’s why the headwind is so feeble in the near term. Between 2012 and 2016, the Census Bureau estimates that the population of voting-age Latinos will increase by 3.9 million people compared to an increase of just 1.8 million whites. But because of their much lower turnout, the expected increase in Latino voters is 9,513 fewer—yes, fewer—than the expected increase in white voters. The only reason that the Democrats can expect even a microscopic 0.3 percentage point increase in the 2016 vote is because of an increase in the black voting-age population.
In the long term, the GOP will indeed face a gale. Much of the explanation for the low turnout of Latinos and Asians is that many are recent immigrants, are not yet citizens, and hence are not eligible to vote. Among citizens, turnout in presidential elections is already around 47% for Latinos and 45% for Asians. As time goes on, it is plausible that the percentage of Latinos and Asians who are citizens will increase, and that the propensity of Latinos and Asians to vote will eventually be about the same as the propensity of other Americans. To be specific, suppose I assume that the overall turnout rate for Latinos and Asians will linearly converge on the African American turnout rate of 57%, reaching that point by 2040. Leaving the other parameters unchanged, the expected Democratic vote for president will rise from 51% in 2012 to 55% in 2024 and to 64% by 2040.
Other changes in the assumptions can expand that predicted gap favoring Democrats. The last few decades have seen the development of a large “marriage gap” in voting, with married couples voting predominantly for Republicans and singles for Democrats—and marriage rates are falling among all groups except Asians. If those marriage rates continue to fall, we can expect the Democrats to reap an additional electoral dividend. But at this point, we are pushing our luck in trying to anticipate what happens. For example, conservatives have more children than liberals, and there is a sizable correlation between the political opinions of parents and their adult children—maybe that will offset a growing marriage gap.
The biggest imponderable, of course, is how the Republican Party evolves. If it continues to be seen as a party dominated by anti-abortion, anti-gay, religious fanatics—and, however unfairly, that’s the way it is seen by broad segments of the non-conservative electorate—it will have gone the way of the Whigs long before 2040. If it becomes identified as the party in favor of free enterprise and free lives, equally opposed to crony capitalism and to bureaucrats who tell us how to live our lives, it could return to power with large majorities. But that is just my strongly felt opinion. My immediate point, with good evidence behind it, is that ethnic changes in the electorate do not make the next few elections nearly as ominous for the GOP as conventional wisdom has it.
Comments are closed.
1150 17th Street, N.W. Washington, D.C. 20036
© 2015 American Enterprise Institute for Public Policy Research