Jump to Any Race
National Politics|By David Wasserman, June 19, 2015

History shows it's extremely difficult for a presidential candidate to succeed a two-term president of the same party, especially one with approval ratings as middling as President Obama's. The "time for change" dynamic in 2016 should no doubt benefit Republicans. However, a common Democratic counter-argument is that the shifting demographics of the American electorate will continue to doom the GOP in the Electoral College.

It's true that Democrats continue to benefit enormously from the rise of non-white voters, who amounted to just 12 percent of all voters in 1980 but comprised 28 percent in 2012. In 1980, Ronald Reagan won 56 percent of white voters and won a huge landslide. In 2012, GOP nominee Mitt Romney took 59 percent of white voters - three points better - but lost by four points overall. President Obama took just 39 percent of whites in 2012, but captured 81 percent of non-whites.

So, what will the American electorate look like in 2016, and what does it mean for both parties' White House hopes? After diving into Census data from the past few years and crunching exit poll numbers from 2012, we can draw some early conclusions.

To answer questions such as "What share of Latino voters does the Republican nominee need to win the White House?" we built a statistical model of the likely electorate in each state in 2016. To do so, we first deconstructed the 2012 electorate by using exit poll data to break down voters into five distinctively behaving groups: 1) whites with a college degree, 2) whites without a college degree, 3) African-Americans, 4) Latinos, and 5) Asians/Others.

Then, to construct a model of the 2016 electorate, we used detailed data from the Census's American Community Survey to assess each group's shift in its share of the eligible voting public over the last four-year period for which data is available. For example, our best estimate is that Latinos comprised 18.8 percent of Nevada's vote 2012. But if the Latino vote rises in sync with the eligible population, Latinos will be 20.5 percent of Nevada's 2016 vote.

Here are our top observations, broken down into good news for Democrats and good news for Republicans, followed by the data we have compiled:

Three Reasons for Democratic Optimism

1. The white share of the electorate is likely to fall about two percent in 2016. In 2012, Edison Research's national exit poll pegged the electorate at 72 percent white, 13 percent African-American, 10 percent Latino, and five percent Asian/Other. If the electorate changes in line with Census estimates for citizens of voting age, the 2016 electorate will be 70 percent white, 13 percent African-American, 11 percent Latino, and six percent Asian/Other.

Of course, a huge question is whether likely Democratic nominee Hillary Clinton can mobilize African-Americans to the same extent President Obama did by virtue of his history-making runs. Although the answer won't be fully known until votes are counted, it's not unreasonable to assume that a small uptick in the black share of the population could offset a small dip in enthusiasm and that the black share of the overall vote will be stable heading into 2016.

2. Demographic shifts alone boost Democrats' national margin about 1.5 percent versus 2012. Based on demographic changes alone, Democrats will begin the 2016 election with a wider advantage than they had in 2012. If President Obama had won the same share of each group's vote but the composition of the electorate were adjusted to 2016's breakdown, his national margin over Mitt Romney would have expanded from 3.85 percent to 5.4 percent.

This wouldn't be enough to swing a state into Democrats' column, but it would come close. If every demographic group in North Carolina were to vote the same way they did in 2012, the GOP's two percent margin from that year would shrink to just 0.3 percent in 2016 on account of population changes alone. Nationally, Republicans would need to win a 2.8 percent higher share of non-whites just to offset the rise in the non-white share of the vote.

3. Even among white voters, rising educational attainment means Republicans' core supporters are in decline. Republicans' demographic woes aren't limited to the rise of Latino and Asian voters. Over the last few decades, Democrats have done better and better with whites who hold college degrees (especially advanced degrees), while whites without a college degree have become a core element of the GOP coalition.

In 2012, each group accounted for 36 percent of the electorate. Obama carried 42 percent of white college graduates, but just 36 percent of non-college whites. In 2016, white college graduates will rise to 37 percent of the electorate. Unfortunately for Republicans, non-college whites - by far their best-performing cohort - are slated to fall three points to 33 percent as more college-educated millennials supplant conservative seniors who didn't attend college.

Three Reasons for GOP Optimism

4. Rising non-white shares in states like Arizona and Georgia probably still aren't enough to put those states in play for Democrats. There's been a lot of talk about how rapidly rising non-white shares in states like Arizona and Georgia could "expand the Electoral College map" for Democrats. But that starts sounding more like hype when that growth is placed within the context of actual margins from 2012.

In Georgia, for example, Romney beat Obama by 7.8 percent in 2012. Although it's true that the white share of Georgia voters could fall from 64 percent in 2012 to 62 percent in 2016, that would only cut the GOP's margin by about 2.5 percent, less than a third of the swing Democrats would need to win the Peach State. The situation is similar in Arizona. Democrats need more than changing demographics in places like these; they need to change minds.

5. There is probably no single "magic number" Republicans need among Latinos - or any other group - to win the White House. The bad news for Republicans is that they got absolutely clobbered with Latinos in 2012, taking just 27 percent of that vote. The bad news for Democrats is that thanks to Latino voters' concentration in safe states like California, Texas, and New York, they are somewhat underrepresented in the swing states that matter.

Nationally, Latinos were 10 percent of all voters in 2012. But Latinos averaged just 6.9 percent of voters in the 12 heavily contested swing states. Even if Romney had won a 10 percent higher share of Latinos in every single state, he would have won only one additional state: Florida. He would have still fallen short in Colorado, Nevada, and New Mexico. So while Republicans badly want to win over more Latinos, the task is not as paramount to their success as often thought.

6. The "target numbers" Republicans need to win among each group to win the White House are achievable for the right candidate. The recipe for a GOP victory in 2016 isn't necessarily winning over large numbers of Latino or Asian voters, though that would certainly help their cause. The more plausible GOP scenario for winning the White House involves making modest improvements with every group of voters.

Nationally, Republicans would need to do three points better with everyone than in 2012 to win the popular vote. That amounts to winning about 65 percent of non-college whites, 59 percent of college-educated whites, nine percent of African-Americans, 30 percent of Latinos, and 34 percent of Asians/Others. Republicans hit nearly all of these benchmarks in 2014, and both former Gov. Jeb Bush and Sen. Marco Rubio have hit similar marks in past races.

At the swing state level, many of the "target numbers" Republicans would need to hit to win Electoral votes are easily imaginable under the right conditions. Republicans might need just 52 percent of college-educated white voters to win Iowa, 57 percent of non-college whites to win Wisconsin, and 37 percent of Latinos to win Virginia. So for all of the long-term trends benefiting Democrats, remember that the shifts the GOP needs in 2016 are still relatively small.

Demographic Breakdown of Voters in 15 States Decided by Less Than 10 Percent in 2012

2012 State College-Educated Whites Non-College Whites Blacks Latinos Asians/
New Hampshire48.0%45.0%2.0%3.0%2.0%
North Carolina33.0%37.0%23.0%4.0%3.0%
United States36.0%36.0%13.0%10.0%5.0%

Projected 2016 Demographic Breakdown of Voters in 15 Selected States

2016 State College-Educated Whites Non-College Whites Blacks Latinos Asians/
New Hampshire49.2%42.0%2.7%3.9%2.3%
North Carolina35.0%33.8%23.5%4.3%3.5%
United States37.4%32.8%13.3%10.9%5.6%

2016 Partisan Shift in 15 Selected States Based on Population Changes Alone

State 2012 Obama 2012 Romney 2016 Dem 2016 GOP% Pro-Dem Margin Shift
New Hampshire52.0%46.4%52.6%45.8%1.2%
North Carolina48.4%50.4%49.2%49.5%1.7%
United States51.1%47.2%51.8%46.4%1.5%

Mitt Romney's Shares in 15 States Decided by Less Than 10 Percent in 2012

2012 State Romney Vote College-Educated Whites Non-College Whites Blacks Latinos Asians/
New Hampshire46.4%45.2%51.2%7.2%28.2%32.2%
North Carolina50.4%59.1%74.1%4.1%31.1%43.1%
United States47.2%55.7%61.7%5.7%26.7%30.7%

Republicans' "Magic Numbers" for Turning Blue States Red in 2016

2016 State GOP Vote College-Educated Whites Non-College Whites Blacks Latinos Asians/
New Hampshire50.1%49.5%55.5%11.5%32.5%36.5%
United States