Demographics
First, a look at the breakdown of the electorates from 1992 thru 2012:
White
1992 - 87%
1996 - 83%
2000 - 81%
2004 - 77%
2008 - 74%
2012 - 72%
African-American:
1992 - 8%
1996 - 10%
2000 - 10%
2004 - 11%
2008 - 13%
2012 - 13%
Hispanic:
1992 - 2%
1996 - 5%
2000 - 7%
2004 - 8%
2008 - 9%
2012 - 11%
Asian:
1992 - 1%
1996 - 1%
2000 - 2%
2004 - 2%
2008 - 2%
2012 - 3%
Demographics of How Groups Voted in the 1992 Presidential Election [1992 thru 2008 data]
Presidential Race - 2012 Election Center - Elections & Politics from CNN.com [2012 data]
Now let's analyze the twenty-year trend to project the 2016 electorate.
2016 projection:
White: 70%
African-American: 13%
Hispanic: 13%
Asian: 3%
Other: 1%
Estiamted 2016 turnout: 55.0%
[Turnout projection based on approximate average of 1992-2008 turnout and preliminary 2012 turnout estimates]
United States population in 2016: ~323,000,000
Approximately 74.3% of the United States populace is eligible to vote
Eligible electorate in 2016 will be approximately 240,000,000
55% turnout assumes 132,000,000 voters
2012 votes by race:
White: Romney 59%, Obama 39%
African-American: Obama 93%, Romney 6%
Hispanic: Obama 71%, Romney 27%
Asian: Obama 73%, Romney 26%
Other: Obama 58%, Romney 38%
Presidential Race - 2012 Election Center - Elections & Politics from CNN.com
Now, let's use the trending electoral breakdown to consider some 2016 outcomes.
Scenario #1 -- Demographic Repeat of 2012
White: R 59%, D 39% (R 54.5m, D 36.0m)
African-American: D 93%, R 6% (D 16.0m, R 1.0m)
Hispanic: D 71%, R 27% (12.2m, R 4.6m)
Asian: D 73%, R 26% (D 2.9m, R 1.0m)
Other: D 58%, R 38% (D 0.8m, R 0.5m)
TOTAL: Democratic nominee 67.9 million, Republican nominee 61.6 million
If the various demographic groups vote the same way, the Democratic candidate's margin will increase from President Obama's current 3.7 million vote lead to 6.3 million votes. That is simply if nothing else changes.
But will we likely get such a repeat? No. So let's look at some other possibilities.
Scenario #2 -- All Numbers Strong for the GOP
White: R 60%, D 38% (R 55.4m, D 35.1m)
African-American: D 88%, R 11% (D 15.1m, R 1.9m)
Hispanic: D 62%, R 35% (D 10.6m, R 6.0m)
Asian: D 63%, R 36% (D 2.5m, R 1.4m)
Other: D 55%, R 41% (D 0.7m, R 0.5m)
TOTAL: Republican nominee 65.2 million, Democratic nominee 64.0 million
In this case, I've assumed that everything will go well for Republicans. I've given them yet another point of the White vote, I've reverted the African-American vote to the Bush/Kerry numbers, I've reverted the Hispanic vote to the Bush/Gore numbers, I've clipped 10% off the Asian numbers and shifted them to the Republicans, and I've even given 3% of the 'Other" category to the GOP (this is a catch-all group that is probably fairly resistant to short-term movement).
This gets the GOP a narrow 1.2 million vote margin, that would have over an 80% change of translating into an Electoral College victory. But in order to get this narrow GOP victory, you have to assume pretty much every demographic trend goes the Republcans' way.
Scenario #3 -- Same as Scenario #2, but 2012 numbers for Whites
White: R 59%, D 39% (R 54.5m, D 36.0m)
African-American: D 88%, R 11% (D 15.1m, R 1.9m)
Hispanic: D 62%, R 35% (D 10.6m, R 6.0m)
Asian: D 63%, R 36% (D 2.5m, R 1.4m)
Other: D 55%, R 41% (D 0.7m, R 0.5m)
TOTAL: Democratic nominee 64.9 million , Republican nominee 64.3 million
This scenario assumes that the GOP has maxed out its share of the White vote. Not a bad assumption; it's hard to see Clinton or Cuomo or O'Malley, or whoever wins the 2016 Democratic nomination, doing any worse among Whites. The rest of the numbers are still very favorable to the GOP.
Still, the GOP loses the popular vote, with probably a 2/3rds chance of also losing the Electoral College.
Scenario #4 -- Same as #2 & #3, but 1% improvement on 2012 for Whites
White: R 58%, D 40% (R 53.6m, D 37.0m)
African-American: D% 88, R 11% (D 15.1m, R 1.9m)
Hispanic: D 62%, R 35% (D 10.6m, R 6.0m)
Asian: D 63%, R 36% (D 2.5m, R 1.4m)
Other: D 55%, R 41% (D 0.7m, R 0.5m)
TOTAL: Democratic nominee 65.9 million, Republican nominee 63.4 million.
These demographic numbers are still pretty good overall for the GOP. It would not be at all surprising to see someone like Clinton improve one percentage point for Democrats among Whites. And if that happens? It's all over, even assuming the GOP still manages to do considerably better with all other demographics. This is what is so important -- even demographic trends that are reasonably GOP-biased translate to a loss in 2016.
Scenario #5 -- Modest GOP gains on minority voting, Bush/Kerry numbers for Whites
White: R 58%, D 41% (R 53.6m, D 37.9m)
Black: D 90%, R 8% (D 15.4m, R 1.4m)
Hispanic: D 66%, R 32% (D 11.3m, R 5.5m)
Asian: D 68%, R 31% (D 2.7m, R 1.2m)
Other: D 55%, R 41% (D 0.7m, R 0.5m)
TOTAL: Democratic nominee 68.0 million, Republican nominee 62.2 million
This final number is what I consider more realistic. For all demographics, I've assumed a reasonable and modest rebound towards 50%. I've given the Bush/Kerry numbers to Whites, and assumed the GOP will peel a few percentages off the other groups. The result? A big Democratic victory.
And if the Democrats can manage to flip more than 1.5% of the White vote, it really becomes apocalyptic for the GOP.
Note that these projections are only for 2016. In 2020, it continues to get progressively worse for Republicans. By then, when the White vote is down to 67% or 68%, with Hispanics up to perhaps 15% and maybe Asians gaining 1%, even the rosiest scenario above for the GOP will result in a clear loss.
This is just some food for thought. I advocate nothing. I am just showing why the Republican Party, in order to win Presidential elections in the future, absolutely must find a way to get non-whites to vote for them in greater numbers. Much greater numbers.