Talk to any Democratic strategist and its clear the Democratic Party is in the midst of a great debate.  To identity politick or not to identity politick?  Democratic strategists such as John Judis have noted how the strategy has failed and is insufficient to establish a Democratic strategy.  So have I here.  Columbia intellectual historian Mark Lilla got into the mix in a recent op-ed.

In the op-ed, Lilla argues the Clinton campaign and many Democrats lost because they embarked on a strategy of reaching out to specific groups, especially minorities, at the expense of developing a broad message.  Certainly this is at least true in the Midwest where Clinton’s fate was sealed by slim margins.  But others, especially progressives, view this as a cave by the Democratic Party and an abandonment of the progressive values they think the party should stand for.

Values aside, the basic problem for Democrats is math.  What does the party need to do to not just win the Presidency but also control Congress (ie. establish a governing majority)?  This is not just a debate in simple arithmetic.  Democrats wasted millions of votes last year and in 2012, when Obama won reelection, he still lost a majority of congressional and legislative districts in the nation.

Democrats have been getting a declining share of the white vote since 2000 but have managed to up their margins among minority voters allowing them to stay competitive.  Progressives argue this means it is only a matter of time before they establish their emerging majority.

However, this tide of Democratic dominance has failed to emerge time and time again.  Republicans would argue it is because the nation is ideologically in-sync with their agenda and supports conservative policies.  They certainly dominated Congressional elections and outran Donald Trump by a combined 5 million votes.  But, other sociological factors could be at play.

First, the forecast of a majority-minority is more problematic than initially assumed.  Secondly, and perhaps even more importantly, the assumption that minorities will always vote Democratic or even 2/3rd Democratic becomes more problematic the further out we go.

Let’s go back to the mathematical problem Democrats face.  The party has consistently lost whites by 20 points in Presidential elections (even more in midterms).  To make up for this massive deficit the party has had to win 80 percent of minorities (as Obama did twice).  But, in an election in which the white vote hovered 1 percent below 2012 levels and the party failed to capture 80 percent of the minority vote, their vote total dropped by over 1.5 million compared to 2012.  Worse, the drop-off in minority support and Trumps strength among white Americans was the crucial Blue Firewall of the Midwest.

But, let’s assume for a second that whites only make up 66 percent of the vote in 2020.  The Democratic share of the white vote remains constant (they lost it by 20).  Whites would only give the party a quarter of the vote.  To even come close to Obama’s 53 percent in 2012 the party would need 80 percent of minorities yet again.  Clinton barely garnered 70 percent for reference according to an analysis of the data by the Washington Post.  To capture Congress Democrats would need in excess of 80 percent of the minority vote.

Going forward though things look better for Democrats right?  Not so fast.  When one looks beyond the immediate data and biased assumptions and actually examines the social processes involved in molding the data the certainty of a Democratic majority gets smaller.

Consider, the forecasts these assumptions rest on may be off.  The Census Bureau predicts whites will be a minority in 2040.  But, the extent of white demographic decline is probably exaggerated.  More so, the prediction of a majority-minority society might not come to pass.  Why?  The reason lies in the way the Census Bureau classifies racial and ethnic groups that come from racial-ethnically mixed backgrounds.

Right now, at least 10 percent of babies are born to racially-ethnically mixed families.  Despite the fact these babies usually have a white parent, the Census Bureau does not classify them as whites or racially mixed but rather as nonwhite.  This becomes important because whites are very likely to behave like other whites in sociological terms.  So, for example, white children will likely have white friends, grow up in majority-white neighborhoods, go to majority-white schools, think of themselves as white and marry whites.  Now, we do not know whether these younger whites will vote the same way as other whites but we will soon find out as they come of voting age in the next decade or so.

Individuals of mixed backgrounds are on the leading edge of assimilation.  For example, the top-tiers of the American workforce are increasingly dominated by Asians and college educated Hispanics.  The Democratic assumption of demographic dominance rests on the assumption these voters back their party because it is the party of minorities, rather than educational status/attainment, income or family status.

There is no reason to assume Democratic dominance will automatically occur.  Indeed, we have two recent examples to look at that make this view dubious.  Catholic ethnics were a large part of the Democratic coalition from 1960 to 1976.  But these voters became many of the Reagan-Democrats of the 80s and last year.  Additionally, in 2012  75 percent of Latinos without a college education backed Obama.  But, among college-educated Latinos which also correlated with higher income, the President’s vote share dropped to 63 percent.

Light skinned Latinos, who share the cultural values of many ethnic Catholics, and identify as Catholic to boot, might be particularly prone to break from the Democratic Party.  Indeed, Democratic dominated California is defined by the black/white progressives views of Northern California and the more business friendly, centrist, socially conservative values of Latino dominated Southern California.  Now, this does not mean these Latino Catholics will vote uniformly Republican.  They might split 50/50 or still vote majority Democratic.  But, due to the math, that is still bad news for the party.

The current Democratic strategy, founded on winning the Obama coalition of minorities, young voters and women, seems eerily similar to the 20th century strategy of party and big city bosses.  Then, these powerbrokers gained and held power by winning over clearly defined racial/ethnic blocks.  A diversifying America does not lend itself to such a strategy.

Systemic racism does still occur, even today, but not to the clear degree as the prior century.  However, other patterns of group melding, including assimilation, are becoming important as well.  Because whites are the dominant group, assimilation usually requires some form of interaction with them.  In a way this actually expands the white majority.  Even among non-whites.

Among Asians this dynamic is particularly clear.  Asians tend to live near or in majority-white neighborhoods, have incomes and educational levels on par with whites, send their kids to similar schools as whites and tend to work in the same occupations as whites.  They interact with whites on a daily basis and to a degree mimic their sociological behaviors and dynamics.  They only do not vote that way.  But, compared to other racial and ethnic groups, Asians are pretty similar to whites.

If minorities, or even just some minorities, behave more as whites and the Census Bureau continues to misclassify racial and ethnically mixed individuals the white majority could persist beyond just Census numbers.  More importantly for electoral politics, if Democrats bet on demographics does not pan out they could be sentenced to a permanent minority in Congress in the same way the GOP was for 40 years.