Why Is Hillary Running Left?
Brian Beutler has a thoughtful piece up at The New Republic discussing Hillary Clinton’s leftward shift vis-à-vis her 2008 run for president (and especially vis-à-vis Bill Clinton’s presidency). The piece is well worth reading in its entirety, but these are the relevant sections for our purposes:
“There’s an ongoing debate in American politics over the extent to which the Obama coalition is unique to Obama, who is himself a unique historical figure. Are the younger, more progressive Democrats who swept him into office ready to do the same for a candidate who doesn’t check all of the same characterological boxes—youth, charisma, diversity? . . .
“Perhaps more importantly, Hillary Clinton also thinks the answer is yes—if, that is, you buy the cynical (but possibly accurate) interpretation of her leftward shift. In fact, this might be the most hopeful interpretation as far as liberals are concerned. Because if Clinton doesn’t have any core convictions, and is only saying whatever she thinks she has to say to win—if indeed she's merely betting that things like campaign finance reform, same-sex marriage, and immigration reform will add up to a winning platform—then it's a nod to her belief that the Obama coalition is stable, loyal, and larger than the Republican electorate.”
I tend to think overall that the 2016 election will be decided much more by the state of the economy than Hillary Clinton’s positioning on marriage equality or ability to increase youth participation rates. In that respect, Wednesday’s GDP report was much worse news for Democrats than anything we might discern about youth turnout in 2016 (though it remains to be seen if there is a summer “bounce-back” for the economy, as there was last year).
But I do think party coalitions matter, however, at least on the margins. In that respect, I think Beutler identifies the most important issue, by several orders of magnitude, for 2016: Is Obama’s coalition transferrable to Clinton (or other Democrats)? I think we need to flesh out this question a bit before we can turn to Beutler’s second question, which actually asks what Hillary Clinton’s positioning tells us about the answer to the first question.
From a demographic standpoint, an election consists of two components: Turnout levels among different demographic groups and levels of support among different demographic groups. Barack Obama won the demographic battle on both fronts: He boosted turnout levels among African-Americans to match those of non-Hispanic whites, and he blew the proverbial roof off in terms of support levels from minorities.
This was critical to his victory in 2012: If Mitt Romney had matched George W. Bush’s level of support among minorities, Romney would have been elected president (even if he had dropped to George W. Bush’s level of support among non-Hispanic whites). Likewise, if black participation had dropped back to 2004 levels, the election would have been too close to call (how you evaluate it would depend on whether you think new African-American voters in 2008 were even more likely to vote for Obama than returning African-American voters).
So is this coalition transferable? Obviously, Obama’s 2012 re-election effort is a very good data point for Democrats on the turnout front. Many analysts, myself included, were skeptical that Obama could re-create the turnout from 2008, much less improve upon it (although much of the demographic change that occurred was due to whites staying home, rather than African-American or Hispanic turnout increasing).
Another positive data point for Democrats is turnout in the 2014 elections. The fact that African-Americans were 12 percent of the electorate in 2014, vs. 10 percent in 2006 and 11 percent in 2010, suggests that many of the marginal voters Obama turned out are no longer marginal. In the battle between voting as a socio-economic phenomenon and voting as habit, what evidence we have suggests that habit may be winning (although turnout was so unusually low in 2014 that it may be tough to generalize from this data).
Things are dicier, however, on the question of support. If we look at support for the parties in the 2004, 2010 and 2014 elections among racial groups, we see an awful lot of stability. Republicans won 59 percent of the non-Hispanic white vote in 2004, 62 percent in 2010 and 61 percent in 2014 (these statistics exclude votes cast for third parties, a common practice when comparing across elections).
Among African-Americans, Republicans won 11 percent, 9 percent and 10 percent in 2004, 2010 and 2014, respectively. For Hispanics, according to the exit polls, Republicans won 45 percent in 2004 (though that is probably an overestimate; the true number is likely closer to 40 percent), 39 percent in 2010 and 37 percent in 2014. Among Asian-Americans, it was 44 percent in 2004, 41 percent in 2010 and 51 percent in 2014. Among “other,” the numbers are 43 percent, 45 percent and 49 percent.
There are differences there, but overall those electorates look roughly alike. This is important, because if the Republican wins the same share of whites, blacks, Hispanics, Asians, and “other” voters in 2016 as Republicans did in any of 2004, 2010 or 2014 elections, the Republican would win the popular vote. This would be true even if the white share of the electorate falls another two points, and the Hispanic and Asian/other shares rise another point apiece. To look at this a shade differently, Republicans just had a landslide win in an electorate that, according to the exit polls, had pretty much the same racial demographics as the one in 2008 (which was supposed to be a more or less unwinnable electorate for Republicans).
So how do we evaluate the “support” data? We have two basic ways of looking at it. Democrats prefer to look at this as a midterm/presidential electorate dynamic. There’s doubtless at least some truth here, although this argument is neutralized somewhat by breaking things down by race: Seen through this lens, you have to demonstrate not just that the voters who stayed home were more Democratic, but that they were more Democratic after controlling for race (see more here).
Republicans, on the other hand, think of this as an Obama coalition vs. a “normal” coalition. The thought is that if you remove the historic nature of Obama’s candidacy, the racial breakdowns will look much more like they did in 2004, 2010 and 2014. There are problems with this view as well – most notably that 2010 and 2014 were wave elections – but the idea that Democrats would have a hard time maintaining 80 percent support among non-whites without the positive stimulus of a non-white presidential candidate isn’t far-fetched. Here’s political scientist John Sides:
“Second, the ‘Obama coalition’ may prove to be exactly that: a coalition specific to Obama. When he is no longer at the top of the ticket, will groups like Latinos and African-Americans turn out in such numbers, and with such strong support for the Democratic candidate? At a University of Denver election panel last week, political scientist Matt Barreto noted that 79% of African-American are “very enthusiastic” about the Democratic Party now, but only 47% say they will be after Obama’s presidency ends (see slide 18). It’s unlikely that African-Americans are going to vote for a Republican candidate in large numbers, but will they turn out in such high numbers for whichever Democrat wants to succeed Obama?”
This brings us back to Beutler’s argument that the decision to shift leftward is made by the Clinton campaign from a position of strength, and provides evidence of a belief that Obama’s coalition is both transferable and larger than the Republican coalition.
I would call this the optimistic interpretation (from a liberal perspective) of her moves, even if you accept that this is simply cynical gamesmanship on her part. The pessimistic interpretation would stem from one of the theses of my book, “The Lost Majority”: that Obama modified Bill Clinton’s coalition into a narrower, deeper one. This enabled Obama to win a victory in 2008 that was almost as large as Clinton’s 1996 win without bringing Appalachians or working-class whites on board. The problem with such a coalition is that it doesn’t allow for much flexibility: At least for now, Democrats have to run up the score with different groups in order to win.
Under the pessimistic take, the Clinton Coalition is simply gone. Bill Clinton had managed to keep Appalachian voters and working-class whites in the Democratic camp through skillful positioning and a bit of luck. But over the course of the next decade, these voters finally broke with the Democrats. This occurred at the presidential level in 2004 and 2008, then occurred at the senatorial level in 2010 and 2014, when relatively conservative Democrats like Mark Pryor and Mary Landrieu found they could no longer win with a Clintonian formula (in fact, they couldn’t come close to winning). This was a break that was a long time coming, but given that it has filtered down to congressional and even state legislative offices, it seems like it will be difficult to reverse.
So from the pessimistic standpoint (again, from the liberal perspective), Clinton is running leftward not because she believes it is the best way to win, but because she believes that she has no choice.