Much has been made of the disparities between presidential and midterm electoral turnout as of late, especially given the twin collapses that have been visited upon the Democratic Party in the past two midterms.
A great deal of the focus, with justification, has been on the racial makeup of the electorate. The status of non-white voters as a critical firewall in the Democratic coalition is absolutely nothing new, and an essential part of the electoral discussion, no matter what election cycle is being examined. Concern about flagging turnout among nonwhite voters (African-American and Latino turnout, in particular) is a constant cause for anxiety in Democratic electoral circles.
That concern is well-placed—but only to a point. If one looks at the exit polling from the past several election cycles, for example, we can see that the 2014 exit polls show a nonwhite proportion of the electorate that is only marginally different from 2008. But even though the racial dynamics of the electorate were little changed, it goes without saying that the outcomes were vastly different.
So, while the Democrats definitely prize a more diverse electorate (and the GOP a decidedly pale one), this alone does not explain victory and defeat for the Democrats. One other subgroup has been of considerable importance to Democrats, and their participation (or lack thereof) has been an under-appreciated piece of the puzzle in explaining the eye-popping lack of consistency in Democratic performance. That group is young voters. Head below the fold for the numbers, and an explanation of why the political whims of voters under 30 matters a great deal for Democrats, and not just in non-presidential years.
Over the past decade of federal elections, one of the most consistent disparities between midterm and presidential elections has been the proportion of young voters who turn out on Election Day. In each of the three midterm election cycles, the drop-off between presidential and midterm turnout among voters aged 18-29 has been either 5 or 6 percentage points.
There are two primary reasons why young voters seem far more unlikely to make their mark than the balance of the electorate. For one thing, younger voters consistently are less inclined to be registered, when compared with their elders. For another, they also underperform their registration at midterm time, relative to the older subset of voters. To visualize this, compare the gap between the green and blue graphs between ages 20-40 (which is comparably wide) and the gap between the two graphs with voters aged 60-80 (which is considerably more narrow). The graph in question was from 2010, but the phenomenon seems likely to be present in every election cycle. For sure, the correlation between age and voter participation is clearly present in this turnout graph for California's recently concluded election cycle. Voters over 65 were roughly four times more likely to turn out than voters under 30.
What's more, the turnout disparities by age transcended ethnicity: The curves for all ethnic groups look close to parallel with one another, with only relatively slight variations.
The net result? Younger voters got outflanked at the polls by nearly a 2-to-1 margin in 2014.
Twenty years ago, that disparity in turnout relative to age might've mattered a bit less. But the complicating factor is that, over the past two decades or so, there have been two marked differences in electoral trajectory. Younger voters (18-29) have moved markedly in the direction of the Democrats. But any positive momentum from that shift has been at least partially offset by a slightly less dramatic erosion of Democratic support among voters 65 and over. It is a phenomenon that David Wasserman noted in a prescient piece in 2013 looking at what he concluded was an institutional disadvantage for Democrats in midterm elections. The statistics, indeed, are pretty startling.
Consider: In the five election cycles between 1994-2002 (which comprised three midterm cycles and two presidential cycles), Democrats averaged 51.2 percent of the vote among younger voters. In the last five cycles (three midterm and two presidential, for consistency in the comparison), Democrats averaged 59.6 percent of the "18-29" vote.
But at the same time, Democrats saw their fortunes with older voters begin to wane. In the 1994-2002 period, Democrats averaged 48.8 percent support with the 65-plus electorate. In the last five cycles, despite two wave cycles for the Democrats in 2006 and 2008 when Democrats actually carried elderly voters, that support has drifted southward, to an average of 45.6 percent of the elder vote. Despite the smaller shift in raw percentages, the impact is magnified, as mentioned before, by the fact that the elder/younger participation gap in midterm elections is vast, when compared with the gap in presidential elections (when the two groups have typically made up a fairly similar share of the electorate).
Of real concern for Democrats: The last three cycles have been the worst in the past two decades in terms of Democratic support among voters over 65. Even in 2012, when the Democrats managed a popular vote victory in the House, they could only manage 44 percent with voters over 65.
Indeed—and this cannot be emphasized enough—this has not merely been a "midterm elections" phenomenon. A look at presidential exit poll numbers shows that Mitt Romney actually did incrementally better among 65-plus voters than did George W. Bush eight years earlier, despite the fact the Romney lost the national popular vote by 4 points, and Bush won it by 3.
Therein lies the dilemma. If this erosion of support among 65-plus voters is a real thing (and, even with three consecutive subpar cycles, it is still a little too early to make that call), Democrats are going to be forced to win by running up the score elsewhere.
Therefore, in order to win nationally, Democrats are going to need both expanded turnout and outsized margins among that younger cohort of voters, even in presidential years. The former (in presidential years, at least) seems plausible. Young voters have been consistently, as seen above, in the upper teens as a share of the electorate in presidential years. The latter, however, seems like a far less certain bet. Democrats reached their high-water mark in 2008, when Obama crushed McCain with young voters, while Democrats did the same down-ballot (winning the subgroup with an almost comical 65 percent of the vote). But while Obama's re-election victory in 2012 was fueled by another rout among the youngest voters, the margin was markedly smaller (60-37). In the midterm years, it has been even worse (in 2014, the Democratic edge trickled down to 54-43).
This leaves the Democrats with some options, but all of them are somewhat tricky. They will either need to reverse the recent trend with older voters (which would relieve the pressure on running up the score elsewhere) or they are going to need a massive youth turnout, and they will need, arguably, no less than 60 percent of them to vote Democratic. Neither are implausible, but both are tall orders. Democrats are better equipped to do so in 2016 (especially if this is anywhere in the realm of reality), but they have four years to figure out how to thread that needle to avoid another midterm meltdown.