Will young people vote in 2020?

A poll of Iowa Democrats entering their caucus locations earlier this month revealed surprising commonality among Democrats of all stripes: their first priority is defeating Trump. By a margin of 62 to 36 percent, Iowa Democrats said they cared more about finding a nominee who could defeat Trump than whether they agreed with that candidate on the issues.

This result matches what we have been seeing this entire election season. A national poll taken nearly a year ago found a similar result, with Democrats caring more about defeating Trump than ideological purity by 55 to 35 percent.

This result is not particularly surprising. Republicans deride us for our “Trump derangement syndrome.” And indeed, Democrats are appalled by Trump as president. We can’t believe that a racist, sexist, ignorant, incompetent is representing our country on the world stage. To us, Trump represents everything that is bad about America. In addition to his policies that outrage us, his tax cuts for the rich, his attacks on healthcare and environmental protections, for instance, he personally disgusts us. We think America is better than he is. Instead, we are all being painted with his brush.

This is especially appalling to Democrats after eight years of Barack Obama, who was a worldly, accomplished intellectual, a dedicated family man, a supporter of the dignity of all people, regardless of their racial, sexual, religious, or other background. We were proud of the image of America he presented to the world.

In some respects, our horror over the Trump presidency is comparable to the horror Trump supporters had over the Obama presidency. Just as we are appalled by the image of America Trump presents to the world, Republicans were appalled by the fact that the world saw America represented by a Black man.

So we want an electable standard-bearer. The problem is that electability is completely amorphous, and we are not so great at identifying it. Who would have thought that a thrice-married “pussy”-grabber would be electable? Similarly, who would have thought that a Black college professor whose middle name is Hussein be electable? And yet, both were elected.

Similarly, John Kerry and Mitt Romney looked pretty electable. Both lost. So much for electability.

The internet has been replete of late with the argument that moderates have a losing record: six losses and five wins in the modern era. As with any statistic, that result will vary based upon where you draw the line. That accounting is true if we start with the 1976 election. However, if we start from the 1992 election, the record changes to four wins and three losses. So much for lies, damn lies, and statistics.

There is a reason, though, that these accountings start after 1972. In that year, Democrats nominated an anti-war progressive, George McGovern. He lost that election to Richard Nixon by a historic margin. So the progressive record is zero wins, one loss; a much worse record than for the moderates.

All of these claims, however, suffer from a serious statistical problem: small sample size. The number of elections we can point to is too small to make any kind of generalizations about electability. Had the smallest number of votes shifted in certain specific places, Al Gore and Hillary Clinton would have been elected president, and history would have been very different. So we can’t draw any conclusions from history.

Analysis of electability is even more challenging given that there are very few unbiased analysts. Take David Frum, the never Trumper conservative who has been raising alarms about the Sanders ascendency. Given that he plans to vote against Trump, he would much rather that he cast a vote for a less progressive Democrat than Bernie. As a result, his analysis skews to emphasize Bernie’s disadvantages and the strengths of more moderate Democrats such as Amy Klobuchar or Joe Biden.

On the other hand, progressive Ibram X. Kendi has argued that a moderate will fail to energize the Democratic base, an essential occurrence for a Democrat to win. So his analysis emphasizes the need to nominate someone matching his ideological inclinations.

In general, it appears that commentators who try to be objective such as David Leonhardt agree that Sanders can win, but it will be a tough race. Any analysis of the 2020 general election at this point, however, is like trying to divine the future with tea leaves.

A central claim regarding the electability of Sanders, however, is that he will energize people who did not vote four years ago to come out and vote for him. The evidence for that claim is shaky at best. Indeed, political scientists pooh-poohed the idea that Sanders could generate enough additional turnout to counterbalance the moderate voters he would turn off. Bernie’s claim regarding turnout, however, is not baseless.

Part of the analysis regarding turnout suffers from the same problems we uncovered looking at the win-loss record of moderate presidential candidates. First, we have a small sample size. As a result, either side can find examples to validate their arguments. Sanders supporters point to the big turnout he generated in New Hampshire. His opponents point to the disappointing turnout in Iowa. So the available data does not help us at all.

Similarly, there is also a problem with cut-offs. Just as the win-loss record of moderates changed if we started the calculation in 1976 or in 1992, young voter turnout changes based upon where we define the end of young people. Do young people include those under 25, under 29, under 35, or older? Which cut-off you choose will change your results.

In one report, the Census Bureau illustrates the problem. In one part of the report, young people are defined as those 18 to 24 years of age. Obviously, the turnout among that group is pathetic. Their peak turnout in recent years was 2008 when Obama was first elected. That year, 44.3 percent of them voted. In the next election, however, turnout dropped to just 38 percent, and in 1996 and 2000, two elections where Democrats could have used their help, turnout among that group was just over 30 percent.

In the same report, when young voters are defined by the more typical 18 to 29 year-old range, their turnout in 2012 was 45 percent, a full seven points higher than when the group only includes those up to 24 years old.

The lesson from this analysis is clear: as you get older, you are more likely to vote.

There is no cut-off at which people start voting more. Eighteen year olds vote at a lower rate than 19 year olds, who vote at a lower rate than 20 year olds, and so on.

This fact makes sense given that 18 year olds tend to have unstable lives. By the time they are getting into their thirties, most people have grown some roots in their community, and so feel a greater connection to the electoral process.

The group to watch among young people are the widely-maligned millennials. They are a frequent target of curmudgeonly old white guys including the one in the White House. But, as I have pointed out, they have some very legitimate grievances regarding their economic status and the environment their elders are leaving them. This group is the most diverse and well-educated generation in our history. All these factors combine to make them extremely negative toward Trump.

Indeed, had only millennials voted in 2016, Clinton would have won in a landslide. Furthermore, a big jump in turnout among those 18 to 44 years-old powered the Democrats to a big win in the 2018 mid-terms. Indeed, this group played an important role in Bernie’s impressive New Hampshire win. So the more millennials who vote, the better it is for the Democrats, especially if they nominate Bernie.

And here’s the rub. According to the Pew Research Center, millennials can be defined as those who were born between 1981 and 1996. In 2016, those voters were aged between 20 and 35. In other words, the youngest were still among that 18 to 24 year-old cohort who are least likely to vote. In this election, however, they will be 24 to 39, for the first time making up almost the entire 30 to 44 year-old group who vote at a much higher rate. Therefore, with each election, anti-Trump millennials are making up a bigger and bigger part of the electorate, and more importantly, a more and more politically engaged group.

I am not the first to point out that we are currently in a period of transition between an electorate dominated by old generation whites into an electorate with a progressive majority. What I believe has been missing from prior analysis, however, is some nuance regarding voter turnout. It is just too simplistic to say that millennials don’t vote because they are young. In reality, millennials are getting older, and as a result, are more likely to vote. What’s more is that the oldest millennials likely vote at a very high rate while younger ones are just starting to vote at an acceptable adult rate. The implications of this transition could be profound in the next election.

Far be it from me to suggest that this analysis means Bernie is our most electable candidate. I really need to get out of the prognosticating business. But at the same time, this data seems to indicate that the argument for Bernie’s electability is not without basis. This fact should give hope to Democrats looking with concern at a primary process that appears to be moving toward the coronation of Bernie Sanders as our presidential nominee.