Iowa assigns Democratic caucus delegates based not on the number of people who show up to the caucus, but on the total number of Democratic votes in the election 2 years prior.¹ But as you can see in the graph above, the proportion of young voters in the electorate falls off in midterm elections. This means that young voters, Bernie Sanders’s strongest demographic, were almost certainly systematically disenfranchised by the Iowa results.
How much was this effect? I made some conservative calculations. I assumed that the average young voter (18-29) was in a precinct with 1.5 times as many young voters (technically speaking: 1.5 times the odds ratio of being young) as the average voter’s precinct. (This is about would happen if, for instance, 1/6 of Iowa 18-29 voters were in college-town precincts where they were the dominant group). Then, I assumed that youth in the 2016 caucuses were about the same fraction of the electorate as in the 2008 general election. This would mean that the average youth vote would have counted 3-4% less than it should have. If that demographic went 83% for Sanders, and it was 17% of the voters, that would mean that the official totals were below the popular vote by about 0.4%; not a huge number, but bigger than Clinton’s margin of victory.
This calculation is pretty crude, as it does not account for the above-parity weight of non-youth voters, only for the below-parity weight of youth. However, I’d bet that Sanders’s popular vote overperforms his official total by 0.25%-0.65%.
In other words, Sanders probably won the popular vote, but by an even smaller margin than Clinton’s reported victory. The margin could be larger if youth turnout beat 2008 levels, their highest point in recent years.
Why does this matter? Am I just using this as an excuse for pie-fighting for my favored candidate?
No, this is bigger than that. Look at the right side of the graph above. That shows the true population breakdown of the electorate, as if all age groups turned out at the same rate (hopefully, something a lot closer to 100%). Clearly, that kind of turnout would be massively favorable to the Democratic party and to democracy as a whole; think of how much better 2008 was than 2010, and then add half that much again. This would be enough to cancel out the Republican gerrymandering advantage, and put the House of Representatives (and many state Houses and Senates) into play.
How can we get that kind of turnout? It’s not going to be easy. But I can tell you what kind of things would help:
- Not using outdated caucus allocations that systematically disenfranchise youth.
- Restoring and updating the Voting Rights act, to roll back onerous voter ID requirements and gerrymandering that disenfranchise People of Color and youth.
- Replace plurality voting and “first past the post” with approval voting and proportional representation so that all votes get counted and all votes matter equally. This would flip the incentives, so that politicians would prosper only by driving their own turnout up, not by driving the other side’s turnout down. (The first link above is to electology.org, a national charity doing good work promoting approval voting. They could use your help.)
- Pass the national popular vote interstate compact, to stop the distortion of the electoral college. Again, this would promote campaigns that care about all voters.
- Nominate the Democratic candidate who clearly does a better job inspiring disaffected voter groups such as youth: Bernie Sanders.
—————
¹I think this is correct. I have looked for a document explaining the allocation in full and I haven’t been able to find it. If I am wrong on this, please correct me.
Note: Data for graph from www.electproject.org and kff.org/… . Since those latter data use different age cutoffs from the former, I used linear interpolation for the “true population percentages” of each age group; this is probably off by a percent or two, but essentially correct.
Note 2: I put some work into gathering and processing the data for this diary. Two good links I didn’t end up using are these interactive maps of Iowa results in the Dem primary and in 2014. If you appreciate the wonkery, please consider recommending.