What a Mess in Iowa

It looks like we will never find out what the vote in Iowa really was. New York Times reporters found arithmetical errors in the sets of numbers set to the Iowa Democratic headquarters from over 200 precincts. But the lawyer for the Iowa Democratic Party says that the original vote tally sheets cannot be changed by law. In any case, it would be impossible now to fix the errors – the caucuses have long dispersed and there is nobody who could be sure how to correct them.

In this peculiar situation, where the results were so close between Bernie Sanders and Pete Buttigieg, the Associated Press declined to call a winner.

We could say, so what? The general results are clear: Bernie and Pete came out on top, with a significant lead over Elizabeth Warren, then further behind Joe Biden, and even further Amy Klobuchar. It is notable that many of the TV reports since then list only the top 4 candidates. Klobuchar’s Iowa race is being widely erased.

One aspect of the results is clear, even if the numbers may be inexact. Sanders won many more votes than anyone else. The caucus voting is a two-step process, because votes for candidates who do not reach 15% in a precinct get redistributed among the candidates who reached that threshold. Sanders received 24.7% of the first votes, against Buttigieg 21.3%, a difference of over 6000 votes. After the first redistribution, the difference had been narrowed: Sanders 26.5%, Buttigieg 25.1%. Most of the votes which had to be redistributed had been given to moderate candidates, like Yang and Steyer, and thus were more likely to go to Buttigieg than to Sanders or Warren.

When these votes were translated in over 1700 precincts into delegates, or as the Iowans have it, state delegate equivalents (SDE), which then get turned into delegates, Sanders and Buttigieg were essentially tied.

It’s easiest to see these shifts in a table:

Why did Sanders’ obvious lead in votes get turned into second place in delegates?

Delegates are apportioned automatically to each county, and rural counties are favored in that apportionment.

Steve Kornacki of NBC pointed out that in rural Shelby County, which went to Buttigieg, every 53 voters got one state delegate equivalent, while in Poweshiek County, home to Grinnell College, where young voters lean toward Sanders , it took 126 votes to get one SDE.

That is a normal feature of Iowa’s political structure. In 2016, in the big university and college precincts, it took over 200 votes to get a delegate, while in rural Fremont County, 45 voters got an equivalent. The smallest counties got the most delegates per person. Although Hillary Clinton was recorded as the winner in Iowa, it is likely that Bernie Sanders got more votes, but nobody knows for sure, because only the SDE’s were tabulated before this year.

Sanders was not the only one who was disadvantaged by this system. Comparing the numbers for the second vote and the SDE’s, Elizabeth Warren lost a couple of percent and Biden gained. She had run a relatively close third place with a big lead over Biden, but in the SDE’s, which is all that most media reported, she is a more distant third and Biden is close to her.

Many commentators, especially those who support the more progressive candidates Sanders and Warren, have noted that Iowa and new Hampshire are among the whitest states in America, and they have an outsized influence on the Democratic primary. They are also among the most rural states, and then Iowa’s system gives even more power to rural voters, who tend to be moderate.

That is part of the American political system. The Senate is the most obvious case of rural bias, where the smallest and most rural states get equal representation with the big urban states, like New York and California. That translates into rural bias in the Electoral College, which is why both Bush and Trump won the Presidency, while losing the popular vote.

In one scary calculation, it would be possible for Trump to lose by 5 million votes to a Democrat, but by very narrowly winning states like Arizona, Florida, North Carolina and Wisconsin, he could win in the Electoral College. Add to this legal, constitutional bias the illegal bias that Republicans have introduced by gerrymandering both Congressional districts and state legislative districts.

It is not possible to create a perfect electoral system, which gives the proper weight to each subgroup of the population: rural vs. urban, black vs. white, conservative vs. moderate vs. progressive. As soon as such a perfect system were created, population movement would throw it out of whack. But America could have a much better system of translating votes into power. The most obvious change would be to get rid of the Electoral College, which has been proposed many times. In 1969, after George Wallace had received 46 electoral votes, Emanuel Celler of New York proposed abolishing the Electoral College in favor of a purely popular vote. That passed the House 339-70, and President Nixon said he supported it. The Senate Judiciary Committee passed the bill to the full Senate by a vote of 11 to 6. But the bill was filibustered by a few Democrats and Republicans, and although it had majority support, there was not a two-thirds majority needed to end the filibuster.

So any candidate whose support lies with more progressive urban voters is at a significant disadvantage. A progressive Democrat could beat Trump handily and still lose.