Donald Trump’s Iowa victory in 2016 was a big surprise. Not that he won. Several polls showed him with a lead, and George W. Bush won the state in 2004 and lost by a sliver in 2000, so it’s not like Iowa was part of the supposed Blue Wall. It was the margin. Nine points, 11 better than Trump’s national popular vote percentage. The best performance by a Republican since Richard Nixon in 1968.
Clearly, Iowa is Trumpy. Or is it? Current polls there are showing him barely ahead of Joe Biden. Should we believe the surveys? Overall, Biden’s polls are ahead of Hillary Clinton’s 2016 results, but among competitive states, his Iowa improvement is about double the average. I think this is legit. And if anything, the numbers are underselling Biden’s odds. Here’s why:
2016 Results: Donald Trump 51.15%, Hillary Clinton 41.74%, Gary Johnson 3.78% (R +11)
As safe as Trump’s margin looks, Jimmy Carter lost 10 points in Iowa between 1976 and 1980. It’s tempting to dismiss this because Carter was so unpopular during his re-election campaign. But Dwight Eisenhower gave up 9 points in Iowa between 1952 and 1956, despite doing better nationwide in his re-election.
FDR went from winning Iowa by 12 in 1936 to losing it by 4 in 1940. His national win was narrower in 1940, but he lost ground faster in Iowa. This state changes it’s mind faster than others. Unlike Hawaii, which tends to take better care of incumbents than challengers, Iowa sours on candidates the longer it gets to know them.
2020 Polls: Trump is currently leading by an average of 1.6 points. The single post-convention survey has him up 2. Each of the surveys since May are showing things between Trump +3 and Even, except for a single outlier (Biden +6.) All evidence points to a close contest.
Key Historical Shifts: Iowa started off red. Very red. From the founding of the GOP in 1856, the state voted Republican in every election until 1912. The first time a Democrat did better in Iowa than the national popular vote was 1972. You can think of Iowa as having three phases.
Phase One (1856-1908): Uniformly pro-Republican. Not only did the GOP candidate win each time, but often by very large margins. In several elections, Republicans easily cleared 60%.
Phase Two (1912-1968): Still a red state, but less consistently so. While Iowans never supported Democrats more than the national popular vote, in several instances (1932, 1948, 1956, and 1964) it was very close to neutral. Some Republicans won by a lot (Warren Harding broke 70% in 1920), but when Democrats won big nationally (1912, 1932, 1936, 1964) they won Iowa too. Harry Truman’s emphasis on the farm vote in 1948 got him a narrow win in a close national election.
Phase Three (1972-2012): A little to decently blue most of the time. Winnable for Republicans in a great year (1972, 1984), or in a close election if the candidate calibrated his message very well (2004.) In 9 of the 11 elections, Democrats did better in Iowa than nationally. The bluest Iowa year was 1988, when Michael Dukakis did 18 points better there.
Two things were universally true over the whole 150+ year stretch. First, Iowa was gradually getting more blue. If trends would continue on that path, by 2050-2070 you’d expect it to be one of the bluest states in the country. Second, frequent shifts of 15 to 20 points from one election to the next, even if one or both of the candidates are the same.
The +11 result for Trump in 2016 was either a shift to a fourth phase or a blip. We’ll know for sure in another couple elections. If the 2020 polls are accurate, is the former. Even if Biden wins, he’s not going to run ahead of his national popular vote in Iowa. It’s possible the state is now red again and is also souring on Trump as it has many previous incumbents.
How Biden Could Improve on Hillary’s 2016 Result: He’s going to. The question is if it’s enough to win the state. Biden is polling better than Hillary did with older white voters without 4 year degrees. Much of Iowa fits that very description. When Iowa turns sharply against the incumbent party, economic reasons are usually key. The post-WW1 collapse in farm prices drove Harding’s almost 50 point winning 1920 margin. The depths of the Depression boosted FDR to a record result for a Democrat in 1932. The anti-Republican vote of 1988 wasn’t just love for Dukakis. The state struggled in the 80s and didn’t enjoy the same Reagan Era prosperity as most of the rest of the country.
The past term hasn’t been a catastrophe for the Hawkeye State, but soybean farmers were harmed by Trump’s mini trade war with China. New manufacturing jobs haven’t appeared, and the state is currently struggling with the worst Covid numbers in the country.
How Trump Can Improve on His 2016 Result: He can’t. If most of Gary Johnson’s voters were Republicans who couldn’t bear to vote for a Democrat, as opposed to specifically unable to vote for Clinton, then maybe Trump gets a few more votes to offset other losses. But he maxed out in Iowa in 2016. Both in his winning margin and how it compared to his national popular vote.
Forecast: FiveThirtyEight thinks Trump is a 2:1 favorite. That’s based in part on the assumption the race will narrow a little, and the rest on Trump’s small polling lead. If Biden wins nationally by the 7 to 8 points he’s ahead now, I think he’ll take Iowa too. Either way it’s going to be close.
I also think Iowa is red again now. Or at least pink.