Wide blue yonder Data clouds Biden’s midterm ray of hope

Since 1978 Ray Fair, professor of Economics at Yale University, has been using economic data to predict US election outcomes. His bare-boned, strictly-bythe-numbers approach has a fairly impressive record, usually coming within 3% of the final tally.

Sadly for Democrats – if Fair’s on track again this time – the Biden administration admin will struggle to keep control contro of Congress in November’s crucial crucia midterm elections.

Elections Ele are noisy events and this year’s y is no different. Recent polling pollin suggests Joe Biden is on a roll, re reclaiming some of the ground he lost los earlier in his presidency. The Democrats Demo have passed major legislation. There T has been a surge in women registering registe to vote after the supreme court overturned Roe v Wade. Abortion rights ri drove voters to the polls in deep-red dee Kansas. Petrol prices, if not overall ov inflation, are falling. In the meantime, meant Donald Trump and the candidates didate he has backed are dominating the headlines he and helping Democrats’ poll nu numbers.

But if Fair is right, we can largely set aside the t personalities and the issues: the economy ec is the signal behind the noise and a Biden is still in trouble.

Using Usin data going back to 1916, Fair’s latest analysis suggests that Democrats will w get 46.7% of the national vote in November Nov – down from the 51.3% in 2020 when w Biden defeated Trump and took c control of the House and a slim majority majori in the Senate.

Fair’s Fair model looks at the national picture, pictur he doesn’t dig down to state battles battle and won’t be drawn into more g granular prognostications. But given the gloomy economic picture in rec recent months, his prediction is unlikely unlike to improve before November and su suggests a loss in the House and a very tough to fight to keep control of the Senate. Senate When Fair’s last prediction was published publish in July, the Democrats’ share of the vote had fallen from 48.99% in October, Octob “due to two fewer strong growth growt quarters and slightly higher inflation”. inflati The economic malaise has only on deepened since then.

“This prediction is based on business as usual,” said Fair. “It’s based on estimations back to 1918, a hundred-plus years of data. In that period what seems to matter, election after election, is inflation, output, growth and the penalty you get for being the incumbent party in the White House.”

Fair will update his model before the election and, given its economic focus, Biden’s percentages are unlikely to improve. Inflation remains close to a 40-year high – soaring prices are now costing the average American household an extra $717 a month.

The US economy has shrunk for two consecutive quarters, a sign taken by many as a harbinger of recession. Interest rates are rising at their sharpest pace since the 1990s as the Federal Reserve fights to tamp down price rises.

The strength of the economic headwinds Biden faces are apparent even in his improving poll numbers. About 69% of Americans think the nation’s economy is getting worse – the highest percentage since 2008 – according to an ABC News/Washington Post poll.

Fair doesn’t think elections are only about the economy. “This is not a perfect story, there’s room for other stories in each election,” he said. Given the equations narrow, economic focus he said it was “reasonable” that people were now looking at what other factors might impact the Democratic vote share in the midterms.

One factor that may have skewed his results in the past, and could do again, is Trump. In 2016 Fair’s model predicted Hillary Clinton would beat Trump. She did win 2.9m more votes than Trump, securing 48.2% of the vote to Trump’s 46.1%. But she lost in the electoral college.

This time too Trump could be a factor, although he is difficult to measure. “There are many reasons why the Democrats may do better. Certainly Trump could be one of them,” said Fair.

But history – or at least the history that Fair measures – suggests for all the recent positive polling, the Democrats face an uphill struggle this November.

“How large is the error I make on average? It’s about 3 percentage points. If the prediction is 47, that would get you up to 50. So it’s a long shot that the Democrats would get more than half,” he said.

‘What seems to matter, election after election, is inflation, output and growth’