Democrats’ chances of gaining control of Congress at US midterms depend on Iran war

In Nate Silver’s forecast model, Democrats are more likely than not to gain control of the US federal Senate at November 3 midterm elections, and strongly favoured to gain control of the

House of Representatives. Donald Trump’s net approval is close to -20.

If disruption caused by the Iran war continues until the midterm elections, with increased petrol prices and inflation, Republicans are likely to lose control of both the House and the Senate, possibly with huge losses.

However, if the disruption eases again and petrol prices fall back, I believe Republicans will do better than they are now and would be favourites to retain control of the Senate.

I can say this because Trump’s net approval and the generic ballot were better for Republicans in early September, before the current fighting began.

Trump’s ratings

In analyst Nate Silver’s aggregate of US national polls, Trump’s net approval is -19.7. Currently, 58.2% disapprove of his performance and 38.5% approve.

Trump’s net approval was -20.4 in my August 12 article. It had improved to -18.7 on September 4 before falling back.

Trump’s net approval is below what any past president since Harry Truman had at this point in their term, with Trump’s first term the closest at -13.1.

On four issues tracked by Silver, Trump’s net approval is -12.2 on immigration, -26.1 on trade, -30.8 on the economy and -46.4 on inflation. Since my August 12 article, Trump’s net approval on trade and inflation have dropped 2–4 points, while his net approval on immigration has been relatively stable. His ratings on the economy had improved before another slump.

Silver also has an aggregate of US support for the Iran war. Currently, 55.7% are opposed to the Iran war while 35.6% support it, for a net support of -20.1, down slightly from -19.6 in my August 12 article.

Midterm elections on November 3

All 435 federal House of Representatives seats and 35 of the 100 senators will be up for election at November 3 midterm elections.

In 2024 elections, Republicans won the House by 220–215 over Democrats and the Senate by 53–47. Owing to Vice President JD Vance’s potential tie-breaking vote, Democrats need at least a four-seat gain to win Senate control.

I previously wrote that, owing to Republican gerrymandering, Democrats need at least a four-point margin in the generic ballot to have a good chance to win the House majority. The Senate is more skewed to Republicans as every state has two senators. Democrats need a double-digit margin in the generic ballot to win the Senate on a uniform swing.

In Silver’s midterm election forecast model, the generic ballot gives Democrats a 50.8–42.7 lead over Republicans, an 8.2-point margin (8.1 points in my August 12 article). Democrats’ lead had dropped to 7.4 points on September 4.

Polls that screen for likely voters usually give Democrats larger leads than polls that ask registered voters. Democrats are more likely to vote at lower turnout elections like the midterms than Republicans, when Trump is not on the ballot.

Silver’s model gives Democrats a 57% chance to win the Senate and an 86% chance to win the House, the same chances as on August 12. Democratic chances to win the Senate had fallen to a low of 50% on September 7, and their chances of winning the House had dropped to 81% on September 3.

While uniform swing indicates that Democrats need a double-digit margin in the generic ballot to win control of the Senate, Silver’s model has Democrats as the favourites to win with a 6–7 point margin in the generic ballot. In the House, Democrats are favourites if they win by 3–4 points.

Democrats have effectively gained a House seat in Missouri after courts annulled a Republican gerrymander. That gerrymander had given Republicans a 7–1 advantage from Missouri’s eight House districts, but Republicans now only have a 6–2 advantage.

Democrats have leads of one to four points in Senate contests in Texas, Ohio and Alaska and are tied at 45.8% each with Republicans in Iowa. Trump won all these states by double-digit margins in 2024. This explains why Democrats are the slight favourites to gain control of the Senate.

In Democratic-held Michigan, left-wing Democrat Abdul El-Sayed trailed Republican Mike Rogers by 1.6 points in my August 12 article, but he now leads by 2.9 points.

The two key items that could go wrong for Democrats are a markedly improved economy by November for most Americans – not just the stock market – and a systematic error in which Republicans overperform the polls.

Economic data

The US August jobs report (released September 4) was good news for Trump. In the establishment survey, 162,000 jobs were added, and June and July numbers were revised up a combined 55,000, turning July from 23,000 jobs lost to 21,000 gained.

In the household survey, the unemployment rate was unchanged since July at 4.1%, but both the participation rate and employment population ratio were up 0.2%, to 61.6% and 59.1% respectively. In July, the employment population ratio had dropped to its lowest since September 2021.

While August was an improvement, the US employment population ratio is still far lower than in Australia, where it was 63.9% in July.

In the August inflation report (released last Friday), overall inflation increased 0.4% and core inflation increased 0.3%, compared with 0.1% and 0.2% increases respectively in July. Annual inflation rates were respectively 3.4% and 2.4%.

Real hourly wages decreased 0.1% in August, but real weekly wages were up 0.2% owing to an increase in weekly hours. In the last year, real hourly wages are down 0.3%, but real weekly wages are up 0.3%.

July was the first month since January where personal income exceeded expenditure, mainly due to a drop in expenditure. The personal savings rate increased 0.3% since June to 3.0% after falling in every month from February to June.

The benchmark US S&P stock market index was up 0.9% in last Friday’s session, breaking a run of four straight losing sessions. The S&P is down 1.8% from its August 13 peak, but up 10.2% from its peak in the week before the Iran war began in late February. It’s up 16.2% from its worst point in the Iran war in late March.

Adrian Beaumont does not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and has disclosed no relevant affiliations beyond their academic appointment.