Philippa Kaur

Available data suggests the Democratic Party will win ten seats and gain control of the House of Representatives in the midterms. More

The resounding victories in recent elections by Democrats Zohran Mamdani in New York, Abigail Spanberger in Virginia and Mikie Sherrill in New Jersey has reinvigorated the party after a dismal year since Donald Trump became president.

The victories were not a mandate for a sharp ideological shift to the left. This may be true for Mamdani, but it is not for Spanberger and Sherrill, since both are mainstream centrist Democrats. The main reason for the victories can be seen in the chart below.

Trends in presidential job approval and Donald Trump’s handling of the economy 2025:

Paul Whiteley/YouGov, Author provided (no reuse)

The data comes from successive polls in the United States conducted by YouGov on behalf of the Economist magazine. All three candidates focused on the issue of the US economy which proved to be a winning strategy since it is clear the economy strongly affects Donald Trump’s job approval ratings.

As the president’s ratings on the economy decline, so does his job approval ratings. The result is that the Republicans took the blame for failing to deal with the issue.

The midterm Congressional elections in the US are due to take place in November 2026. Given the strong relationship between the economy and support for the president, it is interesting to examine how the economy is likely to influence support for the Democrats in those elections.

To investigate this, we can look at elections to the House of Representatives over a long period, given that they occur every two years.

The graph below compares the number of House seats won by the Democrats and economic growth in the US in all 40 House elections since 1946. Economic growth is weighted so that the Democrats benefit from high growth when they control the House but are penalised by this when the Republicans are in control.

This also works in reverse with low growth producing a poor electoral performance for the party when Democrats are in charge and a good performance when the Republicans are in control.

The relationship between economic growth and House seats won by Democrats 1946 to 2024:

Federal Researve Bank of St Louis/Paul Whiteley, Author provided (no reuse)

The impact of the economy on voting in these elections is clearly quite strong, but the number of House seats won declines as the party’s majority gets larger. This is what is known as a “ceiling” effect meaning that when the majority is very large it is difficult to win more seats even in a thriving economy.

But this relationship can nonetheless be used to develop a forecasting model of the seats likely to be captured by the party in midterm elections next year.

When forecasting seats, an additional factor to consider is the inertia of party support over successive elections. If the Democrats did well in one year, they were likely to do well two years later.

For example, in 2008 when Barack Obama won the presidential election, the Democrats captured 233 House seats and the Republicans 202. In the following midterm election in 2010 the party won 257 seats while the Republicans won 178 and so the Democrats retained control of the House.

At the moment the House has a Republican majority of 219 against 213 Democrats. So Republican control is quite vulnerable to a surge in support for the Democrats.

Multiple regression analysis

The forecasting model involves a multiple regression analysis. This uses several variables to predict the behaviour of a specific variable – in this case the number of House seats won by the Democrats.

In addition to the two variables already mentioned, approval ratings and the performance of the economy, the fact that the incumbent president is a Republican is included in the modelling as well since this influences the vote for the Democrats.

We know the number of House seats from the 2024 election and the fact that Trump is a Republican, so to forecast Democrat House seats we need a prediction for economic growth in 2026.

The Federal Reserve Bank of St Louis provides data which forecasts growth in the US economy up to 2028. It predicts that growth in real terms will be 1.8% in 2026 – and when this is included in the modelling, the overall forecast from these variables is 80% accurate.

If a variable is a perfect predictor of House seats it would score 1.0 and if it failed to predict any seats at all it would score 0. The impact of growth on seats when the Democrats controlled the House was 0.75, the inertia effect of past Democrat seats was 0.26 and Trump’s presidency was 0.19.

Low growth boosts Democrats’ prospects

Clearly economic growth dominates the picture showing that low growth rates next year will strengthen the Democrat challenge. This is likely to happen since a recent IMF report suggests that US growth is likely to slow next year.

Actual and predicted House seats in elections 1946 to 2026:

The Presidency Project/Paul Whiteley, Author provided (no reuse)

The third chart shows the relationship between Democratic House seats predicted by the model and the actual number of seats won by the party. The predictions track the actual number of Democrat House seats fairly closely and so the forecast should be reasonably accurate

It should be noted that all forecasting models are subject to significant errors. As the chart shows, the predicted number of seats is not the same as the actual number and if something unforeseen happens the predictions could be wrong. That said, however, the forecast is that the Democrats will win 223 seats – an increase of ten over their performance in 2024. This will give them enough to hand them control of the House. More

More than half of respondents in a new survey said they were concerned about US interference in Australia, a jump of nearly 20 points since 2021. More

Australians remain supportive of the US alliance, but they are viewing it much more critically than before.

And many are more concerned about American behaviour under the Trump administration, while softening their views somewhat on China.

In a new poll of 2,045 people conducted by Australia-China Relations Institute at the University of Technology Sydney, 54% said they were concerned about US interference in Australia, a jump of nearly 20 points since 2021.

This narrows the gap with the level of concern about interference from China and Russia, which has steadied around 64%.

And for the first time in the five years we’ve been conducting this poll, more people think the United States (57%) rather than China (51%) would force Australia to pick sides in the rivalry between the two.

This is a striking shift that shows Australians are aware pressure can come from allies, as well as rivals. Nearly two-thirds of respondents think the second Trump presidency would make conflict with China more likely.

Even views of economic behaviour have flipped. For the first time, more Australians believe the US uses trade to punish countries politically (72%, up from 36% last year) than China (70%).

Rising support for defence spending

Broadly speaking, the poll shows Australians’ views of China have softened since 2021. Concern and mistrust remain widespread, but have eased.

Two-thirds of Australians see China as a security concern, though this is the lowest level in five years. Mistrust of the Chinese government has also fallen, from 76% in 2021 to 64% today.

And yet, regional flashpoints remain a focus. The South China Sea is seen as a major source of tension in the region, with 72% of respondents saying China’s actions there threaten Australia’s interests.

Most back cooperation, including joint patrols, with partners like the Philippines, Japan and the US, to maintain stability.

This heightened sense of risk continues to shape how people think about defence. Support for higher defence spending has reached 72% – the highest its ever been in our poll – though it drops considerably when trade-offs such as health or education spending are mentioned (55%).

Half of Australians think the plan to buy nuclear-powered submarines under AUKUS, the defence pact with the US and the United Kingdom, will make the country safer, while only one in four disagrees.

The USS Minnesota, a Virginia-class fast attack submarine, off the coast of Perth in March 2025.
Collin Murty/AFP Pool/AAP

Views on Taiwan have remained steady. Just 37% of Australians would support sending troops to defend Taiwan if China attacked, with most preferring neutrality or non-military engagement.

In a scenario where the US was drawn into a conflict with China over Taiwan, opinions are evenly split: 50% would favour Australia staying neutral, while 47% would back supporting the US.

China policy influencing more people’s votes

Overall, though, the softening of Australian views towards China signals people are seeing the need to balance the country’s values and interests when it comes to its number one trading partner.

Australians continue to see the economic relationship with China as both vital and risky.

Seven in ten respondents now say Australia should continue building ties with China, up ten points from last year.

And concern about over-dependence on China has dropped from 80% in 2021 to 66%, while those who see the relationship as representing “more risk than opportunity” has fallen from 53% to 39%.

Yet, Australians continue to support certain guardrails to safeguard the economy and national security.

Support for the federal government’s decision to end the Port of Darwin lease, held by the Chinese company Landbridge, is strong at 75%.

And two-thirds of respondents favour limiting all foreign investment in critical minerals, with even higher agreement (74%) when the question refers specifically to China.

An overwhelming majority (82%) draw a clear distinction between their views of the Chinese government and Australians of Chinese heritage. Yet, suspicion persists beneath the surface.

Around four in ten (38%) believe Australians of Chinese background could be mobilised by Beijing to undermine Australia’s interests and social cohesion, while 28% disagree, a durable minority view over five years.

Australians are clearly paying closer attention to how their leaders manage relations with China, too. Foreign policy has rarely shaped how Australians vote, yet 37% of respondents said China policy influenced their vote in the 2025 federal election, up ten points from the previous election in 2022.

Taken together, the findings from our survey show Australians have become more comfortable managing a complex and often tense relationship with China, seeing it as both an economic partner and strategic competitor.

And for the first time this year, this measured outlook extended to the United States, which is now seen as both ally and source of pressure.

The result is a more assured national mood, one that is realistic about risk yet confident in Australia’s ability to steer its own course in a contested world. More