The short version
Americans vote on 3 November in the US midterm elections. Every seat in the House of Representatives is up, a third of the Senate and 39 governorships.
Prediction markets have shifted significantly over the last six months, but currently favour the Democrats to take the House and to narrowly take the Senate.
But those are odds, not results.
Almost everything you’ll read between now and election night is a forecast rather than a fact.
For Australian investors, the result itself matters less than two things that follow it: the impact on shares, in particular the US share market and the impact on the Australian dollar.
The second one is the part most people forget.
What’s actually being voted on
Midterms are the elections held halfway through a president’s four-year term. They are almost always seen as a referendum on the sitting president because the outcome determines how much leverage he has to pass his agenda.
Three things are on the ballot.
The House of Representatives. All 435 seats. The House is what controls all revenue and spending bills. If the opposition controls it, the administration cannot fund new programs without bipartisan support.
The Senate. 35 of the 100 seats. Including two special elections in Florida and Ohio. Senators serve six years, so only about a third face voters each cycle. The Senate confirms appointments, including Supreme Court justices and has to agree with the House for a law to pass.
Governorships. 39 of them. At the State level, they are what control the practical economic decisions investors care about, including whether a new data centre gets approved and what it pays for power. They also set corporate tax credits, right-to-work rules and municipal bond priorities.
The presidency itself isn’t on the ballot. That matters because a lot of what has moved markets over the past two years, such as tariffs, trade deals and deregulation, has come from executive action, not laws. Congress doesn’t have to agree for that to keep happening.
The three outcomes, and what each would change
Republicans currently control both chambers, holding the Senate 53 seats to 47.
As of mid-October, prediction markets put Democrats at around 90% to win the House and around 63% to win the Senate. That Senate figure was 44% in March and 57% in September, so it has climbed steadily. The gap between those two numbers comes down to a handful of toss-up Senate races in Maine, Ohio, Iowa, Kansas and Texas, any of which could land either way. Polling has Democrats ahead by somewhere between 5 and 11 points nationally, and the president’s net approval rating sits at about minus 27.
Which means the market’s base case is no longer gridlock. It’s a Democratic sweep.
A Democratic sweep. Democrats win both chambers, currently the favoured outcome. Attention would shift to healthcare, cost of living and consumer protection, and it would likely bring more scrutiny of data centres and artificial intelligence companies. It would also mean a Republican president facing a Congress controlled entirely by the other party.
A split Congress. Democrats take the House, Republicans hold the Senate. This is what is called a “split or divided government”, one party holds the White House while the other holds at least one chamber of Congress. In plain terms: less new legislation gets passed. Only the essentials get through: budgets, and extensions to things like the farm bill and transport funding. Gridlock also restrains spending, which eases upward pressure on US government bond yields and on inflation, but offers less support to growth and company earnings. You don’t get one without the other.
Republicans hold both. The outcome markets currently rate least likely. It would mean more room to pass the agenda through legislation rather than executive action alone.
Here’s the part that gets less airtime. Of all the possible combinations of president and Congress, a Republican president facing a fully Democratic Congress is the one historically associated with the weakest short-term share returns, roughly 1% below the average three-month return across all periods.
That’s a small effect, over a short window, drawn from very few examples. But it describes the outcome markets now think most likely, and it cuts against the cheerful “markets love gridlock” line you’ll see quoted elsewhere.
What history says, and why it’s a weak guide
Two patterns come up every cycle.
The first: the president’s party usually loses seats. It has happened at almost every midterm since the Second World War, with 1998 and 2002 the exceptions. This cycle looks like a continuation rather than an exception. Independent seat counts have Democrats at or near the 218 needed for a House majority before the toss-up races are even counted.
The second is the one you’ll see quoted everywhere between now and November. Since 1950, the S&P 500 (the index of the 500 largest US listed companies) has been higher 12 months after every midterm election. Nineteen out of nineteen. The average gain has been around 15%.
Each of those nineteen periods had its own interest rate cycle, its own recession or recovery, its own war or boom. Pulling a “midterm effect” out of that is close to impossible. And the caveat that comes with it is well established: political calendars don’t drive markets on their own, and no single statistic should be read in isolation.
But here’s the more important caveat, valuations matter.
None of those nineteen runs began from valuations as stretched as today’s. On one widely used long-run measure, which compares share prices to a decade of earnings, US shares are more expensive than at any time outside the 1999 to 2000 technology bubble. If valuations compress evenly slightly from these levels, that 15% baseline could get dragged down by a couple of percentage points.
That doesn’t mean the historical streak breaks. It means the previous nineteen all started somewhere different from here.
There’s one more unknown worth naming, and it isn’t who wins. Close races can take days or weeks to call, and a delayed or contested result is the scenario most likely to produce short-term volatility. Not because of the outcome, but because markets don’t like not knowing.
One last thing, inflation is largely out of Congress’s hands either way. A lot of the current pressure on prices comes from energy supply after the conflict in the Middle East, and no congressional majority fixes that. Interest rates are set by the US Federal Reserve, which operates independently of whoever controls Congress.
What to watch after the result
Two things will matter in the weeks and months after November 3rd, and the first arrives before the new Congress even sits.
The lame-duck session (November – January)
The weeks between the election and the new Congress being sworn in during January are when the outgoing Congress can still pass legislation. The expectation is a push on defence spending tied to the Middle East conflict, plus agriculture and infrastructure funding, worth up to around 0.5% of US economic output. If that passes, it’s a short term boost to growth. If it stalls, that money doesn’t get spent, The US economy that has been leaning on government spending since the pandemic, faces a drag on growth instead.
The debt ceiling (2027 onwards)
The debt ceiling is the legal cap on how much the US government can borrow. It currently sits at US$41.1 trillion and is expected to be reached sometime in 2027. Raising it needs Congress to agree, and a Democratic House could use that as leverage with the US credit rating sitting in the background. Drawn-out fights over the debt ceiling have moved markets before.
The part most Australians forget
So the election outcome determines policy direction, and the debt ceiling determines whether Congress cooperates or fights. Both move markets. But for Australian investors, the numbers that matter most aren’t decided by either. They’re decided by currency.
When you hold US shares, you hold them in US dollars. So your return has two moving parts: what the share price did and what the exchange rate did. That second part is called currency exposure, and unless it’s hedged, it cuts both ways.
Say Democrats sweep and US equities fall 3% (the historical average for this outcome). But a Democratic sweep means policy uncertainty eases, so the AUD strengthens 2% against USD. Your unhedged US holding is down 5% in AUD terms.
Reverse the scenario: Republicans hold both chambers, equities rise 8%, but the AUD weakens 3% (risk-off sentiment). Your unhedged return is +11% in AUD terms. Same portfolio. Vastly different outcomes. And none of it depends on which outcome you guessed right, it depends on whether equities and currency moved in the same direction.
So what does the election do to the exchange rate? Less than the headlines suggest. One view is that a divided Congress could reduce policy uncertainty and support the US dollar, which would quietly help anyone holding unhedged US shares. But the Reserve Bank, the US Federal Reserve, iron ore and China all have more say in the Australian dollar than the makeup of Congress does.
That’s not a reason to ignore the currency. It’s the opposite. The exchange rate will do more to your US returns over the next year than the election will. It just won’t be the election driving it.
Plenty of Australian investors are already making this call. In a single month earlier this year, they pulled around $460 million out of the main unhedged S&P 500 fund listed on the ASX while putting $232 million into its currency-hedged equivalent. Same index, same 500 companies. The only difference is who carries the currency risk.
That shift had nothing to do with the midterms. It tracks the things that actually move the Australian dollar: commodity prices, the gap between Australian and US interest rates, and what each central bank is expected to do next. Which is rather the point.
That’s the live decision for an Australian with US exposure, and it has nothing to do with who wins on 3 November. Hedged gives you the share return without the currency swing, usually for a slightly higher fee. Unhedged gives you both, which helps when the Australian dollar falls and hurts when it rises.
It’s worth knowing how much of this applies to you. US-listed holdings have been creeping up in Australian portfolios, from 11.9% to 12.7% over the past financial year on one measure, with exchange-traded funds now a bit over a quarter of portfolio value. The pull isn’t really about past returns. Those are the result, not the reason. It’s access: the companies driving the artificial intelligence build-out are almost all listed in the US, and there’s no close equivalent on the ASX.
As for the ASX itself, the read-through is indirect. US spending and trade policy feed into global growth, global growth feeds into commodity demand, and commodity demand feeds into the big miners. That’s a long chain with plenty else attached to it.
Two smaller knock-ons if you hold more than shares. Gold has had political risk priced into it for a while, so a calmer and more predictable Congress could take some of that support away. And for bonds, less political uncertainty helps at the margin, but with US inflation still above target and the government needing to issue a great deal of debt, long-term yields may not fall much either way.
The sectors in the conversation
What gets argued about during a campaign is a fair guide to what gets argued about for the two years afterwards. Four themes are doing most of the work this cycle.
| Theme | What’s being argued about | Where it shows up |
|---|---|---|
| AI and data centres | Electricity prices, water use and local jobs. The governors of Texas and Pennsylvania have both announced measures aimed at speculative projects. These look incremental rather than project-stopping, and new builds may simply drift toward friendlier states. | Power utilities, engineering and construction firms, and the chipmakers supplying the build-out |
| Healthcare | Affordability. This becomes the main focus if Democrats take both chambers. | Health insurers and drug makers |
| Trade and tariffs | Mostly set by executive action, so the election may change less here than the campaigning suggests | Importers and retailers |
| Defence | Possible year-end spending tied to the Middle East conflict | Defence contractors |
Artificial intelligence is the one worth watching closely, because expectations around it are currently a major driver of US share prices and the industry is largely self-regulated. There are four plausible ways it could go from here: a Democratic sweep brings scrutiny of data centres on affordability and safety grounds; an accident or scandal forces regulation sooner; it becomes a 2028 election issue with bipartisan action arriving in 2029; or national security concerns about competing with China win out and the touch stays light. Four paths, no way to pick between them, which is itself the useful thing to know.
A sector being in the news is not the same as a sector being a good investment. By the time a theme is on the front page, a lot of it is already in the share price.
Most of these sectors are reachable from Australia, directly or through a fund. But reachable isn’t the same as worth reaching for, and an election result doesn’t settle that either way.
So where does that leave investors?
An election is one of the few market events with a date printed on it years in advance. Everyone knows 3 November is coming. Traders, fund managers and prediction markets have been positioning for months. Whatever is widely expected is already in prices to a large degree.
That doesn’t mean nothing moves. It means the move comes from the surprise, and nobody knows in advance what the surprise will be.
There’s no universal answer here, but there is a way to work out which question applies to you.
If you’re hedged, the election is largely someone else’s problem. Your US shares move on US earnings alone. You’ve locked in the currency at a specific rate, so election-driven AUD moves don’t touch you.
If you’re unhedged, you’re carrying two positions at once: the shares, and a view on the Australian dollar. The election is a small input to the second. The bigger question is what happens if the result lands outside what’s priced, such as a contested count or a surprise in the Senate. That’s when both parts can move together, in the same direction.
For someone investing over decades, a US midterm is a blip on a chart they won’t remember. For everyone else, 3 November is a reasonable prompt to check which of those two positions you’re actually in.
