There is a habit investors pick up after a long run of easy money, and it makes rising interest rates far more unsettling than they need to be. They start to think of interest rates as a setting rather than a price. Something a central bank chooses, the way you might nudge the thermostat in the office when the room gets a little warm.
For the better part of 13 years after the global financial crisis, that habit was rewarded. Rates were low, then lower, then effectively zero. Every wobble in markets was met with a cut, a bond buying programme or a reassuring speech. It became natural to assume that if borrowing costs ever climbed too far, someone in authority would simply turn the dial back down.
September 2026 has been an uncomfortable lesson in why that assumption deserves a second look.
The Federal Reserve has delivered its first rate hike in three years. The Reserve Bank of Australia has lifted the cash rate to 4.60%, its fourth increase this year. The European Central Bank and the Reserve Bank of New Zealand have both tightened this month. And at the long end of the curve, where central banks have far less direct control, US government borrowing costs have pushed to levels not seen since before the global financial crisis.
The interesting question is not whether rising interest rates hurt investors in the short term. Of course they do. The more useful question is why they are rising, and whether anyone can genuinely make them stop.
The thermostat illusion
Central banks set short term rates. They do not set the price of 10 or 30 year money. That is determined in the market, by the willingness of investors to lend to governments and companies for long periods, and by the return they demand for doing so.
In mid August, the yield on the 30 year US Treasury bond closed above 5.3%, a 19 year high. By mid September, the 10 year yield had pushed to around 5%, its highest level since 2007.
Washington’s response is instructive. On 19 August the US Treasury announced it would at least double the maximum size of its long dated bond buybacks, from US$2 billion per operation to US$4 billion. The idea is simple enough: buy more long bonds, push their prices up, push their yields down. Long rates fell on the announcement and bounced back the following day. When the first enlarged operation was revealed in September, the maximum had been tripled to US$6 billion, and yields still moved higher.
The market, in other words, was not persuaded.
That is worth sitting with. When a government with the deepest bond market in the world steps in to lean on long term rates, and the market shrugs, it tells you something about the forces on the other side of the trade.
Rates are a price, and prices have a job to do
Strip away the jargon and an interest rate is simply the price of money. Like any price, it reflects supply and demand. When more people want to borrow and the pool of willing savers does not grow as quickly, the price goes up.
For most of the past two decades, the supply of capital looked almost unlimited. Global savings were abundant, inflation was low, and central banks were active buyers of government debt. The demand side was comparatively tame.
That balance has shifted. There are now three very large customers queueing at the same counter, and none of them looks likely to leave soon.
Three big borrowers at the same counter
The first is government. The United States is running a budget deficit of roughly 6% of GDP at a time when unemployment sits around 4%. Historically, deficits of that size have been a feature of recessions, not of reasonably healthy economies. US federal debt has now reached around US$40 trillion, and net interest costs are projected to exceed US$1 trillion this year, more than the country spends on defence. On top of rolling over its maturing debt, the Treasury needs to find buyers for roughly US$2 trillion of new net issuance.
Australia is far from that position, but it is worth remembering that very few developed governments are running surpluses. The queue of sovereign borrowers is long and international.
The second is artificial intelligence. McKinsey estimates that more than US$5 trillion will be spent worldwide on AI related data centres through 2030. Some of that will be funded with debt and some with equity, but every dollar draws on the same global pool of savings. Data centres, power stations, transmission lines and chip fabrication plants are capital hungry in a way that software never was. The digital revolution, it turns out, eventually needs concrete, copper and a very large cheque.
The third is energy and security. The conflict in the Middle East has pushed oil sharply higher, with Brent trading above US$107 a barrel in early September. Governments are rebuilding energy security, supply chains and defence capability at the same time. All of it requires capital, and much of it is inflationary in the short run.
Put those three together and you have a lot of demand for money chasing a supply that is not growing as fast. In that light, higher long term rates are not a malfunction. They are the price mechanism doing exactly what it is supposed to do.
Why the last era flattered almost everyone
It is worth being honest about what the period from 2009 to 2021 did for investment returns.
When the price of money falls for a decade, two things happen at once. Assets become more valuable, because future cash flows are discounted at a lower rate. And borrowing becomes cheaper, so anyone using leverage gets a second boost. Property, private equity, infrastructure, long duration growth stocks and highly geared business models all enjoyed this double tailwind.
Many of those returns were genuinely earned through skill. But some were simply the reward for owning long duration assets during a once in a generation fall in interest rates. When the tide of falling rates lifts every boat, it is very hard to tell good sailing from good weather.
The risk today is that investors are still navigating with charts drawn in that period. Valuation models, return targets and capital structures built for a world of 1% money do not automatically work in a world of 5% money.
Talking the price down rarely works for long
Markets have a long memory for governments that try to defend prices against fundamentals. Sometimes it works briefly. Rarely does it work durably.
Buying back long bonds may smooth market functioning and calm nerves for a day or two. But it does not reduce the deficit, it does not lower inflation, and it does not reduce the capital demands of the AI build out. Funding buybacks with more short term borrowing also shortens the maturity profile of the debt, which means refinancing more often at whatever rates prevail.
There is a broader lesson here for investors. When a policymaker treats the symptom rather than the cause, the symptom tends to return. The question to ask is not “will authorities step in?” but “does the step they are taking change the underlying supply and demand?”
That may prove an uncomfortable question for anyone counting on rescue.
What rising interest rates mean for portfolios
None of this is a call to panic. It is a call to recalibrate.
The hurdle rate has gone up. When a government bond offers around 5%, every other investment has to clear a higher bar. Shares, property and private assets need to justify their valuations against a risk free alternative that actually pays something. That favours businesses with real earnings today over those promising earnings a long way into the future.
Balance sheets matter again. Companies that borrowed cheaply on short tenors will have to refinance at higher rates. Those with modest debt, long maturities or net cash have a quiet advantage that does not show up in a single profit result but compounds over time.
Duration risk is everywhere, not just in bonds. Long dated growth stocks, highly geared property and infrastructure assets with floating rate debt all carry interest rate sensitivity. Investors who own a lot of these may be making a bigger interest rate bet than they realise.
Private market valuations deserve scrutiny. Assets that are valued periodically rather than priced daily can look stable while the discount rate underneath them is moving. Stability on a statement is not the same as stability in value.
Cash and fixed income are useful again. For the first time in a long time, defensive assets can do more than sit there. That gives investors patience, optionality and the ability to wait for better prices.
What it does not mean
It would be easy to take this analysis to a dramatic conclusion: sell everything American, buy gold, retreat to cash. That would almost certainly be the wrong lesson.
The problem described here is primarily a fiscal problem, not a corporate one. Many of the world’s best businesses are still listed in the United States, and the ingredients behind that, including deep capital markets, rule of law, innovation and a culture of reinvention, remain largely intact. The US dollar still accounted for around 57% of allocated official reserves in the first quarter of 2026. There is no obvious successor waiting in the wings.
Moving money elsewhere does not remove risk, it changes the type of risk. Other developed economies are running deficits too, many have weaker growth, and some carry more regulation and less scale. Selling quality assets to avoid a problem whose timing nobody knows can look like a mistake for a very long time.
Sensible global diversification is a good idea in any environment. Diversification as an act of fear rarely is.
The risks to this view
Forecasting interest rates is a humbling exercise, and there are several ways this thesis could prove wrong.
AI may turn out to be strongly disinflationary. If it lifts productivity materially, economies could grow faster with less inflation, which would ease pressure on both deficits and rates.
A recession would change the picture quickly. Weaker growth would reduce private demand for capital, central banks would cut, and long term yields could fall sharply, at least for a time.
Policymakers could return to large scale bond buying. Quantitative easing suppressed long term rates for years after 2008, and while today’s inflation backdrop makes that harder to justify, it is not impossible.
And it is worth remembering that confident calls for structurally higher rates were made repeatedly through the 2010s, and they were wrong for a decade. Humility is warranted.
The point is not that rates must keep rising. It is that investors should not build portfolios that only work if they fall.
What should long term investors actually do?
For long term investors, the practical question is not “where will the 10 year yield be at Christmas?” Nobody knows. The better question is “would my portfolio still make sense if money stayed expensive for several years?”
That means looking hard at valuation, preferring businesses that fund themselves, being honest about how much interest rate sensitivity sits across a portfolio, and treating defensive assets as a source of opportunity rather than dead weight.
It also means resisting two opposite temptations. The first is assuming that someone will always step in to make borrowing cheap again. The second is assuming that because the fiscal picture is worrying, disaster is imminent. Problems that are slow to build are usually slow to resolve.
Markets move in hours. Fiscal arithmetic moves in years.
TAMIM Takeaway
Interest rates are rising around the world, and attempts to lean against long term bond yields have so far failed to impress the market. That is because rates are a price, and the demand for capital from government deficits, the AI build out and energy security is enormous.
The market may be underestimating how long this can last, because a generation of investors learned to treat cheap money as the natural state of things. It was not. It was the exception.
For long term investors, the right response is neither panic nor denial. It is discipline: pay sensible prices, favour strong balance sheets, understand where interest rate risk is hiding, and stay globally diversified for the right reasons. Cheap money is unlikely to come back by decree. Portfolios should be built to cope if it does not come back at all.