Australia's neutral rate is going up, and I think it stays up. Two charts.
Chart one: two proxies for nominal r*. The orange line is trend real GDP per capita growth plus my model of inflation expectations, the Wicksellian anchor, now 3.23 and still drifting down. The blue line is the AOFM 5y5y risk-neutral forward, the bond market's guess at where the cash rate settles, now 3.87 and climbing since 2020. They disagree by over half a point and the gap is widening.
Four numbers rarely line up the way they have this year. Growth sits above potential. Unemployment sits below the non-accelerating inflation rate of unemployment (NAIRU). Inflation sits above the target band. And the cash rate sits above the neutral rate, r*, but still short of what a standard Taylor rule would prescribe. Put together, these estimates describe an economy running hot while policy leans against it, but not hard enough.
Headline GDP growth through the year is running at 2.14%, although most of that growth happened in the tail end of 2025. Growth in the first half of 2026 was much more modest.
When a central bank moves the cash rate, it’s acting on a specific model of how the economy works. That model is the New Keynesian synthesis, and it’s the closest thing modern economics has to an operating manual for monetary policy. Every inflation target, every rate decision, every line in a central bank statement about “returning inflation to target” comes out of this framework. Understanding it is the difference between watching the news cycle of rate rises and cuts and actually seeing the logic behind it.
This isn’t a history of how these ideas came together, nor an account of what exactly got synthesised between classical and Keynesian economics. It’s a walk through the intuition: how the pieces of the model fit together, and where they hold up worse than the theory suggests.
Every economy has a speed limit. Economists call it potential output, or Y*. It represents how much an economy can produce when its workers, factories, and capital are all fully but sustainably employed. Actual output (also known as GDP or Y) moves around this limit. The difference between the two, Y minus Y*, is called the output gap. Sometimes it’s negative and Y sits below Y*, and resources go idle. Sometimes it’s positive and Y pushes above Y*, and something has to give.
Money for long-term borrowing moves freely across borders, so liquidity is a global phenomenon. QE bought bonds from asset holders and paid in cash. With this cash they bought other assets, everywhere and in every class. Asset prices rose worldwide, including in countries that never ran a QE programme of their own.
That was inflation, in a market the consumer price index does not cover. And bidding an asset's price up is bidding its yield down, so the same event was a global fall in borrowing costs.
None of this began in 2009. Bernanke saw low rates in 2005 and diagnosed a global savings glut as the cause, favouring a structural explanation over a liquidity one.
Most econometric models that estimate the neutral rate are fitted to consumer prices, output and interest rates. Asset prices are not among the inputs. So the figure they report is the rate that is neutral for groceries, and it has nothing to say about whether borrowing was running ahead of the capacity to service it.
There is a second problem underneath. The models infer the rate from two relationships that a decade of successful stabilisation had flattened, leaving the history of the policy rate itself as the main thing to lean on. Policy rules then took that estimate as an input, and naturally recommended something close to the policy already in place.
Estimation needs variation, and stabilisation exists to remove variation. A central bank that succeeds gradually erases the evidence it needs to keep doing its job.
The liquidity story does the work of several. It explains why asset prices boomed where no programme ran, why consumer prices stayed quiet and then surged in 2021 when the same mechanics reached households instead, and why every country's estimate fell together despite different demographics, deficits and growth.
For a decade, the rate that would have kept borrowing serviceable sat far above the one the models report. That is not a measurement problem. It is a specification problem, and a specification problem returns a confident number to the wrong question.
Today's blog post is very technical. We are talking about which variable should be the focus of a central bank when it sets interest rate policy. Most of the world's central banks use an inflation target, typically around 2 per cent, although Australia has a 2 to 3 per cent target band and within that it targets the 2.5 per cent mid point. An alternative which is often promoted is nominal gross domestic product (nGDP) targeting, where the bank aims to maintain a steady growth rate or path for nGDP.
Australian productivity has gone backwards since 2019. Unit labour costs are running ten points above the United States from a common base. The goods trade balance in trend-terms has fallen from above 13 billion dollars a month at the 2022 peak to barely 3 billion, with the latest print briefly negative. Inflation has reaccelerated above the target band on every measure. By any fundamental read, the Australian dollar should be weakening. It isn't. It has rallied from briefly below 0.60 last year to above 0.72 now and the trend is looking up.
The answer is the interest-rate differential. Everything else is detail.
TL;DR: r-star, the real neutral rate of interest, is one of the unobservable star variables in mainstream macroeconomics. It is notoriously hard to estimate. The Australian IS curve coefficient on the real-rate gap is small relative to macro noise, so the model cannot choose between a growth anchor, a bond-yield anchor, or a blend of the two. Those three approaches, drawing on Wicksell, imply a real r-star between roughly 1.5 and 2.6 per cent, and a nominal neutral cash rate between about 4.0 and 5.1 per cent. With the RBA policy cash rate at 4.35 per cent, policy is not extreme. The data alone does not say whether it is mildly restrictive or mildly accommodative. The post-GFC track record and Governor Bullock’s recent language both lean toward the yield-anchored interpretation, on which 4.35 per cent is mildly restrictive.
After doing additional analytical work that further shaped my views, I have withdrawn my original post.
Please see the replacement post here.
The Monetary Policy Board meets on 4 to 5 May. The cash rate sits at 4.10 per cent after two consecutive 25 basis point hikes in February and March, the first reversals of the easing cycle that ran through 2025. Markets are pricing roughly a 60 per cent chance of a third hike. The case for or against another move depends on what the data say about the underlying inflation problem, not on what borrowers would prefer to hear.
The framework set out in Inflation: Causes, Diagnosis and Cures is a useful discipline here. Diagnose first. Decide second. Skipping the diagnostic step is how public commentary about monetary policy generates more heat than light. So let me work through the diagnosis using the five drivers the framework identifies, then turn to the policy question.
There is a growing chorus of voices arguing that Australia's Consumer Price Index is broken. The CPI, they say, fails to capture the real cost of living because it excludes house prices – the single largest purchase most Australians will ever make. The frustration is genuine. The lived experience of millions of Australians – particularly younger ones, renters, and aspiring first home buyers – diverges sharply from what the headline inflation number suggests. When the Treasurer announces that inflation is under control and real wages are growing, people who cannot afford to buy a home, or who are watching mortgage repayments consume an ever-larger share of their income, are understandably sceptical.
But the argument that the CPI should be reformed to include house prices, land costs, or mortgage repayments rests on a fundamental confusion about what housing affordability actually is. Housing is not one problem. It is three distinct problems, each driven by different forces, each requiring different policy responses, and each moving in different directions in response to the same policy lever. No single price index can combine all three in a way that remains conceptually coherent, policy-usable, and suitable as a monetary target – and trying to force them into one would produce a measure that misleads more than it illuminates, while damaging one of the most important instruments in Australia's macroeconomic framework.
Underlying the whole debate is a confusion between two things that are not the same: a headline indicator of lived economic pressure, and a target variable for monetary policy. The public wants the first. The Reserve Bank needs the second. Forcing a single number to do both jobs is the source of most of the confusion – and it cannot be resolved by reforming the CPI, because the two purposes are structurally in tension.
Let's start with the three housing affordabilities.
The RBA identifies four main transmission channels in its April 2025 Bulletin article on monetary policy transmission through the lens of its MARTIN and DINGO models (Mulqueeney, Ballantyne and Hambur 2025). Christopher Kent, as Assistant Governor (Financial Markets), added a fifth – the credit channel – in his October 2023 address to Bloomberg. These aren't exhaustive, but they're ones the RBA can quantify in its models. I'll come back to some others later.
The original Phillips Curve is a beautiful empirical regularity: lower unemployment meant higher inflation. Policymakers in the 1960s thought they faced a stable menu of choices - trade off a bit more inflation for a bit less unemployment, or vice versa. The current Phillips curve in respect of Australia looks like this.
Markets moved quickly to reprice RBA rate expectations after today's November labour force release. But in their rush to declare the economy sicker than we thought, they may be overlooking the bigger picture.
The labour market is cooling, but not enough to ease inflation pressure. In fact, once you adjust for participation and NAIRU dynamics, the labour market remains tighter than it appears - and inflation is re-broadening. Markets may be reacting to the wrong signal.
Before 2008, nobody argued about the neutral interest rate. It was textbook macroeconomics: the neutral real rate (r) roughly equals trend real economic growth (g). For Australia, that meant r around 3-4%. Add the inflation target of 2.5%, and you get a nominal neutral rate of 5.5-6.5%.
The RBA cash rate averaged about 5.5% from 1993 to 2007, bouncing around neutral as the economy cycled through expansions and contractions. Rates went up when inflation threatened, down when growth weakened, but always gravitating back to that 5-6% centre of gravity.
This wasn't controversial. It was Econ 101.
Then Lehman Brothers collapsed, and everything we thought we knew about interest rates went out the window.
The last graph is not exactly correct as it does not include deposits by offshore lenders. Please see RBA stats B12.2I thought fair enough. This is something I need to look into. So I shot a quick email to the RBA.
Dear Sir/MadamWith impressive alacrity for a government agency, I now have a response from the RBA ...
Table D03 lists various forms of deposits with banks.
Table B12.2 lists the foreign liabilities held by banks
My question is whether the figures from the two tables are fully mutually exclusive, or whether the deposits in D03 include (some or all) of the foreign deposits in table B12.2
Thanks in advance.
MG
Dear MarkSo I need to add table B12.2 to my list of charts ... here it is.
Thank you for your email.
The deposits data in Statistical Table D03 relate to deposits held at Australian-located banks on behalf of residents in all currencies.
The deposits data in Statistical Table B12.2 relate to the International Banking Statistics. In this table, deposits comprises all claims reflecting evidence of deposit – including non-negotiable certificates of deposit – that are not represented by negotiable securities. The data relate to Australian-located banks’ positions with non-residents in all currencies and positions with residents in foreign currency only.
Regards
[etc]
