Speech Australian Financial Conditions – How Do We Judge How Tight or Easy They Are?
Introduction
I would like to thank the CFA Society for the opportunity to speak here today.
A key part of the Monetary Policy Board’s deliberations is to assess whether financial conditions are tight, easy or neutral in terms of their effect on aggregate demand. It then determines whether those conditions are appropriate to achieve its goals of full employment and low and stable inflation, and adjusts the cash rate target if needed.
Today, I’ll speak about three key building blocks we use to assess financial conditions. While the cash rate target is a good starting point, it is not a reliable guide because it does not account for other factors affecting financial conditions, including structural changes in the economy. Hence, it is worthwhile to:
- compare the cash rate with estimates of the neutral interest rate, although these estimates are highly uncertain
- consider a broader set of financial indicators, which also help to track the transmission of monetary policy through the economy
- examine the RBA’s macroeconomic forecasts, which incorporate measures of financial and economic conditions.
These building blocks point to policy having been restrictive from around 2023. More recently, there are signs that restrictiveness has declined following cuts to the cash rate target and with funding readily available to a wide range of household and business borrowers.
Cash rate target
A good starting point for my talk today is the cash rate target, which is a key focus for commentators assessing financial conditions (Graph 1), and rightly so. It has an important bearing on interest rates in the Australian economy and is the instrument the Board sets to influence financial conditions.
But it’s hard to assess financial conditions by looking at the cash rate alone, as its past behaviour is not a reliable basis for comparison.
One reason is that how tight or easy a given cash rate is depends on expectations of both inflation and the cash rate itself (to which I’ll return shortly). For example, the benefit of a high interest rate to a saver is reduced if inflation erodes what their money can buy over time. Similarly, the burden of a high interest rate on a loan is eased if nominal wages or profits are rising quickly. To address this, we can compare the cash rate to estimates of the nominal neutral rate, which adjusts for changes in inflation expectations over time.
The neutral interest rate is also a useful comparator because it can account for changes in global developments and Australia’s economic and financial structures. These can influence how any given level of the cash rate will affect aggregate demand. The neutral cash rate is the rate that is neither expansionary nor contractionary over the long term – balancing investment and savings at levels consistent with full employment and stable inflation (once current shocks fade). Conceptually, at least, comparing the cash rate to the neutral rate helps gauge the restrictiveness of monetary policy.
Neutral interest rate
The RBA estimates the neutral interest rate using several models.1 I’ll focus on the average of the models’ central estimates before turning to key differences between them (Graph 2). The average suggests the neutral cash rate has trended lower over recent decades – a pattern seen in other economies. This trend is likely to reflect structural shifts such as demographic change and slower productivity growth. These shifts can increase savings and reduce investment, in which case lower interest rates would be needed to balance the two.
So while the cash rate has been much lower in recent years than in previous decades, this does not imply a one-to-one easing in financial conditions because the neutral rate has also declined. Indeed, a cash rate of around 4 per cent in recent times may have been just as restrictive as 7 per cent was three decades ago if the neutral rate has fallen by around 3 percentage points since then.
In recent years, central estimates of Australia’s neutral rate have risen by about 1 percentage point on average. Factors contributing to this include rising global public debt, lower saving by retiring baby boomers, and increased public and private investment – including in the green energy transition.2 This rise in the neutral rate implies that any given level of the cash rate is now less restrictive than it would have been otherwise.
So far I have compared the current cash rate with neutral estimates at that time, but we also need to consider cash rate expectations, as they influence longer term interest rates and affect current savings and investment decisions. A declining expected path for the cash rate as shown in Graph 2, for example, implies easier financial conditions than a flat or a rising one.
Limitations of neutral rate estimates
There are limitations to using neutral rate estimates to assess whether financial conditions are tight or easy. The main limitation is that the estimates are imprecise. This has two aspects.
First, the estimates are very uncertain. The span of central estimates across models is wide, but we do not know which model best measures the neutral rate, and each central estimate is derived with uncertainty (Graph 3).
Even so, assuming our models cover the set of reasonable descriptions of the neutral rate, we can have some confidence that cash rates well above the range of central estimates would constrain aggregate demand (and vice versa for rates well below). But we can be less certain for rates closer to or within that range – as is currently the case.
A second limitation is that the models may not capture all the key aspects of financial conditions, or at least not in a timely manner. Indeed, four of our models rely on macroeconomic data, which are only available with some lag and reflect past financial conditions, making them slow to respond to new developments. These are the four models currently showing lower estimates. The three other models are forward looking. They extract estimates of future short rates from bond yields of various maturities and so they are potentially quite responsive to changes affecting the neutral rate; but again, they are estimated with considerable uncertainty.
Given these limitations, neutral rate estimates form only part of our assessment of financial conditions. We also consider a broader set of indicators of financial conditions.
Additional indicators of financial conditions
Financial indicators can help us to track the transmission of monetary policy to the economy. They also can suggest whether financial conditions align with movements in the cash rate or behave in ways that are amplifying or dampening its usual effects. I’ll focus on just a few indicators – though we refer to a broader set in our quarterly Statement on Monetary Policy (SMP).3
Funding cost and interest rate spreads
Changes in the overnight cash rate influence other interest rates, including those affecting banks’ funding costs and lending rates for households and businesses. Graph 4 shows the differences between funding costs and the cash rate, and between loan interest rates and the cash rate. These spreads reflect factors influencing the supply and demand for funding. The top panel shows that the spread between estimates of major banks’ funding costs and the cash rate was very low before the global financial crisis, with depositors receiving low returns relative to the cash rate and bond holders requiring little compensation for a given level of risk. This spread rose sharply during the crisis as credit risk concerns grew and banks shifted from short-term wholesale debt and securitisation to more stable funding sources. During the pandemic, the funding cost spread fell in response to the RBA’s unconventional policies, but it has stayed low since then, reflecting a higher share of at-call deposits and, more recently, low wholesale debt spreads.4
Variations in banks’ funding costs, and their willingness to take on credit risk and compete for borrowers, have underpinned movements in key lending rates to households and businesses, shown as spreads to the cash rate in the middle panel of Graph 4. These spreads have narrowed in recent years. The bottom panel shows the cost for larger businesses to raise funds via bond issuance, with spreads to Australian Government Securities yields currently at very low levels.
The sharp rise in banks’ funding costs and lending spreads during the global financial crisis tightened financial conditions and was one reason the RBA cut the cash rate sharply at the time. However, current loan spreads suggest financial conditions are now less tight than a few years ago for a given level of the cash rate. This is consistent with the rise in neutral rate estimates I just mentioned.
Household finanancial conditions
The cash flow and intertemporal channels influence household savings, consumption and housing investment. These are key channels for monetary policy transmission and the associated indicators are useful for assessing financial conditions.
Mortgage payments
Mortgage payments data offer insight into how these channels operate. While required payments have declined this year as the lower cash rate has passed through to banks’ lending rates, they remain elevated due to interest rates being above pre-pandemic averages (Graph 5).
The bottom panel of Graph 5 shows that mortgagees typically pay more than the minimum required. In response to high mortgage rates, extra payments rose above the pre-pandemic average by the end of 2024 (as a share of household disposable income), consistent with the incentive to save more when interest rates were high. But extra mortgage payments have now declined, which is possibly an early response to the easing in interest rates.5
Household credit
Lending rates can affect household credit growth by influencing housing prices, borrowers’ ability and willingness to take on new debt, and the incentive to repay existing debt. This was evident as interest rates rose from 2022, with the subsequent decline in the ratio of household credit to household disposable incomes consistent with tight monetary policy (Graph 6).
Household credit growth picked up as interest rates declined this year and housing market conditions strengthened, which is consistent with an easing in financial conditions. However, the stock of household credit excluding offset balances is still falling relative to income and, by itself, doesn’t suggest that financial conditions are easy.
Business debt
The ready availability of funding at favourable spreads has supported the rise in business debt in recent years (Graph 7). Strong competition among banks and non-banks, healthy loan books and an improved economic outlook have underpinned the supply of credit to businesses. Large businesses have also benefited from low corporate bond spreads, with non-financial Australian corporations issuing bonds at record levels this year.
Business investment has historically had a weak direct relationship with aggregate business debt, as that investment is mainly internally funded and influenced by factors like profitability and economic conditions.6 Even so, strong business debt growth is consistent with a positive outlook by businesses and lenders. Credit growth also contributes to money supply growth, which can offer a timely – though imprecise – signal of trends in aggregate demand and inflation.7
The indicators I’ve discussed add context and, together with neutral rate estimates, help to assess how tight or easy financial conditions are. However, these two building blocks cannot determine if a policy stance is appropriate for achieving the Board’s goals. This is because they do not account for all the factors that shape the economic outlook, including recent shocks and other cyclical influences. For that, we rely on the RBA’s economic forecasts.
Macroeconomic forecasts
When forecasting, we look at the current state of financial conditions and assume that the cash rate follows the path implied by market pricing. Our forecasts also incorporate a wide range of macroeconomic factors shaping the domestic outlook, such as conditions in major trading partners and Australian governments’ fiscal policies. This approach helps us to assess whether financial conditions are such that the Board’s inflation and employment objectives are likely to be met. If not, it implies that the Board might need to consider a different path for the cash rate than that implied by market pricing.
In addition to the market path for the cash rate, our forecasts assume the Australian dollar trade-weighted exchange rate remains at its current level. Several other financial variables also feed into the forecasts. For example, household consumption is influenced by housing lending rates, credit growth, and equity and housing prices (through wealth effects). The cost of capital, which incorporates business lending rates, feeds into models of non-mining business investment. Despite the inclusion of these measures, the models that form the starting point for our forecasts may not properly capture financial conditions, so judgement about this dimension is required – alongside judgement about macroeconomic factors.
For over a year, forecasts in the SMP have assumed the cash rate would gradually decline through 2025 and early 2026 before stabilising. The forecasts have implicitly reflected an assessment that financial conditions were restrictive and restraining demand. This was bringing demand and potential supply into better balance and easing labour market tightness. As a result, underlying inflation was expected to gradually return towards the midpoint of the 2–3 per cent target range. With the economy approaching balance, policy was expected to move towards a more neutral stance.
We can use model estimates to see how the outlook might change if the cash rate path were to deviate from the baseline. The red shaded area in Graph 8 shows projections for inflation and unemployment if the cash rate was 50 basis points higher or lower than the August SMP baseline.8 For instance, if the cash rate was 50 basis points higher (all else equal), inflation would be expected to fall below 2.5 per cent and be declining by late 2027, while unemployment would be expected to be around 4.5 per cent and rising. In summary, based on what we knew at the time, cash rate paths that deviated too far from the August SMP baseline would have been less likely to meet the Board’s goals for inflation and full employment.
However, our macroeconomic forecasts carry significant uncertainty. This includes uncertainty about our assessment of how tight or easy overall financial conditions are, which has been – and will continue to be – closely scrutinised. Historical forecast errors, illustrated by fan charts, show that the range of potential outcomes for underlying inflation and unemployment is very wide out beyond the near term (Graph 9).
Conclusion
It makes sense to use a number of different methods to assess financial conditions given the considerable uncertainty involved with each.
The first building block – the conceptual cornerstone if you like – is to compare the cash rate with estimates of the nominal neutral rate. Model-based estimates of the neutral rate suggest that financial conditions have been tight, working to restrain aggregate demand. Based on the market path, the cash rate is expected to sit within the wide range of central estimates of neutral over the coming period. However, even that range understates the uncertainty. And while neutral rate estimates are a useful cross-check, they are not a suitable guide to the near-term path of monetary policy.
Nevertheless, what neutral rate estimates suggest accords with a range of indicators of financial conditions. These indicators offer more timely evidence of monetary policy effects than measures like economic activity and inflation. Indeed, several indicators of financial conditions – including mortgage payments and housing credit growth – show early signs of responding to the easing in financial conditions this year.
Finally, the RBA’s macroeconomic forecasts are based on a range of financial indicators, including the expected path of the cash rate implied by market pricing. Our forecasts imply that the tightness in financial conditions has eased, which will help to keep the economy in balance in the period ahead, with full employment and inflation moving toward the centre of the target range. However, this outlook is subject to considerable uncertainty, and we will continue to reassess it in light of what the incoming data mean for the economic outlook and evolving risks.
Endnotes
I thank Indigo Adamson, Joel Findlay, Kristy Guo, Michaela Haderer, Sarah Jennison, Sophie Kelly, Sharon Lai, Patrick Manning, Hamish McLean, Emma Searle, Josh Spiller and Peter Wallis for their excellent assistance in helping me to prepare this speech. *
The RBA uses a suite of models to estimate the real neutral rate for Australia, including semi-structural models, vector autoregressions and financial market-based models. We can convert this to a nominal neutral rate by using a measure of trend inflation. Over the past year or so, we have updated our modelling approach to semi-structural models. This has involved working through how to deal with the COVID-19 pandemic period and how long we modify the models’ usual dynamics. These updates have affected the overall model average of central estimates of the neutral rate. 1
See Kent C (2024), ‘Restrictive Financial Conditions in Australia’, Address to ABA Banking Conference, Melbourne, 26 June, and references therein. 2
I have focused here on debtors in the transmission of monetary policy, but savers also play an important role, not only through the effect of interest rates on their cash flows – which runs counter to the effect on debtors – but also through the intertemporal channel, whereby higher rates, for example, encourage more saving by savers and debtors alike. For a more comprehensive examination of the household cash-flow channel of monetary policy, see Jennison S and M Miller (2025), ‘An Update on the Household Cash-flow Channel of Monetary Policy’, RBA Bulletin, January. 3
See Cole D, V De Zoysa and C Schwartz (2025), ‘Bank Funding in 2024’, RBA Bulletin, April. 4
Also noteworthy in this graph is the sharp rises in extra payments during the pandemic and over the first half of the 2010s. This highlights the fact that these payments can be influenced by things other than interest rates. For example, during the pandemic, there were reduced opportunities for spending and yet incomes were not affected so much (including because of support provided by JobKeeper). 5
For a discussion of the uses of internal and external business funding in Australia, see Kent C (2017), ‘The Availability of Business Finance’, Address to 30th Australasian Finance and Banking Conference, Sydney, 13 December, and references therein. 6
See Doherty E, B Jackman and E Perry (2018), ‘Money in the Australian Economy’, RBA Bulletin, September. 7
These alternative scenarios are generated using the RBA’s MARTIN model. For an overview of MARTIN, see Ballantyne A, T Cusbert, R Evans, R Guttmann, J Hambur, A Hamilton, E Kendall, R McCririck, G Nodari and D Rees (2019), ‘MARTIN Has Its Place: A Macroeconometric Model of the Australian Economy’, RBA Research Discussion Paper No 2019-07. 8