Introduction

In the late 1980s, as the Iron Curtain fell, the US Army War College threw away its old Cold War playbook. In its place, trainee strategists were taught to see the world as Volatile, Uncertain, Complex and Ambiguous: or ‘VUCA’ for short.^[2] The implications were far-reaching. Out went the old certainties. And in came a new approach that stressed the importance of approaching problems from different angles, drawing on multiple perspectives and scenarios, learning from mistakes, making robust decisions, and communicating openly about the uncertainties.

Where the military began, the business world followed: VUCA begat a million Harvard Business Review articles. Inevitably perhaps, it lost some of its shine in the decades that followed. But today it’s back – with a vengeance. The rules of global trade have been turned on their head. New geopolitical realities are dawning. Artificial intelligence, the energy transition, demographic change and the long shadow of COVID-19 are fundamentally changing our concepts of economic activity and work. And Australia, like elsewhere, is seeking new sources of productivity growth.^[3] With the world in flux, companies, households and governments must change how they think, act and plan – just like those army cadets of the 1980s.^[4]

Monetary policy cannot affect these profound changes. But it does have one key job – and that is to ensure that, of all the things people do have to worry about, inflation is not one. High inflation hurts everyone. It hits living standards, particularly for those on low and fixed incomes. And it disrupts households and companies’ plans. The past few years have been a vivid reminder of that. Around the world, core inflation reached multi-decade highs (Graph 1).

Uncertainty rose sharply too. Forecasting prices during the pandemic was harder than at any time in the past quarter of a century: for central banks (Graph 2)^[5] – and for everyone else too.

That left inflation much higher up peoples’ VUCA worry lists than it should be, harming livelihoods and crowding out focus on the economic choices that households and companies should be spending their time on. Our job is to put that into reverse – returning inflation to the background, where it belongs.^[6]

In my remarks today, I want to review progress towards that goal. I’ll start with the good news – inflation is down and employment is up. We are moving on from the narrow path. But monetary policy must always look ahead – and here I want to discuss two decidedly VUCA risks that shape that outlook: the prospects for world trade; and the degree of spare capacity in the Australian labour market. I will conclude with some implications for monetary policy.

Moving on from the narrow path

While Australia saw much the same pickup in inflation as elsewhere, our monetary policy response was different. Interest rates rose significantly – but they never reached the levels seen in many other developed economies (Graph 3).

That was an explicit choice, grounded in our mission: to bring inflation down, but at a pace that helped preserve sustained full employment. An implication of this strategy, clear from the start, was that just as interest rates rose by less, so they would also fall less far – and less quickly.

There were always risks on both sides of this ‘narrow path’ – and people regularly called them out. Some said the RBA should have tightened more to bring inflation down faster and earlier – and clearly we could have. But that would have risked materially higher unemployment. Others said we should have eased more quickly to help kickstart economic activity. And we could have done that too. But it would have risked inflation being higher for even longer. In the Board’s judgment, both alternatives would have left the Australian people worse off.

That is why the latest economic data are encouraging. Year-ended trimmed mean inflation, our preferred measure of underlying price pressures, fell to 3.2 per cent in the December quarter, 0.2 percentage points lower than expected in November. Among other things that reflected lower inflation in new dwelling costs, rents and market services – which had been stubbornly persistent. Measured on a shorter two-quarter annualised basis, trimmed mean inflation was in the 2–3 per cent target range (Graph 4).

While inflation has moderated, employment has continued to grow extraordinarily strongly. That’s true compared both with other developed economies (Graph 5), and with our own history: 64½ per cent of the population now have jobs, the highest on record.

By contrast, economic growth has been much more subdued, particularly in the private sector. But here too there is now cautiously better news, with partial indicators suggesting that household spending picked up in the December quarter. GDP growth is projected to rise back to trend over the forecast period.

So we look to be moving on from the narrow path. But central bankers are paid to worry, not celebrate. And monetary policy works with lags – so it must be set with an eye to the future, not the past. I will now discuss two key uncertainties that shape that outlook.

Key uncertainty 1: Global trade policy – VUC, but especially A?

To the naked eye, the four words in ‘VUCA’ seem just different versions of ‘chaos’. In fact, their meanings are distinct. Volatility and complexity are the simpler concepts. ‘Volatility’ means rapid change, whether predictable or unpredictable – and ‘complexity’ means a world of multiple overlapping causes and effects. Uncertainty and ambiguity are slipperier. ‘Uncertainty’, in the classical sense, means you know the model, but don’t know the parameters. So you have to estimate an imperfect model-based forecast, which you can refine as you get more information. ‘Ambiguity’ means you don’t know the model, so any model-based forecasts will break down, and feeding more information into those same models won’t help. In situations of ambiguity – or ‘Knightian uncertainty’^[7] as economists sometimes call it – judgement and instinct are as important as formal analysis.

These concepts can help us think through the implications for Australia of global trade policy uncertainty – which is at a 50-year high (Graph 6).

As economists, our inclination is to approach this as an analytical problem of classical uncertainty. We might note for example that, from a macroeconomic perspective, Australia’s direct exposure to US tariffs levied on our exports is limited (Graph 7).

Such an analysis might quickly turn, however, to the fact that Australia is heavily integrated into, and reliant on, the global economy more broadly – and particularly China (Graph 8).^[8] Hence the bigger macroeconomic risk for us would be if the imposition of US tariffs on third countries triggered a global trade war that impaired our trade and financial linkages more broadly. As Australia’s long history has shown, we thrive when trade, labour and assets flow freely in the global economy, but we suffer when countries turn inwards.

In principle, it is possible to estimate the quantitative impact of policy alternatives on Australian activity and inflation using macroeconomic models, though the number of assumptions required is daunting. It includes: the scale, scope and persistence of US trade measures globally; the extent of any policy reactions in third countries (including both trade retaliation and domestic stimulus); the reaction in financial markets, including crucially how the exchange rate adjusts; and the responses of global trading firms, including both production and trade diversion.

Our February Statement on Monetary Policy included three stylised scenarios, involving different sets of these assumptions.^[9] These scenarios suggest some downward impact on Australian activity; and an impact on inflation that could be either positive or negative, depending on whether supply or demand effects dominate. But many other alternatives are possible too. Given the large uncertainties at this early stage, only limited changes were made to our central projections for global activity.

Up until very recently, financial markets appeared to be placing little weight on any severe adverse scenario. Measures of implied volatility in equity, bond and most foreign exchange markets were subdued. Estimates of equity risk premia were close to their post-Global Financial Crisis lows (Graph 9).

And equity investors appeared to take out only modest extra downside insurance in response to the early flurry of news about tariffs (Graph 10).

There are several possible reasons for this apparently benign reaction. Investors may have believed tariff threats were being used primarily as a negotiating tool, with relatively limited longer term economic effects. They may have believed other promised US policy initiatives, including fiscal measures and deregulation initiatives, would more than outweigh the impact on global activity. They may have believed that demand in countries outside the US, including Australia, would be insulated by adjustments in exchange rates^[10] and extra stimulus in key overseas markets. Or they may simply have believed that US policymakers would again show limited tolerance for declines in equity prices, as happened in 2018/19.

That confidence has taken a bit of a knock in recent days. Some of that reflects recent US data, and some evolution in the direction of tariff policy. But it may also reflect a growing recognition that, if companies and households come to conclude that trade policy uncertainty has moved on from classical Uncertainty (‘carry on till the fog lifts’) to genuine Ambiguity (‘almost anything could happen’), they may choose to batten down the hatches – postponing planned spending, particularly on longer term capital investment, until things become clearer. Such ‘watchful waiting’ could prove rational individually, but economically damaging in aggregate. As The Economist put it recently, ‘tariff uncertainty can be as ruinous as tariffs themselves’.^[11] The Federal Reserve estimated that heightened uncertainty over trade policy in 2018 reduced global GDP by nearly 1 per cent in 2019^[12] – and Graph 6 suggests the pick-up in policy uncertainty is much larger this time around. The possibility of such an effect played a part in the Board’s policy deliberations in February.

Key uncertainty 2: Capacity in the domestic economy

A second key uncertainty lies closer to home, in the labour market. While the recent strength in employment growth is welcome, it’s also unusual after a period of such subdued GDP growth. The question is what it means for the margin of spare capacity in the economy, and hence for the inflation outlook.

Assessing this issue is harder than it seems. Spare capacity cannot be directly observed. And its sustainable level has no set value, and likely changes over time as the structure of the economy evolves. Some argue this makes the concept meaningless – but that does require you to have an alternative narrative for inflation. At the RBA, we prefer to give it some weight while recognising the pervasive uncertainties, by building up a picture using a wide range of qualitative and quantitative data, and analytical techniques – as well as regularly challenging how we could be wrong.

An obvious place to start when assessing labour market capacity is to look at proxy measures. Two of the most important are unemployment (those looking for work) and underemployment (those in work, but looking to do more hours). As recently as November, we were projecting unemployment to rise to 4¼ per cent by end-2024 and 4½ per cent in late 2025, as past weak activity reduced hiring rates. In fact, unemployment has remained at or around 4 per cent, and underemployment has fallen back to late-2022 levels. A range of other capacity measures have also stabilised or reversed in recent months, including the ratio of vacancies to unemployment, and surveys of firms’ reported labour constraints (Graph 11).

With activity projected to pick up in 2025 as private demand recovers, these developments have caused us to revise down our central projection for unemployment.

But the implications for inflationary pressure depend on where this leaves spare capacity relative to sustainable levels. Two considerations suggest labour market conditions are relatively tight. First, all of the measures in Graph 11 lie some distance above their historical averages – and unemployment remains close to its lowest level at any time in the past 50 years.^[13] But that can’t be the end of the matter – because the levels of nominal and real wage inflation associated with a given level of unemployment have fallen substantially over that period. So the sustainable level must be lower too.^[14] How much lower, no-one can say for sure. But it is possible to back out a range of time-varying estimates from past relationships between unemployment, wage and price inflation, using a suite of statistical methods of varying levels of sophistication. These estimates include the immediate pre-pandemic period, when wage inflation persistently undershot forecasts.

Those analytical approaches all suggest that, while sustainable unemployment levels are likely to have fallen materially in recent decades, current labour market conditions still appear relatively tight. Combined with the lower unemployment projection, that would suggest somewhat greater upward pressure on inflation from the labour market over the medium term. Exercises using the other measures in Graph 11 reach a similar conclusion.

But these are critical judgments – and serious commentators from academia, the financial markets and elsewhere have argued that we may be taking too pessimistic a view. We take those challenges seriously.

Some point out that business surveys of employment intentions have been at, or slightly below, long-run averages. And that is true, but such surveys typically focus on the market sector, where employment growth has been relatively subdued. They tell us less about pressures in the non-market sector, which has accounted for most of the recent strength in aggregate employment (Graph 12).

That leads to a different challenge – that non-market employment has limited influence on aggregate wage and inflation pressure, because it draws on a different labour pool. But it is hard to find support for this in the data. For example, the health care sector – a big contributor to aggregate employment in recent years – has drawn quite materially on workers in other industries (Graph 13), helping to equalise cross-sectoral wage growth.^[15] Discussion with liaison contacts suggest similar mechanisms are at work in other sectors too, including construction.

A third argument against the view that labour market conditions are relatively tight notes that nominal wage growth has been easing (Graph 14). But with measured productivity growth as weak as it has been recently, that still implies elevated growth in companies’ unit labour costs. Some of that apparent strength could reflect under-measurement of productivity growth or a temporary burst of real wage catch-up to past inflation, rather than labour market tightness. But such effects would need to be unusually large to account for the whole of the gap.^[16]

Finally, it is possible that, over and above the impact of labour market conditions, recent disinflation also reflects compression in other aggregate price drivers, including margins and housing costs. In that context it is noteworthy that output-based measures of capacity pressures have continued to fall.

Drawing this all together, our central projection reflects a judgement that labour market conditions will remain relatively tight over the forecast period, and a little tighter than assumed in November. At the same time, we have recognised the risk that recent inflation data may suggest we have overestimated the extent of excess demand in the labour market by applying a little downwards judgement on the inflation profile. And the Statement on Monetary Policy sets out what one would need to believe to justify an even larger downward adjustment, as a risk scenario.^[17]

Implications for the RBA’s monetary policy decision

Graph 15 compares the central projection for trimmed mean inflation in February with that in November. Inflation is slightly lower in the near term, reflecting the downside news on inflation, wages and activity. But it is a little higher further out, stabilising slightly above the midpoint of the target range, reflecting the surprising strength in the labour market.

Why then did the Board cut rates? Did we reject the staff forecasts, as some have claimed? Or did we suddenly and confusingly relax our previously stated intolerance for persistent inflation deviations from target? Nothing of the sort – for me at least, the rationale is relatively simple.

First, the encouraging news on price and wage inflation gave us somewhat greater confidence that underlying inflation is on track to return to the target range in the near term – if anything, a little more rapidly than previously expected. The Board noted that the combination of lower inflation data, and a lower near-term projection, put Australia in a very similar position to many other countries ahead of their first cuts (Graph 16).

Second, however, the Board also recognised that the uncertainties about the outlook for inflation become larger, the further out you go.

One uncertainty relates to future changes in the cash rate. All projections have to assume something about this path, and by convention we assume it follows market expectations. In February, that curve implied up to four 25 basis points cuts over the forecast horizon, at a somewhat more frontloaded pace than in November. In light of the data then available, including the strong labour market, it was not clear that a rate cutting cycle of this depth was likely to return underlying inflation sustainably to the midpoint of the target range. The February projections are consistent with that view.

Third, that did not, however, mean there was no case for a cut at all. To see that, the red swathe in Graph 17 shows an illustrative range of projections for underlying inflation at the time of the February forecast under the alternative assumption of an unchanged cash rate target of 4.35 per cent.

The centre of the swathe lies slightly below the midpoint of the target range, consistent with a bias to cut. But there were good arguments for both a hold and a cut – and the Board discussed them in some detail, as the minutes released earlier this week show.^[18] Foremost in that debate included the issues I have discussed today – the outlook for global activity, and the degree of spare capacity in the labour market.

Some have flagged a concern that the Board’s messaging on rates feels like fine-tuning. It is certainly true that the pervasive uncertainties we will face over the forecast period are orders of magnitude larger than the sorts of differences to the target midpoint I’ve discussed here. But the Statement on the Conduct of Monetary Policy agreed between the Treasurer and the Board is clear: we set monetary policy such that inflation is expected to return to the midpoint of the target range. And we do that because it maximises the chances of inflation remaining sustainably in that range. The rate cut in February reduces the risks of inflation undershooting that midpoint, but the Board does not currently share the market’s confidence that a sequence of further cuts will be required.

That assessment will of course evolve as time proceeds and further data help distinguish between alternative narratives of the economy. Interest rates will go where they need to go to maximise the chances of keeping inflation sustainably in the target band while helping to sustain full employment. Progress towards that target has been good – but it is too soon to declare victory. Many households and companies are continuing to struggle – and the Board will continue to take decisions, meeting by meeting, in the interests of all Australians. In so doing, our goal is to remove inflation from the list of things people have to worry about, leaving them free to focus on navigating an increasingly VUCA world.

Endnotes

I am particularly grateful to Sue Black, Chris Schwartz and Michelle Wright for their help in preparing this speech and to David Bold, Andrea Brischetto, Michele Bullock, Georgia Carney, Natasha Cassidy, Iris Chan, Sally Cray, Boston Dobie, Samuel Evangelinos, Matt Gibson, Jacob Harris, Sarah Hunter, Shan Jayawardhana, Bradley Jones, Christopher Kent, Kevin Lane, Jeremy Lawson, Marcus Miller, Mike Major, Penny Smith, Tim Taylor and Nick West for their comments and contributions. [1]

‘VUCA’ was first coined by Warren Bennis and Burt Nanus in their 1985 book: Bennis W and B Nanus (1985), Leaders: The Strategies for Taking Charge, Harper & Row, New York. [2]

For a recent perspective on the importance of productivity for monetary policy, see Plumb M (2025), ‘Why Productivity Matters’, Speech at the Australian Business Economists Annual Forecasting Conference, Sydney, 27 February. [3]

People didn’t literally stumble across the merits of decision-making under uncertainty in the 1980s, of course. Indeed the RAND Corporation website still links to a 1959 paper by Thomas Schelling, the Nobel prize winning game theorist of the Cold War era: Schelling T (1959), ‘Randomization of Threats and Promises’, RAND Paper No P-1716, 5 June. It has more than a little relevance today. [4]

For a review of inflation forecasting over the pandemic period, see RBA (2022), ‘Box C: What Explains Recent Inflation Forecast Errors?’, Statement on Monetary Policy, November. [5]

The goal of making monetary policy boring is sometimes compared with John Maynard Keynes’ hope that ‘economists could manage to get themselves thought of as humble, competent people on a level with dentists’. As a relatively recent arrival to Australia from the United Kingdom, I have discovered that dentistry is one of the many areas where my new adopted home has a substantial comparative advantage: Crocombe LA, GC Mejia, CR Koster and GD Slade (2009), ‘Comparison of Adult Oral Health in Australia, the USA, Germany and the UK’, Australian Dental Journal, 54(2) pp 147–153. Whether that reads across to relative monetary policy outcomes, I could not possibly say. [6]

The name comes from the Chicago economist of the 1920–1930s, Frank Knight. See, for instance, Knight FH (1933), Risk, Uncertainty and Profit, 1st reprint edition, Houghton Mifflin Co, Boston; or Kay J and M King (2020), Radical Uncertainty: Decision-Making Beyond the Numbers, WW Norton & Co, New York. I also discussed these issues in Hauser A (2024), ‘Beware False Prophets’, Speech to the Economic Society of Australia, Brisbane, 12 August. [7]

Table 1 in my December 2024 speech sets out a range of measures of Australia’s integration with the world economy, and also explains the background to Graph 8 (which is Graph 6 in the earlier speech): Hauser A (2024), ‘The Ghost of Christmas Yet to Come’, Speech at the Australian Business Economists’ Annual Dinner, Sydney, 11 December. [8]

See Key risk #2 in RBA (2025), ‘Chapter 3: Outlook’, Statement on Monetary Policy, February. [9]

Despite the lively external commentary, the Australian dollar has so far adjusted only modestly to news about tariffs: the trade-weighted index fell only 2 per cent between the November 2024 and February 2025 Statements on Monetary Policy. [10]

The Economist (2025), ‘Tariff Uncertainty Can be as Ruinous as Tariffs Themselves’, The Economist, 6 February. [11]

Caldara D, M Iacoviello, P Molligo, A Prestipino and A Raffo (2019), ‘Does Trade Policy Uncertainty Affect Global Economic Activity?’, FEDS Notes, 4 September. [12]

For an analysis of Australian unemployment since 1900, see Borland J and S Kennedy (1998), ‘Dimensions, Structure and History of Australian Unemployment’, Unemployment and the Australian Labour Market, Proceedings of the RBA Annual Conference, Reserve Bank of Australia, Sydney. [13]

See, for instance, Ellis L (2019), ‘Watching the Invisibles’, Speech at the Freebairn Lecture in Public Policy, Melbourne, 12 June. [14]

See RBA (2025), ‘Box C: Health Care Employment and its Impact on Broader Labour Market Conditions’, Statement on Monetary Policy, February. [15]

The models of sustainable labour market capacity used to inform the forecast judgements aim to capture average real wage catch-up through the inclusion of lagged inflation and inflation expectations terms, and adjusting for large award wage increases linked explicitly to high inflation. [16]

See Key risk #1 in RBA (2025), ‘Chapter 3: Outlook’, Statement on Monetary Policy, February. [17]

RBA (2025), ‘Minutes of the Monetary Policy Meeting of the Reserve Bank Board’, 17–18 February. [18]