Watch video: Speech delivered by Andrew Hauser, Deputy Governor, Sir Leslie Melville Lecture, Australian National University, Canberra

Introduction

It’s an honour to give the 24th Sir Leslie Melville Lecture.¹

My theme today is trust. Trust is central to all of our lives. We trust our families and friends. We trust the food we eat. We trust companies with our personal data. We trust governments to keep us safe.²

But few rely on trust as profoundly as central banks. And that’s because money itself is based entirely on a social contract. Banknotes in our wallets or deposits in our bank accounts have no intrinsic value beyond the monetary realm. Their value rests on trust that others will accept them tomorrow at much the same worth. And that trust, in turn, relies on confidence that central banks will keep prices stable. So trust isn’t just a source of legitimacy – it’s integral to how monetary policy actually works.³

That trust must be earned. And the decades following the 1980s were good times for central banks around the globe, as flexible inflation targeting delivered consistently low and stable inflation alongside rising real incomes.⁴ But more recently, and particularly since COVID, things have been rockier: persistent above-target inflation has marred that hard-won track record. The broader climate of trust towards public bodies has deteriorated. And we’ve entered a ‘VUCA’ (Volatile, Uncertain, Complex and Ambiguous) world, amid a fusillade of shocks from tariffs, wars, geopolitical tensions, pandemics, and technological and environmental change.⁵ That means inflation is more likely to be pushed off target; economic forecasts are more likely to be wrong; and policy trade-offs get uglier.

Just as we need it most, trust is under strain.

These challenges are not unique to Australia.⁶ And, whatever the commentary may say, they’re far from unprecedented. Leslie Melville definitely had it worse. He joined the Commonwealth Bank – the RBA’s predecessor – as its first (and initially only) economist in 1931, a year in which Australian GDP fell by nearly 7 per cent, amid the depths of the Great Depression. The following year, unemployment reached nearly 30 per cent – an unbelievable figure to modern eyes.⁷ The financial instability, world war, fierce economic and political conflict and institutional upheaval that followed frankly puts today’s challenges into humbling perspective.⁸

Coming from the United Kingdom, 10,000 miles away, I wouldn’t have expected any personal connection to Melville. But it turns out I have two. The Bank of England, my previous employer, recommended him for the job.⁹ More than 90 years later, as another newcomer to Australian policymaking, I have the joy of working with the RBA’s world-class economics team: a far cry from Melville’s empty office! And while writing this speech, I discovered that the headmaster of my old high school in the United Kingdom in the 1980s – known to us all simply as ‘Mr Melville’ – was in fact Leslie’s son.¹⁰ Tony Melville was born in Adelaide, grew up in Sydney when Leslie joined the Bank, and travelled the opposite way to me: studying first at King’s College in Cambridge before taking up a career in British education. Few, if any, of his pupils knew his extraordinary history.

In my speech today I want to develop the theme of trust – discussing, first, what trust means in everyday life; second, why monetary policy needs it so badly; and third, how that trust is being challenged today around the world. I will finish by setting out how the RBA is responding here in Australia.

How we talk about ‘trust’ in our everyday lives

When we say we trust someone, we usually mean one of two things.

Model 1 is the trust we give to family and close friends. It’s instinctive, heart-felt. Family ties, shared history and values mean we know they will be on our side: honest, open and straight. Not perfect, of course. A trusted friend may get things wrong, be stubborn, even drive us up the wall! But deep down we know they are loyal, and it takes a lot to dislodge that. This is trust rooted in character and relationship. We don’t need a fresh demonstration of it every time we meet: we lean on what we know.

Model 2 describes the trust we place in experts: doctors, lawyers, aircraft pilots. Honesty and integrity still matter. But the key test is competence: do they have the qualifications, skills and tools to do complex, sometimes life- (or livelihood-) threatening, things on our behalf? What if they’re useless, crooked – or both?

Answering this question involves a Catch-22 – because the very thing that creates the need for an expert in the first place (our inability to do the task) makes it impossible to audit them ourselves. And that creates an incentive problem. At one extreme, if experts know we have blind faith in them, there’s no incentive for competence. But, at the other extreme, if we’re perpetually sceptical, nothing ever gets done.¹¹ To address this problem, we demand transparency: show us your credentials, your track record, your reasoning. But sometimes it takes an expert to understand an expert. So we also need accountability – independent expert scrutiny beforehand; assessment of outcomes afterwards.¹² ‘Trust, but verify’ – as Ronald Reagan used to say.

For a public institution, like a central bank, all of these concepts matter (Figure 1):

• We want an institution that’s on our side, like a friend – honest, consistent and reliable, serving our long-term interests not selfish or sectional ones.
• We want an institution that’s up to the job, like a trusted professional – with the right skills, tools and judgment; and a track record of delivering strong outcomes.
• We want an institution that can explain itself clearly and consistently – its objectives, its decisions, the costs and uncertainties, and its mistakes, when they happen.
• And we want an institution that’s accountable for its actions, both ahead of time, and after the event.

These qualities reinforce one another. Competence without integrity or public purpose won’t sustain trust. But neither will good intentions without capability. We need confidence that an institution can explain its actions, and bear the consequences. Transparency and accountability help us judge that.

Why earning public trust is particularly important for monetary policy

This setup applies to many public institutions. So what’s distinctive about monetary policy?

First, it works with a lag: and pain often precedes gain. William McChesney Martin vividly described the central bank as a ‘chaperone who has ordered the punch bowl removed just when the party was really warming up’.¹³ When excessive inflationary pressures build, monetary policy has to be tightened, restraining demand. Over time, weaker demand helps reduce inflation, returning the economy to balance. But while those lags are working through, and people are having to make difficult and costly adjustments, they need assurance that we’re acting competently. Has the central bank demonstrated that it’s acting in the public’s long-term interests? Has it been clear about its reasons for acting – and costs of not acting? And has it held itself open to challenge and scrutiny? That matters even more when the inflationary pressure is associated with adverse supply developments, which brings additional and unavoidable pain. Taking away the punch bowl is rarely a way to win instant popularity. The question is whether people can see why the difficult decision is necessary, even when its immediate effects are unwelcome.

These things aren’t unique to monetary policy: public health measures (e.g. vaccinations), environmental policy or macroprudential adjustment share similar features. But what is distinctive is trust directly affects how well monetary policy works. And that’s because the impact of today’s policy decision depends on whether people trust us to do what’s needed tomorrow too. In simple terms: do they expect us to return the economy to balance, with on-target inflation and full employment, in the long run? In more technical terms: do they understand and trust our ‘reaction function’ (i.e. how we adjust interest rates as the economy evolves in order to achieve those goals)?

To illustrate how beliefs about the future can affect policy effectiveness today, consider a thought experiment using one of the RBA’s models, in which the central bank lifts rates for a few quarters to return inflation to target (Graph 1).¹⁴ The red lines describe a scenario in which people don’t expect the tightening to be persistent – so each period, tighter policy surprises them. Spending (and hence inflation) fall only a little, as people fail to account for future policy. By contrast, the orange lines show a scenario in which people fully understand and believe the central bank’s strategy from the start. They adjust spending today not just in response to today’s policy settings, but in anticipation of future settings too. Although the realised path for interest rates is identical in both cases, output and inflation respond sooner, and by more, when the policy response is trusted.

This is a powerful result – because it means that if the public understands and trusts the central bank’s strategy, the policy response required to deliver a given disinflation can be smaller and less costly in terms of lost output.

To illustrate this with a topical example, Graph 2 uses the same model to trace the effects of a temporary increase in overseas costs, such as oil prices, under different assumptions of trust in the policy response. Both situations show the same path for inflation. What differs is the demand adjustment needed to bring it back to target.

The pink lines show what happens when people do not trust the central bank to return inflation to target in a timely way.¹⁵ Businesses are inclined to pass on more of the cost increases, expecting that a slow policy response will allow cost pressures to persist and demand to keep expanding. And employees are minded to seek higher wages. The temporary cost shock risks becoming embedded in prices and wages. To bring inflation back promptly, the central bank must raise rates quite aggressively, contracting demand sharply.

The blue lines show an alternative case where people do trust the central bank to hit the target promptly. Businesses anticipate policy reducing future cost pressure and demand, so are less inclined to raise prices today. This lower price pressure in turn reduces upward pressure on nominal wages. With the private sector doing some of the work to hold inflationary pressures down, interest rates don’t need to rise so far, and the output cost of achieving the same disinflation is lower.¹⁶

This isn’t just theoretical. Data from the Macquarie University Business Outlook Scenario Survey (BOSS) earlier in the year showed that firms expecting the RBA to return inflation to target in a more timely way planned to pass on less of their Middle East conflict-related costs.¹⁷ Australian households expressing greater trust in the RBA also tend to expect lower near-term inflation (Graph 3). The European Central Bank (ECB) has found the same for the euro area.¹⁸

Central banks also face distinctive accountability mechanisms. Most public policies are judged ultimately at the ballot box. But nine in 10 countries delegate monetary policy to central banks that are operationally independent of government – and are instead held accountable by other means.¹⁹ The near-universal adoption of this model reflects the strong evidence, across countries and time, that greater independence delivers: more stable economic activity; lower and less volatile inflation; and better-anchored inflation expectations, preventing 1970s- and 1980s-style inflationary spirals.²⁰ The strength of this finding is shown by the fact that, despite central banks rightly coming under heightened scrutiny since COVID, almost none have lost independence: 83 per cent are as independent as they were before, and 15 per cent are more independent.²¹

Measuring public trust in central banks

So trust matters. But economic developments are putting it to the test.

Not all the news is bad. GDP growth and employment, explicit or implicit goals for all central banks, have proved resilient in many countries. Despite the sharp tightening in policy since the end of COVID, unemployment is much lower than might have been expected (Graph 4).

But inflation is another story. It’s been above central bank targets for years (Graph 5). And around the world, that’s causing increasing anger. Consumer confidence is exceptionally weak. And households rank inflation as their number one economic concern.²²

Standard economic yardsticks struggle to explain that angst. Bringing inflation down from its post-COVID peaks has involved some of the smallest ‘sacrifice ratios’ (the reduction in output associated with a unit of disinflation) in history.²³ And the so-called ‘misery index’ (inflation plus unemployment) is only a little above its 15-year pre-COVID average (Graph 6).

But these measures miss something crucial: the cumulative increase in prices since COVID has far exceeded that in any other tightening period since inflation targeting began. Graph 7 shows this for Australia; similar pictures can be drawn for the United States and other developed countries.²⁴ A lower inflation rate doesn’t undo that accumulated increase. And it’s that increase, not just the latest rate of change, that seems to be part of what’s driving public anger.

What impact has this had on public trust in central banks?

To help answer this question for Australia, we can turn to our new Community Research Survey, which began in February last year. Four separate waves have drawn on feedback from 12,000 Australians, across all parts of the country and sections of society. The latest was in September.²⁵

The survey shows that around 55 per cent of respondents have ‘high’ or ‘moderately high’ trust in the RBA. That puts us in the upper part of the distribution of Australian public institutions (Graph 8).²⁶

Other central banks using comparable methodologies have similar scores (Graph 9).

Strikingly, our overall trust measure has been broadly stable, or even slightly up, over the past 18 months. That’s a welcome, but frankly quite surprising, result over a period in which interest rates rose, and inflation rose well above the target range.²⁷ Can this really be right? To answer that question we need to look at how the more detailed survey responses have evolved – summarised in Graph 10.

Unsurprisingly, trust in our near-term ability to achieve our objectives, looking one year back and one year forward, has fallen materially (bottom blue bars). Regression analysis finds these outcome measures play a statistically significant role in explaining respondents’ overall trust.²⁸ Australians understand our inflation mandate and hold us accountable when we fail to meet it. Trust does not mean unconditional approval.

But outcomes aren’t the only drivers of trust, as I discussed earlier – and some other drivers have moved the opposite way:

• First, process competence: are our decisions felt to be based on rules, evidence and facts; and do we act promptly when needed? These scores are well into positive territory and have risen a little since February 2025 (upper two blue bars).
• Second, transparency: are we seen as open and honest, and committed to keeping people informed about our decisions and actions; do we provide sufficient information; and do we explain decisions in readily understandable terms? Some of these measures (purple bars) are stronger than others, but all have increased recently – which may have helped offset some of the impact of weaker economic outcomes on overall trust.
• Third, values: are we seen as acting with integrity; do we understand and care about the impact of our decisions; support wellbeing; and maintain appropriate distance from political influence? These measures (green bars) have been more stable, albeit for the most part at a relatively high level. Statistically, these values measures play a particularly strong role in explaining overall trust scores (a result also found for the ECB²⁹). Returning to the concepts in Figure 1, ‘are you on my side?’ matters as much as ‘are you up to the job?’.

These results are encouraging – but they are no grounds for complacency. Our survey has only a short history; confidence in our ability to deliver has declined; and the context in which we operate is becoming more challenging.

Longer-running surveys give some weight to the possibility that trust in central banks has waned over time (Graph 11).

But some of this variation reflects country-specific factors. And the frequently claimed long-run decline in trust in public institutions seems far from universal, at least in Australia.³⁰ While trust in some institutions has declined over time, others (including the civil service, the police and armed forces) have held up (Graph 12). Without a long-term RBA measure, we can’t be sure which group we’re closest to.³¹

Inflation expectations – an alternative, and more longstanding trust measure – also paint a more mixed picture. Longer-term expectations measures derived from financial markets and economists’ forecasts remain well anchored on the 2–3 per cent inflation target here in Australia (Graph 13, lower panel). Other developed countries show broadly the same. But levels aren’t everything. Overseas research suggests that a loss of trust may also be signalled by a tendency for these expectations to respond to changes in monetary policy.³² If markets doubt our determination to bring inflation back to target in a timely way, they may learn about our reaction function from our actions. Reassuringly, Australia’s long-run market expectations have, on average, been quite unresponsive to policy, as in other advanced economies.³³

But domestic prices and wages are set by companies and households, not financial markets – and here the evidence is less reassuring. While we only have a short sample, companies responding to the BOSS survey have become less sure that the inflation target will be achieved over the next 5–10 years (Graph 14). Some contacts in the RBA’s liaison program say similar things. Australia currently has no reliable measure of households’ longer-term expectations,³⁴ but their deep concern about inflation suggests they might feel somewhat similarly.

The third reason for not taking too much comfort is the VUCA world is asking more of us. Volatility means more inflation misses. Uncertainty means bigger expected forecast errors. Complexity means more structural and cyclical drivers of inflation and employment to understand. Ambiguity means genuinely unpredictable geopolitical policy actions. And weaker and more volatile supply makes everything harder still – driving inflation up while activity and employment growth fall.³⁵

A final caution is that trust may be hard won but easily lost. It’s possible that our survey result reflects reserves built up in earlier periods. Those reserves may protect trust against poor economic outcomes for a time. But that may run out: there may be a breaking point we can’t see until we reach it. Our survey can’t identify that threshold – but it isn’t one we want to test. A reserve of trust is not a licence to disappoint. It makes the continuing work of earning trust more pressing, not less.

What is the RBA doing to keep strengthening public trust?

So how are we responding?

Start with values: ‘are they on my side?’ This was a point dear to the heart of Leslie Melville, who helped shape the intellectual foundations for clause 8 of the Commonwealth Bank Act 1945, which charged the Bank with promoting ‘the economic prosperity and welfare of the people of Australia’, and set out the dual mandate of price (originally currency) stability and full employment. That mission remains fundamentally unchanged: it shapes everything we do.³⁶

But delivering on it is never easy. Gains tomorrow may require pain today – but those gains can be hard to quantify, even after the event. And there can be winners and losers in the short run, even if everyone is better off in the long run.³⁷ Against that backdrop, the resilience in the trust scores shown in Graph 10 is cautiously encouraging. But values-based trust is slow to build and quick to fall. So we must continue to listen, through liaison, public engagement and the Community Research Survey itself; learn how our decisions affect different people, especially the vulnerable; and level with people: being open, honest, direct and evidence-based.

But good values aren’t enough: public trust also requires competence – and the persistence of above-target inflation, and the growing threats from a VUCA world, mean central banks have work to do to demonstrate we’re serious about achieving our policy goals. That’s why we’re investing heavily in the skills and tools to navigate uncertainty. Four areas stand out:

• First, understanding future economic challenges: our published research priorities³⁸ describe our key areas of enquiry (including AI, which our chief economist will speak about in coming days). To keep our analysis rigorous and at the frontier, we’re launching a new external engagement program with the research community, publishing more staff work via Insight Notes,³⁹ and establishing a financial markets research and modelling hub.
• Second, investing in our models and frameworks: we’re extending our modelling suite to explore and communicate alternative economic scenarios and their policy implications. The suite includes: MARTIN, our core macroeconometric model; DINGO, a dynamic stochastic general equilibrium model; HALLIE, a next-generation Heterogeneous Agent New Keynesian model; and non-linear structural labour market models.⁴⁰ In financial markets, we have been rolling out a new monetary implementation framework; and our recently published Framework for Additional Monetary Policy Tools explains how we would use our balance sheet if rates again reached the effective lower bound.⁴¹
• Third, better data and intelligence – including WHAM, a new Wealth and Housing Assets Module developed with other government agencies and the Australian Bureau of Statistics;⁴² partnerships with SEEK on labour market data, Macquarie University on BOSS and the Melbourne Institute on household inflation expectations. Our liaison network holds around 900 meetings annually with organisations across Australia: its enormous qualitative and quantitative database can be analysed with increasingly sophisticated AI tools.⁴³ And our Community Research Survey allows us to explore key economic issues for Australian households.
• Fourth, better decisions under uncertainty: we increasingly test alternative scenarios alongside the central forecast. How could the outlook differ – and how should policy respond?⁴⁴ We also regularly review our past forecasts and judgements to learn from them.⁴⁵ The question isn’t simply ‘what do we think will happen?’; it’s also ‘what if things turn out differently?’

But even with better analysis and policy, a VUCA world brings more unexpected inflation misses and uglier trade-offs. And that risks eroding trust in our capacity and commitment, unless we continue to improve our ability to explain ourselves clearly, specifically:

• Why inflation is away from target: unexpected shocks, deliberate tradeoffs, limits in our forecasts and analysis, or the timing and calibration of policy? People need that account to judge whether we’re acting in the public interest, with integrity and competence.
• How we expect the economy to evolve: including the risks and uncertainties, drawing on our analysis of the underlying shocks and the economic mechanisms involved.
• Our strategy for returning inflation to target: how we balance full employment against the speed of disinflation; and how we might respond to material risks that have not yet crystallised. Not a promised rate path – ‘forward guidance’ – but a reaction function that gives people confidence we’ll achieve our goals. That’s an important distinction: explaining how we would respond as circumstances change is different from promising what interest rates will be. We need people to understand the strategy, not mistake it for a guarantee.

Research into how our communications affect different people’s expectations helps guide this work:

• Financial market expectations are important for influencing short-term interest rates and other financial prices, transmitting policy to the economy. Recent research has found that talking about issues like housing risks and wages can shape the weight markets think we’ll place on bringing inflation back to target in a timely way. But sometimes actions speak louder than words.⁴⁶
• Communicating with the public requires different tools. Households interpret economic news and its interest rate implications very differently from markets (Graph 15).⁴⁷ Our survey finds that people pay attention to our actions and communications, and want to know more. But that understanding has important gaps, including on how higher interest rates affect inflation – perhaps monetary policy’s most important single relationship.

So we’re investing in new more targeted communications for households and businesses. The Governor’s media conference is focused on explaining the Board’s decisions, their drivers and what it could mean for people. A more layered approach combines detailed publications like the Statement on Monetary Policy with shorter summaries, website explainers and digital content for different levels of interest and economic knowledge. Recent research finds that even simple explainers can significantly improve accessibility and understanding of key economic relationships, and increase trust.⁴⁸ Experience so far is positive, including better survey scores for openness, honesty and comprehensible explanations. But we must continue to refine our approach: the aim is not to make everyone read the longest publication, it’s to help people find an explanation they can use, without losing the substance of what we’re trying to say.

Accountability brings these elements together. We answer ultimately to the public, including through parliamentary appearances and other public engagement. New five-year reviews of the RBA will strengthen that accountability further, providing periodic assessments of performance and opportunities for improvements. Being independent must never mean being beyond challenge.

These elements must all work together. Clear communication means little unless the actions being taken are pushing us towards good outcomes. Competent action isn’t enough if people don’t understand the need for difficult economic outcomes or trust that we’re trying to do what’s best for the economy. And blind faith in our values cannot by itself sustain trust in a technocratic institution. None of these is a substitute for the others. We need the values, the competence, the explanation and the scrutiny. That’s what makes earning trust a continuing task.

Conclusion

Let me conclude. For a central bank, trust is not a nice-to-have. It’s integral to how monetary policy works. When people trust us to deliver the goals we’ve been asked to pursue, policy is more effective and less costly.

To earn that trust, people must have confidence that we’re on their side, up to the job, and capable of explaining ourselves. Trust in the RBA seems to have held up through a difficult period. But the warnings are clear: some people’s confidence in delivery, particularly on inflation, has fallen. And a VUCA world will keep testing us.

Let me end where I began, with Leslie Melville. In 1931, the Great Depression drove deep fissures into public confidence, and demanded brave and innovative thinking. In response, Melville built what has become the RBA’s Economic Group: establishing the data, analytical frameworks and judgement needed to contribute to the major debates of post-war reconstruction. That combination of public purpose, expertise and evidence helped the Bank earn trust in those difficult years.

There is an irony in the fact that the Bank of England that recommended Melville was led by Montague Norman, whose maxim was ‘never explain, never excuse’. Thankfully, those days are long behind us. Trust is no birthright. We never stop earning it – by continuing to listen, learn and level with the people we serve. To invest in the skills and tools to deliver. Explain our decisions. And welcome scrutiny, including when outcomes fall short.

Melville would have expected nothing less.

Endnotes

* I am extremely grateful to Jonathan Hambur and Peter Rickards, who co-authored this speech with me; and to Sue Black, Anthony Brassil, Michele Bullock, Adam Cagliarini, Anthony Dickman, Samuel Evangelinos, Sarah Hunter, David Jacobs, Brad Jones, Chris Kent, Kevin Lane and Michelle Wright for their comments and suggestions on earlier drafts; and Jack Harden for his help with the survey data and charts.

1 For past lectures, see ANU Research School of Economics, ‘Sir Leslie Melville Lectures’, available at <https://rse.anu.edu.au/seminars-events/sir-leslie-melville-lectures>.

2 For a general discussion of the role of trust in modern life, see O’Neill O (2026), The Reith Lectures – A Question of Trust, BBC Radio 4; for applications to Australian economic policy, see Lowe P (2018), ‘Trust and Prosperity’, Address to CEDA Annual Dinner, Melbourne, 20 November; and Wood D (2026), ‘Growing Trust and Trusting Growth: The Economic Case for Integrity’, Centre for Public Integrity Oration, Sydney, 23 September.

3 See, for example, Carstens A (2023), ‘Trust and Public Policies’, Speech at the High-level Seminar on Central Banking, São Paulo, 19 May. It is sometimes argued that trust plays a similar role in enhancing the efficacy of government policy too (see, for instance, Hetherington MJ (1998), ‘The Political Relevance of Political Trust’, The American Political Science Review, 92(4), pp 791–808), but that is outside the scope of these remarks.

4 As Philip Lowe discussed in his 2019 Melville Lecture: Lowe P (2019), ‘Some Echoes of Melville’, Sir Leslie Melville Lecture, Canberra, 29 October.

5 Hauser A (2025), ‘Monetary Policy in a VUCA World’, Speech at the Australian Financial Review Business Summit, Sydney 5 March; and Bullock M (2026), ‘Monetary Policy in an Era of Shocks’, Speech at the Anika Foundation Fundraising Lunch, Sydney, 28 July.

6 As Loretta Mester discussed in last year’s Melville Lecture, with a focus on the United States: Mester L (2025), ‘The Benefits of Independent Monetary Policymaking in an Uncertain World’, Sir Leslie Melville Lecture, Canberra, 10 July.

7 Bolt J and JL van Zanden (2024), ‘Maddison-style Estimates of the Evolution of the World Economy: A New 2023 Update’, Journal of Economic Surveys, 39(2), pp 631–671; and Keating M (1973), The Labour Force, Employment and Unemployment in Australia, 1911–1966, Australian National University, Canberra.

8 For more on Melville’s influence, see Cornish S (2021), ‘The Central Bank’s First Economist’, RBA Bulletin, December; McFarlane IJ (2022), ‘Sir Leslie Melville: His Contribution to Central Banking in Australia’, Inaugural Sir Leslie Melville Lecture, Canberra, 22 March; and RBA Unreserved, ‘Research Guide: Sir Leslie Melville KBE’.

9 The recommendation to create an economist position, and to appoint Melville to that post, was made by the Bank of England’s Sir Otto Niemeyer and Professor TE Gregory during their 1930 trip to Australia. Sir Ernest Riddle, Governor of the Commonwealth Bank, wrote to Sir Montague Norman, Bank of England Governor, on 25 August 1931: ‘Following on conversations with Sir Otto Niemeyer during his stay in Australia, a recommendation was made by me to my Board of Directors that an economist should be attached to the Bank’s staff, and that the position be offered to Professor Melville of the Adelaide University, of whom Sir Otto and Professor Gregory spoke in very favourable terms. My recommendation was approved by the Board, and the offer of a position on the Staff was accepted by Professor Melville. We have already obtained considerable advantage from his services, and have arranged that he be appointed one of the Committee of Experts to police the carrying out of the Premiers’ Plan by the various Governments.’ See RBA Unreserved, ‘Research Department – General Economic Conditions – Letters to the Reserve Banks – Bank of England Letters – July 1927 – December 1932’.

10 I am grateful to Alex Millmow and Selwyn Cornish for helping me confirm this link to a short obituary of Tony Melville by Donald Smith, available at <https://thehaileyburysociety.org/2024/07/01/anthony-edwin-melville-colvin-house-master/>.

11 In the sociological literature, the first meaning is sometimes called ‘granted trust’ and the second ‘earned trust’.

12 As Onora O’Neill points out in her Reith lectures, transparency and accountability can never fully close the expertise gap, because no matter how many layers of audit and independent assessment are wrapped around an expert endeavour, the non-expert must always rely on some expert’s judgement – whether it be the original actor, or additional experts brought in to assess their reasoning and work. At some point in this potential infinite regress, there must be an act of faith. O’Neill argues that where people choose to draw this line can be better assessed by observing what they do rather than what they say. She notes that people frequently place journalists towards the lower end of trust rankings – yet they devour even poorly sourced press critiques, or gossip, relating to high-profile public figures. Another common tension lies in the contrast between statements that people have little trust in institutions and companies and the trust given to technology providers to protect the types of private data that previous generations would never have dreamed of disclosing. Whether actions or words end up predominating, the bottom line is that a trust-less life is impossible – and technology, far from obviating the need for trust, frequently demands it more than ever.

13 McChesney Martin was chair of the Federal Reserve from 1951–1970; the full speech from which this quotation is taken is McChesney Martin Jr, W (1955), ‘Address before the New York Group of the Investment Bankers Association of America’, New York City, 19 October. As others have pointed out, the metaphor should really be more two-sided: the central bank’s true goal is a sustainable party, not a prematurely short one! If the room is in danger of emptying out too soon, the punch bowl may sometimes actually need topping up, as central banks were required to do in the initial aftermath of the global financial crisis and the onset of COVID. McChesney Martin himself recognised this point when giving testimony to the Senate in 1956: ‘our purpose is to lean against the winds of deflation or inflation, whichever way they are blowing’ (Committee on Banking and Currency (1956), ‘Nomination of William McChesney Martin, Jr.’, Hearings Before the Committee on Banking and Currency, United States Senate, Eighty-Fourth Congress, Second Session, 20 and 27 January). But the original line has endured because it captures an uncomfortable feature of the job.

14 This exercise uses the RBA’s DSGE model DINGO, an updated version of the model laid out in Gibbs C, J Hambur and G Nodari (2018), ‘DSGE Reno: Adding a Housing Block to a Small Open Economy Model’, RBA Research Discussion Paper No 2018-04; and Rees D, P Smith and J Hall (2016), ‘A Multi-Sector Model of the Australian Economy’, Economic Record, 92(298), pp 374–408.

15 Formally, this exercise compares the response to the overseas cost shock in the case where the reaction function has a lower weight on inflation (pink line), and a higher weight (blue line). In the former case, the exercise considers what additional unanticipated policy tightening is required to match the inflation outcomes from the higher weight case.

16 The point here is not to argue that placing a higher weight on inflation is necessarily preferable, as there will be trade-offs with activity. But it highlights that, in trying to achieve a given inflation outcome, it will be less costly if people trust that the central bank is going to achieve that path.

17 These findings are from a special module included in the BOSS as part of a collaboration with the RBA. For more information on the survey, see Macquarie University, ‘Business Outlook Scenarios Survey (BOSS)’.

18 Graph 3 uses results from the RBA’s Community Research Survey, described in more detail below. The ECB results are shown in Figure 6 in Christelis D, D Georgarakos, T Jappelli and M van Rooij (2020), ‘Trust in the Central Bank and Inflation Expectations’, International Journal of Central Banking, 16(6), pp 1–37; and Chart 6 in Dreher F (2024), ‘Trust in the ECB – Insights from the Consumer Expectations Survey’, ECB Economic Bulletin, Issue 3.

19 Haldane A (2020), ‘What Has Central Bank Independence Ever Done for Us?’, Speech to the UCL Economists’ Society Economics Conference, London, 28 November.

20 See, for instance, Alesina A and LH Summers, (1993), ‘Central Bank Independence and Macroeconomic Performance: Some Comparative Evidence’, Journal of Money Credit and Banking, 25(2), pp 151–162; Klomp J and J De Hann (2010), ‘Inflation and Central Bank Independence: A Meta-analysis’, Journal of Economic Surveys, 24(4), pp 593–621; and Haldane, n 19.

21 Based on the CBIE (Central Bank Independence – Extended index) scores of 156 central banks in CBI Data. Relative to pre-COVID, only three central banks have a lower score (Indonesia, Liberia and Iceland); 129 are unchanged; and 24 have a higher score. For further discussion of these measures, see Romelli D (2024), ‘Recent Trends in Central Bank Independence’, CEPR VoxEU.

22 See, for instance, Ipsos (2026), ‘What Worries the World’, June; European Union (2026), ‘Standard Eurobarometer 105 – Spring 2026’, May; Saad L (2026), ‘Affordability Still Dominates Americans’ Financial Worries’, Gallup News, 28 April; and, for Australia, Graph 2 in Harden J, P Rickards and M Wright (2026), ‘Listening to Australians: A New RBA Survey of the Community’, RBA Bulletin, July.

23 See for example Forbes K, J Ha and M Ayhan Kose (2026), ‘Tradeoffs Over Rate Cycles: Activity, Inflation and the Price Level’, NBER Macroeconomics Annual, 40, pp 191–277; Bullock (2025), ‘The RBA’s Dual Mandate – Inflation and Employment’, Speech at the Anika Foundation Fundraising Lunch, Sydney, 24 July.

24 Data for Australia and other countries come with thanks from Kristin Forbes, based on Forbes K (2026), ‘Muted or Mismeasured? Monetary Policy Trade-offs over 50 Years of Rate Cycles’, Proceedings of RBA Annual Conference, Sydney, 8–9 October, using updated data and estimates from Forbes K, J Ha and M Ayhan Kose (2026), ‘Tradeoffs Over Rate Cycles: Activity, Inflation and the Price Level’, NBER Macroeconomics Annual, 40, pp 191–277.

25 The RBA survey is discussed in more detail in Harden, Rickards and Wright, n 22; and the Governor spoke to some of the early findings in her March 2026 speech: Bullock M (2026), ‘Listening to Australians, Interpreting the Data and Setting Monetary Policy’, Speech at the Australian Financial Review Business Summit, 3 March.

26 OECD results taken from OECD (2025), Drivers of Trust in Public Institutions in Australia, OECD Publishing. The OECD trust measurement framework, which the RBA also uses, is set out in Brezzi M, S Gonzalez, D Nguyen and M Prats (2021), ‘An Updated OECD Framework on Drivers of Trust in Public Institutions to Meet Current and Future Challenges’, OECD Working Paper on Public Governance No 48.

27 DemosAU (2026), ‘TrustWatch: Trust in Australia’s National Institutions’, April finds a similar result, suggesting that while trust in most Australian institutions fell between January 2025 and April 2026, trust in the RBA showed a slight rise.

28 Similar results have been found for the ECB: Eickmeier and Petersen show that achieving policy objectives is highly important to an individual’s level of trust in the ECB. See Eickmeier S and L Petersen (2024), ‘Beyond Technocracy: Toward a Holistic Approach to Central Bank Trust’, NBER Working Paper No 32716.

29 Eickmeier S and L Petersen (2024), ‘Beyond Technocracy: Toward a Holistic Approach to Central Bank Trust’, NBER Working Paper No 32716.

30 Trust measures vary somewhat in definition across countries: the Bank of England’s is more focused on inflation outcomes; the Federal Reserve’s on leadership; while the Bank of Canada and European Central Bank measures are more broadly defined.

31 The World Values Survey also suggests that the share of Australians who agree with the statement ‘most people can be trusted’ has risen over the past quarter century, from 39 per cent in 1998 to 49 per cent in 2022 (see World Values Survey Database and Ortiz-Ospina E, M Roser and P Arriagada (2024), ‘Trust’, Our World in Data, April).

32 See, for example, Boccola L, A Dovis, K Jorgensen and R Kirpalani (2025), ‘Monetary Policy without an Anchor’, NBER Working Paper No 34436; and Carvalho C, S Eusepi, E Moench and B Preston (2026), ‘Anchored Inflation Expectations’, American Economic Journal: Macroeconomics, 15(1), pp 1–47.

33 Ryan C (2026), ‘Discussion of Monetary Policy without an Anchor’, Proceedings of RBA Annual Conference, Sydney, 8–9 October.

34 Since February 2026 questions on longer-term inflation expectations have been included in the Community Research Survey. The RBA is also working with the Melbourne Institute to introduce an extension to its current survey to capture longer-run inflation expectations.

35 See also Bullock, n 5; and Hunter S (2026), ‘Understanding Supply Shocks and Their Implication for Monetary Policy’, Address to the Australian Conference of Economists, Canberra, 8 July.

36 The 1945 text and the current text are available from the Federal Register of Legislation.

37 The trust scores in the survey bring this perception out clearly. For example, Australians with a mortgage currently have considerably lower trust in the RBA than those without a mortgage, even after accounting for other factors. In periods where interest rates are declining, it is possible this pattern would reverse. These and other distributional issues were covered in detail at the recent 2026 RBA Annual Conference on Monetary Policy Trade-offs. We have re-doubled our efforts in recent times to communicate the different transmission channels for monetary policy: see, for instance, RBA (2026), ‘How Do Higher Interest Rates Lower Inflation?’.

38 RBA, ‘Our Research Agenda’.

39 RBA, ‘Insights: An RBA Staff Series’.

40 See Iao T and E Shabalina (2026), ‘Monetary Policy Trade-offs Under Imperfect Central Bank Information’, Proceedings of RBA Annual Conference, Sydney, 8–9 October; and Brassil A and C Ryan (2025); ‘A New Keynesian ‘Plucking’ Model of Unemployment and Inflation’, Proceedings of RBA Quantitative Macro Workshop, Sydney, 11–12 December.

41 See Jacobs D (2026), ‘The Road to Ample – Towards a Demand-driven Liquidity Regime’, Speech at the RBA, Sydney, 25 August; Kent C (2026), ‘Additional Monetary Policy Tools: Reflections and a New Framework’, Speech at the RBA, Sydney, 29 June; and RBA (2026), ‘Framework for Additional Monetary Policy Tools at Low Interest Rates’, June.

42 Bradshaw N, J Hambur, M McCarthy and S Nash (2025), ‘Overview and Early Analysis of the New Wealth and Housing Asset Module (WHAM)’, Proceedings of the ABS and RBA Conference, Sydney, 23–24 June. This is a partnership with the Treasury, Department of Social Services, Department of Industry Science and Resources, the Australian Tax Office and the Australian Bureau of Statistics.

43 Gray N, F Lattimore, K McLoughlin and C Windsor (2025), ‘An AI-powered Tool for Central Bank Business Liaisons: Quantitative Indicators and On-demand Insights from Firms’, RBA Research Discussion Paper No 2025-06.

44 For further details, see Hauser, n 5; and Hunter S (2025), ‘Monetary Policy: Forward Looking and Data Dependent in the Face of Uncertainty’, Speech at the Australian Financial Review Banking Summit, Sydney, 18 March. See Chapter 3: Outlook in RBA (2026), Statement on Monetary Policy, May for a public example where two adverse scenarios about the Middle East conflict and oil prices were presented. Chapter 4: Outlook in RBA (2025), Statement on Monetary Policy, May included alternative scenarios of a ‘trade war’.

45 See, for instance, RBA (2025), ‘Box B: Annual Forecast Review’, Statement on Monetary Policy, November.

46 Hambur J and Q Haque (2026), ‘Perceived Monetary Policy Rules: Evidence from Professional Forecasters in Australia’, RBA Research Discussion Paper No 2026-06.

47 Hambur J, M McCarthy, S Munn and E Shaw (2026), ‘Rate and See: Household Interest Rate Expectation Formation in Australia’, CAMA Working Paper No 76/2026.

48 McMahon M, R Rickards and R Rholes (2024), ‘Teach Don’t Give: Coordinating Mental Models With Communication’, Working Paper, mimeo.