RBA Annual Conference – 2018 Introduction John Simon

Twenty-five years ago the Governor of the Reserve Bank of Australia (RBA), Bernie Fraser, gave a speech declaring that ‘My own view is that if the rate of inflation in underlying terms could be held to an average of 2 to 3 per cent over a period of years, that would be a good outcome’ (Fraser 1993, p 2). While this did not mark the formal adoption of inflation targeting in Australia – that would take a few more years – it is as good a point as any to mark the de facto beginning of inflation targeting in Australia.

In the 25 years since, many things have changed. Where recession, high inflation and high unemployment were once common, there has not been a recession in Australia since the start of inflation targeting, inflation has fallen from around 8 per cent to an average of 2–3 per cent and unemployment has fallen from over 10 per cent to average closer to 5 per cent. But, despite these significant macroeconomic improvements, the global financial crisis (GFC) has challenged the practice of central banking.

Around the world, questions are being asked about whether the flexible inflation-targeting framework used by many central banks is the most appropriate framework given the experience of the crisis. Thus, it is a good time to consider whether the inflation-targeting framework that has served Australia so well over the past 25 years is well adapted to the next 25 years.

Reflecting this context, the title of this conference, ‘Central Bank Frameworks: Evolution or Revolution?’, implicitly asks whether the current framework needs to change and, if so, how dramatically. However, while many people think about evolution as a slow process that proceeds gradually and revolution as one that proceeds quickly, the rate of evolutionary change is not necessarily always constant or slow. Within the field of evolutionary biology there are two broad characterisations of the process of evolution: phyletic gradualism, which is most similar to the popular view of evolution as a slow and gradual process, and punctuated equilibria that emphasises the alternation of periods of rapid change with long periods of stasis. The distinction with revolution is not so much the speed of change as the fact that what emerges from evolution is still recognisably similar, while revolution leads to something distinctly different from that which went before.

When thinking about the practice of central banking it seems, at a distance at least, to be characterised by both revolution and punctuated equilibria. There are typically long periods of stability interspersed with relatively rapid change. For example, the gold standard and Bretton Woods systems prevailed for long periods of time before being replaced by very different arrangements. Furthermore, just as evolutionary change is a response to external pressures, so we can also think about the evolution of central bank frameworks. The external pressures on central banking have ebbed and flowed over the years. When those pressures are large, such as around the late 1960s and early 1970s, the frameworks have evolved rapidly. When those pressures are more benign, such as during the Great Moderation, evolution has been slower.

A question this conference is particularly focused on is whether the pressures on inflation targeting over recent years are such that rapid evolution of, or even a revolution in, the monetary policy framework is necessary or whether the current framework is well adapted to the current post-GFC environment.

To answer these questions, the conference was organised in three sections. The first looked at the experiences of New Zealand, Canada and Australia. An important focus of this section was how the various regimes had evolved over the course of the twenty-something years they had been in operation. By understanding what has happened so far, we should be better placed to think about the way things might change in the future. The second section looked at changes in financial markets and the macroeconomy since the introduction of inflation targeting in Australia. By establishing what changes have occurred in the economy, it also provides pointers to ways the inflation-targeting regime might have to adapt. The third section then considered alternatives to the current arrangements. If the framework was going to change, how should it change? The conference concluded with a panel discussion that synthesised the various papers and reflected on what had been learnt over the course of the two days.

The three papers in the first session, looking at the experiences of New Zealand, Canada and Australia, highlight both how inflation targeting has evolved since it was first adopted by New Zealand in 1990 and the differences between the three inflation-targeting regimes. Inflation targeting is not an unchanging framework exactly replicated across countries, but rather a framework that has been adapted to the various country-specific institutional environments it has been used in. An example, perhaps, of niche evolution. For example, while initial formulations of inflation targeting were relatively strict, the dominant form today is ‘flexible inflation targeting’ that places greater emphasis on unemployment or output stabilisation.

The paper on New Zealand by John McDermott and Rebecca Williams highlights that New Zealand's initial choice of a relatively strict inflation-targeting regime grew out of the need to establish the Reserve Bank of New Zealand's (RBNZ's) institutional credibility. It also seems likely that, as the first central bank to adopt inflation targeting, there may have been a broader need to establish the credibility of the regime itself. The framework was designed with four pillars, or stakes, chosen to support the growth of the newly planted regime: operational independence; transparency; a single objective; and a single decision-maker. They note that there have been some changes to the single objective. From a relatively strict objective it has evolved into a more flexible objective and the particular numerical target has also changed over time. In particular, they note that the government of the day has frequently led these changes. That is also true about the most recently announced changes to the RBNZ's framework – replacing the single objective with a dual mandate, adding unemployment to the inflation mandate, and replacing the single decision-maker with a monetary policy committee. Notwithstanding the evolution of the regime, it has delivered low and stable inflation in New Zealand.

The paper on Canada by Thomas J Carter, Rhys Mendes and Lawrence L Schembri emphasises three factors that are seen as central to the success of the Canadian regime: a simple, readily understood specification of the inflation target; the recognition that the government shares the duty of achieving the target and should set non-monetary policies in a way that is coherent with the achievement of the target; and, finally, the regular review of the regime, which has occurred every five years since 2001. As with New Zealand, the regime delivered low and stable inflation. The authors also note that the strength of the regime was demonstrated throughout the GFC. In particular, they emphasise the importance of the joint responsibility for macroeconomic outcomes shared by the central bank and the government – supportive macroprudential policies freed the central bank to focus on macroeconomic risks.

Finally, in the first session, Guy Debelle reflects on Australia's experience. He emphasises the relative stability of the regime and evolutionary nature of Australia's experience. In particular, the Reserve Bank of Australia's mandate has been unchanged since it was first enacted in 1959. The evolution has occurred through the broader policy framework and the way that the mandate has been interpreted and operationalised. Of note, the inflation-targeting regime has been a flexible inflation-targeting regime from the start, in part reflecting the mandate of the RBA, which includes the maintenance of full employment, the stability of the currency, and the welfare of the people of Australia. Debelle also emphasises the fact that, as the regime has evolved, the communication of the RBA has had to evolve along with it. He notes that communication has played an increasingly important role in the operation of the regime given the centrality of inflation expectations to a flexible inflation-targeting regime.

The second session of the conference contained two papers that served as a bridge between the primarily backward-looking first session and the forward-looking third session. The first, by Anthony Brassil, Jon Cheshire and Joseph Muscatello, looked at the transmission of monetary policy through bank balance sheets and how that might have changed during the inflation-targeting period. The second, by Luke Hartigan and James Morley, looked at how the transmission of monetary policy through the macroeconomy might have changed as a result of the adoption of inflation targeting. These papers draw out the way the economic environment has changed since the introduction of inflation targeting. As such, they offer some pointers to ways the inflation-targeting framework might have to evolve.

Brassil, Cheshire and Muscatello consider the way the RBA's policy rate is transmitted to the interest rates Australians pay on their mortgages and receive on their deposits. They document a number of interesting findings, including the fact that, following a reduction in the cash rate, the increase in the relative cost of deposit funding is broadly offset by a reduction in expected loan losses. Their main finding is that, while cash rate changes between 2003 and 2012 were fully passed through to the major banks' lending and deposit rates (in aggregate), pass-through since 2012 has fallen to around 90 per cent as the major banks' return on equity has not fallen with the cash rate. They further highlight the increasing importance of low- or no-interest deposits in bank balance sheets – as interest rates fall these accounts become relatively more expensive, and their share of banks' funding increases at low rates. This points to the possibility that the ‘zero lower bound’ may be a more important consideration for monetary policy frameworks than was the case when inflation targeting was first being developed.

Hartigan and Morley use a factor model of the Australian economy to capture information from a wide range of economic indicators and distil it into a useable form. Having done this, they find that inflation targeting has been associated with a substantial reduction in the common components of economic volatility with little change in the idiosyncratic elements. This finding is highly suggestive that inflation targeting has been particularly successful at stabilising the economy. Importantly, they find that this stabilisation applies not only to nominal aspects of the economy, that one might expect to have been most affected by inflation targeting, but also the real aspects. A second implication of their results, however, is that it is much harder to measure the common component of economic activity (because the idiosyncratic elements are now relatively larger) and so policymakers need to look at a much wider range of indicators in order to correctly judge the state of the economy. A corollary seems to be that communication will be more difficult in this environment because there may be no single indicator policymakers can point to when explaining the reasons for their actions.

The third section of the conference contained three papers. The first, by Warwick J McKibbin and Augustus J Panton, considers whether there might be better frameworks than inflation targeting and proposes nominal income targeting as a superior framework. The second, by Ben Broadbent, considers the relationship between monetary policy and macroprudential policy and whether the experience of the GFC argues for a closer relationship between the regulators. The final paper, by David Archer and Andrew T Levin, considers the appropriate decision-making body for a central bank and, in particular, how a monetary policy committee should be structured.

McKibbin and Panton review a range of alternative monetary policy regimes and compare them with key criteria for a monetary regime. They ask such questions as: How well does the regime handle shocks? Can the target be credibly measured and understood? How transparent is the regime when exceptions are needed? Are prices expectations anchored by the regime? After considering how well each of the alternatives do on these criteria, they suggest that nominal income targeting would be a good regime and one that is superior to the current flexible inflation-targeting framework. An important reason for their conclusion is that, while inflation targeting deals with demand shocks well, it is less well suited to dealing with supply shocks; nominal income targeting, on the other hand, can handle supply shocks better.

Ben Broadbent considers whether macroprudential policy should be conducted jointly with monetary policy, and whether it should be done within the same institution, or separately. He makes the argument that full integration of the two policies could compromise accountability. In particular, he notes that the nature of the objectives of monetary policy and macroprudential policy are quite different. While monetary objectives are clear and verifiable, macroprudential objectives are multiple, opaque and hard to verify. If both these kinds of objectives are merged, there will be an inevitable problem as the clear and verifiable objectives crowd out the opaque objectives. He also notes that the benefits from formal coordination are overstated and, at least in open economies with floating exchange rates, financial stability depends much more on prudential policy than on monetary policy.

Archer and Levin focus on developing a set of robust design principles for monetary policy committees that mitigate the risk of political interference and groupthink. They argue that independence from political interference rests not on statutory independence but on public confidence and the legitimacy of the institution. As such, transparency and public accountability are important to the extent that they support the legitimacy and, thus, independence of the institution. To guard against groupthink they argue that the monetary policy committee should be made up of a diverse group of experts who are individually accountable for their policy decisions.

The conference closed with a panel discussion moderated by Jessica Irvine. The panellists were Philip Lowe, Adam Posen and Sayuri Shirai. The discussion served to draw the various threads of the conference together. In summarising the areas of agreement at the conference, one panellist suggested that participants generally agreed that: regimes matter, flexibility is important, transparency is important, committees bring benefits to the decision-making process and inflation-targeting regimes have enjoyed wide political support. In thinking about Australia's experience, it was felt that Australia's inflation-targeting framework had worked well for the past 25 years and an important part of that was the flexibility it had built in from the start. Looking forward, the discussion considered whether the shocks Australia is likely to face in the future might be different to the shocks it has faced in the past and, if so, there was a possibility that the inflation-targeting regime might have to evolve further. Nonetheless, while it was generally agreed that it was good to consider alternatives to the current regime, and to do so from a position of strength before problems emerged, there was no obvious need for change.

This conference volume collects together the papers presented at the conference, the discussants' comments and a summary of the general discussion that followed each presentation. As the conference is run under the Chatham House Rule, individual participants are not identified in the general discussions.


Fraser BW (1993), ‘Some Aspects of Monetary Policy’, RBA Bulletin, April, pp 1–7.