Speech Monetary Policy in Economic Expansions
The Fifth Chris Higgins Memorial Lecture, the Australian Academy of Science
It is an honour to be invited to deliver the fifth Chris Higgins Memorial Lecture. I knew Chris from 1970 till his death in 1990. He was, in my opinion, the best Australian applied macroeconomist of his era and a policy-maker of distinction. His views on economics would place him amongst those whom Alan Blinder identified as having hard heads but soft hearts. He could have held a Chair at any Australian university and many overseas ones of note if that had been his objective; instead, he had an outstanding career at the OECD and the Australian Treasury, where he rose to become Secretary in 1989. Chris was also an excellent companion – a man of taste, wide reading and wit. It is good to see so many of his friends in the audience tonight.
Chris' death in 1990 robbed us of a good friend, and Australia of an outstanding public servant and economist. I also cannot help but think it was a great shame that Chris did not live to see the 1990s unfold. His professional life was concentrated in the 1970s and 1980s – two periods of relative turmoil – and he did not see many of the things he stood for bear fruit in the 1990s. I would like to take this observation as the theme for my lecture tonight.
Growth over Five Decades
I would like to start by looking at economic growth in Australia over the past half-century in terms of decade averages (Table 1). If we do this, the variation in growth rates is not nearly as great as when we compare different phases of the cycle. This is largely because each of the decades contained a recession (or more than one by some estimates) as well as periods of growth. The 50s and 60s showed the strongest average growth at 4.2 per cent and 5.4 per cent per annum respectively, while the 70s, 80s and 90s showed very similar growth rates in the 3.3 to 3.4 per cent range.
It is disappointing that we could not keep up the 50s and 60s performance into the later decades, but nor could other countries. There was a trend slowdown in growth around the world as the period of post-war reconstruction passed into history. In fact, the fall-off in growth for OECD countries was considerably more pronounced than for Australia, so Australia's relative performance improved towards the end of this fifty-year period. While economic growth in Australia was lower than for the OECD on average over the 50s and 60s, it was higher in the 80s and 90s, with the biggest advantage in Australia's favour occurring in the 90s.
This is a surprise to some people and one response is to query it on the grounds that our recent relatively buoyant economic growth could have been mainly the result of our faster population growth. Table 2 examines this issue by taking out the effect of population growth and looking at real GDP per capita over the same five decades. On this basis, the recent relative improvement is now more pronounced than in the earlier table. The 90s not only shows a pick-up in Australia's growth of real GDP per capita compared with the two previous decades, but it is the only decade in which Australian GDP per capita has grown faster than the OECD area, and by an appreciable amount.
I do not want to suggest that our recent economic performance means that we have returned to a golden age such as we had in the immediate post-war period. Nor do I wish to deny that there is still a backlog of remedial work to be done – our unemployment rate being an obvious example. But, on the other hand, we must be doing something right. Our expansion so far in the 1990s has outlasted its predecessors in the 1970s and 1980s, and it is still going strong. We have withstood a difficult external shock, and should be able to sustain the expansion a good deal longer.
Why the Improvement?
The fact that the annual growth in GDP per capita has increased in the 90s, and that it is now faster than in OECD countries, is a welcome development. The major reason for improvement in GDP per capita is that a closely-related variable, namely labour productivity, has also shown a clear improvement in the 1990s. In addition, multi-factor productivity – which is a little further removed from GDP per capita – has also shown a clear lift compared with earlier decades. Table 3 shows a comparison of these measures of productivity over the three most recent expansions.
|Note: The expansion periods are: March 1975–June 1982, March 1983–June 1990 and June 1991–June 1999.|
The reason for the pronounced rise in productivity is essentially microeconomic. It is because labour is being used more efficiently, capital is being allocated to areas of the highest productivity and innovation is occurring more rapidly than before. Partly this may be due to increased managerial efficiency, but a more likely explanation is to be found in the policy changes designed to make the economy more flexible and competitive – in other words, policies designed to affect the supply side of the economy. Among these I include financial market deregulation, tariff reductions, privatisations, competition policy and decentralisation of the labour market. These have all made a contribution, but it is probably the interaction between them that is more important; in this sense, the total effect is more than the sum of the parts. I have talked on this subject on earlier occasions, so I will not repeat myself tonight, other than to observe that these have brought tangible benefits that are often overlooked by the beneficiaries.
The other distinguishing feature of the 1990s is that the expansion has continued for longer. National accounts data up to the June quarter of 1999 show eight years of expansion at an average rate of 4 per cent per annum, already exceeding the length of the previous two expansions. All forecasters are expecting significant growth in the current financial year, so that, even on the most pessimistic assumptions, the present economic expansion will be a lot longer than its predecessors in the 70s and 80s.
More importantly, in earlier expansions, serious imbalances had built up by this stage. In the early to mid 70s and the early 80s, inflation was in double digits, partly as a result of a surge in wages. In the late 80s, asset prices and credit growth were rising to unsustainably high levels, and consumer price inflation was still excessive. Both the dynamics of the business cycle itself, and the need for tough anti-inflationary monetary policy, pointed to an abrupt end of the expansion. On this occasion, the picture is different and no-one is expecting an abrupt end.
Why has the Expansion been Steadier and Longer this Time?
I would now like to make a few comments on why the expansion of the 90s has been steadier and longer than its two predecessors. Here I think we have to look for essentially macroeconomic explanations.
The first explanation is to be found in a comparison of the rates of inflation this time compared with the two previous expansions. It is curious now to look back to some of the economic debates of the 60s and 70s. In those times, there were many people who thought that you had to tolerate ‘a little bit of extra inflation’ to make sure the economy would grow. Many, if not the majority, thought that getting inflation down again would not be worth the price, because it would result in permanently lower growth. The short-term trade-off (the Phillips Curve) was incorrectly interpreted as being a summary of our medium-term choices.
Now when we look back (Table 4), nothing could be further from the truth. The low inflation decades – 50s, 60s and now the 90s – are the ones where we did well on growth in absolute terms, or in relative terms. The high inflation decades – the 70s and 80s – were the ones where our growth performance was at its poorest.
|(a) If the short-lived Korean War boom is excluded, the average inflation rates in the 1950s would be 2.5 per cent for Australia and 1.7 per cent for the OECD.|
The most reasonable explanation for the lower inflation and hence steadier and longer expansion in the 1990s is the improvement in macroeconomic policies. By this I mean monetary and fiscal policies, or what used to be known as demand management policies.
I will not say very much about fiscal policy, other than to note that medium-term considerations now play a much larger role, something that Chris Higgins consistently argued for. The improvement has occurred in two stages after the disaster of the 1970s. In the late 80s, and again in the mid 90s, the Budget was brought back into surplus where it remains at present. As a result of this fiscal consolidation, the medium-term health of the Government's accounts has improved greatly. One important indicator of this is the stock of government debt to GDP, which is now lower for Australia than for virtually any other OECD country.
On monetary policy, I will say a little more. A similar realignment towards a medium-term approach has certainly occurred in the case of monetary policy. The centrepiece of this has been the inflation target and the Government's reaffirmation of the independence of the Reserve Bank as outlined in our Act. A similar realignment has occurred in a number of other countries, and a model of this type, whether explicit or implicit, is now the mainstream international approach to monetary policy.
This approach has been a major factor behind the low inflation of the 1990s, both here and in most other OECD countries. A central channel through which this approach operates is through maintaining low inflationary expectations. Wage and price setters no longer need to engage in the defensive behaviour they formerly used to protect themselves against future rises in inflation. Similarly, opportunistic tactics aimed at profiting from inflation no longer make sense as a business strategy. And finally, the assumption by businesses that there is no need to resist increasing costs because they can simply be passed on to consumers by raising prices has had to be discarded.
While the inflation target has played an important part in maintaining low inflation, there is more to monetary policy than just setting the target – there also has to be a willingness to act in a timely way. This is best illustrated by events in the period from 1994 to 1996 (Graph 1). Around mid 1994, the economy was growing strongly and there were signs that inflation and wages would soon pick up (this expectation was reflected in a number of places including the yield curve). The first tightening of monetary policy occurred in August 1994 and was followed by two others before the end of the calendar year. It is important to note that at the time the tightening occurred, the current inflation rate was 2 per cent (four-quarter-ended underlying inflation). Thus the tightening was pre-emptive – it was based on an assessment of the outlook rather than current experience. In time, inflation did rise and peaked at 3.3 per cent per annum in the year to March 1996.
Pressures soon abated as the economy slowed and wage claims moderated. By July 1996, the first of a series of monetary policy easings occurred. Again this was pre-emptive because the inflation rate at the time was still above 3 per cent. But, because inflation was forecast to return to the middle of the band, the period of tighter monetary policy that had prevailed since end 1994 was no longer needed, and so interest rates could be taken back to more appropriate levels.
The pre-emptive nature of these policy changes, plus the Government's affirmation of the inflation target and the independence of the Reserve Bank contained in the Statement on the Conduct of Monetary Policy in August 1996, significantly increased the effectiveness of monetary policy and the credibility of the Reserve Bank. This put us in a good position later to handle the contractionary effects of the Asian crisis and the associated period of turbulence in financial markets.
The textbook response to a contractionary external shock such as the Asian crisis is not, in our view, to tighten monetary policy. But the turmoil in financial markets, such as a plunging exchange rate and widening bond spreads (both indications of capital flight), can sometimes only be settled by such a show of forceful action. Most countries in the region, whether developed or emerging markets, chose to, or were forced to, raise interest rates at some stage during the Asian crisis, with subsequent detrimental effects on their economic growth and employment. The fact that we were able to withstand the pressures without doing so (in fact, reducing rates slightly in December last year) is a testimony to the increased credibility of monetary policy in Australia, and to the higher reputation that the Australian economy overall commands in the international market place. It is also an illustration of how timely action, such as in 1994, can ultimately contribute to a longer and more robust expansion.
Before concluding, I would like to make a couple of other observations on this subject. The first is that pre-emptive monetary policy action only refers to being pre-emptive vis-à-vis actual developments in the economy; it does not mean being pre-emptive vis-à-vis the financial markets' assessments. Financial markets are looking at the same data as the Reserve Bank, are making forecasts and calculating the probabilities of monetary policy action. Interest rates on bills and bonds always move in anticipation of monetary policy action. To be pre-emptive vis-à-vis the market would be the same as taking the market by surprise. No-one should regard this as a worthwhile objective, although it will occur from time to time. In the modern world of greater transparency and accountability, such surprises should be rarer and rarer as the market becomes more aware of the central bank's objectives and modus operandi.
The other implication of this new world of monetary policy is that policy changes will probably be a good deal smaller than in the past. Just as the tightening of 1994 was much smaller than its predecessors in the 1980s, it is reasonable to assume that this trend will continue. The past three years have hardly been a placid period for the world economy, yet in successful economies the movements in interest rates have been relatively small (for example, in the United States they have moved through a range of ¾ of a per cent).
The other thing we are seeing is that financial factors are playing a larger role than before. The biggest move in US rates over the second half of the decade was the easing in late 1998 as a result of the ‘credit crunch’ which followed the Russian default and the demise of the hedge fund LTCM. At present, US monetary policy is, of necessity, partly operating through the medium of the stock market as that market moves in anticipation of Fed tightenings. Fortunately in Australia we do not have as highly valued a stock market to contend with, but we cannot ignore these considerations entirely. Our household sector is now a much larger holder of equities and, at the same time, is more highly leveraged than in the past. This is bound to affect the transmission mechanism for monetary policy.
My final comment is that whatever monetary policy does, there will be those who disagree with the decision. This is inevitable and probably healthy. Everyone is entitled to their own view, and has a right to express it. That will always be the case. What has changed over recent years is that, by and large, those who disagree with a decision no longer reach for the ready excuse of claiming that it was ‘only done for political reasons’. This change has been a long time in coming, but now that it has arrived, it is a huge advance. It makes it easier for monetary policy to do what it judges to be right and not be inhibited by fears of public misperception – whether that means temporarily higher interest rates or temporarily lower interest rates – than formerly. In the long run, however, it almost certainly means on average lower and less variable interest rates for the reasons I have outlined above.
Alan S Blinder (1987), ‘Hard Heads, Soft Hearts: Tough-Minded Economics for a Just Society’, Addison-Wesley, Reading, MA. 
EA Boehm identified two recessions in the 1970s, the first in 1974 and the second from August 1976 to October 1977. See ‘The Usefulness and Applications of Economic Indicator Analysis’, Melbourne Institute Working Paper No. 9/89.