Speech Economic Conditions and Prospects
Glenn Stevens
Governor
Address to the Australian Business Economists and the
Economic Society of Australia (NSW Branch)
Annual Forecasting Conference Dinner
Sydney –
PDF version 55KB
It is nice to be back at an ABE-Economic Society function in a new capacity. Thank you for the invitation.
Tonight I will take the opportunity to make some brief remarks about the monetary policy framework, then go on to the current state of the economy and the associated issues for monetary policy. I would like to conclude briefly with a longer-term perspective.
The Monetary Policy Framework
As you know, my predecessor on his appointment in 1996 reached a formal agreement with Treasurer Costello on the conduct of monetary policy.[1] This was updated in 2003 at the time of his reappointment.[2] These Statements featured a numerical inflation target, to be achieved over the medium term, for consumer price inflation, and noted an appropriate degree of flexibility in the conduct of policy over the short term. They emphasised the independence of the Reserve Bank, as provided under legislation, in the conduct of monetary policy. They provided for accountability to Parliament and provision of information to the public. And they recorded the commitment both of the Governor and of the Government to the arrangements.
These arrangements have come to be well understood and widely accepted around the country, and around the world. Preserving the purchasing power of money is the most important contribution that monetary policy can make to sustainable prosperity. Having a medium-term numerical target for inflation – in our case, 2–3 per cent on average – operated by an independent central bank, remains for Australia (and many other countries) the most straightforward way of giving practical effect to that overall goal.
To date, moreover, the system has worked well. For a start, the target has been satisfactorily achieved. In 1995, a colleague[3] and I wrote about Australia's inflation target that:
…if some years from now we can look back and observe that the average rate of inflation has a ‘2’ in front of the decimal place, that will be regarded as a success.
From the vantage point of 2006, we can look back and see that the average inflation rate has indeed had a ‘2’ at the front. The number can be calculated a few different ways, but all give an answer of about 2½ per cent over the past decade.[4] Inflation has been outside the 2–3 per cent range about half the time. That degree of flexibility was always intended because inflation cannot be fine‑tuned over short periods, and shocks occur that will push it away from target. But, importantly, there has been no systematic tendency for the deviation to be one way or the other.
At the same time, variability in real GDP has tended to decline. Several factors have contributed to that but it is important to state that, over the long run, controlling inflation does not harm growth; on the contrary, it leads to an improvement in growth prospects.
With this record of success, the desirability of continuing the system is obvious. At times of changing personnel, moreover, it is worth stressing continuity: there was no case for a discrete change to the system just because a new Governor was being appointed.
The Treasurer and I issued a new Statement on the Conduct of Monetary Policy[5] on 18 September 2006, the day my appointment became effective. The language was identical in almost every respect to the 2003 Statement. The changes were limited to those necessary to update the document and reflect a new incumbent. What this says is that the well-understood framework of inflation targeting, central bank independence and accountability will continue over the years ahead.
I shall turn now to an evaluation of trends and prospects for the global and local economies in turn.
The Global Scene
It is apparent that demand in the US economy is now growing more slowly than it was a year or two ago. Recovery from the shallow 2001 recession is, by 2006, fairly mature and there has been some rise in inflation. So some slowing in demand is welcome and necessary.
At the present time, the debate is over the extent of that slowing – whether it will be enough to take the pressure off prices, or whether it might in fact be too great, resulting in unduly weak economic activity. Observers have had a hard time this year deciding which of these problems was the larger concern. Early in the year, a string of biggish monthly CPI readings had everyone very worried about inflation. More recently, declining forward indicators of housing construction and softness in housing prices have seen markets and economists more concerned about a weak economy. Long-term interest rates have retraced about half their rise in the first part of the year.
With the effects of a buoyant housing market thought to have been an important expansionary force in the US in earlier years, the recent change in sentiment in that market is understandably regarded as significant. Australian experience suggests, as does that of the UK, that the end of a housing price boom can have noticeable effects on aggregate demand. But those experiences also suggest that such effects are manageable. In Australia's case, the resources boom coincided with the housing moderation and helped to dampen its effects, but that has not been the case for the UK, which has had broadly similar economic outcomes to our own. On this basis, one would think that there are reasonable prospects for moderate growth in the US economy in the period ahead. But this is obviously an area of uncertainty, and even a favourable outcome involves slower growth in US aggregate demand in the future than we have tended to see over most of the past decade.
A slowing in the US economy is coinciding with a more positive picture in the euro area and Japan than we have seen over recent years. Japan looks more and more like it is finally escaping the stagnation that followed the excesses of the late 1980s and early 1990s. China has continued to grow with remarkable strength. To the extent that these and other areas are able to generate growth in domestic demand, as opposed to simply being pulled along by the US, the world economy could be expected to continue growing pretty well during 2007, though most likely below the 2006 pace. The consensus of forecasters at present seems to be that such an outcome is the most likely.
Of course, forecasts can be wrong. The US might slow more than expected, perhaps because of larger dampening effects of the housing downturn. Another way it might eventually slow more than expected, as pointed out by the IMF's recent World Economic Outlook, would be if persistent inflation pressures required further monetary tightening. The US remains sufficiently important that, if this occurred, other regions would probably find that their economies slowed too as a result.
These days, we should also contemplate the outlook for China. Periodically people worry about a possible slump in China's growth, understandably given China's impact on the global economy over recent years. But we are also at the point where we probably should give some thought to price pressures in China. Even China must have some limit to how quickly it can grow without causing inflation, and there are certainly anecdotes of rising wage costs in the major coastal industrial centres, on top of the higher costs of energy and raw materials. It appears that Chinese export prices are no longer declining. Were this trend to continue, the rest of the world, hitherto experiencing the effects of deflation in prices for various manufactures, might at some point notice some mild impact on inflation rates.
No discussion of the global economy is complete without some mention of financial trends. Here the main factor at work is still, it seems to me, the search for yield. Although the Fed has more or less normalised the short-term rate structure in the US, short rates in Japan and, to a lesser extent, mainland Europe remain unusually low. Long-term rates remain on the low side as well. Appetite for risk has been a little more variable this year, but overall risk spreads remain pretty low, especially considering some of the events which have occurred over recent years. Of particular note recently has been the marked increase in leveraged buy-out activity around the world. This reflects a combination of low funding costs and high levels of confidence about the potential future productivity and profitability of corporate assets.
Whether or not such confidence is well based remains to be seen. But for the past decade or more, much of the action has been in household balance sheets – with a trend towards larger gross balance sheets and higher levels of debt. If we now are moving to an era in which corporate balance sheet developments are, once again, to the fore, then economists, prudential regulators and other policy-makers will need to be alert to any economic implications that would flow from such a change.
The Australian Economy
After 15 years of more or less continuous expansion, we have an economy which is as fully employed as it has been for a long time. That's a good thing – full employment is one of the objectives of macroeconomic policy after all, and is set down in the Reserve Bank Act. But high rates of resource utilisation affect the conduct of policy: we need to be more alert to the risk of inflation than in periods when the amount of spare capacity was much larger.
The international environment is one of strong growth, and rising costs of materials. Inflation in Australia has risen, and not just because of prices of petrol and bananas. Those are likely to show declines anyway over the next couple of quarters, but measures of consumer price inflation that are not distorted by these factors have picked up. That is not surprising. Input costs have risen across a range of areas. We have a tight labour market, and despite the steadiness recently of official measures of wages growth, there is still pressure on labour costs, including the kinds that do not show up in wage statistics.
At the same time, there are some puzzles in the picture painted by the various pieces of data on the Australian economy. On the one hand, real GDP growth is estimated to have declined to about 2 per cent over the 12-month period to June 2006, after having grown by just under 3 per cent in the preceding year. Growth in domestic demand has moderated from its earlier heady pace, though it still seems to have been running at 3½ to 4 per cent over the year, which is probably a bit above the economy's sustainable capacity to increase production.
Yet growth in employment has remained quite strong, and the rate of unemployment has, if anything, edged down over the past year. This sort of unemployment result would normally suggest that output growth had been at or slightly above trend, which most people would these days put at 3 per cent or a little above. Meanwhile, the rate of growth of tax revenues over the past several years has been well above what historical relationships suggest should have been expected, given the recorded growth of nominal GDP.
At face value, the output and employment results suggest a marked change in the trend in productivity in Australia over the past few years. The various measures of GDP per hour worked suggest there has been approximately zero growth in productivity since the end of 2003. This compares with an average annual pace of growth of 2 per cent or more in the preceding decade.
So what's going on here?
One possibility is that the level of nominal and real GDP is really higher than is being captured at the moment by the statistics. That would mean that growth over the past few years has been higher than, and productivity has not slowed by as much as, the published data suggest. Tax revenues would, in this scenario, look more in line with historical relationships. This outcome would seem more in line as well with the unemployment trends.
A second possibility is that it is the labour market data that are out of line – perhaps due to lags, or sampling effects – and that they will sooner or later come back into line. If that is the story – if the economy really has grown below trend of late – we might expect some rise in the unemployment rate to emerge before much longer. But this story still needs to explain the strength of tax receipts.
Or maybe both productivity and output growth really have slowed, as the published estimates suggest. The question then would be why productivity slowed so much. One hypothesis we hear from time to time is that average productivity might be reduced by adding the workers with the lowest productivity after a long expansion. Perhaps this is not altogether surprising at this stage of the cycle – and if we are now seeing employment of workers whose lesser skills and productivity kept them out of contention in the past, that in itself is to be welcomed.
But this addition of less productive workers isn't enough to explain the extent of the slowdown in overall productivity. If the figures are correct, the productivity growth for the existing workforce must also have declined noticeably. For had it continued at the pace seen over the preceding decade, the workforce that was employed two years ago could have accounted for virtually all the output growth which has occurred since. That would suggest the productivity of the additional workers employed over that period would not just be lower than average, it would actually be (approximately) zero. Surely employers would not have taken on people with productivity that low.
Other hypotheses have been advanced for a slowdown in productivity growth. Among them are that higher levels of labour turnover in a tight market disrupt productivity performance across firms; or that changed regulations (e.g. new accounting standards, tougher requirements to demonstrate appropriate corporate governance and so on) are using resources without adding to output. At this point, however, these don't seem sufficient as an explanation for a productivity slowdown of the magnitude we observe in the data. Hence it appears that, for the moment, we are left with something of a puzzle.
That means that, as usual, monetary policy is being made under conditions of uncertainty. If the GDP data are correct, then the economy grew more slowly than earlier thought in the first half of 2006, potentially with some moderating impact on the outlook for inflation. (Even then, we would still have to ask the further question of whether the slowing was driven mainly by lack of supply or lack of demand. Only in the latter case would it mean that spare capacity in the economy will have been increased.)
If, on the other hand, there really has been more growth than the GDP accounts suggest – more in line with the rise in employment and the trend decline in unemployment – then capacity probably remains pretty tight. In that case, upward pressure on inflation remains a distinct possibility.
Alternatively, if both sets of data are correct, then productivity actually has slowed down considerably. But if that is true, unless it is a temporary phenomenon, then potential GDP growth is not 3 per cent or a bit above any more. It will be less, and our growth aspirations would have to be adjusted accordingly. In this scenario, inflation pressure in the near term could well increase and demand growth may need to be further restrained for inflation to remain under control over time.
In trying to assess which of these possibilities, or which combination of them, is in operation, one of the pieces of evidence to which we will be looking for guidance is the behaviour of prices themselves. An economy with genuinely sub-potential growth over two years ought, other things equal, to start putting some downward pressure on inflation fairly soon. An inflation rate that continued to increase, on the other hand, would presumably raise questions about either the apparent rate of growth of demand and output, or of potential output or both. You do not need me to tell you, therefore, that the price data to be released over the next couple of weeks will be important in evaluating the outlook and the balance of risks facing policy.
Before I leave the current state of the economy, a remark about the differences in performance by region is appropriate. As everyone is aware, spending growth is strongest in Western Australia, as resource producers seek to put in place more capacity and the income gains from the boom are partly spent. But the differences in spending overstate the differences in actual economic performance between the regions. Not all the demand generated in Western Australia is being supplied from there: some of it is being supplied from the rest of the country, and some of it from outside Australia. It is partly for that reason, presumably, that the differences in employment growth and unemployment trends across states are much smaller than the differences in spending. In fact, the extent of the differences in output growth across states, while noticeable, do not appear at present to be unusually large. They also appear to be well within the sorts of differences experienced over time by other comparable countries or regions, like the United States, Canada or the euro area.
At the same time, there is some tendency for labour and capital to move to the resource-intensive areas. That is exactly what is supposed to happen in a flexible economy when relative prices change: labour and capital respond to incentives. Moreover, as the Secretary to the Treasury pointed out a couple of months ago,[6] if the current set of relative prices persists, there will be more such adjustment in the years ahead.
The Longer Term
Before concluding, it would be useful to lift our eyes from the immediate ebb and flow of the short-run data and to ask, taking a medium-term view, what will be the most important task for monetary policy over the period ahead? It should be obvious that, over the next little while, the main job is to ensure that the inflationary pressure we have been experiencing of late is successfully resisted, and that expectations of future inflation remain well anchored. That will be a key part of maintaining an average inflation performance of ‘two point something’.
We could hardly overstate the importance of maintaining that general environment where, as famously characterised by Alan Greenspan, inflation is sufficiently low and stable that it does not materially affect economic decisions by firms and individuals. A stable overall price environment assists in resource allocation and preserves the value of savings.
It also provides monetary policy with more scope to be flexible in the face of shocks. The past decade has seen no shortage of challenges, often of a financial nature: the Asian crisis, the LTCM episode, the dot com mania and subsequent downturn, and so on. When conditions in the real economy were threatened on those occasions by contractionary forces, central banks in a range of countries were able to respond with reductions in interest rates, or by retaining already low rates for an extended period. Some people would be inclined to argue about whether or not such actions were in every case ideal. But the more important point is that, without a background of low and steady inflation and well-anchored expectations, they would not have been feasible at all.
Were the recent higher inflation rates of the past year in Australia and some other countries to persist, and to start affecting behaviour, such a degree of flexibility for monetary policy might not be present in future moments of economic difficulty. Acting as needed to keep inflation in check in the near term, on the other hand, preserves future flexibility.
It is that strategic requirement to which central banks should be, and I believe are, paying close heed. As we do so, of course, we will be bearing in mind the lagged effects of the policy adjustments already made. If we can successfully see off the higher inflation of the past year or so, we will have done a lot to establish the conditions needed for ongoing growth.
Endnotes
Statement on the Conduct of Monetary Policy,
available at
<https://www.rba.gov.au/monetary-policy/framework/stmt-conduct-mp-1-14081996.html>.
[1]
Second Statement on the Conduct of Monetary Policy,
available at
<https://www.rba.gov.au/monetary-policy/framework/stmt-conduct-mp-2-072003.html>.
[2]
Stevens, G. and G. Debelle (1995), “Monetary Policy Goals for Inflation in Australia”, in A.G. Haldane (ed.), Targeting Inflation, Bank of England; also available at <https://www.rba.gov.au/publications/rdp/1995/9503.html>. [3]
The exact figure can be calculated as 2.56 per cent, which is the compound rate of increase in the CPI over the period from mid 1996 until mid 2006, or 2.48 per cent for the same calculation excluding the GST and interest charges (interest charges were included in the CPI until 1998), or 2.45 per cent for the Treasury underlying inflation rate until 1998 and the CPI ex GST thereafter. The logic for the latter is that the Treasury series was the original target variable until 1998, when the target became the published CPI. The wording in the Statement on the Conduct of Monetary Policy, “around the middle of the target band”, covers these minor differences quite neatly. [4]
Third Statement on the Conduct of Monetary Policy,
available at
<https://www.rba.gov.au/monetary-policy/framework/stmt-conduct-mp-3-18092006.html>.
[5]
‘Economic Policies to Address Global Pressures’, speech by Dr Ken Henry, Secretary to the Treasury, to the Australian Industry Group National Forum 2006, Canberra, 14 August 2006; available at <http://www.treasury.gov.au/documents/1140/PDF/AIG_August_2006.pdf>. [6]