Speech Economic Trends Down Under
Talk to the Westpac European Economics Conference
Some people, when they meet a medical practitioner, like to regale the doctor with their current ailments, and seek insights into possible remedies. The corresponding occupational hazard for central bankers is to be interrogated about the outlook for interest rates and exchange rates.
I suspect that the two burning questions on the minds of many people here today are:
- What is the outlook for interest rates in Australia?; and
- How will the Australian dollar move?
I understand this focus, particularly on the part of participants from the financial sector. But such questions do reflect a desire to take short cuts: they slide over a string of questions about more fundamental economic trends and policies. They reflect, too, a preoccupation with the short term – with possible changes over the next month or quarter, rather than the longer time horizons of more stable investors, and certainly of policy makers.
Today, I want to concentrate on the fundamental factors that influence interest and exchange rates in Australia. These should be the proper focus of a central banker – and perhaps of some others as well. These are what will determine living standards and returns on investments in Australia. I have no direct answers to give you on those two burning questions, but I can give you a reading on the underlying trends and policies which will help to determine, inter alia, what happens to interest and exchange rates.
I begin by reminding you that economic recovery in Australia commenced in mid 1991, rather earlier than in the UK and most other OECD countries. Since that time, growth has been higher, and inflation has been lower, than in most OECD countries.
We can look back over 1994 in particular with some satisfaction:
- GDP grew by 5 to 6 per cent;
- expenditure grew even more strongly, underpinned by a welcome surge in business investment, which rose by about 25 per cent;
- employment increased by 3½ per cent and unemployment fell sharply (it is now 8½ per cent);
- productivity continued to grow at an annual rate of about 2 per cent, well above the average in the 1980s; and
- for the third year in a row, underlying inflation remained at 2 per cent.
Even so, the past year has not been a totally worry-free period for policy makers. Demand and cost pressures have emerged, raising questions about the outlook for inflation. The current account deficit also has increased to uncomfortably high levels. I will return often to these two concerns during the course of this talk.
For the moment, I note that the forecasts released last month with the budget suggest that growth, while slowing, will be sustained at a respectable rate in 1995/96 (3¾ per cent); that inflation, while rising, will be contained within manageable bounds; and that a modest start will be made towards unwinding the current account deficit. If it eventuates, this would add up to a good outcome.
Influences on Interest Rates
Interest rates are the key prices in financial markets, and monetary policy operates through interest rates. Our normal response to questions about the outlook for interest rates – that ‘it all depends’ – might therefore seem a bit lame. It depends, first of all, on which interest rates we are talking about.
It is true that some interest rates – at the very short end of the yield curve – are controlled by the central bank, so that we should, at least superficially, be able to say what will happen to these rates. Even here, however, changes reflect judgments about current and prospective trends in domestic activity and inflation, as well as overseas events. And, like other mortals, central bankers are limited in their ability to see into the future.
Further out along the maturity spectrum, the direct influence of the central bank on interest rates diminishes. Longer-term rates reflect world real interest rates (which, in turn, reflect demands for capital and other factors); inflation expectations; and risk and uncertainty premia, which can vary greatly with the market psychology of the moment.
Indirectly, of course, the actions of the monetary authorities are relevant to longer-term rates. In a small country like Australia, we have to take world real interest rates as given, but our domestic policies do affect inflation expectations and the risk and uncertainty premia. In particular, the stronger the market's belief that the authorities will keep inflation down, the lower long-term interest rates are likely to be.
Interest rates charged to borrowers and paid to depositors are subject to some additional influences as well. The extent of competition among financial institutions, for example, means that lending rates and margins can move, to some degree, independently of monetary policy actions. Over the past year, bank lending and deposit rates in Australia have risen by less than the rise in official short-term rates.
We can see these various influences at work during the past 18 months.
At the long end, the sharp rise in rates began well before our monetary policy started to tighten in August 1994. The trigger for this in Australia, as in most other countries, was the rise in bond yields in the US in early 1994. The close linkage between US and Australian bond markets is not always explainable in terms of observable facts, but it has been real nonetheless.
Although they followed similar trends, bond yields in Australia in 1994 rose more than in most countries. This reflected scepticism about our ability to sustain recent low rates of inflation and concerns that we might revert to our earlier, less flattering performances. Most other countries with poor histories of inflation during the 1970s and 1980s also experienced relatively large rises in bond yields in 1994.
Given the historically low levels to which they had fallen in most countries by the end of 1993, some rise in world bond yields was on the cards. The sharp rises which occurred in 1994, however, suggested an over-reaction by investors to the perceived inflation risks posed by the pick-up in economic activity. The fact that inflation has remained subdued in most countries, including Australia, supports this view. The substantial falls in bond yields over recent months suggest that market participants themselves have come to a similar view. Since their peak in late 1994, yields on Australian 10-year bonds have fallen by 175 basis points to under 9 per cent.
In Australia, the bond market has also benefited from the tightening in fiscal policy announced in the May budget. This has eased pressures on the bond market, partly on the grounds that its overall contractionary effect will help to contain inflationary pressures, but mainly because it presages a marked reduction in the Government's own demand for funds. The bond issue program is likely to be no more than $4 billion in 1995/96, compared with about $20 billion in 1994/95.
So much for long rates, where recent experience reminds us just how closely linked we are with overseas financial markets. But it also reminds us that we need to follow credible policies of our own. Credibility is not earned by pious statements of intent, nor by monetary policy alone, but by getting the settings of all policies right and sticking to them. In my view, good progress is being made on the domestic policy front.
In the second half of 1994, when it was emerging that the recovery had a good head of steam, the authorities raised short-term interest rates, even though inflation was still low (2 per cent or less) and unemployment was still high (over 9 per cent). Overnight interest rates were raised almost 3 percentage points in three steps, to 7.5 per cent. These rises are achieving their objectives: they are helping to deliver a sustainable rate of growth in output and employment, while keeping inflation under control. Future changes in monetary policy will be motivated by the same objectives.
The notion that concerns for activity and employment should be part of the Reserve Bank's duty statement is somewhat controversial these days. It should not be. Monetary policy might not be able to do much to raise an economy's long-term growth rate but there is little doubt that it affects output and employment over the course of the business cycle. Indeed, monetary policy operates to lower inflation largely through its influence on the real economy. Helping to keep output and employment on a sustainable growth path, therefore, is not just consistent with low inflation, it is part and parcel of its achievement.
With the economy and employment still growing at a good clip, our main focus at the moment is on keeping inflation in check. Price stability is not a pillar of dogma to be pursued for its own sake. Rather, experience demonstrates that economies work better when prices are relatively stable – when consumers and investors are more inclined to behave in ways conducive to sustaining growth and employment.
Central banks express their inflation objectives in different ways: ours is to hold underlying inflation to an average of 2 to 3 per cent over a run of years. It is a firm and clear commitment. It has been endorsed by the government.
Our formulation allows for the likelihood of some cyclical variation in inflation, but it does not provide for any structural pick-up. In other words, we can accept inflation above 2 to 3 per cent on occasions, provided the necessary policy adjustments are being made to bring it back to that rate within a reasonable period of time. We will judge ourselves successful if underlying inflation is held to ‘2 point something’ over the course of the economic cycle.
The pre-emptive tightenings of monetary policy in the second half of 1994 were designed to put policies in place before inflation actually picked up. It is now 10 months since the initial tightening and underlying inflation is still around 2 per cent. But we expect that it will pick up over the year ahead. The main sources of potential pressure are:
- overly strong levels of demand putting pressure on capacity and resources; and
- cost pressures coming from higher input prices, rising wages or a falling exchange rate.
We are keeping a close watch on these areas, as well as on inflation expectations, which can affect price setting and wage bargaining behaviour. Short-term interest rates are set with each of these pressure points in mind. The pace of economic growth is clearly important but so too are the other potential threats. For the moment, monetary policy is, I believe, hitting its targets.
The pace of growth in demand had to slow from last year's rapid rate to help keep inflation under control. There is no doubt that the economy has slowed, but by how much is less clear. The evidence is mixed. Over recent months, surveys of business and consumer sentiment have generally revealed a more restrained outlook for economic activity, and some indicators have weakened. On the other hand, employment, imports and personal credit have continued to grow quite strongly.
The slowdown that is occurring reflects in part the working out of the dwelling and stock cycles, both of which are expected to detract from growth during 1995/96. But policy is also contributing, led by the monetary policy tightenings in the second half of 1994 and supported by the modest fiscal contraction – equivalent to 0.7 per cent of GDP – provided for in the budget for 1995/96.
Real GDP grew by about 3¾ per cent in the year to the March quarter, and by less than that over the past six months. This suggests a significant slowing which, if sustained, would help to relieve pressure on inflation and the current account.
As always, there are risks in the year ahead and economies rarely grow as smoothly as forecasts imply they will. Two notable areas of uncertainty at this time are the drought and when it might break, and the pace of growth in the world economy, especially in Japan and the United States. Business and consumer sentiment is another unpredictable factor which can shift quite quickly on the back of changes in expectations about interest rates and other developments.
It is part of a prudent central banker's make-up, if not to worry about these things, to at least understand that they can change quickly and turn out differently from what has been forecast. This caution seems well based at present when there is more than usual uncertainty about the course of future demand pressures, but rather less about future cost pressures.
Even if activity continues to slow to a more comfortable pace, cost pressures will need to be watched closely. Problems could come from one or more sources, including higher input prices, further upward drift in wages, and sustained exchange rate depreciation. Prices of some metals and other materials have been rising, but wages and the exchange rate – the two Achilles heels of Australian inflation in the past – are the main areas to watch. Pressures currently in the pipeline suggest that the 2 to 3 per cent inflation objective is in some danger of being exceeded during the year ahead.
Over the past year, growth in ordinary time earnings has moved up to between 4 and 5 per cent (from 3 per cent in 1993/94) while productivity growth has been about 2 per cent. These kinds of numbers remain broadly consistent with the 2 to 3 per cent inflation objective, but this position would be undermined if wages were to grow any more rapidly.
It should not be assumed that further slippage on wages will occur. One reason for optimism is that we have now had a decade of relative wage restraint, and most businesses and employees understand that this has delivered good outcomes for all parties. Employment has grown strongly, profitability has increased and inflation has come down. This experience has changed the attitudes of many Australians and is helping to build a community consensus in favour of responsible wage outcomes and low inflation.
Wage restraint has been underpinned also by changes in wage setting institutional arrangements. One such change is the Accord process, which has given rise to an on-going series of understandings between the Government and the trade union movement. Accord Mark VIII is currently being negotiated, and all the indications are that the unions remain committed to reaching wage outcomes consistent with continuing low inflation. Another change is the shift to enterprise bargaining, which now covers approximately 40 per cent of the workforce and is spreading. Compared with earlier centralised arrangements, enterprise bargaining focuses the attention of employers and employees on productivity, and makes it harder for wage rises in one sector to ‘flow on’ to other sectors.
This new focus on wage moderation and productivity improvement through more flexible work and management practices has been reinforced by other changes. Tariff reductions, deregulation and other measures to open up markets mean that large wage increases cannot easily be passed on to consumers: a welcome discipline is now being imposed on wage setters on both sides of the negotiating table.
These are profound changes. Provided they can be sustained – and there is no reason to believe otherwise – they augur well for keeping inflation under control. The big test will be whether wage increases over the next year or so can be contained to around the current rate of 4 to 5 per cent, in circumstances where employment growth is expected to remain strong.
The other potential near-term threat to inflation is the exchange rate. Depreciation, if sustained, is bad for inflation because it raises the prices of imported goods and materials directly, and reduces the competitive discipline on domestic producers; sustained appreciation has the opposite effects.
In trade-weighted terms, the $A appreciated by over 10 per cent during 1994 but this has been whittled away during the first half of 1995. The pass-through of exchange rate changes to consumer prices usually takes a year or more, and depends on a number of factors, including expectations about future movements, and the strength of demand in the economy. If the recent fall in the Australian dollar was to be sustained, producers and retailers could be expected to pass on a large part of the increase in import prices to consumers, resulting in unwelcome upward pressure on prices over the next year or two – of the order of around ¾ of a percentage point. On the other hand, if the recent fall is unwound over the months ahead, any such effects are likely to be small.
The conclusion I would lead you to draw from all this is that there can be no certainty about the timing or direction of the next move in official interest rates in current circumstances. Any change will be dependent on judgments based on the flow of data about trends in activity, capacity utilisation, prices, wages, exchange rates, overseas developments and other factors. Movements in one particular area are unlikely to be definitive: all relevant pressure points have to be assessed. Monetary policy should be adjusted when those judgments – reached in the light of the twin objectives noted earlier – require it. Ideally, it should be adjusted pre-emptively – in either direction – but not prematurely.
Influences on the $A?
The short answer to our second burning question – what is likely to happen to the Australian dollar? – is that, perhaps more so than with interest rates, we cannot confidently predict how the exchange rate will behave.
Over the years, academic economists and market practitioners have invested heavily in the search for a philosopher's stone of the financial markets: a model that accurately predicts exchange rates. None exists. Many people make predictions but no-one has succeeded in developing a model with any worthwhile predictive power, at least so far as month-to-month or quarter-to-quarter movements are concerned. This is hardly surprising, given that foreign exchange markets are driven largely by expectations of future events, including policy developments. The only consistent and reliable finding that has emerged from all this investment is that, over the long run, exchange rates adjust to reflect differences in inflation and productivity growth among countries. Over shorter periods, other factors like interest rates and market perceptions about policies enter the equation but these effects are especially difficult to quantify.
Notwithstanding these problems, many market participants appear to view the Australian dollar as one of the more predictable exchange rates. This is partly because they see it as a ‘commodity currency’, which rises when commodity prices rise, and vice versa.
This kind of relationship has been evident over the past decade, although the precise linkage is unclear. Commodities still account for 60 per cent of Australia's exports (compared with over 70 per cent 10 years ago), but the fluctuations in export receipts resulting from changes in commodity prices are not sufficiently large, by themselves, to explain fully the swings we have seen in the Australian dollar. Part of the explanation appears to be that many overseas investors and funds managers have a model in their minds that Australian investments are a good buy when commodity prices are rising; they act on this model and their actions tend to become self-fulfilling so far as the Australian dollar is concerned.
Whatever the mechanism, the rise in the Australian dollar during 1994 owed a good deal to signs of rising world industrial activity, which lifted many commodity prices. In 1995, slowing economic activity in the US and continued weakness in Japan have halted this lift and, in some cases, led to falls.
Another market perception of the Australian dollar is that of a ‘dollar bloc currency’. At times, movements in the Australian dollar do seem to be driven largely by movements in the US dollar; some of the recent weakness in the Australian dollar appears to have fed off weakness in the US dollar.
Again, there is no apparent strong rationale for expecting the Australian dollar to follow the US dollar, beyond the fact that (as now) the economic cycles in Australia and the US are often closely synchronised. There are, of course, no institutional linkages between the Australian dollar and the US dollar, in the way that there are between the German mark and many other European currencies.
Although there are times when these market perceptions of the Australian dollar as a commodity currency or ‘dollar bloc’ currency might help to explain day-to-day fluctuations, they cannot explain longer-term movements. The perceptions themselves are not all that soundly based; Australia's economy and export base are diversifying, and its economic ties are increasingly with the Asian region. Here, as in some other areas, market perceptions are lagging behind realities.
Over the longer term, the important influences on the exchange rate will be our performance in containing inflation, improving productivity and raising national saving. In short, in the conduct of economic policy generally. I have said already that I believe good progress is being made on the policy front. The episode of higher than average inflation of the 1970s and 1980s is now behind us, and our productivity performance has picked up. But the policy front keeps moving – and we have to move with it.
The Current Account and National Saving
This leads me to concerns about Australia's current account deficit, which appear to figure prominently in market perceptions of the Australian dollar. Large current account deficits certainly raise the level of market nervousness about the exchange rate. We saw this in early 1995 when the Treasurer indicated that Australia was facing a current account deficit equivalent to almost 6 per cent of GDP. This caused international investors to question the adequacy of domestic economic policies, and contributed significantly to the subsequent decline in the exchange rate.
In part, the blowout in the 1994/95 current account deficit reflects several special factors. Exports of wheat and some other farm products, for example, have been decimated by the drought. At the same time, imports have been boosted by burgeoning domestic spending, especially business investment; capital goods imports rose 20 per cent in the 10 months to April 1995. While these seasonal and cyclical factors can be expected to reverse themselves, the fact remains that Australia's structural, or underlying, current account deficit deteriorated in the early 1980s and has not recovered; it averaged around 2½ per cent of GDP in the 1960s and 1970s, but has averaged 4½ per cent in the 1980s and 1990s.
This change is not, in my view, a reflection of our current international competitiveness. Most of the deterioration occurred in the late 1970s and early 1980s. Since that time, the Australian economy has been transformed: it is now more open and competitive than ever before. Over the past decade, for example:
- the share of trade (imports and exports) in GDP has risen from about 30 per cent to about 40 per cent;
- the share of merchandise exports destined for the fast growing Asian market has grown from 53 per cent to 67 per cent; and
- exports of manufactured goods have grown at an average rate of 17 per cent per annum and now represent 23 per cent of total merchandise exports.
More fundamentally, the current account deterioration reflects an increased reliance on foreign saving to help fund our investment.
This long-term dependence on foreign saving has led to an accumulation of foreign liabilities. But we are coping with these. Net foreign debt, for example, was equivalent to 37 per cent of GDP in March 1995, down from a peak of 43 per cent in September 1993, and below the levels of countries such as Canada, New Zealand, Sweden and Ireland. Debt service payments in the year to March were equivalent to 11 per cent of total export earnings, compared with a peak of 21 per cent in 1989/90.
Although manageable, our on-going dependence on foreign saving does leave us exposed to wild swings in market sentiment against things Australian. Whether they are well based or not (and often they are not), such swings make for greater exchange rate volatility, as well as higher interest rate premia on borrowings. In short, there are some risks in a continuing heavy reliance on foreign capital inflows; certainly, the implications for policy of investors losing confidence in the economy are very sobering.
I believe these risks are now better understood by both the authorities and the community more generally. It is understood, for example, that, given our on-going need to access volatile financial markets, we have additional pressures on us to pursue sound macro policies and structural reforms. And, I am pleased to say, there is a growing appreciation that the best counter of all to this exposure is to reduce our dependence on foreign saving.
Herein lies the real significance of the Government's recent budget. It tackles the current account problem at its source, namely Australia's persistent national saving/investment imbalance. It contains measures to significantly lift national saving over the years ahead through better public and private sector saving efforts. Both sectors have contributed to the secular decline in national saving during the past two decades, with the public sector contributing the bulk of the decline. (As with the current account deterioration, much of the structural decline in national saving occurred in the period from the mid 1970s to the early 1980s.)
Four surpluses in a row were achieved in the late 1980s before the budget moved into deficit, with the onset of the recession of the early 1990s. The budget provides for a small surplus in 1995/96, which is projected to grow in the three following years. Part of the improvement next year reflects the effects of asset sales and some other special transactions, which the Government has been quite up-front about. Adjusting for these transactions (as we should from a national saving perspective), the ‘underlying’ budget deficit is projected to decline from about 3 per cent of GDP this year, to about 1½ per cent in 1995/96. On the same basis, the budget would be in virtual balance in 1996/97, and in surplus thereafter.
This represents substantial progress by international standards. The US, which is ahead of most countries in deficit reduction plans, is hoping to balance its budget by the turn of the century. By the more relevant standards of the stage of the economic cycle in Australia and the extent of our current account problem, it is less impressive – but still very welcome progress.
The budget also announced a major initiative to promote private saving. It will operate through an extension of the superannuation scheme which is designed to encourage employees to save more for their retirement. Measures already are in place to require employers to pay a portion of employees' earnings into superannuation funds; this portion is being increased progressively to 9 per cent per annum by 2002.
From July 1997, these arrangements will be extended to include contributions by employees, building up to 3 per cent per annum. The Government has also undertaken to pay amounts roughly equivalent to these contributions into a superannuation fund on behalf of employees (and self-employed people). All up, it is envisaged that, by 2002, most employees will have about 15 per cent of their earnings going into a retirement fund each year.
How should we rate these various budgetary measures?
It is always possible to argue that more should have been done, but we should also evaluate objectively the measures that have been adopted. Given the starting point, it was always going to take more than one year to eliminate the budget deficit in underlying terms. Real progress has been made and we now have a credible medium-term orientation of fiscal policy. Notwithstanding that progress, I suspect that some further hard pounding lies ahead, particularly in keeping the lid on new spending commitments and in better targeting programs to genuinely needy recipients.
Assuming they are properly implemented (with, for example, effective preservation provisions and protections against ‘double dipping’), the superannuation changes will, in time, deliver higher living standards for many Australians in retirement, as well as making a substantial medium-term structural improvement in national saving.
Taken together, the projected movement to budget surpluses and increased private retirement saving represent a major boost to national saving. It will help to fund the higher levels of investment needed for continued growth, with less reliance on foreign saving.
I have said previously that life would be more comfortable for all of us if the current account deficit were to average around 3 per cent of GDP over the next decade, compared with the 4½ per cent average of the 1980s and early 1990s. There are too many uncertainties to make confident long-term projections but the recent budget measures, properly followed through, do make that more comfortable state a real prospect. To the extent it occurs, it will ease our exposure to external shocks and debt servicing burdens; that should be a plus for many things, including the Australian dollar.
Today I have focused on the fundamental factors and policies which help to determine interest and exchange rate outcomes in Australia. Monetary policy impinges on both, but fiscal and other policies also are clearly important. Success should be judged not in terms of achieving a particular outcome for either interest rates or exchange rates, but rather in terms of economic activity, jobs and inflation.
In the main, monetary (and other) policies have to be based on judgments about prospects for growth and inflation a year or so ahead. These, rather than the often volatile swings in market sentiment, are what determine interest rates and exchange rates in all but the very short term.