The Challenge of Prosperity
Address to the Committee for Economic Development of Australia (CEDA) Annual Dinner
Melbourne - 29 November 2010
Thank you for the invitation to play a part in marking the 50th anniversary of CEDA. It is particularly significant for me to be here because 2010 also marks 50 years since the commencement of central banking operations by the Reserve Bank of Australia. There were in fact a number of significant beginnings in 1960.
It was a time of rising prosperity after a long period of difficulty. Between the depression of the 1890s and the end of World War II, real GDP per capita in Australia had risen by about 35 per cent – or around half a percentage point per year. But in the 15-year period from the end of the war to 1960, it expanded by about 25 per cent – or about 1½ per cent a year.
The long post-war boom would eventually see growing excesses from the late 1960s, which ended in the disastrous instability of the 1970s. But in 1960, the boom still had a long way to run.
So 1960 was a time of optimism. There have been many ups and downs for the Australian economy since then. CEDA has played its role in informing discussion and debate along the way.
Two years ago, when I last addressed this group, optimism was anything but the order of the day. The global financial system was in serious disarray and the global economy was heading into recession. It was obvious Australia would be affected but I suggested that there were good reasons for quiet confidence then about the long-run future of Australia. There still are, two years later.
But we have to turn that confidence into lasting prosperity. So I would like to offer a few observations about some of the things we need to be thinking about. I do not have definitive solutions, but offer these observations as a modest contribution to the discussion.
In so doing, I am not trying to convey anything about recent or prospective monetary policy decisions. Tonight, at an event marking the 50th anniversary of a body devoted to Australia's economic development, it is more useful to lift our gaze beyond the next interest rate decision to look at a broader canvas.
I have one picture to show.
This is a very long-run chart of Australia's terms of trade. You may have noticed the Reserve Bank saying a lot about the terms of trade in the past few years. Before I describe the chart, why is it important?
Our terms of trade have a big bearing on national income. In economic commentary, there is typically a very strong focus on GDP – the value of production – as a summary of national material progress. There is also quite rightly an emphasis on lifting productivity – real GDP per hour worked – as the source of our growth of material living standards.
For open economies, though, our standard of living is affected not just by the physical output we can obtain from our resources of labour and capital, but also by the purchasing power of that output over things we want to have from the rest of the world. This is what the terms of trade is measuring. It is the relative price of our export basket in terms of imports. At the extreme, if the economy were open to the extent that we exported all our production and imported all our consumption, then the price of exports relative to imports would determine our living standards entirely, for any given level of productivity per hour worked. As it is, Australia is not that open, and not as open as many smaller economies, but it is considerably more open than the really large economies like the United States, the euro area or Japan. So the terms of trade matter.
When the terms of trade are high, the international purchasing power of our exports is high. To put it in very (over-) simplified terms, five years ago, a ship load of iron ore was worth about the same as about 2,200 flat screen television sets. Today it is worth about 22,000 flat-screen TV sets – partly due to TV prices falling but more due to the price of iron ore rising by a factor of six. This is of course a trivialised example – we do not want to use the proceeds of exports entirely to purchase TV sets. But the general point is that high terms of trade, all other things equal, will raise living standards, while low terms of trade will reduce them.
Returning to the chart, to my eye there are three key features.
The first is the degree of variability in the terms of trade through the middle parts of the 20th century, from about World War I to the aftermath of the Korean War. This was, of course, a period of considerable instability in the global economy, with the attempt to return to the Gold Standard after the ‘Great War’, followed by the 1930s depression, the Second World War, the post-war expansion and then the Korean War. I might add that, in those days, with the attempt to maintain a fixed exchange rate, these swings were very disruptive to the economy. Typically, a rise in export incomes would result in a rise in money and credit, a boom in economic activity and a rise in inflation. Then the terms of trade would fall back and the whole process would go into a rather painful reverse. The advent of the flexible exchange rate in the early 1980s made a great difference in managing these episodes.
The second feature is the downward trend in the terms of trade, particularly noticeable from the early 1950s to about the mid 1980s. This was the period of resource price pessimism, the ‘Prebisch-Singer hypothesis’ and so on, which held that primary products would tend to decline in price relative to manufactured products. The latter part of this period was the one in which the realisation became widespread that the (apparently) easy gains in living standards of the post-war boom were gone, and in which pessimism about Australia's economic future was probably at its most intense. It was also the period when, under strong political leadership backed by a highly capable bureaucracy and an economically literate media, our determination to press on with various productivity-increasing reforms was greatest. That these two phenomena occurred together was probably not entirely a coincidence.
The third feature is the current level of the terms of trade relative to everything but the all-time peaks over the past century. Measured on a five-year moving average basis, and assuming (as we do) some decline in the terms of trade over the next few years from this year's forecast peak, the terms of trade are as high as anything we have seen since Federation.
To give some perspective on how important this is, let me offer one back-of-the-envelope calculation. The export sector is about one-fifth of the economy. The terms of trade are at present about 60 per cent higher than their average level for the 20th century, and about 80 per cent higher than the outcome would have been had they been on the 100-year trend line. This means that about 12–15 per cent of GDP in additional income is available to this country's producers and/or consumers, each year, compared with what would have occurred under the average or trend set of relative prices over the preceding 100 years (all other things equal). That will continue each year, while the terms of trade remain at this level.
Of course, part of this income accrues to those foreign investors who own substantial stakes in the mineral sector. In this sense, the current boom is a little different from the early-1950s one where most of the income went first to Australian farmers. Nonetheless, a good proportion accrues to local shareholders and employees, and to governments via various taxes. A non-trivial part of it is available to consumers as higher purchasing power over imports, as a result of the high exchange rate.
It does not take much imagination to see that an event of this magnitude is expansionary. Incomes are higher – in some cases a lot higher – and, absent some offsetting force, some of that will be spent. So it has always proved in the past. Moreover, if, as seems very likely, these prices prompt a build-up in investment to supply more of the commodities concerned, there are further expansionary effects. Even applying significant discounts to stated investment intentions, as the Reserve Bank staff have done in their forecasts, there is likely to be a further significant rise in business investment over the next few years, from a level that is already reasonably high as a share of GDP. On all the indications available, we are living through an event that occurs maybe once or twice in a century.
So a very important question for us is: how do we handle all this?
We obviously have to be wary of overheating. The Bank has given its views on this point before and I will say no more about that tonight.
But in fact the issues are broader than that. They extend to how we use the additional income, and how soon, and to questions of structural adjustment.
One difficulty is that it matters a great deal whether the rise in the terms of trade is likely to be permanent or only temporary. Unfortunately, we cannot really answer that question. It is obvious from my chart that past episodes tended not to be permanent, but they sometimes lasted several years and certainly long enough to be very disruptive.
If the rise in income is only temporary, it would be desirable not to raise national consumption by very much. Instead, it would make sense to allow the income gain to flow into a higher stock of saving, which would then be available to fund future consumption (including through periods of temporarily weak terms of trade, which undoubtedly will occur in the future). Moreover, it would probably not make sense for there to be a big increase in investment in resource extraction if that investment could be profitable only at temporarily very high prices (and which could come at the cost of reduced investment in other areas).
If the change is likely to be persistent, then income is likely to be seen as permanently higher. Households and most likely governments will probably see their way clear to lift their consumption permanently, both of traded and non-traded goods and services. Structural economic adjustment will also occur as the sectors whose output prices have risen, now being more profitable, will seek to expand, in the process attracting productive resources – labour and capital – away from other sectors whose output will decline as a share of GDP. Australia's floating exchange rate, which tends to rise in line with the increase in the terms of trade, helps the reallocation of labour and capital by giving price signals to the production sector. The higher exchange rate also speeds the spread of the income gains from the terms of trade rise to sectors other than the resources sector, by directly increasing their purchasing power over imports. The resulting rise in imports spills demand for tradable goods and services abroad, which helps to reduce domestic inflation.
So the shift in the terms of trade will, unless clearly quite temporary, drive shifts in the structure of the economy. It is easy, of course, to speak in the abstract of ‘reallocation of productive resources’, but this means that some businesses and incomes become relatively smaller; jobs growth in some areas slows even as in others it picks up. Some regions struggle more than others. Some sources of government revenue are adversely affected even as other sources see an improvement. This process will be seen, not unreasonably, as costly by those adversely affected, even though the overall outcome is that the country as a whole is considerably better off. (It is also obvious that, if the terms of trade change really is only temporary, it may not be worth paying these adjustment costs from the perspective of the overall economy.) The policy challenge for governments will be whether to help these sectors resist change, or to help them adapt to it.
We can carry out the thought experiment of imagining that, as a society, we wanted to resist these changes completely and seek to preserve the existing structure of the economy. Let me be clear I do not advocate this. But consider what would be involved. We would need, inter alia, to prevent the resources sector from responding to changed prices (preventing any increase in its size). That would probably involve taxing away completely any additional national income resulting from higher prices, and maybe also preventing any additional exploration or capacity expansion to take advantage of strong demand that could be met profitably even at after-tax prices. We would probably need to re-cycle any funds raised overseas, in the process holding down the exchange rate. It is important to note, by the way, that such funds could not be spent at home without adding to aggregate demand and hence risking the inflation we would still be seeking to avoid in this scenario. In the scenario where we want as much as possible to be unchanged, the additional income handed to us by the change in global relative prices all has to be used offshore, one way or another.
If all the above could be achieved – a very big if, when one considers the logistics of what would be required – then the economy's structure could, perhaps, remain as it was. This course would mean forgoing the potential for higher export income by investing more in resource extraction; either those gains would go instead to other resource-supplying countries or, in commodities where Australia is a major producer, our lack of supply response would result in further upward pressure on prices. So we would avoid the disruption of structural change, but overall would be poorer than otherwise would have been the case, as would, perhaps, our trading partners. It is hard to believe such an outcome could be achieved and no less difficult to imagine it being thought desirable.
Realistically, we won't be able to hold the economic structure static in that fashion if there is a major, persistent change in relative prices. Nor, I would argue, should we try. Had we had that approach through our history, we would still be trying to employ 25 per cent of our labour in agriculture and still be trying to ride ‘on the sheep's back’ in chase of a world economy that had moved on to place a much higher value on many things other than wool. We would not have the highly developed services sector that we have today, nor the standard of living we currently enjoy. So if the terms of trade do remain fairly high for a lengthy period, the task is going to be to facilitate structural adjustment so as to make it occur in as low cost a way as possible. But that ought to be feasible given that overall income is considerably higher.
Of course we cannot know whether the terms of trade will be high for a long period. History certainly would counsel caution in this respect. We do know that supply of various resources is set to increase significantly over the years ahead and not just from Australian sources. It is for this reason that we assume some fall in commodity prices over the next several years. The assumption underlying the Bank's forecasts published a few weeks ago is that iron ore prices fall by up to about 30 per cent over the next several years. Even if they do, the terms of trade will remain quite high by the standards of the past 100 years in the near term, as the chart showed.
Is that assumed fall realistic? There is no way of knowing. Larger falls have happened before. In fact they have been the norm. On the other hand, experienced people seem to be saying that something very important – unprecedented even – is occurring in the emergence of very large countries like China and India. If the steel intensity of China's GDP stays where it is already, and China's growth rate remains at 7 or 8 per cent for some years to come, which appears to be the intention of Chinese policy-makers, then the demand for iron ore and metallurgical coal will rise a long way over the next couple of decades. If India's steel intensity goes the same way as most other countries have, that will add further. Even with allowance for supply responses by other producers and considerably lower prices than we see today, that seems to point to a prominent role for the resources sector, broadly defined, over a longish horizon.
So the most prudent assumption to make might be that the terms of trade will be persistently higher than they used to be, by enough that we will need to accommodate structural change in the economy, but not by so much that we shouldn't seek to save the bulk of the surge in national income occurring in the next year or two, at least until it becomes clearer what the long-run prospects for national income might be.
As it happens, there does seem to be a good deal of saving going on, thus far, in the private sector. A little-noticed recent statistical release was the annual national income accounts for the year 2009/10. In that release, the Australian Statistician has made some major revisions to the estimates for household saving (which of course is a residual arising from other major aggregates). The revision lifted estimated household saving by $45 billion, or about 5 percentage points of income, from the previous estimates. The net saving rate is now seen at some 9–10 per cent of income over the past year or two, up from about −1 per cent five years ago.
In all the circumstances, considering what has happened around the world in recent years, more cautious behaviour by households is not surprising. Nor, I would argue, is it unwelcome. With the stimulus from the terms of trade and the likely investment build-up, the economy can cope with more saving by households for a time. On the other hand, to expect it to absorb a major surge in consumption at the same time as an historic increase in investment is also occurring would be rather ambitious. In fact, we probably need private saving to remain on a higher trajectory, and we will also need public saving to rise, as scheduled.
In the longer term, the economy's increased exposure to large emerging economies like China and India (these two now accounting for over a quarter of exports) – assuming that continues – may also pose important questions. If these and other emerging economies continue to grow strongly on average, but also, as with every other country, still have business cycles, the result may be the Australian export sector, and therefore the Australian economy, having a potential path of expansion characterised by faster average growth in income, but with more variability. That possibility has been noted by some observers. It is worth recording that such concentration would hardly be unprecedented – think about the dominance of Japan in Australia's trade in the 1970s and 1980s, or the dominance of the United Kingdom in an earlier era. Nonetheless, the degree of concentration could be higher than we have seen in the past decade or more, which was a time of considerable stability for the Australian economy overall.
We can't know whether this scenario of higher but more variable income growth will come to pass. But if it did, how should we respond?
We could simply accept higher variability, if that comes, as the price of higher average income growth. That would see higher variability in demand in the economy, which would have its own implications, not least that it could make it harder for macroeconomic policies to foster stability.
Another approach would be to reflect the higher income variability in our saving and portfolio behaviour rather than our spending behaviour. We could seek to smooth our consumption – responding less to rises or falls in income with changes in spending and allowing the effects to be reflected in fluctuations in saving. In the most ambitious version of this approach, we could seek to hold those savings in assets that provided some sort of natural hedge against the variability of trading partners, or whose returns were at least were uncorrelated with them. Of course, such assets might be hard to find – the international choice of quality assets with reasonable returns these days is a good deal more limited than it used to be.
It is possible that this behaviour might be managed through the decisions of private savers. There might also be a case for some of it occurring through the public finances. That would mean accepting considerably larger cyclical variation in the budget position, and especially considerably larger surpluses in the upswings of future cycles, than those to which we have been accustomed in the past. There would also be issues of governance and management of any net asset positions accumulated by the government as part of such an approach, including whether it should be, as some have suggested, in a stabilisation fund of some sort. These are pretty big questions and addressing them would not be straightforward, so I am not going to attempt that tonight. The point simply is that, in the face of what appears to be a very big event in our terms of trade, these issues are deserving of consideration – perhaps by CEDA, among others, as you enter your sixth decade.
As I said at the outset, we have grounds for confidence in the future of our country, just as at CEDA's beginning 50 years ago. Recent performance, not to mention the economic opportunities in our time zone, has helped to strengthen our confidence.
But it would be a mistake to rest on recent achievements, as significant as they have been, and to fail to press on in our efforts to do better. Past periods of apparently easy affluence, conferred by favourable international conditions, probably lessened the sharpness of our focus on the other element of raising living standards, namely productivity. It was subsequently a long and difficult grind when we realised that international conditions had become less favourable. So while I have not talked about productivity this evening, I do not wish my focus on the terms of trade to be interpreted as implying that lifting productivity is unimportant. On the contrary, while our terms of trade are handed to us, for better or worse, by international relative prices, the efficiency with which we work is a variable we can actually do something about.
A prudent approach might be to use the current period of exceptionally favourable international prices to raise our saving, while maintaining a disciplined approach to ensuring there are no impediments to lifting productivity. Consumption deferred – private or public – can easily be enjoyed in the future; consumption we get used to today is harder to wind back in the future if circumstances change. These issues, and the associated structural adjustment issues, no doubt will pose a challenge. But that's the challenge of prosperity – and not a bad challenge to have.
It is sometimes said that Australia manages adversity well but prosperity badly. There will never be a better opportunity than now to show otherwise.
In 1960, some saw that care was needed. In the Economic Record in August
of that year, there appeared the following statement:
‘In July 1960 the Australian economy is producing more, and expanding its production faster … than at any earlier date. … One outstanding economic problem seems to remain, and it has been frequently discussed academically and in public debate: can the boom be sustained without a dangerous degree of inflation?’ (Bowen 1960)
- ‘The Economic Situation’, address to CEDA Annual Dinner, Melbourne, 19 November 2008. Available at: <http://www.rba.gov.au/publications/bulletin/2008/dec/pdf/bu-1208-3.pdf>.
- In fact, fitting a trend to the data for the 20th century shows a statistically significant downward trend of about 0.2 per cent per annum on average.
- Prebisch (1950) and Singer (1950).
Bowen I (1960), ‘The Australian Economy, July 1960’, The Economic Record, 36 (75), pp 307–355.
Gillitzer C and J Kearns (2005), ‘Long-term Patterns in Australia's Terms of Trade’, RBA Research Discussion Paper No 2005-01.
Prebisch R (1950), The Economic Development of Latin America and Its Principal Problems, United Nations, New York.
Singer H (1950), ‘The Distribution of Gains between Investing and Borrowing Countries’, American Economic Review, 40, pp 473–485.