Do Oil Prices still Matter for the World Economy ?

There was a time when a large and relatively sudden increase in the price of petroleum in world markets could bring the world economy to its knees, throwing the developed industrial centres into disarray and severely affecting macroeconomic stability in oil-importing developing countries. Indeed, the stagflation of the world economy in the 1970s was widely attributed to the cost-push inflationary pressures released by the OPEC price hikes of 1973-74.
 
Even the subsequent oil price hike of 1979 was powerful in terms of its negative effects on output growth and inflation across most regions of the world economy, and for a long time thereafter oil price increases became identified in the public imagination with upward pressure on the general price level as well as recessionary implications for output.
 
However, recent trends indicate that the world economy currently is rather different, and that oil price increases need not have quite the same impact that they have had in the past. Between the beginning of 1999 and May 2000, suddenly and unexpectedly, international prices of petrol more than doubled. Despite this, the world economy - which was not exactly performing spectacularly before then - did not show any significant deviation from trend in terms of the low to moderate economic growth rates which were evident in that year.
 
In fact, to the extent that various regions - such as Western Europe and East Asia - were showing recovery from the earlier slump, such recovery was not affected by the rise in oil prices. The US economy remained as buoyant as it has been for a remarkable eight years, fuelled by apparently endlessly bullish investor sentiment and the inflow of savings from the rest of the world. And, perhaps even more surprisingly, world trade prices were only minimally affected, and the rise in petrol prices did not lead to a cost-push spiral in other sectors, so overall inflation remained low.
 
This is not just qualitatively different from the experience of the 1970s and early 1980s. It also suggests processes that may appear counterintuitive to an understanding of the mechanisms of the world economy which is informed by that earlier experience. So it is worth investigating what exactly has changed, and whether it would be accurate to conclude that oil prices no longer matter very much to the major players in the world economy, or to world trade and output in general.
 
One response to this particular tendency has been the celebratory one of seeing this as confirming the focus on monetarist macroeconomic policies in many countries. Thus, the attempts to regulate domestic credit and tighten interest rates, as well as so-called "prudent" fiscal policies which inhibit spending by the state, have been cited as reasons why economies have been better able to handle sudden shocks such as oil price increases.
 
Also, the technological changes which induced substitution away from oil in OECD countries over the 1980s are supposed to have reduced dependence on oil as a crucial input in most material production. These factors are supposed to explain why there has been no inflationary consequence and why real trends in output and employment appear to be generally unaffected by the price shock. Such trends in output and world trade growth are indicated in Chart 2.

A closer inspection, however, reveals that such a view is not really a complete or even accurate explanation of the processes currently at work. To understand whether oil prices remain of continuing significance for the world economy or not, it is first necessary to examine the long term trends in this price relative to those of other important trade commodities.
 
Chart 1 plots these trends in the world trade prices of oil, non-oil commodities and manufactured goods. The prices refer to index numbers of nominal dollar prices with 1990 as the base year. The first point that needs to be noted is that the most recent hike has come after a long period of slump in oil prices, even as manufactured goods prices have been gradually increasing. Between 1986 and 1998, oil prices remained low if unstable, and even after the increase, by May 2000 they had not even recovered to the level achieved in 1979, and were about 25 per cent lower than their peak in 1980. (It should be noted that all the data for 2000 pertain to January-May of this year.)
 
So the increase in oil prices, sharp as it may appear, may be seen in more long run terms as a corrective to the very low levels that have prevailed for the past decade and a half. What is more significant in terms of international inflationary pressures, is that such a rise in petroleum prices has not been accompanied by any companion increase in the prices of non-oil commodities.
 
This is unlike the pattern of the 1970s, when oil price hikes effectively set the tone for more generalised rises in international commodity prices, which peaked in 1976 and 1980. The recent increase has been associated with a continued slump in non-oil commodities, with no apparent indication of reversal of such a tendency.
 
Not only has the recent increase occurred after a prolonged stagnation in price and not been associated with other commodity price increases, but even in itself it has not actually been as sharp and significant as its more famous predecessors. It is evident that the oil price hike of 1973-74, which effected a fivefold increase in dollar prices, as well as the subsequent shock of 1978-79, which amounted to a further more-than-tripling of the price level, were quantitatively far more significant than the most recent increase.
 
This is highlighted in Chart 3, which shows the per cent change in oil and manufactured prices in the three periods of 1972-74, 1978-80 and 1998-2000. Chart 3 also indicates that the first two shocks were also coterminous with (many would say, causally related) quite substantial increases in the dollar prices of traded manufactured goods as well, whereas as the most recent increase has been associated with a zero inflation rate for manufactures.

These are all reasons why the latest period has been associated with little change in terms of overall output and prices : because the oil prices themselves have not changed as much as they did earlier, and because this increase in turn has been less associated with simultaneous rises in other commodity prices and manufactured goods.
 
Indeed, in relative terms, oil prices have hardly increased with respect to manufactured goods prices despite the very significant nominal spurt. Chart 4 plots the change in nominal oil prices relative to manufactured goods prices. It shows that even after the recent hike, the relative price of oil is still substantially lower than it was in 1985, and really seems closer to the depressed levels of the rest of the decade of the 1990s. The closest comparison is with the periods 1976-77 or 1990, and does not suggest a sharp upward movement in relative price.

But of course, for those with a shorter time horizon, the recent price rise still appears very significant. Increasingly, agents in the world economy appear to suffer from long-term amnesia, in terms of not looking more than at most a decade back. Seen only in this light, it is true that the latest OPEC-induced supply limitation has indeed been effective in causing world oil prices to rise very sharply. Chart 2 show how this appears in the overall context of the 1990s.
 
What is interesting for the nineties, of course, is that the oil price rise is associated not just with no setback, but even with a recovery in world trade and output growth (albeit a relatively weak one) from 1998 onwards. In fact, in the second half of the decade, the movement of oil prices has been in the same direction as that of output and export volumes, in sharp contrast to the experience of the 1970s and early 1980s.
 
For this, we need to find a further explanation beyond that of the comparative size of this particular price rise. As mentioned earlier, the price increase itself reflects the sudden ability of OPEC, after years of relative incapacity, to determine levels of output and adhere to the supply discipline necessary to make an impact. Such discipline on the part of all OPEC members need not last for long, as the recent decision of Saudi Arabia to increase its output, and the resulting immediate declines in prices per barrel, indicate. But if they are maintained, then they may in fact result in firmer prices for a more prolonged period.
 
There are several answers to the question of why the rise in oil prices appears to have a much more limited impact on other prices. The first, and possibly the most significant, is that nowadays in most countries, the inflationary impact tends to be limited to the energy components of general price indices and to some of the industries which are more energy-intensive. They generally do not have second round effects on wages and other costs. This is a reflection of the greatly reduced bargaining power of workers generally, and the inability to ensure any meaningful indexation in most countries, after a decade of "globalisation".
 
Because the rise in oil prices typically does not enter wage costs in the second round, it allows other prices to be more or less maintained. But there may be another aspect to this as well. The recent period has seen abnormally high rates of return on private capital in the developed industrial countries, to the point where the OECD estimates that average rates of return on business investment are in excess of 16 per cent (compared to around 12.5 per cent in the 1980s).
 
Such high rates of return may allow capitalists to suffer a slight drop in mark-up if demand conditions are anyway not very buoyant, which is the case everywhere outside the United States. Thus, even when energy costs increase, there may be a tendency to repress consequent price increases and take a small decline in mark up simply to maintain demand for their products. Of course, this is not likely to be a very long-lived practice, and so if the oil price hike is prolonged or even if it stabilises at this level, we may witness a more generalised increase in other prices.
 
One reason why both workers and capitalists may be more willing to accept some loss of indexation (quite aside from the obvious issue of lower bargaining strength of workers generally) is because of possible expectation that such an oil price hike may be short-lived. Indeed, one of the most apparent lessons of the experience since the early 1970s is the volatility of the world oil market, and the tendency of booms to be followed by long period of slump in price.
 
A look at Chart 1 will reveal that over the six cycles that can be discerned in oil price movements over the period, nearly three-fourths of the period was spent in what could be called slump phases. Also, booms have generally been of shorter duration than slumps, and before 1999 the most recent experience was of very prolonged slump of nearly five years duration.
 
It is true that this does not necessarily mean that the future pattern of oil prices will follow the same course. But it may well affect expectations and therefore the behaviour of economic agents who would set their own pricing decisions and wage demands accordingly. This experience was not available to economic agents in the 1970s, who therefore tended to react with much stronger inflationary expectations which eventually became self-fulfilling.
 
The role of technological change which has led to a decline in the energy intensity of material production is obviously also important. But its importance should not be overplayed to the extent of seeing this crucial input as no longer relevant for production and pricing decisions. At the margin, oil prices can still mean very great changes in overall input costs and can affect production substantially. And there is no question that the western powers themselves are strongly aware of how important it is for the stability of their economies to ensure a continuous supply of a relatively cheap energy source.
 
What all this suggests, then, is that while there are certain forces in the international economy which have limited the ripple effects of the latest and rather muted oil price shock, this does not mean that a more intense or prolonged shock will not have adverse effects on inflation, output and employment.
 
So much for the effect on the world economy; what of the effect on the oil exporters themselves ? It has been one of the unfortunate ironies of the OPEC attempts to increase its member countries' share of world income, that oil exports have only in rare cases ensured either development or even growth on par with non-oil exporting countries.
 
One of the important reasons has been because most developing country oil exporters have used their oil resources as implicit collateral to access external capital in the form of external commercial borrowings, and the interest rates they have had to pay have usually been moving along with the movement of oil prices themselves. This is evident from Charts 5 and 6, which show that an oil-exporting borrowing country would have experienced rather little advantage of oil prices changes relative to interest charges on debt.

Chart 7 shows the comparative performance of developing country oil exporters, compared to non-oil exporting developing countries, in terms of GDP growth over the period 1980 to 1997. Apart from Sub-Saharan Africa, all the other major regions show that GDP growth was significantly higher for non-oil exporters.

Some of the reason for this may lie in the poorer performance in terms of investment, of the oil-exporting developing countries. As Chart 8 shows, once again in West Asia and North Africa, as well as in Latin America and the Caribbean, oil exporting countries had significantly lower (and sometimes negative) rates of growth of investment over this period. Once again, Sub-Saharan Africa shows the opposite tendency, but only to a limited extent.

Investment rates have turned out to be so low and falling, not only because of poor performance in terms of domestic savings. Interestingly, net capital flows - which should ideally be countercyclical in order to smoothen adjustment in oil-exporting countries - have exhibited pronounced pro-cyclical tendencies. This is illustrated in Table 1. In periods when the oil price has changed substantially upwards, there net capital inflows have also been the highest. And when the prices has fallen, there has been net outflow of capital. Even the amounts seem to vary proportionately with the oil price change.

So it is not just that domestically oil-exporting countries have not been able to use savings rates changes to smoothen out or stabilise domestic income and absorption over the cycle. It is also that foreign capital itself has responded in a highly pro-cyclical manner, thereby accentuating the booms and slumps.
 
In conclusion, it should be emphasised that, despite the premature celebration in many western capital, OPEC still has the potential for altering oil price levels, and therefore making a significant on inflation, output and employment if such changes are sustained. The saddest part of this story is that thus far, despite this potential power, it has not really been able to ensure a feasible and sustainable development trajectory for most of its members.