Nevertheless, the volatility of commodity prices - especially energy prices - in the past 18 months has been so pronounced that it has been hard for markets to ignore completely their impact. From late 1997 to early 1999, the Goldman Sachs Commodity Price Index (GSCI) fell by about 40 per cent, which was by far the largest decline seen in the Nineties.
This was driven by a collapse in oil prices from around $21/barrel to under $11/barrel at the low point early this year. However, other components of the GSCI also fell sharply - for example, the Agricultural Index dropped by about 30 per cent, while the Industrial Metals Index fell by about 35 per cent.
This slump in commodity prices during 1998 may have had a rather larger beneficial impact on the global economy than has been generally recognised. For the major developed economies, headline consumer price inflation (including energy) fell from 2.2 per cent in late 1997 to a low point of 1.2 per cent in February 1999. Meanwhile, the core CPI (excluding energy) fell from 2.2 per cent to around 1.6 per cent over the same period. This suggests that the decline in oil prices directly reduced global inflation by around 0.4 per cent last year.
Furthermore, declines in other commodity prices probably added slightly to this disinflationary effect, making an estimated total impact of around 0.5 per cent from the commodity shock on the global CPI.
The immediate impact of this disinflationary shock on real GDP was also highly favourable. With wages largely unaffected by the drop in consumer prices during 1998, the commodity shock boosted real household income by around 0.5 per cent last year, increasing consumers' expenditure by around 0.4 per cent, and implying a boost (including multiplier effects) to real GDP in the OECD area of about the same amount. In summary, then, the impact of the commodity shock in 1998 was probably to reduce global inflation by around 0.5 per cent, while boosting global GDP growth by 0.4 per cent.
Of course, these economic effects have undoubtedly impacted both short and long-term interest rates during the recent past. We can use the well- known Taylor Rule for assessing the possible impact on global short rates. The decline in consumer price inflation of 0.5 per cent would, according to the Taylor Rule, have reduced global short rates by 0.75 per cent, while the boost to real GDP of 0.4 per cent would have increased global short rates by 0.2 per cent, making an overall net effect on global short rates of minus 0.55 per cent.
According to the Goldman Sachs bond model, these shock effects are likely to have reduced 10- year global bond yields by around 0.45 per cent compared with what they would otherwise have been. Since global bond yields fell by around 1.2 per cent from late 1997 to early 1999, this implies that the commodity price shock accounted for about one-third of last year's global bond rally.
And, because lower bond yields were the main driving force behind higher share prices last year, the commodity shock also contributed substantially to the equity bull market.
In recent weeks, however, the slump in commodity prices has started to reverse. So far, the recovery in agricultural and metal prices has been minimal, but the oil price has rebounded by over 50 per cent from its $10-11/barrel low point.
With energy prices accounting for about a 48 per cent weight in the GSCI, this has been sufficient to lead to a rebound of almost 25 per cent in overall commodity prices in the past twomonths. The key issue for 1999 is to what extent last year's favourable economic effects are likely to be reversed, given the commodity price rebound we have now seen.
Let us look first at the likely adverse effect on inflation. Goldman Sachs economists in the US, Japan and Euroland have estimated the impact on producer and consumer prices of three different alternative scenarios for the oil price. These scenarios are shown in the accompanying table - scenario two ($17/barrel) is currently the central case, with scenario one showing a low oil price variant, and scenario three a high oil price variant.
The cumulative impact on OECD consumer prices over about two quarters would be plus-0.6 per cent if oil prices were stable at $17/barrel throughout the summer, as compared with remaining at only $11/barrel. The effect on producer prices would, of course, be much larger at plus-2.3 per cent. There is some variation in these effects between individual countries, with the inflation impact in Japan apparently being considerably smaller than that in Euroland, and the US coming somewhere in between.
This increase in inflation will reverse all of the benefits from lower commodity prices which were enjoyed last year. In fact, it will involve a drag on real GDP of around 0.5 per cent in 1999, reflecting the reduction in real household disposable income which higher energy prices will entail. This is certainly not a negligible hit to the world economy - to put it in context, it may be only a little smaller than the drag which occurred as a result of the Asian shock in 1997-1998.
With global inflation higher than it otherwise would have been, and global GDP growth lower than otherwise, we can use the reasoning outlined above to calculate the impact of the oil price rebound on the expected path for interest rates.
According to the Taylor Rule, global short rates are likely to be around 0.65 per cent higher than they otherwise would have been, which in practice means that the scope for any further cuts in global interest rates following last week's 0.5 per cent cut announced by the European Central Bank (ECB) now looks very limited. In fact, it is unlikely that either the Federal Reserve or the ECB will cut rates again this year, and at least in the United States there are growing fears that rates may have to rise.
Fortunately, with the global economy still operating with a large margin of spare capacity, there are still strong disinflationary effects stemming from the manufacturing sector. But it now looks as if the rebound in oil prices will be sufficient to almost exactly offset the improvement in inflation stemming from these other sources. This will leave the global inflation rate unchanged over a 12-month horizon instead of falling further, as it would have done if oil prices had remained at $11/barrel.
Furthermore, global bond yields are likely to be around 0.5 per cent higher as a result of the rebound in oil prices. Interestingly, this is almost exactly the extent of the rebound in nominal global bond yields which has occurred over the past several months, so on this basis it seems likely that the bond markets have already absorbed most of the bad news emanating from the rise in oil prices seen so far.
Commodity prices, despite their recent rebound, still stand about 10 per cent lower than a year ago, and remain at 30-year lows against OECD consumer prices. They would have to rise sharply further to pose a fatal threat to the bull market in bonds or equities this year, and this seems improbable. But the margin of safety for world inflation now looks much thinner than before.