>COMMODITIES: Negative growth feedback to constrain oil prices in 2010

The interdependency of our growth and oil-price forecasts
In light of our relatively optimistic outlook for global growth, see Global Scenarios, it is natural to ask why our forecasts leave little in the way of further upside in oil? While our economists look for the US economy to grow 3.2% y/y in 2010, we expect oil prices to average USD83 this year (see charts).

There are three main reasons for this view. First of all, the significant stock overhang currently in place will take time to be worked off, notwithstanding the expected pick-up in demand. Secondly, during the past year, commodity prices appear to have risen largely on the back of expectations of a recovery in demand, which remains yet to be fully confirmed. Thus, the commodity sector seems to have risen rather early in the cycle this time around, contrasting with its late-cycle performance historically. Finally, a negative feedback loop with economic growth should constrain oil prices to some degree this year. We consider the latter effect in depth below.

Simulating feedback effects between oil and growth
As economic activity improves, demand for commodities – not least for energy – usually increases as well, driving e.g. oil prices higher. However, a rising oil price would in itself dampen growth to some degree, at least for an oil-importing country such as the US.

A Fed paper, see Aggregate Disturbances, Monetary Policy, and the Macroeconomy (1999), estimates that a permanent increase in the oil price in the US relative to the rest of the world cumulating to USD10 per barrel over four quarters leads to a decline in real GDP of 0.2% in the first year after the shock relative to the situation where no shock occurs and a decline of 0.4% in the second year. Consumer-price inflation is found to end 0.5% higher in year one and then come down to end a mere 0.2% higher in the second year after the shock has hit. The Fed would in turn react to the changing inflation-growth outlook by adjusting the short-term interest rate.

To illustrate these feedback mechanisms, we provide here some simulation results from a simplified model of the US economy. Our quarterly model based on data for the period 1986 to 2006, that is, leaving out the crisis period, allows for simultaneous modelling of US real GDP, PCE inflation, the three-month USD Libor rate, a ten-year government benchmark bond rate, the real effective USD exchange rate and the nominal WTI oil price.

To read the full report: COMMODITIES