BARRELL AND KIRBY
OIL PRICES AND GROWTH 39
OIL PRICES AND GROWTH Ray Barrell and Simon Kirby
Over the past fifteen months oil prices have steadily risen from around $60 a barrel to over $100, as we can see from Figure 1. The projections in our forecast are based on information from futures markets and on projections by the US Energy Information Administration, and as prices have risen there has been increasing evidence that they are considered to be overshooting a sustainable position, as can be seen from our projections for oil prices. Oil prices are expected to fall back to $90 a barrel by the end of 2009. Hence the current oil price shock can be seen as a combination of a temporary increase and a permanent increase in oil prices. The impact of increases in oil prices on growth depends in part on the reasons for the increase, with the effects of supply reduction induced increases in prices being more negative than demand induced increases in prices. A supply reduction involves a loss of revenue for Figure 1. Oil price forecasts since January 2007
producers as compared to a fixed volume demand induced rise in oil prices, and hence a supply reduction would result in lower levels of imports by oil producers. We have assumed in our forecast that the increase in oil prices is driven by increasing demand from China, India and elsewhere as their strong growth increases their energy use. There are also supply constraints in Russia, Latin America and the North Sea, which, along with Saudi reluctance to increase output, mean that supply is not increasing as much as we might expect. Oil producers’ revenues will be buoyant, and this is allowed for in our forecast, with countries that trade a lot with oil producers potentially benefiting from increases in their exports. We undertake two diagnostic simulations on our global econometric model, NiGEM, to investigate the potential impacts of increases in oil prices. In the first we increase oil prices permanently by $10 a barrel with our default policy rules in place. In the second we increase oil prices by $20 a barrel for two years and then return them to baseline, again with default policy rules in place. There are many factors that pattern the impacts of oil prices and these are discussed in Barrell and Pomerantz (2004), who stress the importance of policy rules in determining the impacts on oil prices. The default policy rule in NiGEM feeds back on inflation and a nominal magnitude which stabilises the price level, as discussed in Barrell, Hall and Hurst (2006). In the permanent increase scenario US interest rates rise by 0.4 percentage points initially and slowly return to base over seven years. Real long-term interest rates have to rise by 0.1 percentage points in the US, and by similar amounts elsewhere. The oil importing countries need to free up real resources to pay the increased oil bill, and higher real interest rates are one of the driving factors behind lower sustainable absorption. Increases in nominal interest rates in the Euro Area and the UK are smaller, at 0.25 and 0.2 percentage points respectively, despite the fact that very similar policy rules are used. This reflects the smaller impact of oil prices on inflation in these less energy intensive economies.1 If oil prices rise by more,
110 100 90 $ per barrel 80 70 60 50 40 2006Q1 2007Q1 2008Q1 2009Q1 2010Q1 2011Q1 2012Q1 2013Q1 2014Q1
April, 2008 October, 2007 April, 2007
January, 2008 July, 2007 January, 2007
40 NATIONAL INSTITUTE ECONOMIC REVIEW No. 204 APRIL 2008
Table 1. Output effects of a permanent $10 rise in oil prices (% difference from base)
Euro Area 2008 2009 2010 2011 Average 2012–2018 –0.07 –0.18 –0.28 –0.36 –0.47 France –0.09 –0.20 –0.27 –0.32 –0.43 Germany –0.09 –0.21 –0.28 –0.31 –0.32 Greece –0.12 –0.27 –0.38 –0.47 –0.65 Italy –0.03 –0.13 –0.28 –0.42 –0.64 Neths –0.05 –0.18 –0.33 –0.45 –0.59 Spain –0.07 –0.24 –0.38 –0.49 –0.59 Japan –0.07 –0.20 –0.32 –0.42 –0.57 UK –0.03 –0.09 –0.15 –0.21 –0.34 US –0.11 –0.27 –0.38 –0.47 –0.63
Table 2. Output effects of a temporary two-year $20 rise in oil prices (% difference from base)
Euro Area 2008 2009 2010 2011 –0.08 –0.22 –0.25 –0.22 France –0.09 –0.19 –0.15 –0.08 Germany –0.14 –0.30 –0.24 –0.12 Greece –0.11 –0.22 –0.20 –0.20 Italy –0.01 –0.10 –0.30 –0.42 Neths –0.08 –0.29 –0.43 –0.44 Spain –0.07 –0.26 –0.35 –0.28 Japan –0.13 –0.35 –0.41 –0.37 UK –0.06 –0.14 –0.14 –0.13 US –0.13 –0.30 –0.33 –0.30
Table 3a. Inflation rate effects of a permanent $10 rise in oil prices (difference from base)
Euro Area 2008 2009 2010 2011 0.30 0.23 0.16 0.09 France 0.22 0.17 0.14 0.10 Germany 0.26 0.16 0.08 0.03 Greece 0.26 0.21 0.17 0.12 Italy 0.47 0.39 0.22 0.10 Neths 0.25 0.25 0.26 0.21 Spain 0.33 0.28 0.18 0.08 Japan 0.13 0.18 0.13 0.10 UK 0.22 0.17 0.13 0.09 US 0.38 0.33 0.19 0.11
Table 3b. Inflation rate effects of a temporary two-year $20 rise in oil prices (difference from base)
Euro Area 2008 2009 2010 2011 0.57 0.46 –0.24 –0.29 France 0.42 0.33 –0.14 –0.14 Germany 0.50 0.32 –0.30 –0.26 Greece 0.50 0.42 –0.12 –0.14 Italy 0.90 0.77 –0.42 –0.57 Neths 0.48 0.50 0.07 –0.10 Spain 0.64 0.56 –0.25 –0.38 Japan 0.22 0.30 0.02 –0.12 UK 0.38 0.32 –0.07 –0.18 US 0.79 0.77 –0.28 –0.41
but only for two years, then interest rates will also increase by more. The US short-term interest rate rises by 0.9 percentage points for two years, whilst the interest rate in the Euro Area rises by almost 0.6 percentage points, and in the UK it would rise by about 0.4 percentage points. Exchange rate ‘jumps’ in the first period in our rational expectations model mirror these change in expected interest rates. A permanent rise in oil prices reduces output everywhere, except in oil producing countries such as Norway, Russia and the OPEC member states. The increased oil revenue is spent in these countries and, as in Australia in the current commodity boom, output rises in response, as do imports. Both Russia and Norway have oil revenue saving schemes, and we assume that in
a temporary oil price shock much of the windfall revenue is saved and not spent.2 The major output effects from a temporary oil price increase come from the impacts of the change in the terms of trade affecting costs of production. The effects are larger in the US than in the Euro Area on average, and than in the UK, which remains a relatively large oil producer, but less than in Japan which depends heavily on oil and coal imports. The effects of oil price increases on inflation depend in the first year almost entirely on their impact on costs, and, as we can see from table 3, in all countries these are almost twice as large in a temporary $20 shock as in a permanent $10 shock. If oil prices rise by $10 a barrel, then inflation will increase by about a quarter of a percentage point, with the impact being about 30 per
BARRELL AND KIRBY
OIL PRICES AND GROWTH 41
Figure 2. Euro Area output growth with different paths for the oil price
3.5 3.0 Per cent per annum 2.5 2.0 1.5 1.0 2006 2007 2008 2009 2010 2011 2012 2013 2014
Figure 3. US output growth with different paths for the oil price
3.5 3.0 Per cent per annum 2.5 2.0 1.5 1.0 2006 2007 2008 2009 2010 2011 2012 2013 2014
Figure 4. UK output growth with different paths for the oil price
3.5 3.0 Per cent per annum
Figure 5. Japanese output growth with different paths for the oil price
3.5 3.0 Per cent per annum 2.5 2.0 1.5 1.0
2.5 2.0 1.5 1.0 2006 2007 2008 2009 2010 2011 2012 2013 2014
2006
2007
2008
2009
2010
2011
2012
2013
April 2008 forecast April 2008 forecast less 2008q1 oil shock April 2008 forecast less 2008q1 to 2007q4 oil shocks April 2008 forecast less 2008q1 to 2007q3 oil shocks April 2008 forecast less 2008q1 to 2007q2 oil shocks April 2008 forecast less 2008q1 to 2007q1 oil shocks
2014
42 NATIONAL INSTITUTE ECONOMIC REVIEW No. 204 APRIL 2008
cent bigger in the US than in the Euro Area, and marginally lower in the UK because it remains relatively self sufficient in oil, and there are offsetting effects on the exchange rate. The longer-term effects of oil prices on inflation depend on the degree of inertia in the wage– price system, on the credibility of monetary policy, and most importantly on the effectiveness with which the central bank combats inflation. Labour market reforms and improvements in central banking are the major defence against a repeat of the explosion of inflation seen with higher oil prices in the 1970s. It is possible to evaluate the impacts of the increase in oil prices we have seen over the past fifteen months by progressively stripping off the shocks and creating a counterfactual historical sequence. The shock can be seen as a combination of a $10 temporary increase in oil prices and a $34 or so permanent increase in oil prices. We strip it off as a sequence of five quarters of a $9 increase followed by a permanent $7 a barrel increase, and the experiment is repeated for each quarter back to January 2007. We report the impacts on output growth in the US, the UK, the Euro Area and Japan in figures 2– 5. In the absence of the oil price increases since early 2007, output growth in 2008 in the Euro Area and Japan would be around half a per cent stronger than on our April 2008 forecast. The impacts would have been about half this in the UK, and the more oil intensive US would have seen growth of around 2 per cent this year, rather than around 1.3 per cent. The impacts of oil prices on inflation prospects for 2008 in these economies is relatively easy to calculate from our tables, but has been overlaid by significant changes in exchange rates. If only oil prices had changed, we would have revised up our Euro Area inflation forecast
by around 1.0 percentage point between January 2007 and April 2008, but the 12 per cent effective appreciation of the euro over this period has offset many of the effects, and we do not expect Euro Area inflation to remain at the levels we currently see. Conversely the dollar and the pound have fallen by around 10 per cent in effective terms and this will have added to inflationary pressures. The oil price increase alone should have added more than 1.0 percentage points to US inflation, but we have raised our projection for 2008 by half a per cent more than this since January 2007 because we have also to take account of the exchange rate change. The yen has appreciated by around 7 per cent, and this is almost enough to offset the impact of oil prices on inflation in 2008.
NOTES
1 2 Barrell and Pomerantz (2008) look at fossil energy intensity and demonstrate how it has fallen everywhere, but remains noticeably higher in the US than in Europe. To ensure this we endogenously reduce domestic demand growth to a quarter of what it would otherwise have been. If we do not do this, then output rises in countries that trade closely with these oil producers. Their trade is heavily weighted toward Denmark, Sweden, Finland and the Baltic countries
REFERENCES
Barrell, R., Hall, S. and Hurst, A.I. (2006), ‘Evaluating policy feedback rules using the joint density function of a stochastic model’, Economics Letters, 93 (1), October, pp. 1–5. Barrell, R. and Pomerantz, O. (2004), ‘Oil prices and the world economy’, Focus on European Economic Integration, Oesterreichische Nationalbank, 1/2004. — (2008), ‘Oil prices and world inflation’, National Institute Economic Review, 203, January, pp. 31–4.