Oil prices and economic recovery

Jan 25, 2003 01:00 AM

by John Gault

Nearly every news article speculating about the pace of world economic recovery includes the caveat “unless oil prices suddenly rise”. Such concern increasingly is related to a possible oil price jump if and when the US carries out its threat to invade Iraq.
The International Monetary Fund (IMF) recently expressed the concern in these terms: “… the possibility of further sharp increases in oil prices arising from a deterioration in the Middle East security situation remains an important downside risk to the global outlook.” [1]
Exactly how an oil price increase would affect economic activity and the magnitude of any such impact deserves careful attention. Neither the two large oil price increases of 1973-74 and 1979-80 with their subsequent decelerations in world economic growth, nor the experience of the 1990-91 Gulf War, are adequate precedents for understanding the present situation.

An exceptional situation
A comparison of the current economic slump withthose in 1974-75, 1979-82 and 1990-91 indicates some important differences. One noticeable difference is that each previous slow-down began almost simultaneously with an oil price run-up, while the current deceleration commenced long after oil prices had risen.
The lag in the US, the engine of economic expansion throughout the 1990s, was remarkable; economic growth continued for seven quarters after oil prices began to rise in March 1999. From the first quarter of 1999 to the second quarter of 2000 the price of crude oil [2] more than doubled, but real GDP growth in the US continued to expand at a blistering pace. [3] Clearly, the current economic slump is different. Almost no one claims that oil prices triggered the economic stagnation experienced since mid 2000. [4]

Indeed, there are numerous reasons why oil prices now have much less impact on the economic health of the industrialized world than in the past:
1. The oil and energy intensities of industrialized economies have greatly diminished since the early 1970s. The OECD countries utilized 1.43 barrels of oil per $ 1,000 of GDP in 1970, compared with 0.74 barrel in 2000. The same group of industrialized countries utilized 2.82 boe of commercial energy in 1970, and only 1.69 boe in 2000.[5]

2. Many oil importing countries, especially European countries, have increased excise taxes on petroleum products, particularly gasoline (Table 1).[6] A substantial portion of these taxes is not ad valorum. Therefore the percentage impact on retail prices of any given change in crude oil prices will be muted.

Table 1.
Taxes as a percentage of retail prices in selected countries
Gasoline automotive diesel
1978 2002 1978 2002

France 63.2 73.0 54.2 59.3
Germany 57.9 72.0 53.1 59.9
Italy 71.2 67.7 17.4 56.3
Spain 22.3 61.5 16.5 48.9
UK 50.3 76.6 45.0 85.2
Japan 39.1 56.5 n/a 43.0
US 19.2 27.4 25.9 31.8
Canada 29.4 41.2 n/a 33.5

3. Oil prices in real terms are far below levels of the late 1970s/early 1980s. General price levels in the advanced economies approximately doubled between 1980 and 2001, so that the average price of Brent crude oil in 1980, $ 37.89/bbl, would correspond to a price of over $ 70/bbl today. [7]
Some of the adaptations to higher real energy prices made in the 1980s, such as improved building insulation standards and more efficient industrial process engineering, will not need to be repeated even if oil prices were to increase from today’s levels.

4. Active futures markets for crude oil and petroleum products, not available during the oil price increases of 1973-74 and 1979-80, today insulate many refiners and large fuel users from the immediate impact of oil price swings. Already in 1990 some hedgers benefited from long futures positions during the run-up to Desert Storm. The backwardation of the market throughout the 1990-91 crisis signalled a consensus that high spot prices were temporary, thereby helping to calm markets.
The above considerations would lead one to expect that the impact of any given price increase today would be substantially smaller than a corresponding increase might have been in the past. Indeed, calculations prepared by the IMF in December 2000 reflected such a diminishing impact (Table 2). [8]
The projected nominal oil price increase from 1999 to 2001 of $ 11.10/bbl would shift an additional $ 96 bn from the advanced countries [9] to the oil exporters, but this amount would correspond to only 0.4 % of the same countries’ combined GDP. This may be compared with the nominal oil price increase of $ 8.40/bbl from 1973 to 1974 and the consequent income transfer of $ 88 bn, constituting 2.6 % of GDP for the same group of advanced countries.

If one modifies the IMF calculations shown in Table 2 to account for the actual variations in oil prices during the period from early 1999 to the last quarter of 2002 (the price movement has not been only upward), and if one uses a 1998 price base (approximately $ 12.50/bbl for IEA crude oil imports, rather than the $ 17.90/bbl used in Table 2), then the direct impact on the net trade balance of the advanced countries is even smaller than the 0.4 % of GDP indicated in Table 2. [10]
The direct impact on net trade balances measures the first-round transfer of income from the advanced economies to the oil exporters, but several caveats must be borne in mind: The impact is not the same on each of the advanced economies. Some, such as oil-exporter Norway, will experience a positive direct impact on its net trade balance from an oil price increase. The indirect impacts of an oil price increase will to some extent offset the direct impact. As oil exporters’ incomes rise, they import more goods and services from the advanced economies. [11]

The estimates shown in Table 2 are therefore upper bound estimates of the final direct and indirect impacts on the net trade balance of the advanced countries.
Table 2.
Direct impact of nominal oil price hikes [12] ($, unless otherwise stated)
Direct impact on net oil prices
Balance of advanced countries
Episode Pre-hike (1) Post-hike (2) Change ($ bn) (% of GDP)
1973 to 1974 3.2 11.6 8.4 -88 -2.6
1978 to 1980 13.3 36.6 23.3 -232 -3.7
1989 to 1990 17.9 28.3 10.4 -38 -0.2
1999 to 2001 17.9 29.0 11.1 -96 -0.4

Source: IMF staff estimates.
(1) The average oil price in the first year of each episode.
(2) The average oil price in the last year of each episode, except for 1990 and 2001. For 1990, it is the average price for the second half of the year. For 2001, the price is projected using futures markets data.

On the other hand, the first-round income transfer will have secondary multiplier effects on total demand in the oil importing countries. Macroeconomic models take into consideration these and other indirect effects such as changes in inflation and interest rates.
The IMF’s model suggested (at the end of 2000) that the overall impact of a permanent increase in oil price of $ 5/bbl would be to reduce real GDP in industrial countries by between 0.2 % and 0.4 % in each of the first four years following the price hike, with diminishing impact thereafter. [13]

The OECD’s somewhat different model predicts a smaller impact. For example, the OECD model predicted in 1999 that an oil price increase of $ 10/bbl would reduce real GDP by 0.2 % in the USA and Europe, and by 0.4 % to 0.5 % in Japan, during the initial years following the price “shock”. [14]
While these models utilize differing assumptions, and hence yield different ultimate impacts, a review of the impacts estimated by the OECD model since the early 1980s confirms the impression given by Table 2: The impact of any given oil price change on the economies of the industrialized, oil-importing economies will be significantly smaller now than it would have been on past occasions. [15]

If the ultimate impact of an oil price increase on the industrialized world has so greatly diminished, are the concerns about a future oil price hike, caused perhaps by supply interruptions related to an invasion of Iraq, justified? Two areas of remaining concern may be considered:
-- A “last straw” effect: Under present circumstances, could the psychological effect of an oil price leap on consumers exaggerate the expected impact?
-- Conflicting goals: Under present circumstances, would monetary and/or fiscal authorities react to an oil price leap in a manner that worsens the expected impact? We will examine these possibilities using the US as an example, because the US was the driver of the world economy for such a long period prior to the current slump, and because the US has appeared to be leading, tentatively, the world recovery. A setback for the US at this time would be a setback for the entire world.

The last straw
The present economic slump is unusual in that consumer spending has not yet declined. In previous recessions, real consumption expenditures declined and did not return to pre-recession levels until at least the third quarter. Present consumer behaviour is all the more remarkable because consumers’ wealth has shrunk due to stock market losses since the bursting of the technology bubble early in 2000.
From a peak of $ 19 t in August 2000, the combined market value of all stocks listed on the three principal US exchanges fell to $ 12 t in September 2002, a loss of $ 7 tn. Such a dramatic decline in wealth “should” have brought about a retrenching by consumers. A recent review of the literature by the IMF concluded that the decline in household spending in the US would be in the order of $ 0.03 to $ 0.05 per dollar of lost wealth, all other factors held constant. [16]
This would correspond to a retrenchment of $ 210 bn to $ 350 bn in US consumer spending on the assumption that all of the $ 7 tn decline on US stock markets was borne by US households.

Of course, this assumption is a simplification; some of the shares were held by non-US owners, and some US households own shares on stock markets outside the United States. Stock markets in Europe and Japan, however, also have fallen sharply. Whatever the limitations of the assumption, the as-yet-unrealised wealth impact on consumer spending is potentially enormous.
Such a retrenchment would dwarf the impact of a $ 10/bbl increase in oil prices (0.2 % to 0.4 % of GDP depending upon which macroeconomic model one prefers). Applied to a US GDP of about $ 10.4 tn, a $ 10/bbl oil price jump would yield a decline in GDP of between $ 21 bn and $ 42 bn in the first two years and diminishing amounts thereafter, all other factors held constant. [17]
This is an order of magnitude smaller than the hypothetical wealth effect. Various reasons have been proposed to explain the persistence of consumer spending in the US despite the current economic downturn and the decline in stock market valuations. Americans, it is said, have refinanced their residential mortgages as interest rates have declined and housing values have risen, thereby monetising their higher homeowners’ equity.

Americans have continued to purchase automobiles thanks to zero-interest-rate financing offered by US auto manufacturers. Or, possibly, the largest losses on the stock exchanges have been borne by individuals of above-average wealth, whose marginal propensity to consume out of wealth is below average.
Whatever the reason, the persistence of consumer spending suggests that some relatively minor event -- something much smaller than the $ 7 tn decline in stock market values -- could be sufficient to trigger a long-overdue retrenchment. A new oil price jump, even if only temporary and even if oil futures markets remain in backwardation, could be viewed as a potential trigger.
The blow to consumer confidence when hostilities in the Middle East begin is very likely to depress spending. If so, would it be fair to assign the resulting double-dip in the recession predominantly to oil prices?

Conflicting goals
While consumption expenditures have failed to drop, investment outlays in the US declined for four consecutive quarters but have since begun to recover. The recovery has not been as rapid as in 1980 or 1990, but has been more rapid than after the 1974 recession.
One factor contributing to this tentative recovery has been low interest rates, especially in the US. If oil prices should experience a sudden upward jump, this would cause an immediate, first-round increase in the rate of inflation. Based upon past experience, the Federal Reserve may be tempted either to increase interest rates or to postpone further reductions in interest rates to minimize the secondary inflationary effects (increases in the prices of labour and other inputs) following the oil price jump.

In the wake of the 1973-74 oil price “shock”, the Federal Reserve and central banks in some European countries reduced interest rates and thus followed a generally accommodative policy. In retrospect, this is widely viewed as having been misguided because inflation rebounded strongly. During subsequent oil price “shocks”, in 1979-80 and in 1990-91, central banks focused instead on combating the inflationary impacts.
In recent years central banks have become even more single-mindedly focused on achieving inflation targets. Although most central bankers profess that the first round inflationary impacts of a supply shock should be accommodated, in practice it is impossible to differentiate clearly between first and second round effects in the data. Therefore it is plausible to expect that a future oil price increase may lead to a moderation or even reversal of the Fed’s current expansionary monetary policy.

A further decline in the US net trade balance caused by an oil price increase could compound the pressure on the Federal Reserve to raise interest rates. The US is running a record current account deficit, much higher than during previous oil price run-ups. The US current account deficit is larger than that of any other major industrialized country and is widely viewed as unsustainable. A sudden increase in the size of the deficit (a first-round effect of an oil price jump) may provoke a decline in the value of the dollar, adding to US inflationary pressures and provoking a Fed decision to tighten monetary policy.
If this should happen and the US were to lead the world into the second “dip” of a double-dip recession, would it be reasonable to blame oil prices? Rising oil prices have been far from the only factor driving the US deeper into a negative trade balance. Nearly the entire deficit is attributable to automobiles, automobile parts, consumer durables, and other non-food consumer products. All that one may conclude is that an oil price jump would put the Fed into a difficult dilemma, given the fragility of the present economic recovery.

The role of OPEC
It would appear, therefore, that over the long run the impact of oil prices on the industrialized economies has greatly diminished, but present weak economic circumstances magnify the ability of an oil price jump to trigger undesirable consequences.
If such a price jump should occur, it almost certainly will not be due to an action by OPEC. OPEC has declared its intention to keep its basket price within the $ 22-$ 28/bbl band. Individual members with idle production capacity -- particularly Saudi Arabia -- may be expected to increase production to moderate any price excursion caused by military action against Iraq, as was done in 1990.

OPEC’s production decisions in recent years have been highly compatible with the central banks’ focus on inflation. At times when central banks have increased interest rates, OPEC has raised its production ceiling thereby moderating oil price increases and accommodating higher oil demand. [18]
OPEC has reduced its production ceiling only at times when central banks were lowering interest rates, indicating a slackening of (macroeconomic) demand and a lessened concern about inflation in the industrialized economies. This is not to suggest that OPEC has consciously coordinated its policies with Western central banks, but rather that OPEC’s pursuit of oil price targets has not been entirely inconsistent with central banks’ broader objectives.

Oil prices undoubtedly will leap upward at the onset of any military conflict in Iraq, as happened in August-October 1990. Whether the price increase endures depends in part on the duration of the military conflict and in part on the willingness and ability of other OPEC countries to expand production.
Whether the price increase provokes a double-dip worldwide recession depends on the spending behaviour of consumers and on policy responses of the major oil importing countries. Whatever happens, news articles undoubtedly will continue to place more of the blame on oil prices and on oil exporting countries than either deserves.

References:[1] International Monetary Fund (IMF), World Economic Outlook, September 2002, p. 58.
[2] The OPEC basket price rose from $ 11.08/bbl average for the first quarter of 1999 to $ 26.38/bbl average for the second quarter of 2000, an increase of 138 %.
[3] Real GDP in the US, in billions of chained 1996 dollars and at seasonally adjusted annual rates, rose from $ 8,396.3 in the first quarter of 1999 to $ 9,205.7 in the second quarter of 2000, or 9.6 %. US Department of Commerce, Bureau of Economic Analysis, Survey of Current Business, October 2002.
[4] An IMF study at the end of 2000 found that “while it is still too early to make a final judgment, the latest data suggest that the impact [of the recent $ 5/bbl oil price increase] on core inflation in advanced countries has been relatively modest to date and there is little sign of feed through into wage claims. Although there has been a decline in consumer and business confidence, they so far remain relatively strong and although stock prices have fallen, the decline appears to be much more due to non-oil factors.” IMF, “The Impact of Higher Oil Prices on the Global Economy”, 8 December 2000, p. 33-35
[5] GDP is measured in constant 1990 dollars on a purchasing-power-parity (PPP) basis. OPEC Review, March 2002, p. 66.
[6] Percentages shown in Table 1 are calculated from data found in IEA, Energy Prices and Taxes, third quarter 1984, and IEA Monthly Oil Market Report, 11 October 2002.
[7] The average GDP deflator for the advanced economies (as defined by the IMF) for the period 1980-2001 was 3.27 %. Inflation was higher in the first half of this period than in more recent years. See IMF, World Economic Outlook, various issues. For the price of Brent crude oil in 1980, see Middle East Economic Survey, 21 January 2002, p. A11.
[8] IMF, “The Impact of Higher Oil Prices on the Global Economy”, op cit. The crude oil prices in Table 2 are IMF “reference prices”, an arithmetic average of Brent, West Texas Intermediate (WTI), and Dubai.
[9] The term “advanced countries” as used here and in Table 2 follows the IMF terminology.
[10] Using IMF figures for GDP and IEA figures for prices and volumes of net crude oil imports into the IEA countries, and extrapolating the fourth quarter of 2002 based upon the first three quarters of the year.
[11] Thomas R. Stauffer, “Oil and the US Balance of Payments”, Middle East Economic Survey, 19 February 1979, Supplement.
[12] Table 2 is a reproduction of Table A1 in IMF, “The Impact of Higher Oil Prices on the Global Economy”, op cit, p. 41.
[13] IMF, “The Impact of Higher Oil Prices on the Global Economy”, op cit., p. 15-17. The IMF estimates assume a pass-through of the oil price increase in the form of increases in nominal wages and profits. The impact, of course, would be smaller if one assumed flexible labour and capital markets and incomplete pass-through, as shown in an alternative scenario at p. 19.
[14] “The Rise in Oil Prices: A Cause for Concern?”, OECD Economic Outlook No. 66, December 1999, p. 8-9.
[15] Compare “A Ready Reckoner of a Hypothetical Change of Energy Prices on the OECD Economy”, OECD Economic Outlook No. 39, May 1986, p. 164, with “The Rise in Oil Prices: A Cause for Concern?” in OECD Economic Outlook No. 66, December 1999, p. 9. The hypothetical $ 3/bbl price change in the earlier estimate is shown to have approximately the same impact on GDP growth as a $ 10/bbl change in the second estimate. Earlier estimates yielded even higher impacts: see OECD Economic Outlook No. 27, July 1980, p. 128-130 and OECD Economic Outlook No. 31, July 1982, p. 139-140.
[16] “Much of the existing literature focuses on the US, where the effects of equity prices on consumption are in the range of 3-5 cents per dollar, with the effect taking one to three years to materialize. Spending out of housing wealth is somewhat higher, close to 4-6 cents per dollar, also taking one to three years to materialize.” IMF, “Is Wealth Increasingly Driving Consumption”, Chapter II in World Economic Outlook, April 2002. See also “Asset Prices and the Business Cycle” in World Economic Outlook, May 2000.
[17] See the discussion of oil price impacts above.
[18] Sources of data underlying the graphic illustration: US Federal Reserve (discount rate), European Central Bank (marginal lending facility), Bank of England (interest rate), and Middle East Economic Survey (OPEC-10 ceilings and production).

John Gault, is an economist and consultant based in Geneva, Switzerland. He contributed this analysis to the Middle East Economic Survey.

Source: MEES
Alexander's Commentary

Change of face - change of phase

In the period of July 20 till August 3, 2015, Alexander will be out of the office and the site will not or only irreg

read more ...
« August 2020 »
1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30

Register to announce Your Event

View All Events