The Regional Economist • January 2001

After falling sharply during the Asian financial crisis in late 1998, market prices of crude oiland other energy products climbed in November 2000 to their highest levels since the GulfWar. The oil price shock, as economists have coined it, occurred as monetary policy-makersacted to keep the economy from overheating. This combination of events has raised a fewred flags in certain quarters, since nearly all post-World War II recessions were preceded by

RISING OIL PRICES ANDECONOMIC luHMOIL

MUST THEY ALWAYS GO HAND IN HAND?

by Kevin L. Kliesen

higher oil prices and a restrictive monetary policy. Unlike other oil price shocks since the1970s, however, the current run-up in energy prices has not yet raised the alarm one mighthave expected. Indeed, while the pace of U.S. economic activity appeared to slip noticeablyduring the fourth quarter of 2000, most forecasters continue to project solid growth andmoderate inflation for the next two years. Is the U.S. economy better positioned today towithstand the body blows of sharply higher oil prices than it was in the past?

Oil Price Shocks in U.S. Economic History

An oil price shock is one of several possible disturbances to a country's aggregate pricelevel. Its significance reflects the fact that crude oil is an important energy source for mostindustrialized countries, who use energy as a direct or indirect input in the production ofmost goods and services. Because higher oil prices tend to raise the prices of petroleum-based products and alternative sources of energy, such as natural gas, the aggregate pricelevel will rise unless the prices of all other goods and services fall to offset the rise in oilprices. Of course, oil price shocks are not always one-sided. Prices of crude oil can alsofall in significant fashion, as they did in 1986 and 1998. On balance, though, the publicand policy-makers tend to think of oil price shocks as energy price increases, hence, theadverse connotation.

In the early post-World War II period, the run-up in energy prices prior to economicdownturns was comparatively mild. During the four quarters preceding the onset of the1948-49, 1953-54, 1957-58, 1960-61 and 1969-70 contractions, the relative (real) price ofenergy increased a little more than 1.5 percent, on average. These increases, however,were tempered by the fact that the relative price of energy actually declined 2.4 percentand 1 percent prior to the 1960-61 and 1969-70 contractions, respectively—thoughenergy prices subsequently rose sharply in the midst of these two recessions.

[5]

The figure below shows that energyprice changes have been much morepronounced (relative to output prices)since the early 1970s. The averageincrease in real energy prices priorto the onset of the four recessions dur-ing this period—1973-75,1980,1981-82and 1990-91—was 17.5 percent, muchgreater than in earlier recessions. Exceptfor the 1990-91 recession, one of themildest since World War II, the contrac-tions of the 1970s and 1980s were thedeepest since the Great Depression.Though U.S. economic growth remainedsolid through the third quarter of 2000,the 23 percent increase in real oil pricesfrom four quarters earlier is neverthelesslarge in historical terms.

Rising Oil Prices Have UsuallyPreceded Most Recessions

Increases in the Price of Energy Relative to the Price of Output

NOTE: Producer Price Index for Fuels and Related Products and Power (not seasonallyadjusted) divided by the price index for Business Sector Output. Shaded areas areperiods of recessions, as defined by the National Bureau of Economic Research.

SOURCE: U.S. Bureau of Labor Statistics.

As the figure above shows,nearly all post-World War IIrecessions in the United Stateswere preceded by, or accom-panied by, a sharp increase inenergy prices relative to theaggregate price level. As aresult, oil price shocks tendto be viewed with alarm bymacroeconomists, marketsand public policy-makers.This heightened sense ofconcern is largely an outcrop-ping of the 1970s and 1980s.

One reason why oil prices have beenmore volatile since the 1970s has beenthe advent of the Organization of Petro-leum Exporting Countries (OPEC).Economically, OPEC is a cartel—a typeof oligopoly that attempts to keep pricesabove their competitive levels by activelyadjusting crude oil supplies. Last year,for example, OPEC oil ministersannounced that they intended to alterproduction to keep the market price ofcrude oil between $22 and $28 a barrel.OPEC has also tried to manipulateprices for political reasons. The best-known example was the 1973 Arab oil"embargo," which sought to punish theUnited States for its support of Israelduring the Arab-Israeli War.

Ultimately, OPEC's ability to manipu-late market prices is limited, since oil is areadily fungible commodity—that is, it istraded internationally and its U.S. pricedepends not only on domestic supplyand demand, but also on foreign supply

and demand. Moreover, its power is fur-ther limited because many of the largestoil-producing countries—such as theUnited States, Russia, Mexico and Norway—are not OPEC members.

Nevertheless, the cartel has consider-able influence on crude oil prices simplybecause OPEC countries control roughly40 percent of world production. ButOPEC is not the only reason for increasedvolatility of world energy prices duringthe past 30 years or so. Another possiblereason is the deregulation of U.S. energymarkets in the late 1970s and early1980s. Because price determination isnow largely a function of market forces,economic or political developments cansometimes engender wide swings in oilprices if markets are caught unaware.

A salient feature of most oil priceshocks since the 1970s has been a changein the amount of oil supplied, or theexpectation of a disruption in the oil sup-ply (such as during the Gulf War). Butoil price shocks can also occur when thedemand for oil rises or falls relative to thesupply.1 Indeed, the price of importedoil plummeted to below $10 a barrel inlate 1998 because of a marked slowingin the growth of world oil consumption—reflecting, importantly, the significantdeclines in economic activity in South-east Asia. Even though the subsequentrun-up in prices during the past twoyears occurred while OPEC was tryingto limit output to boost prices, steadilyfaster rates of worldwide economicgrowth suggest that demand-sideinfluences have also been importantin pushing up the price of oil.

Forecasting the future pace of realGDP growth is a high-risk venture evenin the best of times. But when the econ-omy is hit with an oil price shock and amodestly restrictive monetary policy, thedegree of difficulty increases by severalmagnitudes. At present, higher energyprices seem not to have rattled consu-mer confidence. This could obviouslychange if events take a turn for theunexpected or, as government forecast-ers expect, household energy expendi-tures climb markedly higher this winterthan they did during last year's relativelybenign winter. And whereas the coststructure of firms in industries that arefairly intensive users of energy, likemanufacturers, transportation firms andagriculture, also tend to be hit dispropor-tionately when energy prices turn higher,most firms—and by extension, the U.S.economy—have managed to cope withthe latest oil price shock rather well.

In view of this development, is theresomething fundamentally different inthe structure of the economy that explainsthis development, or have policy-makers,particularly the Federal Reserve, done a

[6]

Vie Regional Economist* [anuai y 2001

better job of coping with the current oilprice shock?

What's Different This Time?

In December 1998, the price ofimported oil plunged to around $9 abarrel. In real terms, this was its lowestlevel since late 1973. All else equal, thispositive price shock should have beengood for the economy. But with consid-erable uncertainty hanging in the airbecause of the Asian financial crisis,Russian debt devaluation, and the failureof Long Term Capital Management, amajority of U.S. forecasters were predict-ing a downshift in the pace of U.S. realGDP growth in 1999 to just over 2 per-cent; they also expected CPI inflation torise to about 2.3 percent.2 When the firstforecasts for 2000 were publishedin January 1999, Blue Chip forecastersexpected continued moderate real GDPgrowth (2.4 percent) and inflation (2.5percent), and not much change in theunemployment rate (4.9 percent).3

Needless to say, these forecastsmissed their mark by a wide margin.Real GDP growth turned out to be5 percent in 1999, the strongest growthsince 1984, while the unemploymentrate dropped to a shade over 4 percent.The brisk pace of economic activity con-tinued into 2000, with real GDP growingat a 4.2 percent annual rate during thefirst three quarters, and the civilianunemployment rate staying near its30-year low of 4 percent. CPI inflationtracked higher in 1999 and 2000, asforecasters expected, but largely not forthe reasons they cited. And while the oil-induced acceleration in consumer pricesin 2000 (3.5 percent through November)is more than most forecasters expected—and, rather troubling when viewedagainst the relatively good inflationperformance from 1992 to 1998—there appears to be little erosion inlong-term inflation expectations bythe public and most forecasters. Indeed,judging from its continued strong per-formance, the economy is better posi-tioned to weather this combination ofoil price shock and monetary tighteningthan it was previously.4

On the surface, the recent poor per-formance of forecasters and—thus far,at least—the ability of the U.S. economyto weather the recent oil price stormappear unconnected.5 But in fact, threedevelopments suggest a closer corre-spondence. Two developments arelargely an outcropping of ongoing struc-tural improvements in the economystemming from the upsurge in spendingon high-tech capital goods. The thirddevelopment, though somewhat harderto quantify, is better monetary policy,

THE ECONOMICS OF AN OIL PRICE SHOCK

attempted to learn how oil (or energy) price disturbancesaffect the growth of output, employment, productivity andinflation.1 To help answer this question, economists typi-cally use models to explain how the economy operates.There are numerous competing models out there, and noone model can account for every last wiggle in the data.That said, one well-known model is a standard aggregateproduction function framework. In this approach, aneconomy's output (real GDP) is determined by the quantityand quality of its capital and labor inputs.2 Thus, how fastthe economy ultimately grows depends on its rate ofinvestment (increase in capital), its population growth(increase in labor) and its productivity growth (increase inthe efficiency of capital and labor arising from technologi-cal improvements). Inflation is a product of the moneymarket: An excess supply of money—determined by mone-tary policy-relative to demand causes inflation to accelerate.

In this simple model, an increase in oil prices reducesthe relative efficiency of older capital (machines andequipment); newer capital uses energy more efficiently.Firms react by striving to economize on their use of energyin the production process by scrapping older vintages ofcapital goods. Thus, capital obsolescence results in lesscapital available per worker, causing a reduction in laborproductivity, a decline in output and an increase in theprice level. Although unemployment tends to increase, thefall in the real wage may spur firms to change the mix oftheir inputs—substituting the relatively less expensivelabor for the more expensive capital. For a decrease in oilprices, the effect is just the opposite: Productivity andoutput increase, while inflation subsides.

Ultimately, higher inflation results from too muchmoney chasing too few goods. Whether higher inflationresults from an increase in energy prices depends, first, onwhether the increase in oil prices is temporary or perma-nent and, second, on whether the Federal Reserve choosesto increase the supply of money to counteract the adverseeffect on output and employment. Economists now gener-ally agree that the Fed was too accommodative duringmost of the 1970s, when the relative price of oil increasedsteadily. That is, it actually fueled higher inflation byincreasing the supply of money too much (more on thislater). During the Gulf War-induced jump in oil prices,which was short-lived, the Fed was more successful, as evi-denced by the subsequent deceleration in inflation andrebound in economic activity.

which has resulted in the achievementof near price stability.

Explanation #1: A MoreProductive Economy

Perhaps the largest difference betweenthe macroeconomic effects of this oilprice shock and previous shocks is theunderlying performance of the U.S.economy. In particular, it appears thatthere has been a marked upswing in theeconomy's capacity to produce goodsand services with a given level of inputslike labor, capital and energy. As inputs

Forecasters Expect Much Better Inflation Performancein the Aftermath of the Latest Oil Shock

GDP Inflation Expectations Before and After Major U.S. Oil Price Shocks

Number of Quarter* Before/After the Oil Price Shock

SOURCE: Survey of Professional Forecasters.

The figure above plots market

expectations of inflation (GDP

prices) eight quarters before and

after four major oil price shocks

since the early 1970s. It shows

another reason why economic

activity has not been appreciably

hampered by the current oil price

shock: Not only were inflation

expectations prior to the oil price

surge in the first quarter of 1999

fairly low and stable, but inflation

expectations were falling as well.

A similar pattern was seen before

the Gulf War, though expected

inflation was somewhat higher

and modestly more volatile. In

contrast, prior to the 1973 and

1979 shocks, expected inflation

was not only higher than in

1999, but it was also rising.

become more efficient, they earn ahigher rate of return: Real wages oflabor increase, as do the returns to capi-tal. Invariably, these efficiency gainsboost the nation's living standards.Since 1995, per capita real GDP hasgrown at a 3.2 percent annual rate, farexceeding the 1.8 percent rate of gainfrom 1973 to 1995. In fact, per capitaoutput growth during the past five yearssurpasses even the heyday of strongproductivity growth from 1947 to 1973,when living standards grew at a 2.4 per-cent annual rate.

Most forecasters and many academicresearchers seem to believe that thisimprovement is permanent. They haveconcluded that the U.S. economy cannow grow much faster on a sustainedbasis without sparking an accelerationin inflation—the potential rate of out-put. For example, in a survey of recentresearch on this topic, the consensuswas that gains in labor productivitystemming from the technological inno-vations and other advancements haveboosted the economy's potential rate ofgrowth by about 1 percentage pointduring the past five years.6 As a result,compared with the period of slow

growth that existed from 1973 to 1995,when most economists figured thatpotential output growth was around 2to 2.5 percent, many economists nowbelieve that the economy can grow byaround 3.5 percent per year.

The primary reason for this producti-vity upswing has been the proliferationof computers, advances in telecommu-nications equipment and the Internet,all of which have dramatically loweredthe cost of information-gathering andretooled production and distributionprocesses. These innovations are embod-ied in the equipment, software andbusiness practices of firms, enabling agreater amount goods and services tobe produced and distributed with pro-portionately fewer inputs.

One obvious example is in the pro-duction of crude oil. Innovations suchas horizontal or directional oil drillingtechniques, and 3-D and 4-D seismicimaging, today allow energy producersto lift oil from the ground in areas pre-viously considered inaccessible orinfeasible. Other innovations includethe development of deep-sea drillingand floating production platforms.Soon, the maximum feasible depth ofuseful production is expected to drop to7,500 feet, more than triple what it wasin the late 1980s.

In the parlance of economists, theeconomy's production possibilitiesfrontier has been shifting outward fromthe origin. The ability to produce pro-portionately more output, accordingly,has enabled firms to keep labor andnonlabor costs in check. At a timewhen margins are being squeezed byhigher labor costs and, more recently,by the sharp increase in energy prices,firms have been able to post significantprofits, which have, accordingly, boost-ed equity prices and engendered othereconomy-wide benefits.7

Explanation #2: The U.S. Economyis More Energy Efficient

Since energy is an important inputinto the production of goods and ser-vices, efficiency gains arising fromtechnological innovations have neces-sarily meant that, over time, the UnitedStates uses less energy per unit of out-put. For example, since the early1970s, the average efficiency of a newrefrigerator has nearly tripled, whilethe average fuel economy of vehiclessold in the United States has doubled.Further improvements are on the hori-zon. The inevitable switch from thecurrent 14-volt electrical system inautomobiles to the more powerful42-volt system is expected to boostfuture fuel efficiency by as much as10 percent.

[8]

Whether bushels of corn or wheat,tons of steel, or millions of board feet oflumber, the energy component of U.S.output has been dwindling over time.In 1973, for example, a little more than18,000 BTUs (British Thermal Units)were required to produce $1 of realGDP in the United States (measured in1996 dollars). By 1983, the number ofBTUs required to produce $1 of realGDP was cut by nearly one-fourth toabout 13,750. By 1999, only about10,500 BTUs were needed to produce$1 of real GDP—more than 40 percentbelow the 1973 requirements.8

From a competitive standpoint,firms and industries that have mostsuccessfully adopted energy-efficientproduction processes should have anadvantage. One way to see howchanges in energy efficiency affect eco-nomic performance is by consideringenergy usage by state. In general, statesthat have above-average increases inenergy-efficiency gains—measured asthe growth of gross state product rela-tive to total energy consumption from1977 to 1997—have also experiencedfaster growth in employment. Con-versely, states whose industries haveexperienced below-average increasesin energy efficiency have had modestlyslower employment growth.

Explanation #3: Better Monetary PolicyThe price system plays an important

role in a market economy like theUnited States. Scarce resources tend toflow to those areas where after-taxreturns are the highest (or opportunitycosts are the smallest). Crucial in thisregard is a low and stable inflationrate—because it helps to minimize riskand uncertainty about future real returns.Indeed, the contours of our record-setting business expansion suggest thatprice stability has been a key ingredientin firms'decisions to add to their pro-ductivity-enhancing stock of capitalgoods. Firms will thus be less apt toalter investment plans when the believethat rampant inflation will not be theultimate outcome of an oil price shock.

A key lesson learned by policy-makers during the 1970s was that theamount of money supplied in responseto an oil price shock will eventuallydetermine the course of inflation. If themoney supply increases relative to thefewer amount of goods and servicesproduced after the shock, then the rela-tive value of money will fall (too muchmoney chasing too few goods). If thiscontinues, the Fed runs the risk of fuel-ing even more inflation, possibly leadingpeople to revise upward their expecta-tion of future inflation. With higheractual inflation and expectations for

future inflation now greater than before,the degree of action required by theFed to bring inflation under control—as the early 1980s demonstrated—willbe stronger than if it had acted vigilantlyin the first place.

As the figure at left shows, monetarypolicy-makers failed to quell the higherinflation expectations that followed the1973 and 1979 oil price shocks—that is,they were too accommodative. In theaftermath of those events, expectationsof higher inflation were built intoplanned adjustments in wages andprices, further eroding the purchas-ing power of the consumer's dollar.Although expected inflation hasclimbed noticeably since the first quar-ter of 1999, markets appear confidentthat the Federal Reserve will not letinflation expectations and, hence, actualinflation, get out of hand—somethingthey appeared to do successfully follow-ing the ephemeral 1990 oil price shock.In short, current monetary policy perfor-mance has been much better than itwas in prior episodes.

Are Oil Price ShocksNow Off Our Backs?

Oil price shocks have periodicallydotted the post-World War II economiclandscape. Thus far, however, the severeeconomic disruption and dislocationthat usually followed the spike in oilprices during the 1970s and early 1980shas not come to pass. Indeed, the cur-rent outlook for the U.S. economy sug-gests that oil-induced turbulence mightbe a relic of the past. The torrent ofinvestment in information and commu-nications technology equipment inrecent years appears to have led to sig-nificant structural improvements in U.S.labor productivity and energy efficiency,allowing firms to offset higher energyand non-energy costs. But perhapsmore important, since the Great Infla-tion of the 1970s and early 1980s, theFederal Reserve has implemented, andlargely achieved, a policy designed tomaintain price stability. As long asthese factors persist, current forecastscalling for continued solid growth andmoderate inflation may turn out to beright for a change.

Kevin L. Kliesen is an economist at the FederalReserve Bank of St. Louis. Thomas A. Pollmannprovided research assistance.

Tiic Regional Economist • January 2001

ENDNOTES

1 The supply of oil tends to be fixed inthe short term, since it takes time toexplore for new reserves or tap pro-ven reserves. In economist lingo, thesupply curve for oil is inelastic in theshort run. Accordingly, an increaseor decrease in the demand for oil willengender a disproportionate responseon the price side.

2 Real GDP grew at about a 4.25 per-cent annual rate through the firstthree quarters of 1998; CPI inflationaveraged 1.5 percent during thesame period.

3 Blue Chip Economic Indicators datedDecember 10,1998, and January 10,1999, respectively.

4 The Experimental Recession Indexdeveloped by Harvard ProfessorJames Stock and Princeton Univer-sity Professor Mark Watson, whichattempts to predict turning points inthe business expansion, showed thatas of November 2000, the probabilityof the U.S. economy being in reces-sion in May 2001 was only 7 percent.The report can be obtained athttp://ksghome.harvard.edu/~JStock.Academic.Ksg/xri/INDEX.HTM.

5 In recent years, the "average" fore-caster has tended to underpredictthe strength of U.S. economic activity(real GDP growth and unemploy-ment rates) and overpredict the run-up in inflation. See Kliesen (2000).

6 International Monetary Fund (2000),pp. 48-52.

7 For example, realized capital gainshave boosted consumer incomes andthus spending (the wealth effect); inaddition, increases in realized capitalgains have caused individual tax pay-ments to the Treasury to skyrocket,helping to produce record federalbudget surpluses.

8 U.S. energy use per unit of real GDPdeclined at annual rates of 2.7 per-cent from 1973 to 1983; 1.4 percenta year from 1983 to 1995, and2.6 percent a year from 1995 to 1999.

REFERENCESAuerbach, Alan J., and Laurence J.

Kotlikoff. Macroeconomics: AnIntegrated Approach (Cambridge,Mass.: The MIT Press, 1998).

Blue Chip Economic Indicators. AspenPublishers, Inc.

Hamilton, James D. "Oil and theMacroeconomy Since World War II,"Journal of Political Economy (April1983), pp. 228-48.

International Monetary Fund. WorldEconomic Outlook, Washington, D.C.,October 2000.

Kliesen, Kevin L. "The Economic Out-look for 2000: Bulls on Parade?"Federal Reserve Bank of St. LouisNational Economic Trends (January2000).

Rasche, Robert H., and John A. Tatom."The Effects of the New EnergyRegime on Economic Capacity,Production and Prices," Review,Federal Reserve Bank of St. Louis(May 1977), pp. 2-12.

. "Energy Price Shocks,Aggregate Supply, and MonetaryPolicy: The Theory and the Inter-national Evidence," Supply Shocks,Incentives and National Wealth,Carnegie-Rochester Conference Serieson Public Policy, volume 14, KarlBrunner and Allan H. Meltzer, eds.,Amsterdam: North-Holland (Spring1981), pp. 9-94.

[9]