Has oil lost the capacity to shock?

David Walton1

Abstract. Recent high oil prices have neither triggered a recession nor causedinflation. This is in contrast to the `stagflation' of the 1970s. The dissimilaritycan be explained by increased labour market flexibility, central bankindependence and the role of monetary policy.

Despite a doubling in oil prices between the end of 2003 andthe summer of 2005, the UK economy appears, so far atleast, to have emerged relatively unscathed. QuarterlyGDP growth fell below its long-run average of 2ÃÙÄ per centa year during the second half of 2004, reaching a trough inearly 2005 and recovering gradually back to around trendat the end of last year. Consumer price inflation rose abovetheGovernment's 2 per cent target during the secondhalf of2005, reachingapeakof 2.5 per cent in September, but it hassince fallen back to around the target.

These developments are in marked contrast to theeconomy's performance after several previous oil priceshocks; periods that were often characterised by both risinginflation and recession. In the following, I investigate thereasons for this much more benign outcome and considerwhat we have learned from oil shocks, past and present, forthe conduct of monetary policy.

The size and nature of the shock

Part of the reason for the difference in economicperformance may relate to the size and nature of the shock.

· Including the most recent episode, there have been fivesignificant periods of rising oil prices since 1970: 1973^74,1978^79,1990,1999^2000 and 2004^05.The doubling in realsterling oil prices on this occasion, though comparable insize to other oil shocks, has taken much longer than usualto unfold.

· The economy is less dependent on oil than during the1970s, reducing the impact of any given oil price rise on theeconomy. Energy use was a little less than 1.5 per cent ofnon-oil gross final expenditure in 2003, down from a peakof 3.5 per cent in the early 1980s. The share of householdspending on fuels has also declined steadily over the pasttwenty years.

·Oil prices have risen as a consequence of strong globaldemand for oil rather than as a result of supply disruptionsassociated withwars.

The size and nature of the oil price shock are important.But so too are the mechanisms by which the shock getspropagated through the economy. Among other things, thiswill depend on the state of the economy at the time theshock hits, the extent of rigidities in the economy, themonetary policy framework and the monetary policyresponse. It is to each of these that I now want to turn.

The state of the economy at the time each oil shockhit

For some episodes, while the oil shock might haveexacerbated the economic downturn and helped to boostinflation, it is hard to argue that it was the fundamentalcause of either. For the first three episodes, there is plentyof evidence of excess demand in the economy prior to theoil shock that monetary policy was initially slow to respondto. According to the CBI Industrial Trends Survey, skilledlabour shortages went through the roof during 1973 andthey were well above normal in 1978^79 and 1988^89.Capacity utilisation was extremely high during 1988.Inflationwas ona rising trendprior to each of these oil priceshocks.

By contrast, statistical estimates of the output gap andsurvey measures of capacity utilisation and labour short-ages suggest that the economy has been operating fairlyclose to potential in recent years. Inflation expectations areless likely to be destabilised if an oil shock hits when theeconomy is operating close to normal capacity than if therewere already considerable inflationary pressure in thesystem.

Real wage rigidity hinders the economy's adjustmentto an oil shock

In the face of higher oil prices which raise firms' costs, thereal consumption wage (i.e. the post-tax wage paid toworkers deflated by consumer prices) must fall if firms are

DOI: 10.1111/j.1752-5209.2006.0004.x

Oxonomics 1 (2006) 9±12.ß 2006 The Author. Journal compilationß 2006 The Oxonomics SocietyPublished by Blackwell Publishing Ltd 9

1 DavidWaltonwasavisitingProfessoratOxfordUniversity'sDepartmentofEconomics, but sadlydiedaftera short illness on21June 2006.

tomaintain their profit share, andmaintain employment. Ifworkers resist the fall in the real consumption wage bybidding for higher nominal wages, thereby raising the realproduct wage in value added terms (ie the full cost of labourto firms dividedby the price firms get for their product), theend result will be lower employment. The decline inemployment will be reinforced if monetary policy istightened in response to these so-called second round effectsonwages.

The extent to which the real consumptionwagemust falldepends on the size of the oil change, the shares of oil andlabour in gross output and the degree of complementaritybetween factors of production. Estimates of the required fallin the real consumption wage to maintain employmentrange from1per cent to 2ÃÙÄ per cent.

There is evidence of realwage rigidity in the1970s, partlybecause of the unfortunate timing of two wage accords.

· At precisely the time that the impact of the first oilshock was working through, Edward Heath's Conservativegovernment introduced an incomes policy under whichwages would be permitted to rise in strict proportion toincreases in the cost of living above a specified threshold.Thiswas the first time that an incomes policy had containeda formal link between increases in pay and prices.

· In August 1979, the newly elected Conservativegovernment accepted the Clegg Commission recommenda-tion for staged pay increases for public sector employees ofup to 26 per cent. A further wedge was driven between thereal product wage and the real consumptionwage when thegovernment financed a significant reduction in income taxbya large increase inVAT, adding 3ÃÙÄ to 4 percentage pointsto retail price inflation. Both factors contributed to a rapidrise in whole economy average earnings growth whichpeaked at over 20 per cent in1980.

In the first three episodes, the level of the realconsumption wage was far higher than that consistent withkeeping the real product wage share of employees at itsmarket-clearing level, given trend movements in indirecttaxes and benefits, terms of trade changes and productivity.Not surprisingly, unemployment was on a rising trend. Bycontrast, there is no evidence of excessive real wages in themore recent episodes and unemployment has remainedbroadly stable.

The monetary framework matters

Is it a coincidence that the inflationary effects of oil priceshocks have been significantly smaller since the intro-duction of inflation targeting? There are several reasons tobelieve not.

First, and foremost, in the current monetary framework,the Monetary Policy Committee (MPC) will fight anydeviation in consumer price inflation from the 2 per cent

target that threatens to be persistent. If inflation moves bymore than one percentage point away from target, theGovernor is required to write an open letter to theChancellor explaining the reasons for this, the policyactions taken tobring inflationback to target and theperiodwithin which this is expected to occur. In a highlytransparent inflation targeting framework such as this, it ishard to see why inflation shocks should be highly persistent.

Second, if the monetary framework is credible, inflationexpectations are less likely to be dislodged in the event of acost shock. Real wage resistance might be reduced ifworkers realise that the MPC will react to any attempt byworkers to be compensated for the unavoidable loss ofspending power in the event of an oil shock. Both factorslower the likelihood of second round effects on wages,reducing the persistence of the inflation shock. There willalso be a smaller hit to output if the MPC does not have toraise interest rates to rein in second round effects.

Third, because the monetary framework recognises thatexcess demand is the key cause of persistent inflationarypressure, it is less likely that the economy will be running asignificant positive output gap when an oil shock hits.

How should monetary policy respond to an oilshock?

Under the Bank of England's remit, monetary policy has tostabilise consumer price inflation at 2 per cent while tryingto avoid unnecessary fluctuations in economic activity.Following any shock that moves inflation away from targetand output from its normal level, the MPC's job is to workout how best ^ notably over what time period ^ to bringinflation back to target without causing undesirablevolatility in output.TheMPC sets interest rates accordinglyand provides a justification for its actions in the publishedminutes of MPC meetings, and a more detailedexplanation, including projections for GDP growth andinflation, in the quarterly Inflation Report.

Higher oil prices pose a challenge for the MPC since, ashas been evident over the past year, they tend simulta-neously to push up inflation and depress economic activity.In essence, monetary policy needs to be sufficiently tight'toprevent the inflationary impulse from the rise in oil pricesbecoming entrenched through second round effects onwages and inflation expectations. At the same time,monetary policy needs to be sufficiently easy' to avoidunnecessary negative effects on demand and output.

Higher oil prices have a fairly immediate impact onconsumer price inflation working through higher petrolprices, heating bills and transport services. Whether theMPC needs to respond directly to this depends on thecredibility of the monetary policy framework. If inflationexpectations rise, nominal interest rates would need to riseby at least as much to maintain the stance of monetary

David Walton Has oil lost the capacity to shock?

10 Oxonomics 1 (2006) 9±12.ß 2006 The Author. Journal compilationß 2006 The Oxonomics Society

policy. But if firms and households believe that the MPCwill do whatever is necessary to keep inflation, on average,at 2 per cent, there is little obvious need for the MPC toreact to the first round impact of higher oil prices since thisis just a shift in relative prices. Encouragingly, inflationexpectations ^ whether measured from surveys, financialmarkets or pay settlements ^ have remained very stable overthe past two years.

Of course, monetary policy will only remain credible ifthe MPC consistently behaves in a manner that keepsinflation on target. An oil shock can have effects on supplyand demand in the economy, both in the short run and thelong run. It is the balance between supply and demand thatmatters most for inflation and, hence, interest rates.

There are several mechanisms by which demand mightbe reduced by higher oil prices, at least in the short term,without any help frommonetary policy:

·High oil prices act like a tax, transferring money fromconsumers to oil producers, leaving many of us worse off.Since the price elasticity of demand for energy is fairly low,households will need to direct a greater proportion of theirincome to energy-related items and cut back on discre-tionary spending. Over time, this will be offset to someextent by higher spending by oil producing countries onUKexports. Reflecting theUK's position as anoil producer,households will benefit from higher dividends by oilcompanies. The UK government is also a beneficiary ofhigher oil prices; higher oil-related taxes can be used to cuttaxes elsewhere or boost public spending, providing supportfor demand.

·A large increase in oil prices may generate uncertaintyboth about the future outlook for oil prices and the economyas awhole. If the future is uncertain, households may decideto postpone spending on consumer durables and businessesmay decide to postpone investment. Over time, though,high energy prices may act as a spur to spending onconsumer durables and investment, as households andcompanies invest in cars and capital equipment that useenergy more efficiently.

There are also several mechanisms by which potentialoutput might be reduced by higher oil prices.

· An increase in oil prices raises companies' costs and,hence, the prices at which they are willing to supply theirproducts. At higher prices therewill be less demand, leadingto lower output andemployment.This cyclical loss of outputcould become permanent if workers refused to accept alower real consumption wage and, instead, bid for highernominal wages to offset the rise in final goods prices.

·Higher oil pricesmaymake some of the existing capitalstock redundant or reduce the utilisation of capital. Thiswould reduce for a time the measured growth of total factorproductivity.

·Higher oil pricesmay lead to a reallocation of resourceswithin the economy.This may result in an under-utilisationof resources and higher unemployment during theadjustment process.

· If firms postpone investment decisions because ofincreased uncertainty, this will reduce temporarily thegrowth of the capital stock and the growth of potentialoutput.

The magnitude and timing of these demand and supplyeffects are very hard to determine. Nevertheless, they arelikely to be very important for the profile of the output gapand, hence, for inflation and interest rates. There is nomechanical formula that can be applied to tell the MPChow to adjust interest rates to deal with higher oil prices.The appropriate monetary policy response will depend onthe size and nature of the shock and how households andbusinesses react.There is awide range of possible outcomes.Demand could conceivably soften too much if business andconsumer confidence are damaged. Inflation expectationscould become destabilised if inflation moves too far awayfrom target. And some productive capacity could be lostpermanently. Each of these will affect the chances ofmeeting the inflation target and, hence, the appropriatelevel of interest rates.

Conclusions

I have offered a number of reasons why the economy seemsto have emerged relatively unscathed from a doubling in oilprices since the end of 2003. In summary:

·The size and nature of the shock havebeen different.Relative toprevious episodes, the shock has taken longer to unfold.Theeconomy is also less dependent on oil than during the1970s.And oil prices have risen as a consequence of strong globaldemand rather than as a result of supply disruptionsassociated withwars.

·The UKeconomy has been better placed to absorb the current oilprice shock.TheMPC's constant focus on inflationmeant thatthere were few inflationary pressures in the economy whenoil prices rose sharply, unlike on some previous occasionswhen there was clearly evidence of excess demand at thetime the oil shock hit.There has also been little sign of realwage resistance, no doubt reflecting structural changes thathave increased the flexibility of the labour market.This hasbeen in marked contrast to the 1970s when, on occasions,real wage resistance seemed unwittingly to have become anobjective of government policy.

· The monetary policy framework has played an important role.Inflation targeting has helped to anchor inflationexpectations, yet it has allowed theMPC to respond flexiblyto the oil shock.The stability of wage growthmeant that theMPC did not have to respond directly to the first roundeffects of higher oil prices on consumer price inflation.With

Has oil lost the capacity to shock? David Walton

Oxonomics 1 (2006) 9±12.ß 2006 The Author. Journal compilationß 2006 The Oxonomics Society 11

inflation expectations remaining stable, the MPC was ableto pay more attention initially to the negative consequencesfor demand: interest rates were cut 25bp in August 2005. Butthe Committee also needs towatch carefully for any signs ofan adverse impact on supply, particularly now that GDPgrowth appears to have returned to around its long-runaverage rate. At all times, the Committee's focus will be onachieving the appropriate balance between demand andsupply to keep consumer price inflation on track to hit thegovernment's target of 2 per cent.

David Walton was a Visiting Research Professor atOxford University, and joined the Monetary PolicyCommittee of the Bank of England in 2005. He startedhis career working as an economist at the Treasuryfrom 1984 to 1986, before joining the investment bankGoldman Sachs in 1987, where he became ChiefEuropean Economist. DavidWalton died after a shortillness on 21June 2006.

David Walton Has oil lost the capacity to shock?

12 Oxonomics 1 (2006) 9±12.ß 2006 The Author. Journal compilationß 2006 The Oxonomics Society