London, 9 March 2011

Policymakers and markets are struggling to get to grips with chaos in Libya and what it means for oil prices and the economy. Recent statements by officials and commentators betray confusion and incoherent thinking. Nowhere was the confusion more evident than contrasting comments made yesterday by Richard Fisher and Dennis Lockhart -- respectively the presidents of the Federal Reserve Banks of Dallas and Atlanta.

Reflecting his stance as an inflation hawk and sceptic about the merits of a second round of quantitative easing (QE2), Fisher warned he was watching carefully to see how much of the increase in oil prices will be passed through to customers. He cautioned "the liquidity tanks are full, it not brimming over." The Fed has done all it can. Now it is up to private business to convert liquidity into investment and jobs:

"As a voting member of the FOMC this year, I have made clear within the meeting room and in public speeches that, barring some frightful development, I will vote against any program that might seek to extend or enlarge the substantial monetary accommodation we have already provided".

In contrast, Lockhart observed the Fed would have to respond by easing monetary policy further if rising oil prices threatened to push the U.S. economy back into recession. Lockhart argued the United States could absorb a moderate increase in oil prices over a relatively short period of time, and $120 per barrel would probably be manageable.

But a further rise in prices, though not his base case, could not be ruled out, and at $150 would create more serious problems.

MOVING INFLECTION POINTS

It has become fashionable to try to identify an "inflection point" at which rising prices will damage global growth. In 2010, when prices were ranging between $70 and $85, the International Energy Agency pegged the danger zone above $100 per barrel. But as prices have hit $100-105 (WTI) and $115-120 (Brent), first on strong demand and then on the Libyan disruption, observers have pushed the inflection point back to $120, $120-130 or even $150 per barrel.

There are several problems with this approach. Marketmakers, central banks and consumer-country governments have an obvious interest in downplaying the impact of rising oil prices on inflation, growth and profits. Stretching inflection points, like stretching of sustainable price/earnings ratios, may be an example of wishful forecasting.

In any case, the concept of inflection points makes little sense. Prices are not benign at $119 and then suddenly become a problem at $121. Each $1 increase is likely to have some impact on inflation, growth, profits and equity prices, depending on factors such as the impact on real incomes, margins, and petrodollar recycling.

The impact may not be not linear. But every dollar makes a difference. The fascination with picking ever-higher inflection points seems arbitrary and poorly supported by statistics.

TANGLED UP IN ELASTICITIES

Many forecasters argue the tipping point is higher than in the 1970s because of improvements in energy efficiency; the smaller share of energy consumption in GDP; the low sensitivity of oil demand to a change in prices especially in the developed economies (low price elasticity of demand); and its high sensitivity to rising income especially in emerging markets (high income elasticity of demand).

In this view, prices would need to rise a long way before there was significant demand destruction. Again this is bad mathematics. Price elasticity of demand measures the sensitivity of consumption to "small" changes in prices assuming everything else is held constant.

Sensitivities are not constant for large price changes. More important, large changes have an effect on income distribution so there are income effects as well. Once prices rise $20, $30 or even $80, the impact needs to be modelled in a dynamic framework rather than a static one, and the parameters become highly uncertain.

Even if we accept the conventional view there will be a limited consumption response, this is a double-edged sword. In a limited-demand response environment, the likelihood of a very large increase in prices is much higher. But if risk of a bigger spike is higher, Lockhart might need to assign more weight to it, and it makes it more likely that the Fed will have to ease further, perhaps preemptively.

It should be clear the inflection point/price sensitivity analysis is circular. If consumption is sensitive to prices and the economic impact is large, the danger zone is lower. There is little likelihood of a "super-spike" to $150, but the Fed could find itself having to act at just $115-120 per barrel.

If consumption and growth are not very sensitive, the danger zone may be pushed back to $150, but there is a much greater chance prices will test this level, and the Fed will eventually be forced to act anyway.

IS THIS A SUPPLY SHOCK?

Should the Fed intervene at all to offset the impact of oil prices on the economy, and should it respond by tightening policy to dampen inflationary pressure, or provide more stimulus to support jobs and growth?

Policymakers and market watchers are as usual divided about whether rising oil prices reflect cheap credit, excess demand and the lack of a global output gap; or manifest microeconomic supply-demand fundamentals, relative demand, and a negative supply shock that worsens the whole growth/inflation trade off.

The European Central Bank, the People's Bank of China, rate-setter Andrew Sentance at the Bank of England, and Thomas Hoenig and Richard Fisher at the Fed incline to the former view. Fed Chairman Ben Bernanke, Bank of England Governor Mervyn King and many other central bankers incline to the latter.

In fact there is evidence to support both. Prices were already rising steeply in Q4 and at the start of 2011, pointing to excess demand, while markets surged higher in response to the mounting unrest across North Africa and the Middle East, pointing to a supply shock.

Bernanke and King have been careful to argue soaring oil and commodity prices are a relative price change or unfavourable supply shock, to which monetary policy should not respond. But if rising prices are truly a supply shock, monetary policy has no role offsetting its impact on growth. Stimulating demand in response would simply worsen inflation without adding output or employment. Of course, if rising oil prices are partly due to strong demand and monetary expansion, launching a new round of quantitative easing would be an even worse mistake.

SPIKE OR PLATEAU?

Obviously, the full impact of prices depends how long they stay at any given level. Here, too, policymakers are confused. Some imply prices are likely to fall moderately once problems in Libya are resolved. Others view prices as having reached a plateau and are unlikely to rise further. A few dare wonder how much higher prices could soar if the loss of Libyan output proves permanent and demand continues to grow in line with forecasts at the start of the year.

In formulating a response, investors and policymakers (both those in charge of monetary policy and those deciding whether to release strategic stocks) need to answer three questions:

(1) Is the loss of Libyan output expected to be temporary or permanent? Temporary output losses would justify a stock release and possibly monetary accommodation. A permanent supply shock would render stock releases and accommodation futile, even counterproductive.

(2) How will rising oil prices impact on growth and consumption? If rising prices have little impact, releases are unjustified and monetary responses are both unnecessary and dangerously inflationary. If prices are expected to have a big impact, accommodation might seem more justified, but might succeed only in generating more oil-driven inflation rather than real growth and jobs.

(3) Is the surge in oil prices a supply shock or response to excess demand? If it is a supply shock, monetary policy easing cannot help and might make matters worse. If it is a sign of excess demand, accommodation would be a disaster, and the correct response is to begin gradually normalising credit conditions.

Ends --

The views expressed here are his own.