Goldman On Why The Fed Can't Have Its Low Unemployment, And Eat Cheap Oil Too

Tyler Durden's picture

Jan Hatzius' economic team finally comes out with a report that bears presenting as it aptly discloses one of the core conundrums facing the Fed: you can have low unemployment (eventually... courtesy of many years of ZIRP and QE in an environment of "fiscal adjustment", or Goldman's term for Congressional "austerity"), or you can have low gas prices. But you can't have both. To wit: "The combination of tight energy markets and high unemployment poses a dilemma for monetary policy. If policy is kept easy to boost growth, unemployment will decline but the oil market is at risk of overheating. But if policy is tightened to confront the pressure from higher oil prices on (headline) inflation, unemployment is likely to remain  far above desirable levels for a long time to come." And while the price of gas can be (very briefly) controlled by various volatility enhancing margin moves by the exchanges (which for those confused are nothing but self-reinforcing loops - increased vol leading to a margin hike, leads to more vol, leading to more margin hikes, etc). Too bad the CME can't just lower margin on unemployment to -100%. But it can't. Which is why very soon the Fed will be forced to admit to the whole world that "ultimately, a return to equilibrium in both the oil and labor market is likely to require an increase in the real price of oil. In theory, policymakers could react to this by  targeting either a combination of temporarily higher headline inflation with stable core inflation, or stable headline with lower—and in an extreme case negative—core inflation."  And here Goldman throws a stunner: when debating the implications for fiscal policy (we all know what monetary policy will look like: QE 3 through N), the firm proposes the following: "one complement to a low interest rate policy could be a higher energy tax. If one believes that higher real energy prices will be needed in coming years, an energy tax would promote that shift and also capture some of the surplus that would otherwise have gone to foreign producers." Is the government about to unleash some EPS destruction in the E&P and refining space? It appears Goldman has already given the green light which is really all it takes.

Full Hatzius/Tilton note:

Monetary Policy When Oil Is Scarce

  • The combination of tight energy markets and high unemployment poses a dilemma for monetary policy. If policy is kept easy to boost growth, unemployment will decline but the oil market is at risk of overheating. But if policy is tightened to confront the pressure from higher oil prices on (headline) inflation, unemployment is likely to remain far above desirable levels for a long time to come.
  • Ultimately, a return to equilibrium in both the oil and labor market is likely to require an increase in the real price of oil. In theory, policymakers could react to this by targeting either a combination of temporarily higher headline inflation with stable core inflation, or stable headline with lower—and in an extreme case negative—core inflation.
  • The higher headline/stable core combination is likely to be much less painful for economic activity. Moreover, if long-term inflation expectations remain stable, the “sacrifice ratio”—the ultimate output loss needed to bring inflation back down—after real oil prices have reached equilibrium levels is likely to be negligible. Should inflation become embedded in higher inflation expectations, the sacrifice ratio would increase and tighter monetary policy may become unavoidable—but we see no evidence for this so far.
  • One common objection against a low interest rate policy in response to a rising oil price is that it will shift the real income distribution from lower- to higher-income households. However, the employment opportunities of lower-income workers are far more sensitive to overall labor market  performance, which is likely to be significantly stronger under a low interest rate policy.

I. Monetary Policy When Oil Is Scarce

Crude oil prices have eased over the past few weeks, as our commodity strategy team had anticipated. The July Brent crude future stands at $110/barrel as of this writing, down $15 from the late April peak. Likewise, gasoline prices have fallen about 40c/gallon.

But that does not mean that the threat from higher oil prices to economic activity and inflation is behind us. For one thing, prices remain significantly higher than they were six months earlier, and our analysis still suggests a negative impact on growth in the next couple of quarters even if prices stay near current levels. Moreover, the fundamental story of increased oil scarcity is unchanged, and our commodity strategists now see distinct upside risks to their current forecast of $120/barrel for Brent crude by late 2012. So the impact of scarcer oil and higher oil prices on
economic activity remains at the top of our list of worries.

Oil Scarcity Complicates Policy

We have argued repeatedly that there is still ample slack in the world economy. Despite rapid growth that has brought output near potential in many emerging markets, more sluggish recoveries in developed economies have kept the global output gap in the neighborhood of 3%.

Nowhere is slack more visible than in the US labor market. At present, the US unemployment rate stands at 9%, the highest level since the early  1980s other than the last couple of years. Moreover, the evidence suggests that most of the increase in the unemployment is due to cyclical  weakness in the economy rather than “mismatch” between the available workers and the available jobs.

At the aggregate level, the US capital stock also seems relatively ample. Vacancies in the residential and commercial real estate sector are high, and utilization in most of the service sector also remains below average. Exhibit 1 illustrates a series of capacity measures in different sectors of the economy, with the right-hand column expressing these in a normalized format (standard deviations from average).

However, capacity in some areas is a lot more strained. Most significantly, the global oil market is very tight. Exhibit 2 shows the estimates from  our commodity strategy team on the actual production and production capacity of crude petroleum worldwide. They estimate that effective spare  capacity will be exhausted at some point in 2012 (depending to a large extent on the availability of Libyan supply).

Exhibit 3 shows how unusual  the current combination of low (US) labor utilization and high (global) oil utilization is by historical standards. For comparison purposes, we express the US labor gap and effective spare capacity in oil production in standard deviations relative to their historical averages. The US labor gap is two standard deviations below the historical average while oil capacity utilization is one standard deviation above the historical average.

This stark contrast in resource utilization between different factors of production is problematic from the perspective of US policymakers. If policy is kept easy and this translates into above-trend growth, the labor gap will shrink but the oil market will overheat. This could put severe upward pressure on (headline) inflation and therefore downward pressure on real income, especially among lower- and middle-income households. But if policy is tightened to confront the inflationary pressure, unemployment could remain far above desirable levels for a long time to come.

The low rate of growth in oil supply is likely to exacerbate this tension in coming years. Our commodity strategists expect oil production capacity to grow only about 1% per year (see Exhibit 4). Meanwhile, the growth rate of “oil productivity”— dollars of real GDP per barrel of oil—has averaged about 2% over the long term (Exhibit 5). If these trends continue, then the supply of oil can support a global growth pace of only about 3% without generating upward pressure on oil prices. For global growth of better than 4%—consistent with our forecasts and the “consensus” outlook—oil productivity and/or oil production need to accelerate. The most plausible reason for either to occur would be an increase in real oil prices, as the charts make clear.

 

A “Crude” Framework for Policy Tradeoffs

We illustrate US policymakers’ dilemma schematically in Exhibit 6, which shows the equilibrium—defined as a normal level of capacity utilization—in the labor and oil markets as a function of the price of oil and the real interest rate.

Schedule L denotes equilibrium in the labor market, i.e. full employment. It slopes downward because a higher real interest rate needs to be offset by a decline in oil prices for overall economic activity to stay unchanged, and thus for full employment to be maintained. Points to the right of  schedule L indicate a slack labor market (because of too-tight monetary policy and/or too-high oil prices), while points to the left indicate a tight labor market.

Schedule O denotes equilibrium in the oil market. It also slopes downward, again because a higher real  interest rate needs to be offset by a decline in oil prices for full oil utilization to be maintained. Crucially, however, the slope of schedule O is steeper  than that of schedule L. In other words, a given increase in oil prices has a larger effect on oil utilization than on labor utilization. The reason is that higher oil prices not only lower the overall level of economic activity but also induce some degree of substitution away from oil use and toward other factors of production.

Point A approximates the current situation. The oil market is in (uneasy) equilibrium but the labor market is clearly well below full employment.  Labor market weakness gives the Fed an incentive to lower real interest rates—e.g., by keeping the funds rate lower for longer than generally expected or, more radically, by returning to quantitative easing. Over time, schedule O is likely to shift up to O’ if our commodity strategists’ view of the supply trend is correct. This increases the equilibrium oil price and, at a given real interest rate, the labor shortfall (point B). An easy monetary policy could help bring the labor market back to equilibrium over time (point C) but at the cost of even more upward pressure on oil  prices and headline inflation. So what should policymakers do?

Core Deflation Is a Very Painful Way of Raising Real Oil Prices

Ultimately, the only way to bring both the labor market and the oil market into equilibrium is likely to be through a further increase in the real oil price. This would presumably increase oil supply by making exploration and production more attractive, and reduce oil demand by increasing energy efficiency.

Real oil prices can increase in one of two ways. Either the nominal price of oil rises, or the nominal price of everything else declines. Put differently, policymakers could react to oil scarcity either by accommodating higher headline inflation for a time while trying to keep core inflation on target, or focus on keeping headline inflation at target while pushing core inflation much lower. Neither path is particularly pleasant, since a relative increase in energy prices implies lower living standards for energy consumers. The question for policymakers is then how to minimize the costs of the adjustment in relative prices.

As long as long-term inflation expectations remain stable, there is a strong case for facilitating the adjustment via temporarily higher headline  inflation instead of lower core inflation. If inflation expectations are stable, the “sacrifice ratio”—the output cost of bringing inflation back down to  the target—after oil prices have reached their equilibrium level is likely to be quite low. In contrast, a policy that focused on keeping headline inflation at or below 2% would almost certainly be very painful. This is because of downward price rigidities that make it costly in terms of output losses to push core inflation or wage growth too close to—let alone below—zero.

If higher headline inflation gets embedded in longterm inflation expectations and higher nominal wage growth, however, the cost of a loose policy rises significantly. This is because the “sacrifice ratio” for bringing inflation back down again would then increase substantially.

Fortunately, there is little sign that the oil price increases to date have led to a serious increase in inflation expectations. This is shown in Exhibit 7, which plots two key measures of long-term inflation expectations: the 5-year forward breakeven inflation rate discounted in the Treasury bond market and the 4-9 year 1-year forward inflation expectation of consumers according to the University of Michigan consumer sentiment survey. Both measures are close to their averages over the past decade.

Our conclusion is close to the bottom line from our analysis using our version of the so-called Taylor rule. This says that in the anchored inflation
expectations regime of the past 25 years, increases in oil prices are more likely to lead to a cut in the funds rate—i.e. an attempt by Fed officials to loosen financial conditions—than to a hike.

Implications of Low Rates for the Income Distribution

One common objection against a low-interest rate policy in response to a rising oil price is that it will shift the real income distribution from lower-  to higher-income consumers [our emphasis]. And on the face of it, there is some evidence for this contention. According to the
Labor Department, the bottom 20% of the income distribution devote 4.6% of their total spending to gasoline purchases, against 3.5% for the top 20%. This implies that a 10% increase in gasoline prices cuts about 0.1% more from the real income of lower-income households than higher-income ones.

However, it is important to evaluate the effects of low rates not only via the impact on oil prices but also via the impact on the labor market. This is because the costs of keeping unemployment higher for longer—or in terms of Exhibit 6, staying at point A or B rather than moving to point C—disproportionately fall on lower-income workers. Exhibit 8 suggests that an extra 1 percentage point of unemployment among college graduates implies about 3 points of extra unemployment among those with less than a highschool education. Because their fortunes are so sensitive to overall labor market developments, we estimate that it takes a drop of only about 0.1 percentage point in the aggregate unemployment rate to offset the impact of a 10% rise in oil prices on the income distribution.

This suggests that the commonly heard argument that lower-income workers as a whole are hurt by a policy of keeping interest rates low in the  face of higher oil prices is somewhat misleading. With no change in jobs or wages, higher oil prices clearly do hurt real incomes—more so for the typical lower-income household. But insofar as lower rates bring about labor market improvement, this will lead to significant gains in real income for a subset of households drawn disproportionately from the lower end of the income distribution.

Implications of (and for) Fiscal Policy

Policymakers’ range of options for dealing with underemployment is limited not only by the tight oil market, but also by government finances, which are on an unsustainable path. The US requires a large fiscal adjustment which will impose a meaningful drag on growth in coming years and exacerbate the shortfall in labor demand.

In this context, one complement to a low interest rate policy could be a higher energy tax. If one believes that higher real energy prices will be  needed in coming years, an energy tax would promote that shift and also capture some of the surplus that would otherwise have gone to foreign producers. (In the first quarter, US household spending on energy goods, mostly gasoline, was $70bn higher than the 2010 average—a run rate of 0.5% of GDP.) Higher energy taxes could either be offset by reductions to other taxes, or used to help reduce the federal deficit. In any case, whether it occurs through energy taxes, other taxes, or spending cuts, fiscal adjustment will only reinforce the case for a long period of easy monetary policy.