December 16, 2012
Unemployment, Inflation, and the Fed
Unemployment, Inflation, and the Fed—Posner
As Becker points out, the unemployment rate is a misleading figure because it ignores discouraged workers, who have left the labor force (the denominator in calculating the unemployment rate, where the numerator is the number of full-time employed), and workers involuntarily working part time. But that is to say that the Fed's choice of 6.5 percent unemployment as the level below which the Fed will stop flooding the economy with money is conservative, in the sense that at that rate the underutilization of the labor force is considerable and warrants extraordinary measures. Similarly, an inflation rate of 2.5 percent is low, suggesting that the Fed will stop flooding the economy with money well before inflation reaches a dangerous level.
I am not a macroeconomist, but my sense is that macroeconomics is so complex, in the sense of involving so many interactive forces, that no one understands it fully. For example, it may seem obvious that "hand outs," such as unemployment insurance, food stamps, and Medicaid, increase unemployment by reducing the financial pressure on an unemployed person to seek work. On this ground, Casey Mulligan, whom Becker cites, has argued that the hand-outs have prevented poverty from increasing during the current depression, and that this is a bad thing because poverty makes unemployed people search harder for jobs. The other side of this coin, however, is that the hand-outs increase the amount of money that people have and can spend on consumption, and consumption drives production, which in turn drives employment. If there are no jobs to be had, searching for a job has no payoff. Of course there may be methods of increasing consumer incomes that do not involve making work less attractive, such as Keynesian public projects that increase employment directly, or income tax reductions. Those are fiscal measures impeded by the dysfunctional character of our political system, from which the Federal Reserve is largely immune; and so the burden of speeding the economy's recovery has fallen on the Fed.
The Federal Reserve's easy money policy, adopted expressly to speed recovery, keeps interest rates very low, which reduces both the burden of debt and the cost of borrowing and therefore stimulates consumption and therefore production and so employment. Inflation, which is stimulated by an increase in the ratio of money to goods, also reduces the cost of debt. As debt falls, people save less and spend more.
It makes sense, therefore, for the Federal Reserve to increase the supply of money when employment (however measured) is abnormally high and inflation very low, which is what it is doing. And by not only doing these things but committing to continue doing them (until the 6.5 percent and 2.5 percent trigger points are reached), which is a departure from the traditional taciturnity of central bankers, the Federal Reserve is reducing uncertainty. This is important because economic uncertainty tends to impede action, whether it is hiring people or buying consumer goods. Freezing is a common and rational reaction to uncertainty—one freezes in the hope that the uncertainty will soon dissipate and then one can act with greater certitude about the consequences of one's actions.
The Federal Reserve expands the money supply basically by buying government securities for cash. Should inflation loom, the Fed would try to reduce it by selling government securities for cash, and retiring the cash it received, thus reducing the money supply. The danger is that by reducing the money supply and thus forcing up interest rates, the Fed's action in response to inflation could cause a recession. But today the continued weakness of the economy, given the large deficit, seems a more urgent concern, justifying the Fed's action.
The Unemployment Rate and Central Bank Policy-Becker
Low inflation and "full" employment have been statutory goals of the Federal Reserve for the past 35 years. Often, however, inflation received the most attention, as when former Fed chairman Paul Volcker in the early 1980s sharply raised interest rates and put the economy in recession in order to wring inflationary expectations out of the system.
On December 12th, Ben Bernanke, the chairman of the Fed, indicated that the Fed would pursue what one might think is simply a variant of the full employment target by keeping nominal interest rates close to zero until the US unemployment rate dips below 6.5%-it is currently 7.7%- or until inflation is forecast to exceed 2.5%. The challenge facing this proposal is that while an unemployment rate target may seem to be just the flip side of the full employment target, unemployment can be nudged by other government policies in ways that have little effect on employment.
The present high level of unemployment in the US in good measure reflects the slow rate of recovery of real GDP and employment from its recession levels. According to "Okun's Law", the recovery in employment from a recession is simply related to the recovery in real GDP (see the discussion of Okun's Law in my blog post on 11/4/2012). Okun's Law implies that a central bank can use the recovery in real GDP as a proxy for the recovery in employment toward a full employment goal.
If the change in the unemployment rate were simply the mirror image of the change in the employment rate, "low" unemployment would be as good a target as full employment. That would imply that Okun's Law could be stated as a relation between the decline in unemployment from recession levels and the recovery in real GDP from recession levels. Then the recovery in unemployment to its "full employment" level could meet the Fed's goal of maintaining full employment (along with controlling inflation), as in Bernanke's recent announced goal of no more than 6.5% unemployment.
The complication is that changes in unemployment rates during business cycles are not just mirror images of changes in employment rates. This has been especially the case during the Great Recession. By definition, the unemployed equals the difference between the number of persons in the labor force and the number of persons working. The unemployment rate is then defined as the number of individuals who are unemployed as a fraction of the labor force. It follows from the definition of unemployment that the unemployment rate equals one minus the employment rate (the ratio of the number of persons employed to the number of persons in the labor force). This relation shows that changes in the unemployment rate would be equal to but opposite in sign to changes in the employment rate only as long as the labor force remained fixed.
During business cycles, the employment rate and the unemployment rate do move in opposite directions, but the relation is far from one to one, especially during severe recessions. The reason is that the labor force also changes over the course of a business cycle. Especially during severe recessions, some workers get discouraged about finding jobs and leave the labor force. This would tend artificially to reduce the unemployment rate even when both employment and unemployment did not change. This is why the official unemployment rate is usually supplemented with measures of the "total" unemployment rate that include both individuals who got discouraged and withdraw from the labor force, as well as those working part time because they could not find full time jobs. This total unemployment rate now stands at 14.4%, much above the 7.7% official rate.
The official unemployment rate is also affected by public policies that encourage individuals to either leave the labor force or work only part time. These are partly policies that increase eligibility for means tested welfare programs, such as food stamps, subsidies on mortgage payments, and Medicaid benefits. These and other means-tested programs greatly expanded during the past four years (for a good discussion of the evidence on this, see Casey Mulligan's recent book "The Redistribution Recession: How Labor Market Distortions Contracted the Economy").
The unemployment rate is also affected by policies that affect eligibility for unemployment compensation, such as the extension of unemployment benefits during this recession to 99 weeks. Such an extension increases unemployment because it encourages individuals to become or remain unemployed in order to collect unemployment benefits for a longer time. The net effect of extensions in unemployment benefits is to increase the unemployment rate differently from any decline in the employment rate.
This discussion implies that a shift by the Fed from a full employment target to an unemployment target could lead to some shift in the Fed's goals. In addition, and most important, some of the causes of changes in the unemployment rate are not amenable to manipulation by the Fed through very low interest rates and the purchase of bonds or other assets. Unemployment rates will remain higher than in the past, regardless of Fed policies, as long as the extensions of various means-tested programs and of unemployment benefits during the past few years remain in effect.
A major risk of trying to implement an unemployment target through present Fed policies is that the inflation rate could increase in a futile attempt to bring down further the unemployment rate to a targeted rate, as happened in the 1970s. To its credit, the Fed protected against this possibility by setting its target at a relatively high unemployment rate of 6.5%, even though the rate prior to the onset of the recession in 2007 was well under 5%. The Fed also directly faced the risk of creating excessive inflation by setting its target as no more than 6.5% unemployment only as long as the inflation rate does not rise about 2.5%, a modest rate of inflation.