Dean Baker and Jared Bernstein. Getting Back to Full Employment: A Better Bargain for Working People. Washington, DC: Center for Economic and Policy Research, 2013.
In this short but cogent book, Baker and Bernstein make their case for focusing public policy more on the goal of achieving full employment. That goal seems strangely out of fashion, considering how high unemployment has been since the financial crisis of 2008. Reducing the federal deficit and holding down inflation are the goals that preoccupy policymakers, but the authors see that as economically counterproductive.
As economists use the term, “full employment” doesn’t mean that everyone who would like a job is actually working. The labor force always includes some people who are currently between jobs (the “frictionally” unemployed) and some who lack the ability or skill for the existing jobs (the “structurally” unemployed). An economy does not have to employ those groups to be considered at full employment, but it does have to have enough jobs to eliminate “cyclical unemployment,” unemployment resulting from a weak demand for labor in general. While structural unemployment is a problem of labor supply, requiring improvements in the qualifications of workers, cyclical unemployment is a problem of labor demand, potentially responsive to stimulatory fiscal and monetary policy. To the degree that government can stimulate demand through such measures as cutting interest rates or increasing public spending, it can reduce cyclical unemployment and move the economy toward full employment.
If, however, the economy is already near full employment, and most of the remaining unemployment is structural, then policies intended to promote additional employment could mainly generate inflation instead. If the job-seekers are mostly underqualified for existing jobs, then employers who wish to hire must either bid up the price of the qualified workers or pay the unqualified more than the value of what they can actually produce. Either puts upward pressure on wages and/or prices. Theoretically, there is a rate of unemployment at which cyclical–but not structural–unemployment is absent, and below which unemployment cannot go without increasing the rate of inflation. Economists call that the “non-accelerating inflation rate of unemployment,” or NAIRU. It is the theoretical sweet spot of “full employment and stable inflation.”
In practice, economists have trouble agreeing on the NAIRU rate, or on the inflationary cost of trying to get unemployment even lower once that rate has been reached. Baker and Bernstein are concerned that an excessive fear of hitting an inflation threshold makes policymakers too timid about fighting today’s high unemployment, unemployment at a rate they consider well above NAIRU.
To acknowledge this relationship between low unemployment and price pressure is common sense. But there is a huge difference between acknowledging the relationship and believing that public policy must avoid full employment because it will cause inflation, or that it must tolerate a cruelly high level of unemployment simply to avoid a slight risk of inflation.
In the early 1990s, most economists thought that unemployment couldn’t go below 6% without triggering more inflation, but it actually came down to 4% by the end of the decade, while wages remained stable. Today, the Congressional Budget Office estimates NAIRU at 5.5%, and a few economists regard unemployment rates over 7% as mostly structural rather than cyclical. That view blames unemployment primarily on the poor qualifications of the workers and discourages efforts to stimulate economic demand by warning of inflationary consequences.
Baker and Bernstein disagree. They reason that if employers generally were having trouble finding qualified workers, they would be offering higher pay or lengthening the workweek for existing workers. That’s true for a few jobs, but not for the economy generally. The authors also point out that unemployment that appears structural could actually be cyclical, since employers are much more likely to upgrade worker qualifications through training when the demand for labor is high.
I would add that a lack of high-paying jobs has a structural dimension, since it is at least partly due to a lack of education or skills in a particular population. But as Arne Kalleberg argues in Good Jobs, Bad Jobs, it also reflects how employers choose to organize work. Some companies deliberately create jobs that can be performed by cheap, low-skill labor, and such jobs have proliferated in our “post-industrial” economy, especially in service industries. When labor demand is high, skill requirements are not as significant an obstacle to employment as all the talk about hi-tech industries would suggest.
Believing as they do that today’s very high unemployment is mostly cyclical, Baker and Bernstein argue that the benefits of reducing it far outweigh any inflationary costs of doing so. “If the unemployment rate could in fact fall to 4.0 percent, and possibly lower, without leading to accelerating inflation, then the result of a policy that kept it higher would be the needless unemployment of millions of workers and lower wages for tens of millions.”
When the demand for labor is low, as it has been since the financial crisis, that creates unemployment for some and low wages for many more. Workers are in a much stronger bargaining position when labor markets are tight. Long-term unemployment has negative effects on earnings that last for many years beyond the actual period of unemployment. Since these effects are greatest for workers at the low end of the income scale to begin with, slack labor markets tend to widen the income gap between rich and poor.
Moreover, the damage of job loss extends beyond earnings and hours worked, as job losers have been found more likely to experience a number of noneconomic negative impacts, including increased rates of stroke and heart attack, higher rates of divorce, lower rates of home ownership, and even lower life expectancy. Generational effects have also been found as the children of parents facing long-term unemployment are more likely to have lower test scores and reduced earnings as adults than similarly placed children whose parents avoid long jobless spells.
Beyond the effects on workers and their families, economic downturns that drag on needlessly have huge costs for the entire country through lost productivity. Since 2008, United States has produced $6 trillion less than the CBO projected it would before the recession. “This is a large cost that dwarfs the estimates of the losses associated with modestly higher rates of inflation.” Unemployment also affects the federal budget deficit, since it reduces tax revenue and increases the number of people qualifying for government assistance. For example, workers with chronic health problems who could find jobs when labor demand is high have been applying for disability benefits since demand declined.
Baker and Bernstein see the costs of inflation as much smaller than all that. Even if stimulatory economic policies were to push the unemployment rate a full percentage point below NAIRU–and they believe that it is well above that point now–the research indicates that the rise in inflation would only be in the range of 0.3 to 0.5 percent. The costs in terms of economic growth would be less than those of persistently high unemployment. Although some studies have found a negative impact of inflation on economic growth, “the size of the estimated impact of inflation on growth in the studies that found an effect is certainly well within the range that could be explained by measurement error.” (Here they discuss a particular type of measurement error that plagues these studies. If, as many economists believe, conventional measures often overstate inflation, such overstatements would both increase measured inflation and decrease measured GDP growth, since real growth is nominal growth minus inflation. Such cases could skew the results so that the same countries appeared to have both higher inflation and lower real growth.)
A modest amount of inflation can even help with some policy issues. Suppose the Federal Reserve reduces the federal funds rate to 2%, in order to stimulate the economy by making borrowing more affordable. If inflation is 1%, the real federal funds rate (nominal rate minus inflation) also becomes 1%. But a higher inflation rate can bring the real rate down to 0% or even below zero, making borrowing really pay.
The authors are especially critical of central bankers who make low inflation their overriding policy objective, especially when the central bank has a legal mandate to pursue both price stability and high employment, as it does in the United States. They summarize:
The evidence used to support inflation-targeting policy is dubious. If the general public and even most politicians fully understood the costs and risks associated with the inflation policy pursued by central banks, few would agree that it is appropriate to keep millions out of work and deny wage growth to tens of millions simply to reduce the risk of modestly higher inflation.
While Baker and Bernstein focus on the economics rather than the politics of unemployment, one may observe how easily full employment gets relegated to the back burner. Although high unemployment hurts the entire economy, the costs fall disproportionately on the have-nots, the workers whose labor is in least demand. Inflation, on the other hand, erodes the value of whatever wealth and income people already have. In a society with extreme inequality and minimal restrictions on political spending, the haves can use their influence to create a systemic bias toward fighting inflation instead of unemployment. Prevailing policies may not be optimal for economic growth, but they help perpetuate the unequal distribution of the benefits that flow from whatever growth occurs. Much of the resulting unemployment may not be structural in the economic sense, that is, resulting from the workers’ inability to do today’s jobs. But it could be structural in a more political sense, resulting from the weak position in the power structure of workers in general and unemployed workers in particular. That would explain how unemployment could be technically cyclical as Baker and Bernstein argue, and yet so persistent.