Why Fed Has Failed to Lower U.S. Unemployment
The Federal Reserve has a dual mandate to promote price stability and ensure full employment. It has yet to achieve either. It has set an annual inflation target of 2 percent, yet the consumer price index rose only 0.1 percent in May after falling 0.4 percent in April, and it was up only 1.4 percent in the 12 months ending in May.
The Fed’s preferred metric, the personal consumption expenditures price index excluding food and energy, rose just 1.1 percent in June from a year earlier.
The central bank hasn’t defined what it would consider full employment, even when it pledged to continue buying $85 billion in securities each month “until the outlook for the labor market has improved substantially.”
After record-low interest rates failed to induce banks to lend and creditworthy borrowers to borrow, the Fed and other major central banks embarked on a novel, and by their own admission, uncertain, course of stimulus known as quantitative easing, involving huge purchases of government and other securities. It had been tried by the Bank of Japan for years without notable success, but Western central banks have become increasingly desperate as they cast around for ways to create jobs.
Central banks can raise or lower short-term interest rates, and buy or sell securities. That’s it. Those actions are a long way from creating more jobs. In contrast, fiscal policy can be surgically precise, aiding the unemployed by extending and fattening unemployment benefits.
The Fed’s program to spark job-creation is a five-step process that is supposed to work like this: First, the central bank buys Treasuries or mortgage-related securities. Second, the sellers reinvest the proceeds in stocks, commodities, real estate and other assets, pushing up prices. Third, higher asset prices have a wealth effect by making their owners feel richer. Fourth, consumers and businesses spend on goods and services and capital equipment. Fifth, that spending spurs production and demand for labor.
Nevertheless, the Fed’s efforts haven’t worked very efficiently when it comes to the last three steps. The U.S. unemployment rate, at 7.6 percent in June, remains high by historical standards. Despite a recovery, payroll employment remains well below the previous peak in January 2008.
Furthermore, there have been virtually no follow-on effects from the Fed’s creation of member bank reserves. When the central bank buys securities, the proceeds clear through the seller’s bank and end up as additions to that bank’s reserves at the Fed. In normal times, those reserves are lent out by the bank with successive relending in the fractional reserve system, and each dollar in reserves created by the Fed turns into $70 of M2 money. But since August 2008, when the crisis started, the multiplier has been only $1.5. As a result, the unused cash has piled up in the Fed’s member bank accounts, and the excess reserves -- the difference between total and required reserves - - now exceed $1.9 trillion.
Fed Chairman Ben Bernanke continues to believe that more aggregate demand is all that is needed to put back to work the unemployed, underemployed and discouraged. About 1.65 million more people would have jobs if the unemployment rate dropped to the Fed’s trigger level of 6.5 percent. Even though the Fed hasn’t specified a full-employment rate, past discussions suggest it believes that 5 percent to 6 percent joblessness is possible without straining labor markets. A 5.5 percent rate would add an additional 1.6 million employees, for a total of 3.2 million.
Then there are those on the sidelines because of the huge decline since 2000 in the labor-participation rate, the percentage of the total population age 16 and above who are in the labor force, meaning they are employed or have looked for a job within the past four months. If the labor force participation rate were the same as in February 2000, the pool would be larger by 9.5 million people.
Because the unemployment rate in 2000 was 4.1 percent, this implies 9.11 million more jobholders. And at that rate, an additional 5.4 million people would be employed, for a total of 14.5 million. That’s a lot of workers to absorb, even with the annual real gross domestic product growth of about 3.5 percent that should resume in five years or so when global deleveraging is completed. And that pace of growth surpasses recent economic performance. Since deleveraging began in the fourth quarter of 2007, annual growth has averaged just 0.6 percent; it has been 2.1 percent since the recovery began in the second quarter of 2009, and 1.7 percent since the previous cyclical peak in the first quarter of 2001.
If real annual GDP growth averages about 3.5 percent in the years ahead and productivity growth averages the 2.5 percent of the past decade, employment would increase about 1 percent a year, or by 1.44 million, at current levels. But the pool of 14.5 million available workers would be augmented by the continuing increase in the working-age population of about 2.2 million annually, exceeding the 1.44 million new jobs. By these calculations, the pool of people who could work, by early 2000 standards, would chronically rise.
Just since the December 2007 peak in household survey employment, the civilian population has increased by 12.4 million, though the labor force has grown by only 1.9 million. The number of those not employed or actively looking for a job rose 10.1 million, and of those, 2.6 million -- more than the labor force increase -- say they want to work. Total employment in the household survey is down by 1.8 million since the December 2007 peak.
These measures of excess labor are from the Bureau of Labor Statistics’ household survey, which calculates the unemployment rate. The Bureau’s monthly survey of nonfarm payroll employment tells a similar story. From the peak of 138.1 million payroll jobs in January 2008, 8.7 million, or 6.3 percent, were lost by the time of the trough in February 2010. Since then, 6.6 million jobs have been added, or 5.1 percent, but that still leaves payroll employment below the 2008 peak by 2.2 million, or 1.6 percent.
Before the 1990-1991 recession, manufacturing and similar jobs accounted for a declining -- but still dominant -- share of the economy. Those who were laid off in recessions went back to their old jobs a few months later as inventory liquidation was completed and production resumed. But in the last three decades, many of those jobs have moved to Asia; as a result, a greater share of Americans now holds service and technical jobs that tend to suffer longer stretches of unemployment. Also, more people are changing fields when they get new jobs, adding to delays. This explains the slower recoveries in payroll employment after the 1990-1991, 2001 and 2007-2009 recessions.
Nevertheless, the jobs rebound after this recession has been unusually slow, even though the job loss was much greater than in earlier downturns. At the end of the 2007-2009 recession, many didn’t realize that we had entered what I call the Age of Deleveraging, which is characterized by slow economic and job growth. Instead, they believed that the swoons in real GDP and employment would be followed by rapid rebounds. In June, more than three years after the February 2010 trough, payroll employment still was 1.6 percent shy of the previous peak. In past recoveries, such cyclical peaks were exceeded much more rapidly.
U.S. business has had little ability to raise prices in this recovery. Furthermore, sales volume has been severely limited by slow economic growth in the U.S. and around the world, and the strengthening dollar also pressures profits. That means cost-cutting has been the route to pushing profit margins to all-time highs and, as a result, employees have suffered both limited job growth and declining real wages.
Month-to-month changes in payroll employment are volatile. The total rose by 195,000 jobs in June, more than expected, and May’s number was revised to a gain of 195,000 from 175,000. April’s figures have changed from an initial gain of 165,000 that was then lower to 149,000 before being revised to 199,000. The BLS reports that its monthly total can vary by plus or minus 90,000. So at that level of confidence, the number of added jobs for June -- which will undoubtedly be revised next month and again in September -- could be between 105,000 and 285,000.
Furthermore, the excellent analysis by the Liscio Report showed that the June payroll increase was slightly below the average of the previous six months. That means it would take 11 months to get back to the January 2008 peak. The leisure and hospitality category showed gains of 75,000 in June -- more than a third of the total increase.
These are low-quality jobs that often are part-time, with low pay. Similarly, most of the upward revisions in April and May were linked to government, or the leisure and hospitality sectors.
(A. Gary Shilling is a Bloomberg View columnist and president of A. Gary Shilling & Co. He is the author of “The Age of Deleveraging: Investment Strategies for a Decade of Slow Growth and Deflation.” This is the first in a three-part series. Read Part 2.)
To contact the writer of this column: A. Gary Shilling at firstname.lastname@example.org.
To contact the editor responsible for this column: Max Berley at email@example.com.