EconoMonitor

Dan Alpert's Two Cents

Unhappy Days are Near Again

On the occasion of President Obama’s inauguration, we are publishing pieces on the two elements of the economy that we believe require the most urgent attention by the incoming administration – the repair of the banking system and, in this report, the unemployment crisis.

Highlights

  • Much has been written on the similarities between the economic afflictions of the present day and those of the Great Depression. A frequent refrain sung in such contrasts references unemployment rates of the earlier era, followed immediately by the happy verse “of course, we don’t see anything near those levels today.”
  • Reassuring as this conclusion may seem, today’s unemployment picture may be significantly closer to that of the 1930s, owing fully to the fact that present methods of calculating unemployment data differ markedly from Depression-era formulas.
  • Beginning in the Reagan/Bush(I) years, some in government voiced concerns that unemployment statistics were underreporting the job growth during the last debt-driven expansion of the mid- to late 1980s.  The changes that were subsequently made to the reporting of unemployment and job creation/destruction nearly all had the politically expedient result of increasing the numerator and decreasing the denominator in calculating the number of Americans employed and, conversely, reducing the percentage of those reported as unemployed.
  • Examining the government’s modern calculation methods, and comparing today’s employment statistics to those of the Great Depression has led us to believe that we are far closer to the unhappy days of the 1930’s than most may think.
  • Today’s announcement of 589,000 new jobless claims during the second week of January, and December’s 524,000 net job losses, is bringing the reality of the situation into sharper focus.  Emergency employment measures, being advanced by the Obama administration, will be critical in 2009.

Overview

The imagery of this week’s inaugural ceremonies invariably recalls March 4, 1933, when another great man summoned the nation to overcome its fears and begin the process of resurrecting itself. Much has been written on the similarities between the economic afflictions of the present day and those of the Great Depression. A frequent refrain sung in such contrasts references unemployment rates of the earlier era, followed immediately by the happy verse “of course, we don’t see anything near those levels today.”

Reassuring as this conclusion may seem, today’s unemployment picture may be significantly closer to that of the 1930s, owing fully to the fact that present methods of calculating unemployment data differ markedly from Depression-era formulas.

This past fall’s revised jobless numbers, coupled with the 524,000 lost jobs recorded in December, have generated an official unemployment rate of 7.2%, with a rising trend. As bad as this may be in comparison to the bubble-obscured unemployment levels of +/-4.5% just two years ago, it represents but a fraction of real unemployment when you unravel the complex algorithms that have minimized the headline unemployment rate and masked real changes in the nature of our economy.

What Has Our Government Actually Been Reporting About Employment

Beginning in the Reagan/Bush(I) years, some in government voiced concerns that unemployment statistics were underreporting the job growth during the last debt-driven expansion of the mid- to late 1980s.  The Bureau of Labor Statistics (BLS) was encouraged to promulgate changes to its methods used for monthly information-gathering and calculation of unemployment and job creation/destruction. While implementation of many of these changes failed to occur until 1994 (with further adjustments in 2003), nearly all had the politically expedient result of increasing the numerator and decreasing the denominator in calculating the number of Americans employed and, conversely, reducing the percentage of those reported as unemployed.

The calculation of the official unemployment rate starts with a U.S. population of approximately 304 million. Almost 70 million are 16 years or younger, in prison or otherwise institutionalized, or in the military; they are subtracted from the population to arrive (for December) at the roughly 235 million members of what the BLS calls the Civilian Noninstitutional Population.

The BLS eliminates approximately 80.5 million adults from the civilian population who are classified as “not in the labor force”—some 34% of the total, driving the denominator for employment statistics down to about 154.4 million. As a result, the so-called Official Unemployment Rate, or U-3, presents a highly filtered assessment of employment and marginal employment:

  • Workers who say they are available for and want employment, but have not looked for a job in four weeks, are considered not in the labor force (“marginally attached workers”).
  • People working only part time, for as little as one hour or even  without pay in some cases because they cannot find a full-time job, are considered fully employed (“part time for economic reasons”).
  • Workers who are on strike without pay are considered employed.
  • Workers who are enrolled in job training or other courses are considered not in the labor force.
  • Workers on temporary leave from their jobs, regardless of whether they are being paid, are considered to be employed.
  • Older Americans and other workers who would like to find employment, but feel they are too old or otherwise unqualified to be attractive to employers, are considered not in the labor force.

Among the statistics the BLS releases each month is the alternative unemployment rate, or U-6, which is somewhat closer to reality. U-6—as of December a whopping 13.5% (up from 12.6% the month before and 8.7% at the beginning of the recession)—adds back to the labor pool the marginally attached workers described above and does not consider employed those who work part time for economic reasons when they would otherwise seek full-time work. The least adulterated statistic, and the one closest to pre-1994 calculations of “official” unemployment, U-6 is still well below the generally understood average unemployment rate during the 1930s of 18.2%—but even U-6 is not directly comparable with joblessness measures from the Depression era.

image002_512_37.gif  

This isn’t Your Grandfather’s Labor Force

During the 1930s, women comprised only 21.9% of the workforce. For social and economic reasons, the vast majority contributed their labor to society in uncompensated ways (and one can persuasively argue that an economy in which a larger number of people needn’t work is delivering a higher per-worker standard of living, but that’s a story for another time). Today, by contrast, 46.8% of the workforce is female. The overall workforce as a percentage of the adult Civilian Noninstitutional Population stood at about 55% in the 1930s, while today it is 66.5% (including the 1.9 million marginally attached workers).

Consequently, the unemployment rate during the Depression was a smaller percentage of the adult population than an equivalent rate would be today—in reverse, yielding an adjustment that would increase U-6 by 21% in an apples-to-apples comparison. So, this takes us from a December U-6 unemployment rate of 13.5% to 16.3% in Depression-era terms. Of course, today’s longer life spans swell the ranks of older adults who are truly unavailable for work, in comparison to the 1930s, when few but the well-to-do “retired.” To the extent we view longer life spans and retirement as delivering a higher standard of living/quality of life, some reduction to the above result may be appropriate.

What Actually Happened During the Depression?

On the other side of the comparison, the reported levels of unemployment during the Depression also warrant examination. As it turns out, unemployment statistics from this era, topping out at a frightening 24.9% in 1933, track only private-sector and ordinary government employment, ignoring employment fostered by the New Deal’s emergency employment measure. All of the dam and bridge builders, conservation workers, cultural-institution creators and others employed by the likes of the WPA are not included as employed in Depression-era numbers, even though they drew wages and went to work every day. Actual unemployment during the Depression, after we factor in emergency employment, topped out at around 23% in the year FDR assumed the presidency and averaged not 18.2%, but actually closer to 14.5% over the entirety of the 1930s—a lot closer to the U-6 result at year-end 2008.  The solid line on the graph below shows the raw unemployment statistics during the 1930’s, while the dashed line includes emergency workers among those employed.

image004_67.gif

How Jobs Grow While They are Lost

So far, this piece has covered issues relating to the BLS’ calculation of unemployment through its monthly Current Population Survey of about 60,000 households. The other BLS statistical element that bears mention is the monthly Current Employment Survey (CES) of about 150,000 businesses and government agencies, representing approximately 390,000 individual worksites, which is used to determine levels of job creation and destruction, wages and work hours. While the CES uses broad sampling, BLS statisticians determined in the late 1990s that the data-gathering (even with previous bias adjustments) inadequately reflected job activity in businesses too small to be regularly sampled or too recently created to be included. Accordingly, the Net Birth/Death Model was implemented in 2000 to cover this perceived sampling deficit. For 2008, adjustments to job growth/destruction statistics generated by the Net Birth/Death Model were as follows:

image006_512_13.gif What’s interesting about the Net Birth/Death Model adjustments to the CES statistics over the past year is their decidedly positive outlook in the face of rapidly increasing levels of overall unemployment. Note that in the last reported month (December 2008), the model reported a gain of 72,000 jobs versus the overall loss of 524,000 jobs on a seasonally adjusted basis (or a loss of 954,000 jobs on a nonseasonally adjusted basis, which is how the model is calculated). The algorithms that generate these rosy results are, no doubt, highly defensible by BLS statisticians. But one may venture to ask where, as just one example, 18,000 otherwise undetectable jobs were created in the financial sector last month in the face of overwhelming reductions in force throughout that industry?

Conclusion

Whatever the analysis, we continue to believe much of this decade’s economic pain comes from the U.S. economy’s inability to create jobs in the face of a tidal wave of low-wage competition from the nearly 2.5 billion people who came into the capitalist labor force after the Iron and Bamboo Curtains parted from the political-economic world. The effects of nearly doubling the competitive labor force were masked for a time by technological and productivity leaps in the 1990s and the need for emerging countries to build the necessary physical and technological infrastructure to communicate with, and economically move product to, the rest of the world. By the turn of the millennium, labor in emerging countries began to erode our domestic-jobs base. But we were able to maintain the mirage of prosperity for a time, based on unprecedented levels of borrowing and debt-induced asset inflation. We inevitably hit the wall and our assets are now falling in value, debt defaulting and employment (as it would have years ago in the absence of our debauchery of debt and over-consumption) is suffering from our inability to compete globally while maintaining the standards of living to which we have become accustomed.

Lest we miss another important point, while we have been busy outsourcing menial (and even less-menial) labor to developing nations, we have failed to maintain our educational system at a level comparable to the educational product being turned out by our competitors.  They have learned their lesson on the importance of an educated population, even as we were forgetting ours.

Watching today’s announcement of 589,000 new jobless claims last week and a continuing jobless roll number at historic highs, we nevertheless suppose there will be some (with any luck, perhaps ourselves) who will be insulated from the realities of what’s unfolding before our eyes. It’s critical, however, that our government and business leaders have their eyes wide open and plan for a future with an excess of workers, both in the United States and in other parts of the world now able to compete with American labor on an unprecedented scale. The need for emergency employment actions will become acute this year, as the measures of unemployment we have engineered over the past two decades will no longer mask the reality of economic challenges unseen for more than seven, but with far more complex, longer-term solutions required this time around.

Comments are closed.

Most Read | Featured | Popular

Blogger Spotlight

Dan Steinbock

Dr Dan Steinbock is a recognized expert of the multipolar world. He focuses on international business, international relations, investment and risk among the major advanced economies (G7) and large emerging economies (BRICS and beyond).

Economics Blog Aggregator

Our favorite economics blogs aggregated.