American Unemployment: Past, Present, and Future

Published by EH.Net (January 2021)

Frank Stricker, American Unemployment: Past, Present, and Future. Urbana, IL: University of Illinois Press, 2020. x + 267 pp. $20 (paperback), ISBN: 978-0-252-08502-4.

Reviewed for EH.Net by Lee Ohanian, Department of Economics, UCLA.

There is perhaps no more challenging issue to write about in macroeconomics than unemployment. Measurement issues, theoretical issues, even conceptual issues make unemployment like quicksilver. We want to get our hands on it, but it invariably slips through our fingers.

Consider how many different interpretations/definitions exist in the literature. Old school Keynesians created “NAIRU” — the non-accelerating inflation rate of unemployment — which is based on the Phillips Curve. Neoclassical economists, including Milton Friedman, Edmund Phelps, and Armen Alchian created the “Natural Rate of Unemployment,” which was front and center in Robert Lucas’s misperceptions model that gave birth to the rational expectations business cycle literature.

Search theorists take a different approach, decomposing unemployment into “frictional unemployment,” which is the time it takes an individual to find a job that is a good match for their skills, and “structural unemployment,” which refers to those who remain unemployed long after search and matching frictions can reasonably be operative.

Conceptual struggles are not just academic. The Federal Reserve uses six different concepts of unemployment (known as U1-U6), ranging from the plain vanilla definition, to measures that incorporate long-term unemployment, temporary jobs, and the discouraged worker effect.

Some economists eschew unemployment and instead study alternative labor market indicators such as employment and/or market hours worked as a share of the working age population. The benefit of focusing on how much work is being done is that measurement is conceptually simple. It doesn’t require knowing an individual’s subjective attachment to the labor force or how much time they are devoting to searching if unemployed.

Understanding unemployment is a tall order. Despite this challenge, Frank Stricker — professor emeritus of history at California State University, Dominguez Hills — studies the history of American unemployment from roughly the postbellum period through the 2007-08 financial crisis and its aftermath in American Unemployment: Past, Present, and Future.

As a non-economist, Stricker brings a different eye towards the problem of unemployment. He is a passionate advocate for “full employment” and creating high paying jobs. He focuses the book on whether the shifting American model of a free enterprise economy can create achieve these goals. His answer is a clear “no,” and the latter chapters of the book present an explanation for his views, and present an argument for how government can be the solution to the problems he sees as ailing America’s labor market.

Stricker’s passion for struggling individuals throughout America’s history dominates his writing, and this is an important, positive feature of the book. The primary theme is that the U.S. labor market doesn’t function efficiently, thus leaving a lot of human resources underutilized much of the time. His perception is that the U.S. economy rarely achieves full employment, that compensation is limited by the chronic presence of the unemployed, that it has been a mistake to try to fight inflation using unemployment, and that government jobs policies would lead to significantly better economic outcomes.

While I disagree with Stricker’s characterization of the chronic lack of efficiency of the U.S. labor market, I agree with his perspective about the Phillips Curve and the failed logic of using unemployment to fight inflation. I also agree that there are labor market policy reforms available that could be beneficial in principle, including jobs policies. I discuss these issues below.

The book’s major limitation is the lack of a formal economic model to guide thinking about the U.S. labor market and to interpret labor market data. The term “full employment,” which is the centerpiece of the book, means different things to different people and fundamentally requires a model-based definition. Stricker refers to an unemployment rate of 4 percent as a benchmark, but that number is largely a relic from a time when our understanding of the labor market was much more limited.

The book begins with a discussion of the labor market between 1873 to 1920, a period of significant economic growth, including the technological revolutions of electrification and the internal combustion engine. Stricker discusses the period as one of chronic capitalist crises occurring in the 1870s, as well as the panics of 1893, 1907, and weak economic activity around 1914 and the focus of the chapter is the inability of the American economy to maintain full employment. Stanley Lebergott’s unemployment estimates, which are available from 1900 onwards, shed some light on this issue (though only beginning in 1900). The mean rate of unemployment between 1900 and 1920 was about 4.5 percent.

My view is that an average unemployment rate of 4.5 percent is remarkably low. Keep in mind that this period is one in which the American economy was poor and technologically backwards, which means that information costs were very high and mobility costs were very high, both of which increased unemployment. Moreover, new technologies were changing the nature and scope of work, sharply reducing the demand for agricultural workers and for workers employed in formerly labor-intensive technologies. And rural labor markets were not very competitive at that time. The biggest employer in a small town likely had significant monopsony power, which reduces hiring. I see the 1900-1920 U.S. labor market as a glass 4/5th full, rather than a glass that is 1/5th empty, which seems to be what Professor Stricker sees.

Stricker then moves to the 1920s and 1930s, the latter of which encompasses the Great Depression. This is a period in which government policies played an enormous role in impacting the economy. Stricker rightfully focuses on the role of government in this chapter, though he paints with broad strokes, setting this up as either a “laissez-faire” economy or one with government interventions.

But some 1930s government labor market interventions were destructive, including the National Industrial Recovery Act and the Wagner Act, the latter which de fatco permitted the use and threatened use of the sit-down strike. By impeding the normal forces of competition, the NIRA and the Wagner Act created substantial unemployment by raising real wages in manufacturing and other goods-producing sectors far above market-clearing levels. These are the types of policies that should concern labor activists because they create market inefficiencies by preventing mutually advantageous trades between the suppliers and demanders of labor services.

On the other hand, his positive description of government infrastructure projects in the 1930s (think Hoover Dam) is powerfully told, and the economics of these programs is clearly on his side. There is no better time to build infrastructure than when there is low demand for the factors of production that build that infrastructure. Sadly, our ability to do this today is probably much less efficient than it was 85 years ago. Understanding deeply why this is the case would be a wonderful contribution to American economic history.

The book transitions to three different periods that are demarcated by the observed patterns of unemployment and inflation. There is 1940-1974, which he refers to as “Full Employment: Experiments and Battles”; 1975-2000, which he refers to as “Low Unemployment + Low Inflation: Can’t Be Done, Is Done”; and then 2000-2018, which he refers to as “Low Pay, Great Recession.”

The guiding theme of these three chapters is the chronic conflict between labor and capital, and the evolution of economic policies that are both influenced by and shape these battles. The interpretation is that jobs have been limited for much of this period, and these limitations have depressed worker compensation.

The most compelling point discussed in these chapters is the failed policy of using the Phillips Curve to try and manage inflation by tolerating high unemployment. The Phillips Curve has been an extremely costly policy detour that remains in some policy quarters, despite that it empirically has failed for nearly forty years. Stricker correctly argues that this policy has harmed the economy and is one that should have been abandoned long ago.

An important postwar labor development is that labor conflict and frequent strikes, particularly in Rust Belt industries, sowed the seeds for future weak job growth and compensation growth by reducing investment and the adoption of the latest technologies in these industries. The remarkable increase in steel and auto imports reflects foreign producers who became much better at producing these products because of technological superiority and peaceful relations with their workers. Rust Belt industry management deserves some of this blame for the Rust Belt’s decline, but so do the United Auto Workers and the United Steel Workers, both of whom have admitted that they should have been more cooperative and less combative. What is not recognized in American Unemployment is that the labor movement of the 1950s-1970s damaged future workers.

An important misperception in American Unemployment that is particularly relevant for the recent economy is that wage levels and wage growth are a good proxy for compensation. This was accurate for the 1960s and before, when fringe benefits were a very small fraction of compensation. But today, fringe benefits are nearly 30 percent of compensation.

Competitive models of the labor market predict that real compensation will move closely with worker productivity, but this comparison requires better measurement than using just wages. Worker productivity has increased by about 180 percent since 1965, while compensation has increased by about 125 percent. This is a compensation shortfall of about 30 percent over the 55-year period, which amounts to about a 0.4 annualized percentage point difference. Not perfect, but certainly in the right direction as predicted by the theory. In contrast, measuring compensation using just wages, and deflating by the CPI rather than by the GDP deflator (which is used to construct productivity) shows only a 15 percent increase since 1965. No wonder that Stricker sees workers getting the short end of the stick based on this flawed measure.

The final chapter offers Stricker’s policy ideas. I support in principle his idea of government-provided jobs program, particularly as an alternative to long-term unemployment benefits. A positive externality of this policy is that this expands the ability of otherwise idled workers to contribute to their communities and to enhance their skills. Other policy ideas from Stricker, such as a $20 minimum wage, may not be nearly as productive as he believes, particularly in relatively poor labor markets. Currently, the median Mississippi worker earns less than $15 per hour. The downside of high minimum wages is that they price the least experienced workers, including immigrants, out of the labor market, and essentially force them into public assistance.

What is missing from the policy recommendations section is a detailed assessment of broader forces that will shape the U.S. labor market, including how continued globalization, automation, changing gender roles, and a deficient American K-12 education system will impact future jobs. Almost certainly, the most important policy reform is to improve our schools so that more of our students are adequately trained in STEM disciplines and in written and oral communication, all of which are skills that will be increasingly in demand.

In contrast to Stricker’s views, standard measures of labor market performance show that the American labor market is remarkably efficient. Before the pandemic, nearly six million jobs turned over each month, reflecting large changes in supply and demand across sectors and industries. This works out to be a turnover rate of about 40 percent of jobs annually. In terms of the speed of matching workers, two-thirds of the unemployed found a new match within 14 weeks of unemployment prior to the pandemic. And the percentage of employed prime-age workers (25-54 years old) was 80.5 percent, just below the peak of about 81 percent in the late 1990s. This is not a labor market that is broken.

The problem with our labor market is not that it doesn’t effectively match workers with businesses at compensation levels that are reasonable. Rather, any economy’s ability to create lots of high-paying jobs, which is Stricker’s understandable goal, requires lots of workers with high human capital. For Stricker, unemployed workers are a valuable human resource sitting idle, losing their ability to provide for themselves and their families, and losing the dignity that goes with having a decent job. While most of us agree with Stricker on this point, there is a deeper reason why these workers have been idled that goes far beyond politics and the potential deficiencies of a capitalist economy. It has to do with skills and why more American workers haven’t accumulated more skills to enhance their employability. This is perhaps the most important lesson that the history of labor economics provides, and shows why any discussion of unemployment and the labor market must dig more deeply than Stricker’s engaging and provocative book.

Lee E. Ohanian is Professor of Economics at UCLA and is a Senior Fellow at the Hoover Institution, Stanford University. He is the author of “New Deal Policies and the Persistence of the Great Depression.”

Copyright (c) 2021 by EH.Net. All rights reserved. This work may be copied for non-profit educational uses if proper credit is given to the author and the list. For other permission, please contact the EH.Net Administrator (administrator@eh.net). Published by EH.Net (January 2021). All EH.Net reviews are archived at http://www.eh.net/BookReview.