robots welding cars
What are the major explanations for the decline in employment among Americans? Robots, as well as growth in imports from China, according to Katharine Abraham and Melissa Kearney. They also found that the growth in disability benefits, increasing minimum wage, and rising levels of incarceration each play smaller, but still significant roles.

Ban the Box may be counterproductive

And four other lessons in full employment from a day at the Minneapolis Fed

Just a week after the announcement that the nation’s unemployment rate had hit its lowest level since late 2000, I spent a day immersed in some of the latest and greatest research ultimately aimed addressing labor market weaknesses.

The day was organized by the Minneapolis Federal Reserve Bank’s Opportunity and Inclusive Growth Institute, launched just over a year ago with the mission to “conduct and promote research that will increase economic opportunity and inclusive growth and help the Federal Reserve achieve its maximum employment mandate.”

Toward that end, every presenter focused on one or both of the following:

  • Our nation’s large and persistent racial employment gaps

  • A longer-term decline in the proportion of all people age 16 or older who are employed for pay

Contrary to recent trends in the unemployment rate, which does not consider those who have opted out of the labor force, the more comprehensive Employment-to-Population Ratio (EPOP) remains lower today than it has since the mid-1980s, a time when women’s workforce participation was still far lower than that of men.

Employment-to-Population Ratio (EPOP), January 1960-April 2018

EPOP 1960-2018
APM Research Lab
Source: Bureau of Labor Statistics. Note: The EPOP is calculated by dividing the age 16+ employed population by the entire 16+ civilian, non-institutionalized population.

Among the many key takeaways:

  1. Criminal records place a drag on employment, especially among populations of color.
    Presentations by both the University of Minnesota’s Aaron Sojourner and Rutgers’ Amanda Agan drew from recent ground-breaking work which goes beyond measuring the number of people in jail at a single point-in-time, to estimating the growth and prevalence of felony convictions that can impact employment opportunities for a lifetime. According to this article, eight percent of the entire U.S. adult population—over 19 million individuals—have felony records, including one-third of all African American men.

    The rapid growth in numbers of people with convictions in recent years adds up to measurable impacts on the national labor market. The solution, then, might be to implement restrictions on employers’ ability to ask about applicants’ criminal histories on job applications—the so-called “Ban the Box” policies. But see # 5 below.

  2. Some other things you might think might hurt employment don’t. But robots do.
    University of Maryland’s Katharine Abraham, presented her recent paper with Melissa Kearney, “Explaining the Decline in the U.S. Employment-to-Population Ratio.” Major explanations? Growth in imports from China, and robots. They also found that the growth in disability benefits, increasing minimum wage, and growth in incarceration each play smaller, but still significant roles in the downward trend in the EPOP.

    Contrary to some popular speculation on things that keep people sidelined from working, they found no strong evidence that government assistance (beyond disability), spousal employment, immigration, or unionization had anything to do with the decline in employment.

    And, no, the retirement of the Baby Boom generation does not explain away this pattern. Since late 2000, employment has gone down for every age group under age 55.

  3. In the international context, the U.S. labor market is not as strong as you might expect.
    In the day’s keynote, Harvard Economist Jason Furman, who served on the Council of Economic Advisors under both Presidents Clinton and Obama, placed our labor force in international perspective. He argued that the United States had struck a “grand bargain” where a free labor market, coupled with a tolerance for higher levels of income inequality, is supposed to result in strong employment.

    Using OECD data, he demonstrated that in comparison to other developed nations, the United States does indeed have less regulated labor markets with lower minimum wages, as well as high levels of income inequality. The strong employment markets, however, are not so forthcoming. At least among prime-age workers (age 25-54), employment is very middling in the United States, and not all countries have experienced the decline in the EPOP that he and others dissected throughout the day.

  4. Apparently, there are no easy answers.
    William Darity discussed a proposal for “Baby Bonds” that he had developed with fellow panelist Darrick Hamilton to address wealth inequality (the cumulative effect of employment disparity plus our nation’s historical legacies of slavery, Jim Crow, and other racial injustices). Whatever the estimated benefits may be, the estimated $18 trillion price tag attached to this proposal presents a definite challenge.

    Hamilton himself made a strong argument for a federal employment guarantee. An idea that dates at least back to President Roosevelt’s 1944 State of the Union address and carries a $650 billion price tag.  

    Jason Furman questioned those sorts of “big ideas” as expensive and potentially undermining to competition and productivity. He does, however, see some promise in more targeted and dynamic wage subsidies, as well as the cumulative impact of smaller reforms that would for example, increase workplace flexibility and worker mobility. Perhaps he was being appropriately cautious in a room full of economists, but if a presidential advisor cannot conclusively identify the preferred policy solutions, you know there are no easy answers!

  5. Ban the Box may be counterproductive.
    What about the policy prescription that would seem most obvious and actionable, given the specific issue of criminal records? Rutgers Economist Amanda Agon, offered what might have been the day’s biggest lesson in an economist’s favorite topic: unintended consequences.  

    Agon recently published the results of an experiment she had conducted along with Sonja Starr. It involved submitting 15,000 fictitious online job applications to real employers in New Jersey and New York City. Each fictitious applicant had “a distinctly Black or distinctly White name,” and applications were submitted both before and after each jurisdiction had adopted Ban the Box policies restricting employers from asking about criminal histories in job applications.  

    Agon’s key slide showed that prior to Ban the Box policies, the employers they tested discriminated little by race, but a lot based on criminal record. After Ban the Box was in place, employers were significantly more likely to call back White than Black applicants. Unintended consequences indeed!

Callback Rates by Race, Criminal Record, and Period

ban the box graph2
APM Research Lab
Source: Graph adapted by APM Research Lab from Amanda Agon and Sonja Starr, “Ban the Box, Criminal Records, and Racial Discrimination: A Field Experiment,” Quarterly Journal of Economics (2017;133(1):191-235).

While the Institute does not specialize in easy answers, it does offer honest attempts to factually address important policy questions. There was much more to the day than I can summarize here, and I would encourage you to check out the full proceedings online.

While the conference was more focused on the academic than the practical it was grounded to some degree by the presence of many members of the Institute’s newly-formed Community Advisory Board. Among those in attendance was Tawanna Black, who came to the meeting directly from her launch of the Center for Economic Inclusion earlier that morning. Perhaps future Institute conferences will provide more opportunity for practitioners like Ms. Black to bring their grounded experience to bear on Institute’s impressive research agenda.

-Craig (@c_helmstetter)

Reactions? Please email us your thoughts or respond on Twitter or Facebook.