Wednesday, April 15, 2020

[NJFAC] The Gig Economy: How Big, How Bad? Part I: The Numbers

            There's been much discussion about the gig economy, about whether it is the wave of the future, and whether new kinds of gig jobs are good for human beings.  
            The simplest definition of a gig worker is someone who is self-employed and/or an independent contractor. That means someone who is not legally considered the employee of an employer. Modern examples are Uber and Lyft drivers. Some authors focus mainly on gig jobs mediated by on-line platforms. But others include all kinds of independent contractors and self-employed workers. Some claim that a third of all jobholders are gig workers.
            It is not uncommon for bloggers to write as though there were no gig workers until the 2010s. They may ignore retro-giggers: house cleaners who work for themselves, lawn mowers, people who regularly sell at local flea markets, people who walk dogs and house-sit for a fee, house painters who have no or few employees. And jazz musicians who don't have a steady job. (They gave us the gig-term, didn't they?) If gig workers are those who are not considered employees of those for whom they provide products and services, there have always been gig workers, and much about the gig phenomenon is not new.
            But are there more gig workers than ever? Counting gig workers, old and new, is something that the Bureau of Labor Statistics and the Census Bureau do. Together they are responsible for tons of employment data, and, most importantly, the monthly unemployment rate. But they have rarely surveyed people in "alternative work arrangements"--which includes independent contractors--and contingent jobs. The survey is called the Contingent Worker Supplement (CWS). But when neo-gigs came into view in the 2010s, there had not been a new CWS since 2005.
            Princeton Economist Alan Kreuger and Harvard Economist Lawrence Katz aimed to fill the vacuum with a survey carried out by the Rand Corporation in the fall of 2015. It was called "The Rise and Nature of Alternative Work Arrangements in the United States, 1995-2015," and it is available from the National Bureau of Economic Research. Like CWS, it focused on people's main or sole jobs. Perhaps the most striking conclusions were these:
1. People in alternative work arrangements, which included independent contractors and people hired by temp agencies and labor contractors, increased their share of the work force between 2010 and the fall of 2015 from 10.7% to 15.8%. That is substantial.
2. Of all net new jobs, 94% were gig jobs. Seems like a revolution. But was it?
3. Almost a fifth (19.4%) of all job holders reported that they sold goods and services directly to customers. A minority used old-time intermediaries like Avon. Relatively few used innovative tools. For 2015, the share of the work force that found their main or sole jobs through on-line intermediaries was 0.5%. Not much of a revolution.  K and K showed that many gig workers were not working for Uber, Taskrabbit, or similar companies. Many were in sales, health and educational services, construction, and other apparently retro jobs.
            Two years after K and K's pioneering study, the Bureau of Labor Statistics was able to fund a new Contingent Worker Supplement. The results were obtained in May of 2017 and published on June 7, 2018. They showed that there was little change in the share of workers with alternative work arrangements.  The category of independent contractors, including independent consultants and freelance workers, totaled 10.6 million people. But that represented a smaller share of the labor force at 6.9% than in 2005 (7.4%). So no revolution?
            Careful scholars and journalists found reasons for skepticism.
-Neil Irwin, in "Maybe We're Not All Going to Be Gig Workers" (New York Times, September 15, 2019), provided a concise survey of research that suggested that the neo-gig economy was a niche arrangement in a few sectors and mostly provided a side hustle for people whose regular jobs did not pay enough.
-Irwin mentioned the research of  Dmitri Koustas, who found that many people's earnings from their regular jobs fell off just before they started as gig workers.
-The Federal Reserve Report on the Economic Well-Being of U.S. Households in 2018 showed that 3 in 10 adults engaged in a gig activity in the month before the survey, but the number who were doing things like driving for Uber and Lyft was smaller than the number who sold stuff at flea markets and about the same number as those who walked dogs for pay. The enterprise that involved the single largest share of adults--10%--was online selling. Digital, yes, but not earth-shattering. And did online selling yield much income?
-Researchers at the JPMorgan Chase Institute studied 39 million Chase checking accounts and found that 2.3 million accounts received at least one payment from online platforms between October of 2012 and March of 2018. Millions, yes, but not so many over five years. In March of 2018, the state and the city with the highest share of participators were Nevada and San Francisco, but in each case just 2.8% of the families were generating platform earnings.
-On May 15, 2018, Lawrence Mishel of the Economic Policy Institute provided a clever way to measure the employment impact of Uber and similar companies. While 833,000 people drove for Uber in a year, most did not work a full week or year-round. The total hours worked by all Uber drivers was the equivalent of 90,521 full-time, full-year workers.
            Some of these reports and articles and also the 2017 BLS Contingent Worker Supplement raised questions about Kreuger and Katz's 2016 study, in particular that the gig economy was surging. In a 2019 report, K and K walked back some arguments. They admitted that there were  many new irregular jobs in the 2010s not because of deep structural changes, but because the Great Recession and very high unemployment made workers desperate for cash. Some became "independent contractors," but that could have meant, as was reported at the time, that they were offering to clean out your garage, mow your lawn, or assemble your Ikea furniture for $10 an hour. Much of the increase in the number of independent contractors was a cyclical event rather than a permanent shift. (K and K, I believe, devoted just one paragraph in their 2015 study to the cyclical explanation.)
            But K and K were not giving up. They pointed out that the CWS--like their 2015 study--only studied people's main or sole job. Perhaps the gig revolution was hidden in people's second and third jobs. They had a point. As we've seen, most people who drive for Uber in a single year are not full-time, year-round workers. For many the job provides supplemental income to help pay off debt, pad retirement funds, or surmount an income shock in their regular jobs. However, federal statistics on multiple job holders showed no significant increases. End of story? Nope. K and K claimed that the Census Bureau and the Bureau of Labor Statistics were missing many multiple job holders. Perhaps so. But who many?
            What we can say about the gig revolution is that the neo-part has often been exaggerated. We can admit that flexibility of hours and easy entry into gig jobs can be useful. For example, you are a single parent and cannot work a standard schedule; or you've retired and need a little more money. But the big picture is that Uber, Lyft, Instacart, and GrubHub are adding a lot of bad jobs in America. Pay rates are subject to capricious changes by the de facto employer; the companies usually offer no benefits and pay nothing into unemployment, Social Security, and Medicare funds. These companies are parasites living off the fact that many other jobs don't provide flexible hours and adequate pay and benefits. But they don't provide the latter either.
             In some states, notably California, workers and liberal politicians have had enough, and they have begun to redefine Uber drivers and other freelancers as employees of the concerns that they work for. On-demand drivers are angry, as Bryce Covert shows in the March-April, 2020 issue The American Prospect. It is also noteworthy that the March, 2020 coronavirus stimulus bill, known as CARES, allows gig workers to apply for unemployment benefits, although neither they nor the companies they worked for have contributed to state unemployment funds. Does that imply that gig workers and their de facto employers need to contribute? Yes. The Uber model of the no-benefits employer has failed miserably. But more about these issues in Part II.         
Frank Stricker is a board member of the National Jobs for All Network and emeritus history and labor studies professor at California State University, Dominguez Hills. The views in this article are not necessarily those of his organizations.

This list is only for announcements, so you may not post. To contact the list manager, write to njfac [at]
You received this message because you are subscribed to the Google Groups "goodjobs" group.
To unsubscribe from this group and stop receiving emails from it, send an email to
To view this discussion on the web visit