On Friday, September 28, 2018, the Department of Labor’s Bureau of Labor Statistics (BLS) issued its first estimate of how many U.S. workers participate in electronically mediated (online platform) work, and concluded that “electronically mediated workers” accounted for 1.0% of total employment in May 2017.

By way of background, in May 2017, BLS added four questions to its Contingent Worker Supplement (CWS), a supplemental set of questions that are periodically included in BLS’s monthly Current Population Survey.  The purpose of these questions was to identify the number of workers who secure work through a website or mobile app, as opposed to a staffing agency or other traditional methods of seeking employment.

In June 2018, BLS released a summary of its Contingent and Alternative Employment Arrangements survey, and concluded that 3.8% of workers (roughly 5.9 million people) held contingent jobs, down from 4.1% of workers in February 2005.  The findings from the four questions relating to online platform work, however, were not released in that report.  As BLS explained last Friday, after extensive review, the agency concluded that the questions asked in May 2017 had not worked as intended, and resulted in a large number of “false positives” or incorrect “yes” answers.  In response to these concerns, BLS went back and used workers’ responses to other questions in the survey to manually recode the data and produce more accurate results.

Using this revised data, BLS concluded that in May 2017, there were approximately 1.6 million workers using an online platform for work, accounting for 1% of total employment.  This figure included workers who obtain jobs or “gig” tasks through websites or mobile apps.  Of the 1.6 million electronically mediated workers, the survey results show that 0.6% of workers did electronically mediated work in person, while 0.5% performed electronically mediated work entirely online.

The survey also included “which job” questions that examined whether workers performed electronically mediated work as a primary source of income.  The survey results indicated that, of workers who performed electronically mediated work, 72.4% did so as their primary job, 14.3% did it as a second job, and 12.1% did it as additional work for pay.  BLS cautioned, however, that the estimates of workers who hold such jobs as a second job or as additional work for pay may still include a number of false positives.  This possibility arises because the other CWS questions did not examine jobs other than the worker’s primary job, which made it impossible for BLS to verify whether a worker’s report regarding their “second job” was consistent with other known information about the worker.

These new CWS results are largely consistent with those of a survey that Lawrence Katz and Alan Krueger conducted in 2015 for the RAND American Life Panel.  The Katz/Krueger study, which followed the same basic methodology as the CWS, estimated that workers who provide services through online intermediaries constituted 0.5% of the workforce in 2015, and other estimates of the size of the online “gig economy” in 2014 and 2015 likewise arrived at figures ranging from 0.4% to 0.6%.  It is not clear whether the slightly higher estimate of the CWS reflects a real increase over the past two years in the number of workers doing work through online platforms, or if it is instead due to other factors such as the wording of the survey questions.  But Littler has found that the rise of the sharing economy is already having greater impacts on the labor market and companies’ business plans than their seemingly small numbers suggest.

BLS’s announcement of the new data, and a summary of its findings, can be found here.