Yesterday’s release of the 2017 Contingent Worker Supplement (CWS) provides the first government numbers on contingent and alternative work arrangements in thirteen years. The findings? The percent of workers in alternative work arrangements dropped slightly from 10.7 percent in 2005 to 10.1 in 2017. This translates to 15.5 million workers who rely on on-call, temp-agency, or contract-firm jobs or independent contracting as their main job. Below are five top takeaways from these new numbers.
1. The CWS only tells part of the story — other data sources are necessary.
In order to gain a comprehensive understanding of the labor market, we need to consider the range of ways in which people work. The CWS is a nationally representative, rigorous survey. It gives us important information about the number of workers who rely on a particular set of work arrangements for their main job, and the relative stability of that number over time. The supplement asks specifically about workers’ main job–the one for which they work the most hours. People work in a lot of different ways and looking at their main job only provides part of the picture. Supplemental alternative work is very common. Surveys that ask about it find that many of those engaged in alternative and independent arrangements do so to supplement other income. For example, McKinsey Global Institute found that over 50 percent of independent workers were supplementing other income sources.
The updated CWS adds to a larger landscape of knowledge on work, which also includes supplemental, occasional, and informal work. If people are increasingly having to rely on alternative work as a supplemental income source, there are substantial implications for the economy and policy. When people talk about the “gig economy,” they need to specify the types of work and work arrangements they mean; the BLS clarifies that the CWS does not purport to approximate the “gig economy.” Those wanting to know more about the scope of knowledge on alternative and independent work arrangements should explore the Gig Economy Data Hub, a collaborative project of the Future of Work Initiative and Cornell University’s ILR School that seeks to consolidate and clarify available data on this topic.
2. The portion of the workforce engaged in alternative work as their main job appears to be relatively stable over time.
Over time, the CWS data on these arrangements has been strikingly stable. In 1995, the first year the supplement was administered, 9.9 percent of workers were in alternative arrangements. That share has never varied by more than 0.8 percentage points any year it was administered.
Given how infrequently the data has been collected, though, it is difficult to state what the long-term trend of alternative work looks like. Since the 2005 CWS, the U.S. economy experienced a recession and extended recovery. The most recent release comes in the midst of a historically strong labor market that may provide workers who had previously been in alternative work with opportunities to move into traditional work. In order to understand if and how economic cycles relate to people’s reliance on alternative work, we need more consistent data collection.
3. We’ll have to wait until September to learn more about participation in online platform work.
Some work is found through online platforms, which connect workers to clients via apps or websites, and some estimates show this type of work has grown rapidly. Online platform work, though, is not directly addressed by the numbers released yesterday. The 2017 CWS did ask four questions about work through online platforms and online tasks on both a primary or supplementary basis, but those numbers are not yet out. BLS has committed to releasing them by the end of September. Increasing numbers of alternative workers may be using online platforms to find work, but there is no way to differentiate them from offline alternative workers in this data.
There is, however, one piece of evidence in the CWS data released June 7 that suggests online platform work may be growing. The data show that one of the only occupational groups to see an increase in the frequency of alternative work was transportation and material moving, from 9.7 percent to 12.2 percent. That growth was largely among independent contractors. Although more data is needed to know conclusively, rideshare drivers, like those for Uber and Lyft, may be behind this jump.
4. We see variation between demographic and occupational groups.
The stable trend of alternative work arrangements in the CWS is not uniform across demographic groups or occupations. For example, while the prevalence of alternative work fell slightly between 2005 and 2017, it rose for both black and Hispanic workers, and was increasingly concentrated in independent contracting. The new data also show a decrease in the prevalence of full-time alternative work as a main job, but an increase in part-time alternative work as a main job. That is, those workers whose main job is part time are slightly more likely to be in an alternative arrangement than they were in 2005.
The trend in alternative work also varies by occupation. While most occupational groups have slightly lower rates of alternative work, both transportation workers (mentioned above) and service workers show higher rates, largely concentrated among independent contractors. Overall, these differences are small, and what they mean for the experience of work is hard to infer from the summary tables. Researchers will be able to further examine differences and analyze their implications when BLS releases the full datasets soon.
5. There are still substantial gaps in access to benefits for alternative workers.
One of the most striking changes since 2005 is the increase in health coverage rates across all alternative work arrangements. The largest gains were among those in contingent and alternative arrangements who did not access care through their employer. The Affordable Care Act (ACA), which expanded access to health coverage for millions of workers, likely played a large role — in fact, past research has shown that independent workers are three times more likely to use ACA coverage than traditional workers.
However, there are still substantial discrepancies between alternative and traditional workers. Those outside of traditional arrangements are still less likely to have health insurance, and they are particularly less likely to access that insurance through their employer. Traditional workers are over twice as likely to receive employer-provided health insurance as the wage and salary alternative workers, and 60 percent more likely to access employer-provided retirement plans. This highlights the conceptual difference between traditional work, in which the employer often assumes responsibility for the economic security of their workers, and alternative work arrangements, where employers often do not. We need to develop more portable benefits systems, like the ACA, that can increase economic security among workers across arrangements.
Although the numbers have remained stable over the past 13 years, the economy has not. With such sporadic and infrequent data collection, it is difficult to relate structural trends in the economy to these numbers. Conducting a regular, detailed analysis of work arrangements would allow us better understand the ways in which Americans work.
In addition, we need to look beyond self-reported surveys and consider administrative data, like tax returns and employment records. Research has shown that the way people identify their work arrangements on surveys often contradicts the way they identify on their tax returns, suggesting that self-reported surveys alone are not enough to understand this complex segment of the labor market. We also have little information directly from employers, which could improve knowledge on some arrangements, especially subcontracted labor.
Administering nationally representative surveys and undertaking rigorous research takes significant resources, but is necessary in order to identify the challenges of today’s workers and develop policies to best ensure their economic security.
 Data from the Federal Reserve’s SHED finds that 39 percent of independent workers are doing informal work to supplement income from their main job. Jonathan Hall and Alan Krueger estimate 68 percent of Uber’s driver-partners did not rely on Uber as their primary source of income.
For example, the JPMorgan Chase Institute found that the number of people using platforms more than doubled each year from 2013 to 2016. Uber reports its number of drivers doubled every six months from 2012 to 2015.
 Independent contractors do not have an employer and so are excluded from these calculations.