Executive Summary

This report compares the 2017 Contingent Worker Survey (CWS) to previous reports in order to identify shifting trends in alternative work arrangements (more popularly known as the “gig economy”) over the last decade. The CWS indicates that growth in alternative work arrangements has stagnated over the last decade, contradicting previous research and common assumptions. In particular:

• The prevalence of alternative work declined from 10.7 percent in 2005 to 10.1 percent in 2017;
• The annual growth rate in alternative workers declined from 2.1 percent between 1995 and 2005 to just 0.3 percent between 2005 and 2017;
• This stagnation is mainly due to the number of full-time alternative workers declining between 2005 and 2017; and
• In general, since 2005, real median weekly earnings of alternative workers performed better than real earnings of the entire employment population.

Introduction

Today the Bureau of Labor Statistics (BLS) released the 2017 Contingent Worker Survey (CWS). The CWS measures the prevalence of workers in alternative work arrangements, which is commonly referred to as the “gig economy.” This survey marks the first time BLS has conducted the CWS since 2005. Over the past few years, the assumed rise of the gig economy has captured the attention of researchers and policymakers, with many speculating about its growth. Actual data on the gig economy, however, has been very limited. While researchers have pulled from a variety of surveys and data sources to get a sense of the direction of the gig economy, none are as comprehensive or dependable as the CWS. Hence, today’s release of the 2017 CWS provides important information on the growth of the gig economy and its workers’ characteristics. In this report, we compare the results from the 2017 CWS to those from 1995 and 2005, analyzing trends in overall growth, differences between full-time and part-time workers, and growth in weekly earnings.

Total Growth in Alternative Work

Alternative work arrangements consist of independent contractors, on-call workers, temp-agency workers, and contract-company workers.

Table 1 contains each category of alternative work as a percent of total employment in 1995, 2005, and 2017.

Table 1: Alternative Work Arrangements, 1995-2017

Worker Category	1995	2005	2017
Total Alternative Workers	9.9%	10.7%	10.1%
Independent Contractors	6.7%	7.4%	6.9%
On-Call	1.7%	1.8%	1.7%
Temp Agency	1.0%	0.9%	0.9%
Contract Company	0.5%	0.6%	0.6%

Overall, there appears to be little growth in alternative work between 1995 and 2017. In 2017, 10.1 percent of all employed people fell into one of the four categories of alternative work arrangements, down from 10.7 percent in 2005 and slightly higher than 9.9 percent in 1995. Across the four categories, most of the variation occurs with independent contractors, which declined from 7.4 percent in 2005 to 6.9 percent in 2017. On-call, temp agency, and contract workers are far less prevalent and are little changed in the past two decades. Stagnation in the growth of contract company workers is particularly noteworthy, given the popularity of the so-called “fissured workplace” hypothesis, which speculates that the number of contract-company workers has grown substantially in recent decades.

The decline in the prevalence of alternative work between 2005 and 2017 indicates that the growth in alternative work has slowed considerably since 2005. Using these data and BLS total employment figures, we calculate the compounded annual growth rates of total employment, alternative work, and traditional work from 1995 to 2005 and from 2005 to 2017. Table 2 contains the results.

Table 2: Annual Growth in Alternative Work, 1995-2005 and 2005-2017

Worker Category	1995-2005	2005-2017
Total Employment	1.3%	0.8%
Alternative	2.1%	0.3%
Traditional	1.2%	0.8%

Over the past decade, growth in alternative work has stagnated. In particular, the annual growth rate of alternative work dropped from 2.1 percent between 1995 and 2005 to just 0.3 percent between 2005 and 2017. With the Great Recession occurring in this period, growth in traditional jobs slowed as well, but not by as large of a degree. Traditional employment rose 1.2 percent per year between 1995 and 2005, and then grew 0.8 percent per year going forward.

These findings in part contradict the results of several reports over the past few years. Economists Lawrence Katz and Alan Krueger estimated “a significant rise in the incidence of alternative work arrangements in the U.S. economy from 2005 to 2015.” Specifically, their survey suggested that alternative work arrangements increased from 10.7 percent in February 2005 to 15.8 percent in late 2015.

A few reasons could account for the differences between the Katz and Krueger findings and those of the 2017 CWS. On one hand, the estimates in the Katz and Krueger report are based on a survey with a substantially smaller sample size than the CWS, making their results less reliable than today’s figures. On the other hand, however, the differences could be explained by stronger economic and labor-market growth over the last few years. Research suggests alternative work may be countercyclical, growing during recessions and declining during periods of growth. While it is only a small part of all alternative work arrangements, the JPMorgan Chase Institute found that growth in the online platform economy peaked in 2014 and then fell by 2016. It is possible that stronger economic growth over the last few years could have also resulted in tapering growth in all alternative work, which could at least partly explain the difference between the 2017 CWS figures and the 2015 Katz and Krueger figures.

As AAF noted in 2017, shifts in labor patterns are sensitive to the definition of alternative work arrangements and timeframes. This 2017 study offered three different definitions of the gig economy, a much broader term for alternative work arrangements, and found that these gig economy workers accounted for between 14.1 percent to 20.5 percent of all workers in the country in 2014. Cognizant that the data was collected just as the labor market was recovering, the report suggested, “One interpretation is that workers in recent years have turned to gig economy work after losing a job and facing a decline in family income.” This study suggested a dynamic labor market, since the gig economy labor force could have grown as much as 11.1 percent or contracted by 0.02 percent from 2010 to 2014, depending on the definition.

At the most, the totality of these other sources in conjunction with this new 2017 CWS advance a counter-cyclical narrative regarding alternative work arrangements. As the economy gets worse, people look toward different forms of work, but move back into a typical 9-to-5 job when the economy recovers.

Full-Time vs Part-Time Status

The 2017 CWS indicates that the slowdown in alternative work’s growth was almost entirely caused by a decline in full-time alternative workers. Once again, we use the CWS estimates and BLS total employment figures to compare annual growth rates for full-time and part-time workers. Table 3 contains the resulting figures for alternative work and traditional work.

Table 3: Annual Growth in Full-Time and Part-Time Workers, 1995-2005 and 2005-2017

Worker Category	1995-2005		2005-2017
	Full-Time	Part-Time	Full-Time	Part-Time
Alternative	2.4%	1.3%	-0.1%	1.2%
Traditional	1.4%	0.4%	0.8%	0.8%

From 2005 to 2017, the number of full-time alternative workers declined at a 0.1 percent annual rate after growing at a 2.4 percent rate in the prior decade. Meanwhile, the annual growth rates in part-time alternative workers were similar during the two periods, decreasing slightly from 1.3 percent in the first period to 1.2 percent in the second.

The growth rate in full-time traditional work also decreased, although not as much. Annual growth in full-time traditional employment slowed from 1.4 percent to 0.8 percent. For part-time workers, traditional jobs accelerated from growing 0.4 percent annually to 0.8 percent annually.

Due to the decline in full-time alternative work, there was a noticeable shift in the distribution of full-time and part-time workers. Table 4 contains the distribution of alternative and traditional work by full-time and part-time status.

Table 4: Full-Time and Part-Time Distribution, 1995-2017

Worker Category	1995	2005	2017
Alternative
Full-Time	70.5%	72.7%	69.4%
Part-Time	29.5%	27.3%	30.6%
Traditional
Full-Time	81.7%	83.1%	83.1%
Part-Time	18.3%	16.9%	16.9%

While the distribution of traditional workers remained steady between 2005 and 2017, alternative workers became more likely to be part-time. Between the 2005 and 2017, the portion of alternative workers who were full-time fell from 72.7 percent to 69.4 percent. Meanwhile, the portion who were part-time rose from 27.3 percent to 30.6 percent.

Weekly Earnings

The CWS also measures the median weekly earnings of each category of alternative workers by full-time and part-time status. Table 5 contains the median weekly earnings of each alternative-worker category and of total employment.

Table 5: Median Weekly Earnings, 2017

Worker Category	Full-Time	Part-Time
Independent Contractor	$851	$333
On Call	$797	$229
Temp Agency	$521	$263
Contract Company	$1,077	$322
Traditional	$884	$255

In 2017, the median weekly earnings of workers in alternative arrangements varied considerably, ranging from $521 for temp agency workers to $1,077 for contract-company workers. The median weekly earnings of traditional workers, at $884 per week, fell toward the higher end of the range for alternative workers.

Part-time alternative workers saw similar variation in their earnings, as their median weekly earnings ranged from $229 for on-call workers to $333 for independent contractors. Median weekly earnings among part-time traditional workers was toward the lower end of the range for alternative workers at $255 per week.

The Consumer Price Index allows the calculation of the real (inflation-adjusted) median earnings from the previous surveys. The table below then, using these figures, presents the annual growth rate in real median weekly earnings from 1997 (the first year the CWS collected earnings data) to 2005 and from 2005 to 2017. Table 6 contains the results.

Table 6: Annual Growth in Real Median Weekly Pay, 1997-2005 and 2005-2017

Worker Category	1997-2005		2005-2017
	Full-Time	Part-Time	Full-Time	Part-Time
Independent Contractor	1.6%	1.3%	-0.5%	0.3%
On Call	0.0%	2.4%	1.6%	0.4%
Temp Agency	0.5%	2.9%	-0.1%	-0.6%
Contract Company	0.2%	2.0%	1.0%	1.8%
All Employed Workers	0.9%	1.4%	-1.5%	0.2%

Overall, real earnings growth for both full-time and part-time workers varied considerably. In general, however, since 2005 real median weekly earnings performed better among alternative workers than among the total employment population.[1]

From 2005 to 2017, real median weekly earnings of all employed full-time workers declined at a 1.5 percent annual rate. During the same period, real median weekly earnings of alternative workers ranged from declining 0.5 percent annually for independent contractors to growing 1.6 percent annually for on-call workers. Meanwhile, real median weekly earnings for all part-time workers only grew 0.2 percent annually. While real median earnings for part-time temp-agency workers declined, earnings grew faster for independent contractors, on-call workers, and, particularly, contract-company workers than for the entire part-time employment population. Of note, for most worker categories, growth in real median weekly earnings slowed between the 1997 to 2005 period and the 2005 to 2017 period. The two main exceptions are full-time on-call and contract-company workers, whose real earnings accelerated substantially.

These results are particularly informative, given that policymakers and commentators often express concern about the wages of workers in alternative work arrangements. These results also call into question the popular “fissured workplace” hypothesis that suggests workers are suffering from lower wages due to the growth of contract companies. This theory guided Department of Labor (DOL) rulemaking and National Labor Relations Board (NLRB) decisions during the Obama Administration, including the DOL joint employer administrative interpretation and the NLRB’s 2015 decision to broaden the joint employer standard in its Browning-Ferris Industries case.

Not only does this CWS indicate that the number of contract-company workers has been steady over the past two decades (as detailed in table 1), but it also suggests that these workers earn higher wages than those in traditional jobs. In 2017, median weekly earnings of contract company workers was $1,077, compared to $884 for traditional workers. Additionally, real earnings growth has been more durable for contract-company workers relative to the the entire employment population. Real median weekly earnings of full-time contract company workers accelerated from growing 0.2 percent annually between 1997 and 2005 to 1 percent annually between 2005 and 2017. Meanwhile, real median weekly earnings of all full-time workers went from growing 0.9 percent annually to declining at a 1.5 percent annual rate. Moreover, the annual growth rate of part-time contract-company workers’ real earnings only decreased slightly from 2 percent to 1.8 percent. Yet, for all part-time workers, the annual growth rate declined from 1.4 percent to 0.2 percent.

Conclusion

Over the past few years, many have speculated that alternative work arrangements, commonly referred to as the gig economy, have grown rapidly. Yet, the 2017 CWS suggests that the growth in alternative work has stagnated since 2005. The portion of workers in alternative arrangements declined from 10.7 percent in 2005 to 10.1 percent in 2017. These figures suggest that the annual growth rate in alternative work declined from 2.1 percent before 2005 to just 0.3 percent after. For comparison, the annual growth rate in traditional jobs declined from 1.2 percent to 0.8 percent. Meanwhile, earnings among alternative workers generally performed better than those for traditional workers.

These findings call into question common perceptions and previous research on alternative work arrangements. It is possible that alternative work grew in response to the Great Recession and then tapered during the recovery. The long-run trend, however, appears to have not changed in the last decade.

[1] Because the CWS did not collect median earnings among traditional workers in 2005, we are unable to calculate the shift in growth for those workers specifically.