WASHINGTON – U.S. Sens. Mark R. Warner (D-VA) and Sherrod Brown (D-OH) today urged the U.S. Government Accountability Office (GAO) to report on companies’ increased use of subcontracted and contingent workers for work that used to be performed by direct employees. This leads to the creation of ‘fissured workplaces’ – environments where primary employers outsource “non-core” business functions to subcontracted firms but still maintain tight control over the outcomes of those subcontractors. This has primarily affected occupations in payroll, accounting, janitorial services, facilities maintenance, security, food preparation, among others.

In a letter, the Senators pressed Comptroller General Gene Dodaro for more information to better understand how companies’ efforts to cut costs are affecting the long term resiliency and stability of workers, and how this trend is affecting the U.S. workforce and its paths for upward mobility. 

“We suspect that fissured workplaces could be leading to an erosion of the American social contract. Employers’ use of contract and contingent workers, including subcontracted workers, independent contractors, and temporary workers, has likely contributed to the decline in employer-sponsored training,  and led to ambiguity in terms of who is responsible for providing workplace protections and who must be at the bargaining table when workers form a union,” wrote the Senators. “The implications for fewer benefits for contract workers are far reaching: for example, studies in multiple states find that temp workers experienced twice the rate of injury as permanent workers perhaps due to the fact that they are nearly twice as likely as permanent workers to have never received safety training.”

“The rise of the fissured workplace may also partially explain the significant decline in the large firm wage premium,” they continued. “Researchers find that this change is largely a consequence of large firms with over 1000 employees no longer paying above market salaries to their workers.  What’s more, others note that this decline has occurred exclusively for those at the lower and middle of the wage distribution, with no change in the wage premium for higher income earners.”

Currently, public companies are only required to report the number of full-time workers to the Securities and Exchange Commission (SEC), but are not required to report the number of contract or contingent workers, making it difficult for lawmakers to understand the extent to which public companies are now relying on these workers in order to save costs. 

This lack of public knowledge is concerning, given the benefit and wage disparities that exist between contingent workers and company employees.

“We know that fissured workplaces can lead to compensation penalties for workers in certain subcontracted occupations. For example, by the year 2000, 45% of janitors and over 70% of security guards worked as subcontractors instead of in-house employees.  While studying this phenomenon, researchers found that contracted janitors earned about 15% less than in-house janitors and, similarly, contracted security guards earned 17% less than in-house guards.  This trend is troubling and all the more important following the onset of the COVID-19 crisis – clean and safe places of work are essential and all workers are worthy of fair and competitive compensation,” the Senators noted. “Current research suggests that workers in standard work arrangements are 10% more likely to have access to health insurance than independent contractors.  That same research suggests that over 46% of full-time employees have access to a retirement plan, while only 2.3% of independent contractors and 38.3% of workers provided by a contract firm do. “

In their letter, the Senators also expressed concern about the rise in hiring of contract and contingent workers for particularly vulnerable occupations and tasks. They specifically highlighted social media companies’ high rates of reliance on these workers for work that could have significant long-term mental health consequences such as content moderation – the process of reviewing user content for adherence to company policy, a process that entails viewing disturbing online content like graphic violence, sexual predation and assault, child pornography, or other traumatic content. 

In order to understand the prevalence and address the impacts of fissured workplaces in the economy, the Senators posed the following questions for Comptroller General Dodaro:

1. Since 1980, to what extent have companies increased their use of temporary workers, subcontracted workers, independent contractors, and franchises for work that was typically done by direct employees? Is this use of alternative work arrangements typically permanent or long-term?  
   
   
2. What are the demographic characteristics of contract and contingent workers by sector? Are there any sectors in which these employment shifts have been the most pronounced? If so, what are the most common business justifications given for the use of these alternative work arrangements in those sectors and how do those justifications compare across sectors?  
   
   
3. How do the pay, benefits, schedules, access to workforce training opportunities, and job security of employees in these different work arrangements compare to the pay, benefits, schedules, access to workforce training opportunities, and job security of employees that are direct employees of the companies? How have these differences changed over time and to what extent have they contributed to reduced paths to upward mobility in the American economy?
   
   
4. For contingent workers at large employers, how long do typical contingent workers work in each “gig” and how long does it typically take workers to find another “gig”? Are there typically pathways for workers who are temporary staff, subcontracted, or independent contractors to become full-time, direct employees of the companies? 
   
   
5. What are the estimated business savings to a company who uses temporary workers, subcontracted workers, and independent contractors?

6. What are the long-term effects of contingent work in areas such as content moderation on workers’ future employment, psychological well-being, and physical health and does this work create costs or externalities borne by state, local, or federal governments? What are some potential long-term consequences for economic security for workers in contract and contingent work arrangements?
   
   
7. Are there comprehensive federally available data to track companies’ use of alternative work arrangements? If so, what are the data sets, how often are the data collected, how comprehensive are the data sets, and what gaps remain?  Does GAO have any policy recommendations for improving data collection on these employment trends?
   
   
8. What private data are available to track these trends?
   
   
9. In the technology sector, please use any private data, including private data provided by companies in that sector, to provide additional analysis of the use of these alternative work arrangements and assessment on the impact to employees’ pay, benefits, schedules, and job security. Additionally, to what extent do workers take on contract-based work with a hiring preference for full-time direct employment and an expectation that contract-based work will turn into a direct employee opportunity? 
   
   
10. Have these types of workers faced any additional economic hurdles brought about by COVID-19, such as greater likelihood of being laid off, furloughed, or reduced access to benefits, including employer-provided health insurance?

Sen. Warner has been an outspoken advocate of investing in workers and ensuring they are adequately equipped to participate in the 21st century labor force. Earlier this year, he called on the SEC to establish an Environmental, Social, and Governance (ESG) Task Force following a GAO report that revealed investors pursue reporting of non-financial indicators covering a company’s environmental, social, and governance practices. The report was a direct result of Sen. Warner’s efforts to get more details on the extent to which firms currently report on ESG issues, and whether Congress and the SEC should act to require such disclosures. 

Last year, the SEC announced a proposed Regulation S-K rule following advocacy by Sen. Warner, who previously urged the Commission to heed the calls of investors and utilize its rulemaking authority to require companies across the board to provide further details relating to human capital management. Sen. Warner has also sent a letter requesting that the SEC require companies to disclose specific metrics in addition to human capital resources, measures, and objectives. In May, Sen. Warner along with U.S. Rep. Cindy Axne (D-IA) urged the SEC to require that human capital management information is made publicly available in a timely and accurate manner to help determine whether a company will be successfully able to weather risks following the COVID-19 crisis – a critical issue for investors and the overall economy.  

A copy of the letter can be found here and text is available below.

Dear Mr. Dodaro:            

In the past several decades, large companies have faced increased pressure from short-term oriented shareholders and executives to cut costs and increase efficiency. As the onset of the COVID-19 crisis has demonstrated, this focus on hyper efficiency has come at the cost of long term resiliency for many. David Weil at Brandeis University notes that companies have – in many cases – responded to short-term pressures by increasingly focusing on “core business” functions, outsourcing competencies perceived as less essential or valuable to subcontractors.^[i] In practice, this meant that certain tasks and occupations in payroll, accounting, janitorial services, facilities maintenance, security, food preparation, and others, would no longer be done or performed by in-house employees of the company. Instead, these tasks might still be completed by workers in the same building as core company employees, but – in reality – their compensation and benefits would be provided by a subcontractor. This type of arrangement is known as a “fissured workplace” – an environment where a primary employer outsources non-core business functions to subcontracted firms but still maintains tight control over the outcomes of those subcontractors.^[ii] It has been the subject of many articles over the years, most notably Neil Irwin’s in the New York Times titled, “To Understand Rising Inequality, Consider the Janitors at Two Top Companies, Then and Now,” outlining the stark differences in paths for upward mobility between a janitor at Kodak in the 1980s and a janitor at Apple today.^[iii] 

We know that fissured workplaces can lead to compensation penalties for workers in certain subcontracted occupations. For example, by the year 2000, 45% of janitors and over 70% of security guards worked as subcontractors instead of in-house employees.^[iv] While studying this phenomenon, researchers found that contracted janitors earned about 15% less than in-house janitors and, similarly, contracted security guards earned 17% less than in-house guards.^[v] This trend is troubling and all the more important following the onset of the COVID-19 crisis – clean and safe places of work are essential and all workers are worthy of fair and competitive compensation. We should know more about the extent to which this disparity in compensation is prevalent in other subcontracted occupations and across sectors.  

 We suspect that fissured workplaces could be leading to an erosion of the American social contract. Employers’ use of contract and contingent workers, including subcontracted workers, independent contractors, and temporary workers, has likely contributed to the decline in employer-sponsored training,^[vi] and led to ambiguity in terms of who is responsible for providing workplace protections and who must be at the bargaining table when workers form a union. Current research suggests that workers in standard work arrangements are 10% more likely to have access to health insurance than independent contractors.^[vii] That same research suggests that over 46% of full-time employees have access to a retirement plan, while only 2.3% of independent contractors and 38.3% of workers provided by a contract firm do.^{[viii]; [ix]} Additional reports from the National Employment Law Project suggest that the differences are stark for certain types of contract and contingent workers: potentially only 12.8% of temporary help agency workers, 28.2% of on-call workers, and 41.3% of workers provided by contract firms have access to employer-provided health insurance compared to a majority of direct employees.^[x] The implications for fewer benefits for contract workers are far reaching: for example, studies in multiple states find that temp workers experienced twice the rate of injury as permanent workers perhaps due to the fact that they are nearly twice as likely as permanent workers to have never received safety training.^[xi]

The rise of the fissured workplace may also partially explain the significant decline in the large firm wage premium. For most of the 20^th century, large firms paid higher wages than smaller firms, even after controlling for the quality of the worker, across many countries.^[xii] In fact, they were an important way to mitigate labor market wage inequality – less-educated workers received a higher wage premium from working for large firms than more-educated workers.^[xiii] In the past few decades, however, the average worker earnings premium for a large firm has fallen from 47% in 1980 to 20% in 2013.^[xiv] Researchers find that this change is largely a consequence of large firms with over 1000 employees no longer paying above market salaries to their workers.^[xv] What’s more, others note that this decline has occurred exclusively for those at the lower and middle of the wage distribution, with no change in the wage premium for higher income earners.^[xvi] We don’t know why this is occurring, but researchers suggest this may be a result of changing norms^[xvii] around acceptable compensation for workers at the lower end of the wage distribution and fissured workplaces.^[xviii] Since large firms are not required to disclose information about workers that are not considered employees, it is possible that the large firm wage premium is even lower than what scholars have found.

Increasingly, essential functions routinely completed by full-time employees are being relegated to contract and contingent work. Companies such as Google rely on lower-paid contractors even for software development.^[xix]According to investigative reports and accounts from those with access to internal company databases, the total number of contractors at Google outnumbered direct employees in 2018.^[xx] Those same reports suggest that, in some cases, those lower-paid contractors were performing work often indistinguishable from functions completed by direct employees^[xxi] without access to the same benefits and compensation. Contractors are also reportedly prevalent at other tech giants, such as Apple, Facebook, and YouTube, similarly without access to the same worker benefits as traditional employees.

While reports indicate that tasks typically associated with skilled labor, such as programming, are increasingly going to contractors, technology companies frequently seem to hire contract and contingent workers for particularly vulnerable occupations and tasks. At social media companies, reports find that contract workers are used at high rates for content moderation – the process of reviewing user content for adherence to company policy, which entails viewing disturbing online content that can include graphic violence, sexual predation and assault, child pornography, or other traumatic content.^[xxii] These positions frequently involve repetitive tasks, queue- and rate-based decision-making, and daily quotas.^[xxiii] While this type of work could have significant long-term mental health consequences, there are reports that moderators are often required to sign a non-disclosure agreement that prevents them from seeking outside counseling.^[xxiv] Similar to other types of subcontracted workers in other industries, reports find that content moderators suffer a wage penalty and lack the employment benefits of their in-house counterparts.^[xxv] At Facebook, for example, starting salary for a content moderator is only 14% of the median full-time employee salary.^[xxvi]

Understanding the full extent of the fissured workplace is a difficult task. Though public companies are required to report the number of full-time workers to the Securities and Exchange Commission, they are not currently required to report the number of contract or contingent workers. Often in household surveys, workers will incorrectly classify themselves as employees of the primary firm when, in reality, they work for a subcontractor or as independent contractors.^[xxvii] This misunderstanding may help partially explain why scholars note an increase in self-employment through tax filings data while standard measures of self-identification in the Bureau of Labor Statistics’ surveys have not seen an increase in contract or contingent work.^[xxviii]

For policymakers to address the impacts of fissured workplaces in the economy, we need to understand how prevalent the situation is for American workers. For example, research suggests that Black and Hispanic workers are overrepresented in contract-based work, ^[xxix] which could have economic ramifications for upward mobility and financial security. To more thoroughly study the fissured workplace, scholars call for marrying business transaction data with worker data.^[xxx] Current conservative estimates suggest around 19% of the private sector workforce operates in industries with fissured workplaces.^[xxxi]Estimates that include industries with mixed use of practices could double this figure, which would suggest that working in fissured workplaces is more common today than being a member of a union was at its peak of 34% in 1956.^[xxxii]

In light of these issues, we would like GAO to report on the following:

1. Since 1980, to what extent have companies increased their use of temporary workers, subcontracted workers, independent contractors, and franchises for work that was typically done by direct employees? Is this use of alternative work arrangements typically permanent or long-term?  
2. What are the demographic characteristics of contract and contingent workers by sector? Are there any sectors in which these employment shifts have been the most pronounced? If so, what are the most common business justifications given for the use of these alternative work arrangements in those sectors and how do those justifications compare across sectors?  
3. How do the pay, benefits, schedules, access to workforce training opportunities, and job security of employees in these different work arrangements compare to the pay, benefits, schedules, access to workforce training opportunities, and job security of employees that are direct employees of the companies? How have these differences changed over time and to what extent have they contributed to reduced paths to upward mobility in the American economy?
4. For contingent workers at large employers, how long do typical contingent workers work in each “gig” and how long does it typically take workers to find another “gig”? Are there typically pathways for workers who are temporary staff, subcontracted, or independent contractors to become full-time, direct employees of the companies? 
5. What are the estimated business savings to a company who uses temporary workers, subcontracted workers, and independent contractors? 

6. What are the long-term effects of contingent work in areas such as content moderation on workers’ future employment, psychological well-being, and physical health and does this work create costs or externalities borne by state, local, or federal governments? What are some potential long-term consequences for economic security for workers in contract and contingent work arrangements?
7. Are there comprehensive federally available data to track companies’ use of alternative work arrangements? If so, what are the data sets, how often are the data collected, how comprehensive are the data sets, and what gaps remain?  Does GAO have any policy recommendations for improving data collection on these employment trends?
8. What private data are available to track these trends?
9. In the technology sector, please use any private data, including private data provided by companies in that sector, to provide additional analysis of the use of these alternative work arrangements and assessment on the impact to employees’ pay, benefits, schedules, and job security. Additionally, to what extent do workers take on contract-based work with a hiring preference for full-time direct employment and an expectation that contract-based work will turn into a direct employee opportunity? 
10. Have these types of workers faced any additional economic hurdles brought about by COVID-19, such as greater likelihood of being laid off, furloughed, or reduced access to benefits, including employer-provided health insurance?

Sincerely,

###