In the world of paid work, employment law recognizes two types of workers: employees and independent contractors. Most workers are the former—they are hired by a company to do a set task, in a certain way, usually in a set location, and are paid an hourly wage or a salary. Other workers, though, are independent contractors—they operate like independent businesses and contract out their labor and skills to a range of companies and individuals, usually for a set or negotiated fee, and have autonomy on how, when, and where they do their work.

Unfortunately, sometimes companies hire workers that by definition should be employees, yet deliberately pay them and treat them as if they were independent contractors. This pernicious practice—called misclassification—is used by unscrupulous employers to avoid paying their share of payroll taxes and to evade certain responsibilities and liabilities covered by employment law. In some industries, misclassification has become so extreme that it has gone underground, with employers not even formally classifying many workers even as independent contractors, but rather paying them through cash-only, off-the-books transactions or payments to intermediaries and shell companies.1

The costs of worker misclassification and off-the-books payments are enormous, particularly in the construction industry. This report finds that, nationally, 1.1 to 2.1 million construction workers are estimated to be misclassified or paid off-the-books, and there are reasons to believe this may undercount the true extent of the problem. These workers lose key legal rights when employers misclassify them or pay them off the books. Moreover, misclassification and off-the-books payments have significant costs to individual workers through lost worker’s compensation insurance, unemployment insurance (UI), and lost overtime pay. State and federal government social insurance programs also rely on employer taxes for funding. By misclassifying workers or paying them off the books, employers avoid paying their share of taxes to programs such as Social Security and unemployment insurance. This report estimates that unscrupulous employers in the construction industry that misclassify or resort to off-the-books payments are underpaying workers and shortchanging payments toward legally required benefits (such as Social Security, unemployment insurance, and workers’ compensation) by over $12 billion per year, costing taxpayers between $5 and $10 billion per year.

This report looks at the problem of worker misclassification and off-the-books payments in the construction industry, and is the first to evaluate the level and costs of misclassification in the construction sector for all fifty states and the District of Columbia. It begins with an overview of how and why companies pursue misclassification and off-the-books payments before quantifying the extent of malfeasance that occurs and the damage it does. This report then covers progress being made against worker misclassification and off-the-books payments as well as steps for future action.

Overview of Misclassification and Off-the-Books Payments in the Construction Industry

The term misclassification makes the false categorization of employees as independent contractors sound like a benign problem or the result of a harmless and unintentional clerical error. But worker misclassification is hardly benign and it’s often quite intentional. A business’s decision to falsely employ someone as an independent contractor is often a conscious decision to break—or conveniently ignore—labor law to absolve the employer of their tax burden and all the additional costs of hiring.

Unfortunately, misclassification is an especially egregious illegal labor practice that has far-reaching consequences. Most directly, it denies workers their rights under state and federal law, including the right to overtime pay, minimum wage, certain workplace safety protections, paid family and medical leave and other mandated benefits, a discrimination-and-sexual-harassment-free workplace, and collective bargaining. Workers misclassified as independent contractors are also denied access to critical social insurance benefits, such as workers’ compensation insurance coverage if they are hurt on the job and unemployment insurance benefits if they are laid off. And their employer’s decision to misclassify them means that the business effectively pushes their Social Security and Medicare tax burden—the “employer share”—from their books onto the backs of the workers themselves. Beyond this, the use of off-the-books payments removes workers entirely from many of the rights and benefits that employees or independent contractors have. Meanwhile, law-abiding businesses often find it difficult to compete with unscrupulous employers that reduce their labor costs substantially via illegal means, and taxpayers are out billions of dollars due to what is essentially an implicit subsidy of public money to employers that knowingly violate the law.

The issue of worker misclassification has received significant notoriety in recent years, given the rise of app-based companies such as Uber, Lyft, DoorDash, and the like. These companies—and many other titans in the “gig” economy—have hired the workers powering their businesses as independent contractors even though these companies set or control nearly all of the terms of the work, including assignments and pay. Some state and city agencies have sued app-based companies for misclassifying their workers as independent contractors. In response, these companies are lobbying in states and in Congress to enshrine their workers as independent contractors under state and federal law, which would mean these workers would be legally excluded from the rights, benefits, and protections they would be entitled to as employees. But the public attention paid to presumed misclassification in platform-based work has caused many to overlook its pervasiveness in what one scholar has dubbed “the original gig economy”: the construction industry.2

For Many Construction Companies, Worker Misclassification and Off-the-Books Payments Are the Business Model

Decades of reporting from across the country suggest that the U.S. construction sector is awash in worker misclassification.3 But the problem goes even deeper than the decision of whether to hire workers as independent contractors rather than employees. Instead, a substantial portion of the construction workforce now operates in the underground economy, with many employers hiring them through cash-only, off-the-books employment relationships that cast aside any pretense of the formality (or documentation) of traditional independent contracting. This process of misclassification and off-the-books payments is aided by the presence of “labor brokers,” who are effectively unregistered middlemen capable of recruiting and employing legions of construction workers on a jobsite on short notice and who operate in the world of cash payments and check-cashing services.

Tom Juravich, a professor at University of Massachusetts–Amherst, conducted several in-depth interviews with misclassified workers on construction sites in Massachusetts. One interview with a construction worker, Nuno, succinctly explains how off-the-books payments are facilitated by labor brokers:

In terms of the labor brokers who worked with Metro Walls [the subcontractor], Nuno reports that the workers on the job knew nothing about them. “A first name, and where to meet him so you could get paid, sometimes not even the real name, it’s a nickname. I’ve talked to workers that have been working for the same guy for almost two years, and all they know about him is his name and that he pays them every week.” Carlos is a young carpenter who got his start hanging drywall in the non-union residential housing industry. When it came time to be paid, he tells us, “They would either come to our house or we’d go to their house or meet somewhere. It would be the most underground thing ever. Always cash.” When asked if he ever saw the labor broker on the job, he replied, “Never, they’re never there.”4

Many construction workers operating in this underground cash economy are low-paid immigrant workers. This is unsurprising, considering that an estimated 23 percent of the construction industry is composed of undocumented workers.5 Since undocumented workers are less likely to report labor violations due to their status and fear of retaliation, certain segments of the construction industry have become notorious for rampant worker misclassification and exploitation.6

The general disregard for labor law in many parts of the U.S. construction sector has led one observer to deem the industry “the Wild West.”7 But misclassification and off-the-books employment often puts workers in incredibly precarious positions. Construction is a notoriously dangerous vocation, and workers are regularly injured on the job. Absent the security provided by employer-sponsored health insurance or workers’ compensation coverage, injured workers and their families too often must rely on public services for medical care and financial survival. A 2022 study from the UC Berkeley Center for Labor Research and Education found that not only are construction workers three times more likely to lack health insurance than all workers, but their families are also more likely to rely on public safety net programs due to the deterioration of job quality in the industry.8 In sum, the study estimated that 39 percent of families of construction workers are enrolled in at least one safety net program, costing taxpayers nearly $28 billion per year.

The cash-only nature of employment has also made the construction sector a hothouse for wage theft. In many cases, this means requirements that workers operate “off the clock.9 In other cases, workers often report working for weeks only to have the employer or labor broker simply refuse to pay the promised amount of money.10 In extreme cases, it is not unheard of for the employer or labor broker to simply skip town with their workers’ money, never to be heard from again. And all of this occurs as law-abiding employers struggle to remain competitive in some corners of the industry, and taxpayers implicitly subsidize the unscrupulous employers making millions through their unethical and illegal actions.

The Extent of Misclassification and Off-the-Books Payments in the Construction Industry

The persistence of misclassification and off-the-books payments as a way to commit payroll fraud in the construction industry occurs for numerous reasons including, but not limited to, under-resourced enforcement agencies, insufficient penalties, the use of shell companies and other complex schemes by employers to avoid detection, increased subcontracting and the absence of joint liability affecting developers and general contractors, and legislative aversion to workers’ rights issues.11 But one significant obstacle to public policy initiatives has been the difficulty in quantifying the incidence and economic costs of worker misclassification. In other words, motivating political action requires assessment of how many construction workers are directly affected and how much the practice of misclassification harms workers, taxpayers, and law-abiding contractors.

The most direct evidence of the extent of worker misclassification in the skilled construction trades comes from the publication of the results of employer audits conducted by a state government’s labor agency in their oversight of state unemployment insurance programs. These audits explicitly attempt to identify employers’ actions to misclassify workers and underreport worker income to state agencies. For example, two recent reports examined UI audit data provided by state agencies to conclude that 8.4 percent of the construction workforce in Rhode Island was misclassified, while that number was 6.6 percent in Massachusetts.12 Meanwhile, reports in these states and elsewhere across the country in the 2000s and 2010s regularly suggested that worker misclassification in the construction industry was well above 10 percent of the sector’s workforce.13

Presumed declining rates of worker misclassification during a time when industry stakeholders are nearly unanimous that the problem is worsening—a conclusion recently echoed by the U.S. Treasury Office—underline the weaknesses of UI audits and why they, at best, represent lower-bound estimates of payroll fraud.14 Most prominently, state departments of labor only audit employers who are registered in the state’s UI system, which means that the influence of unregistered labor brokers is systematically overlooked. Operating as employment “middlemen,” labor brokers are capable of recruiting vast numbers of workers to a construction site and often compensate these workers via cash or check-cashing services; their role in the industry has grown to the point where the state of Pennsylvania called them the “key contributors to the problem of misclassification.”15 In effect, as misclassification in construction has increasingly gone underground—from simply misclassifying employees as independent contractors (think 1099-MISC forms) to paying them through cash-only, off-the-books payments to workers and legitimate-looking payments to intermediaries and shell companies—it has become harder and harder for audits to identify the paper trail that leads to findings of worker misclassification.

How bad could it be? Surveys that explicitly query workers about their employment relationship on construction jobsites offer some perspective. A 2021 study published by the Catholic Labor Network demonstrated that nearly half (47 percent) of the seventy-nine workers surveyed on various large commercial or public construction sites in Washington, D.C., were either paid in cash or via a check without payroll deductions taken out.16 These results echo three older but much larger jobsite surveys administered by the Workers Defense Project. Surveying hundreds of workers, researchers discovered a 32 percent misclassification rate across six Southern cities (2017 report), a 41 percent rate on jobsites across Texas (2013), and a 38 percent rate in Austin, Texas (2009).17 While these studies offer similar conclusions—that payroll fraud is rampant in the construction industry—researchers cannot be sure that the jobsites, workers involved, and jurisdictions chosen for the surveys are representative of the larger industry, making it difficult to confidently project regional or national totals from these subsets of workers.

Three other studies help provide a potential baseline. A 2002 study by an economist with the U.S. Census Bureau matched data from household surveys to income records with the Social Security Administration (data that is unavailable to nearly all researchers) to discover that 6.7 percent of workers across all industries were likely misclassified, including working off-the-books, in the mid-1990s, with double-digit rates among many construction occupations (such as carpenters, laborers, painters, and carpet installers).18 Relying on unique data sets, a pair of recent studies by scholars at the University of Maryland and the Upjohn Institute suggested that between 10 percent and 14 percent of people who claim to be an employee on household surveys across all industries turn out to later identify as independent contractors or freelancers, given sufficient follow-up questions.19 While inconsistent answers on surveys do not prove those workers to be misclassified, the reports offer a benchmark for the potential scope of the problem across the entire U.S. labor market.

As notable as these studies are in identifying worker misclassification, they also highlight researchers’ frustration with available data sources to accurately evaluate its extent. Unlike most outcomes in the labor market, there are no publicly available data sources that provide direct evidence of misclassification, and other avenues to generate such data feature significant obstacles (such as the need for government cooperation, or the huge costs of conducting large and representative jobsite surveys). This leaves researchers without great options to directly measure the extent of misclassification in a state or region. As a result, numerous researchers have developed indirect methods of estimating the full extent of worker misclassification and off-the-books employment in the construction industry.

While indirect approaches to estimate misclassification in the construction sector differ slightly, many start with two numbers that are readily available: the total number of workers in the industry (via the American Community Survey (ACS) administered by the Census Bureau) and the total number of legal wage-and-salary workers (via unemployment insurance records). The difference between these two numbers represents the number of self-employed workers in the industry; that includes (a) the legally self-employed, (b) workers that should be considered employees but are misclassified as independent contractors, and (c) those working in the underground construction economy. As summarized in a 2020 study by the Institute for Construction Employment Research (ICERES), there is no perfect way of distinguishing between these groups in the data, causing scholars to have to estimate the proportion of this group that are misclassified or working off-the-books; this introduces uncertainty and, thus, a margin of error to the estimates.20

The analysis done for this report uses estimates of income underreporting to tax agencies by self-employed construction workers as a proxy for the extent of illegal employment, an approach fully described in the 2020 ICERES report and a follow-up 2021 ICERES report on worker misclassification in Massachusetts.21 Using this methodology results in a projection that there were likely between 1.1 and 2.1 million construction workers that were either misclassified or working in the underground economy on a national basis in 2021; that equates to between 10 percent and 19 percent of the industry workforce. (See appendix A for more detail.)

While this method involves some margin for error, these projections are validated by the fact that they are (a) between the lower-bound results from UI audit studies and the presumed upper-bound estimates from jobsite surveys in Washington, D.C., Texas, and Southern urban areas; (b) consistent with the results of state-specific studies that used slightly similar methodologies;22 (c) supported by the national results offered by alternative methods explored by the authors of this report;23 and (d) in line with other proposed methods of identifying misclassified workers in the industry.

As an example of the last point, researchers at the Center for New York City Affairs contend that the proportion of construction workers who identify as “unincorporated, self-employed” on Census surveys is another good proxy for misclassification.24 In the 2021 American Community Survey (ACS), this amounted to 14.7 percent of the national construction workforce, which is directly in the middle of this study’s projected range.25

Despite this validation, the authors have reasons to believe that the estimates for this report possibly—if not likely—undercount the extent of worker misclassification in the construction industry. Many of these reasons are methodological. In the absence of direct evidence of misclassification, the authors chose to be conservative in how they estimated the extent of illegality whenever possible. However, less conservative interpretations of the data from the American Community Survey—which was used as the basis for estimating total employment in the construction industry in this study—would support much higher estimates of the industry’s misclassification rate. This includes, but is not limited to, emerging academic research suggesting that the ACS is possibly, if not likely, (a) undercounting total employment due to respondents’ failure to provide a full accounting of their work situation to survey takers and (b) undercounting the number of immigrants from Central America in the United States. A full discussion of these issues and their effects on the estimated industry misclassification rate is offered in Appendix A.

Finally, it is important to note that the methodology used for this report is an attempt to estimate misclassification and off-the-books payments in the construction industry, which includes both blue-collar tradesworkers as well as white-collar professionals and support staff. As such, these results should be seen as lower-bound estimates of rates for tradesworkers, given the presumption that misclassification occurs more frequently with blue-collar workers than other workers in the industry. Even within the skilled trades, there are huge disparities in the extent of illegality. For example, while there are reports of misclassification and off-the-books payments affecting every type of construction (industrial, commercial, and so on) a 2021 study revealed that worker misclassification and off-the-books payments in the construction industry is more heavily concentrated in residential construction and in four key occupations: painters, laborers, roofers, and carpenters.26 It would not be surprising if rates of misclassification in these trades, in some regions, were found to be multiples of the estimated national industry average.

Map 1

Worker misclassification and off-the-books construction employment is a widespread national problem. But the issues in the industry are more extensive in some regions of the country. As highlighted in Map 1, analysis for this report finds that misclassification rates are highest in the South and the Northeast. Higher rates in the South are unsurprising, given previous studies by the Workers Defense Project, relatively weaker labor rights policies, and higher volumes of vulnerable workers on construction sites; misclassification rates are especially high in Mississippi, Arkansas, Georgia, Tennessee, Texas, and Alabama.

Higher rates in the Northeast have been less documented in the research, but are likely due to a number of factors, headlined by greater urban density in the region. Another reason is likely the cost of workers’ compensation insurance. As misclassification is fundamentally a process to evade required taxes and insurance, it is revealing that the three Northeast states (Vermont, Connecticut, and New Jersey) with the highest estimated rates of misclassification are also in the top five in the country in terms of the cost of workers’ compensation insurance premiums. This result is unsurprising, given (a) the high price of this insurance due to the dangers of construction work and (b) research on the connection between payroll fraud and high workers’ compensation insurance premium costs established in a 2007 report by University of California-Berkeley researchers.27 (See our section below on Progress on Fighting Misclassification to learn more about what states and Congress can do to curb this practice.)

Some of the state-to-state variability in rates of misclassification and off-the-books payments may also be due to data limitations. First, the reliability of state estimates is lower in small-population states due to smaller sample sizes in those states; this may explain there being small-population states with unexpectedly high (for example, Vermont) and low (for example, Utah and Nevada) estimated rates of misclassification. Second, a 2022 study published in the Journal on Migration and Human Security suggests that the American Community Survey–the primary data source used in this study–may suffer from substantial undercounting of immigrants, especially from Central America and Cuba. Considering the composition of the U.S. construction workforce, extensive undercounting of immigrants from these regions (and others) in the ACS would mean that the estimates of total employment–and thus, the number of misclassified workers–presented in this study are likely too low.28 This would be true on a national basis, and would be especially informative in suggesting that misclassification rates may be especially too low in states (e.g., Florida, Texas, California) with relatively large immigrant populations from these regions of the world.

Finally, estimates are based on the location of the headquarters of the employer, not the actual jobsites where workers perform their labor; this is unavoidable, due to limitations in the data. Given that many construction workers travel long distances to a jobsite—frequently crossing state lines—this can influence rates. For example, numerous reports have identified high rates of illegal labor practices on construction sites in New York City; if such workers are employed by companies based in neighboring Connecticut and New Jersey, this would also partially explain the relatively high misclassification in those states.29 Similarly, if a substantial portion of construction in Las Vegas is completed by misclassified workers hired by companies based in California and Arizona, this would partially explain the unexpectedly lower misclassification rates in Nevada.

The Cost of Misclassification and Off-the-Books Payments in the Construction Industry

The economic costs of worker misclassification and off-the-books payments are significant, and they ripple through society in ways both seen and unseen. The direct costs of misclassified employment relationships are incurred by the workers themselves who are denied their rights to overtime pay, workers’ compensation, and unemployment insurance, and who absorb the tax burden of the “employer share” of Social Security and Medicare passed to them by the selfish actions of their employer.30 The costs are also incurred by taxpayers, as the immediate act of misclassification leads to underfunded unemployment and workers’ compensation funds, lost income taxes and Social Security and Medicare funds resulting from the lack of automatic payroll deductions, and, ultimately, income underreporting on the part of some workers.

Estimating the direct costs of worker misclassification in construction is complicated by the fact that there is no reasonable way of accurately estimating the amount of money changing hands—via cash and check-cashing services—in the underground economy. Instead, this study evaluates costs by considering a per-worker cost of misclassification—based on what affected workers would have been expected to earn had they been legally employed—and then multiplying that cost by the number of affected workers from each state identified earlier in this report; for simplicity, this study uses the midpoint of the employment estimates for each state identified in the method described in the previous section. As misclassified workers are often—but not always—concentrated in lower-paying trades, the authors assumed that misclassified and off-the-books workers would have earned the equivalent of the twenty-fifth percentile of earnings of legally employed tradespeople in each state’s construction industry.31 On a national basis, that equates to expected earnings as a legal employee of $39,720 in 2021.

Using national data on overtime usage and state-specific data on tax rates, unemployment insurance, and workers’ compensation insurance premiums, the aggregate costs of worker misclassification and off-the-books employment in the construction industry for 2021 are presented in Map 2.32 Considering that worker misclassification and off-the-books employment allow contractors to evade the legal and tax obligations of regularized employment, the results suggest that, on a national basis, these illegal labor practices result in workers losing $1.9 billion in overtime pay, $5.0 billion in workers’ compensation premiums being unpaid to insurers, and state unemployment insurance funds losing $791 million in 2021. Further, misclassification allowed employers to offload their required share of Social Security and Medicare taxes—amounting to an estimated $5.1 billion—onto the backs of workers. Altogether, construction employers are projected to have reduced their aggregate payroll by $12.8 billion using these assumptions in 2021.

Map 2

Download detailed state-by-state cost estimates here.

While this number is large, there are reasons to believe that these estimates are reasonable, if not conservative. First, a 2022 report by the Coalition Against Insurance Fraud contends that employers annually dodge $25 billion in workers’ compensation insurance premiums.33 The construction industry accounting for roughly 20 percent of this total would be unsurprising, and a 2021 ICERES paper in Massachusetts offers evidence that the true industry proportion of unpaid premiums may be considerably higher.34 Second, considering the sheer size of America’s construction industry as measured by its legal payroll, projections that worker misclassification and off-the-books payments shaves off $12.8 billion seem potentially conservative, given the extent of the practice in the industry that reveals itself statistically and in our conversations with stakeholders from coast to coast. As such, while the underground nature of the practice makes it difficult to prove, it is possible the true extent of the costs are higher than provided in this study.

An underappreciated effect of worker misclassification is that employers are not required to deduct taxes and social contributions from the paychecks of independent contractors. Combined with illegal hiring that occurs entirely off-the-books, these actions lead to income underreporting on state and federal tax filings. Applying a range of estimates on income underreporting, this study projects that Social Security and Medicare lost between $3.3 billion and $6.6 billion in 2021 due to underreporting by construction workers made possible by unscrupulous employers. Finally, using a set of simplistic assumptions in the absence of worker-level data, the results suggest a loss of between $1.3 billion and $2.5 billion in federal income taxes and $486 and $973 million in state income taxes in 2021; given the conservative assumptions, it is likely that both sets of numbers are considerable underestimates of the true tax loss that results from misclassification in construction.35 Predictably, Map 2 reflects that states with larger populations, higher rates of misclassification, and higher required tax contributions of employers feature higher costs.

It is acknowledged that cost estimates involve some margin of error given that the methodology is based on assumptions of what workers would earn in the legal labor market and not what they actually earned via illicit cash or funneled check-cashing services. But if anything, these estimates are very likely to undercount the total costs of misclassification and off-the-books payments, which includes boundless indirect costs that are difficult to quantify. For example, workers operating on a cash-only basis are more vulnerable to wage theft, or the explicit nonpayment of wages. But estimates of wage theft on a per-worker basis are limited.36 Misclassified workers who get hurt on the job or go without employment for a spell—both common occurrences in the construction industry—often must turn to public assistance programs for medical costs and financial survival.37 Unfortunately, it is not clear what proportion of construction workers on public assistance are misclassified versus those who are legally employed but simply lowly paid. And all of this ignores the psycho-social strain on misclassified workers and their families in the absence of security deriving from regularized employment and the rights and benefits that accompany it.

Other important indirect costs are those imposed on law-abiding contractors. The actions of unscrupulous contractors to misclassify workers and thus reduce labor costs help them win bids on construction projects. These winning bids can come at the expense of fair-minded contractors, which often puts these honest employers between a rock and a hard place: match their competitors’ actions to reduce labor costs, or go out of business.

One such law-abiding contractor explains this problem clearly:

Contractors using this business model can always submit a lower bid than a law-abiding contractor while knowing they will still pocket enormous profits. This is because in developing their bids these contractors do not worry about how many hours the workers will labor to complete the project or whether they will work over forty hours a week to get the job done on time. They do not worry about the costs of workers’ compensation or unemployment insurance…These contractors do not pay one penny more for the work when the men and women doing it labor over 40 hours a week to finish on schedule.38

But, while there is plenty of anecdotal evidence, like the quote shared above, generating reliable estimates of the revenue lost by law-abiding contractors at the hands of employers engaging in illegal labor practices is not possible with current data available to researchers.

Worker misclassification also leaves social insurance programs vulnerable and underfunded. For state unemployment insurance programs this translates to trust funds that are unable to provide the necessary benefits in future economic downturns. In 2023, only thirteen states met the U.S. Department of Labor’s recommended funding levels for unemployment insurance trust funds. Thirty-two states were deemed to be solvent but would be unable to pay out benefits in the event of a recession or severe economic downturn, and five states were considered insolvent.39 The concerning solvency of state UI trust funds comes despite the influx of federal pandemic relief funding which many states used to supplement their depleted trust funds after the spike in unemployment during the COVID-19 pandemic.40 Without historical investment from the federal government, state programs would be in much worse shape due to the underfunding of UI trust funds that is in some part due to worker misclassification.

Along those same lines, unpaid workers’ compensation insurance premiums affect more than workers and insurance carriers: they also lead to underfunded state programs. Many states collect surcharges from insurance carriers based on the amount of workers’ compensation insurance premiums they collect. These monies fund administration and programming at the state level, often centered on workers’ compensation “trust funds” that exist to provide insurance benefits to workers who are injured on the job but whose employer did not have valid insurance coverage. As a result, the billions of dollars in unpaid premiums projected in this study means that state trust funds and other programs are losing out on millions of dollars of revenue annually because of fraud in the construction industry. The outsized funding loss attributable to the actions of unscrupulous construction employers is especially painful considering that construction workers are disproportionately the ones seeking assistance from state trust funds; for example, a 2021 report in Massachusetts highlighted that 47 percent of approved trust fund cases in the state were paid out to injured construction workers.41

Finally, worker misclassification—in construction and elsewhere in the economy—is typically an overlooked reason for the declining revenues of the Social Security and Medicare programs in the United States. Automatic payroll deductions are the law of the land when a company hires an employee, and this ensures that Social Security and Medicare taxes—from both the employer and employee—are paid in full. But worker misclassification removes these automatic payroll deductions from workers’ paychecks; instead, workers are responsible for both shares of the tax when filing their income taxes every April. But IRS research reflects that many self-employed workers—especially those working in the cash economy—do not fully report their income (and may not report income at all).42 As a result, worker misclassification across all industries effectively opens the door for both employers and workers to evade their responsibilities under the law, leading to massive shortfalls in Social Security and Medicare funding as highlighted in this report.

Progress on Fighting Misclassification and Off-the-Books Payments

The employer–employee relationship is foundational to individuals’ access to bedrock rights and protections in the workplace. Nearly all workers’ rights and protections—including the right to minimum wage and overtime, the right to a discrimination and harassment free workplace, the right to collectively bargain, the right to have union representation, the right to family and medical leave—apply only to employees, not independent contractors. Similarly, employers are only required to pay employment taxes that fund social insurance benefits like Medicare, Social Security, and unemployment insurance pursuant to the Federal Insurance Contributions Act (FICA) and the Federal Unemployment Tax Act for their employees. Independent contractors, in contrast, pay both the employer and employee share of these taxes. The Fair Labor Standards Act (FLSA), for example, establishes employees’ fundamental right to minimum wage and overtime pay for working more than forty hours a week. A key component underpinning these rights established by the FLSA is the definition of an “employee.” Employers do not need to provide these rights to hired independent contractors.

The U.S. Department of Labor’s Wage and Hour Division (WHD) has long been tackling the issue of worker misclassification. In 2015, the WHD interpreted the definition of employment under the Fair Labor Standards Act broadly, both textually and in the context of economic reality tests that courts often use to determine whether a worker is an employee.43 More recently, in October 2022, the U.S. Department of Labor proposed a rule to provide guidance to employers under the Fair Labor Standards Act to determine who is an employee (and therefore covered by the FLSA) and who is an independent contractor (and is not).44 The proposed rule comes after a 2021 rule on the same topic that was considered inconsistent with the FLSA and case law.45 The 2022 guidance proposes a six-factor “economic reality” test that gets to the central issue: Is the worker in business for themself, or do they depend on finding work in the business of others? The Department of Labor’s proposed final rule would clarify how each factor should be considered and would provide clarity and focus to help workers and businesses know who is considered an employee or an independent contractor. Thus, the rule would clarify that 1.1 million to 2.1 million construction workers estimated to be misclassified or paid off the books in this study should be covered by the Fair Labor Standards Act and therefore deserve the same minimum wage and overtime rights that any other employee in the United States has under the FLSA.46

Additionally, whether a worker has the right to organize and be represented by a union in her workplace depends on whether she is a covered employee under the National Labor Relations Act. Independent contractors, by contrast, have no right to organize a union and could be in violation of antitrust law if they engage in concerted action with fellow independent contractors to set the price of or withhold their labor.

Over the past several years, Congress has repeatedly introduced the Protecting the Right to Organize (PRO) Act, which focuses on strengthening the right to organize under the National Labor Relations Act (NLRA). The bill revises the definition of employee under the NLRA by adopting the clear and broad ABC test.47

ABC tests use three criteria to determine whether a worker is an employee or not. A worker is presumed to be an employee unless the hiring entity/employer can show all three factors are met:

  • (A) The work is done without the direction and control of the employer.
  • (B) The work is performed outside the usual course of the employer’s business.
  • (C) The work is done by someone who has their own, independent business or trade doing that kind of work.48

Currently, twenty states and the District of Columbia have an ABC test that covers construction workers to determine who is an employee under state employment laws, such as wage-and-hour laws or unemployment insurance laws. The states that use ABC tests are: Alaska, California, Connecticut, Delaware, the District of Columbia, Hawaii, Illinois, Indiana, Maine, Maryland, Massachusetts, Nebraska, Nevada, New Hampshire, New Jersey, New Mexico, New York, Vermont, and Washington.49 Some states’ ABC tests, such as New York’s, only cover certain industries (such as construction), while other states like Massachusetts, have more expansive ABC tests that cover workers regardless of industry.50

However, there is much more that states, the Biden administration, and Congress can do. First, Congress should pass the PRO Act to ensure that the NLRA is interpreted expansively so that all working people—including workers improperly classified as independent contractors—have the right to organize and join a union. Second, the U.S. Department of Labor must finalize its independent contractor rule so that all individuals working for someone else are covered by our bedrock federal minimum and overtime wage requirements. Congress should also eliminate “safe harbors,” an IRS exemption that enables worker misclassification for employers that operate in industries with a long-standing practice of using independent contractors.51 Lastly, misclassification itself is not currently a violation of the law, only the improper tax and labor treatments that result from it. As misclassification is a gateway to denying workers the full gamut of their rights, enforcement agencies should not have to wait for the many consequent labor violations to ripen before acting. A promising development on this front is a complaint filed against a company last year by the National Labor Relations Board (NLRB) General Counsel alleging that the act itself of misclassifying employees as independent contractors constitutes a violation of the NLRA, as the misclassification chills the exercise of employee rights by misinforming employees about the scope of their rights. But whether this interpretation of the statute holds will depend upon the Board and courts.52 Federal and state legislators should therefore carefully consider policies that explicitly prohibit worker misclassification—thereby allowing enforcement to act promptly—with penalties accruing to willful or repeated violators.

Enforcement is another critical component to ensure that employers are following the letter of the law as it exists now. Sufficient funding for the IRS, the U.S. Attorney’s offices, and the federal Wage and Hour Division is crucial for these agencies to be able to investigate incidences of worker misclassification. State enforcement agencies, from state Wage and Hour Divisions to prosecutors and district attorneys offices, also need sufficient funding to investigate these issues at the state level. Vastly increasing the penalties to employers for misclassification is also critical to reducing employer’s financial incentive to misclassify workers.53

Expanding joint employer liability, in particular at the state level, is key to ensuring that all employers that may be involved in a contract and subcontracts are held accountable for misclassification.54 Joint employment refers to instances where multiple employers are responsible for paying worker wages. Joint employer liability is especially critical within the context of the construction industry because of the prevalence of subcontracting for different elements of construction projects. Establishing joint employment liability ensures that all relevant employers, from general contractors to subcontractors, are held liable for violations of the FLSA. Relatedly, states must adopt joint and several liability for upper-tier contractors for the wage theft committed by subcontractors. Doing so would help ensure that the general or higher-tier contractors that hire subcontractors—who are often the ones misclassifying workers—are jointly liable for misclassification and are therefore incentivized to hold subcontractors accountable for labor violations. Thus, expanding joint employer liability and adopting joint and several liability is essential to ensuring that all responsible actors are held accountable for illegal worker misclassification.

Effective enforcement also depends upon the willingness of workers to file complaints and participate in investigations. Given that many misclassified workers are also undocumented, it is critical that these workers be encouraged to blow the whistle or speak to investigators to support enforcement efforts. Instead, these workers are vulnerable to employers’ threats of deportation if they complain. Federal agencies have taken important steps to address this problem. Comprehensive immigration reform that can lead to normalized status for undocumented workers would directly address this disincentive to whistleblowing. In the meantime, federal agencies have taken important steps to ameliorate the problem. In May 2022, the General Counsel’s Office of the National Labor Relations Board issued a memorandum advising field offices to inform witnesses that immigration status is irrelevant to NLRA violations, that information that witnesses provide is protected, and complainants and witnesses can ask the board to seek immigration relief to protect witnesses participating in a case before the NLRB or exercising their rights under the NLRA.55 In January 2023, the U.S. Department of Homeland Security announced a streamlined process whereby undocumented workers may ask authorities to defer enforcement action on their immigration status if they are participating in an investigation of labor law violations or are victims of such violations.56 The U.S. Department of Labor has also issued additional guidance about the deferred action process in recent years.57 Importantly, the DHS deferred-action process is available to workers involved with state and local labor law agencies, as well as federal ones. Whether workers are made aware of this protection from employer retaliation and whether agencies support the workers’ applications for deferred action will help determine whether this option improves enforcement efforts against worker misclassification.

Looking Ahead

When corporations misclassify their workers as independent contractors, they illegally evade their obligations to these workers and to the public. But while public attention has been trained on the political actions by Uber, Lyft, and other technology-based labor platforms to treat their workers as independent contractors for company gain, it has ignored that the original gig economy—the construction industry—is awash in misclassification and off-the-books employment. While there are many reasons to presume that the estimates in this report are conservative, the incidence and cost of misclassification—to both workers and taxpayers—are enormous and require urgent policy action. Policy gains are critical to ensuring employers pay their fair share into social insurance programs and for building a vibrant middle-class by securing blue-collar workers better pay and their legally earned rights under American labor law.

Acknowledgments: The authors would like to thank the United Brotherhood of Carpenters and Joiners of America for funding this project. The authors would also like to thank David Weil, Laura Padin, and James Parrott for their input and review.

Appendix A: Estimating the Number of Misclassified Workers by State

This report draws directly from the indirect method of estimating the number of misclassified workers originally described in a 2020 ICERES report and updated in a 2021 ICERES study of misclassification in Massachusetts.58 Like many indirect methods of estimation, it relies on two basic elements. First, the American Community Survey—a large, nationally representative household survey administered by the Census Bureau—offers precise estimates of the total number of workers operating in the construction industry regardless of their employment type; in 2021, that amounted to 11,061,935 workers, after some minor adjustments.59 Second, unemployment insurance filings by employers available from the Bureau of Economic Analysis establish the total number of legal wage-and-salary employees (think W-2) in the sector; that represented 7,598,000 workers in 2021 after similar adjustments.

The difference between these two numbers represents the number of self-employed workers in the industry; in 2021, that amounted to 3,336,945 workers nationally. This group includes (a) the legally self-employed, (b) workers misclassified as independent contractors and (c) those working in the underground construction economy. This step is important in quantitatively narrowing the pool of workers who may be in a fraudulent employment relationship. Separating the legal from the illegal at an aggregated level, however, is the fundamental challenge, as there is no perfect way of distinguishing affected individuals in household surveys or employer records; this introduces some degree of uncertainty into the projections.

In the absence of direct evidence, researchers have differed slightly in their methods to distinguish the “legal” from “illegal” self-employed in this group and defensible arguments could be made for different approaches given this uncertainty. The method advanced in the 2020 and 2021 papers by ICERES—headlined by scholars from Allegheny College, Michigan State University, Harvard University, and the University of Massachusetts—uses estimated rates of income underreporting by self-employed construction workers on tax filings as a conservative proxy of the amount of illegal employment amongst this group.

While there is a range of potential estimates of tax evasion of self-employed workers from numerous studies, this report relies on two key endpoints.60 First, a 2021 study revealed that 32 percent of non-employee compensation listed on 1099-MISC forms provided by Massachusetts construction employers to the state’s residents was not reported on state income tax filings that year.61 Second, a 2016 IRS report highlighted that 64 percent of nonfarm proprietor income—which is subject to “little to no information reporting”—is unreported on tax forms across all industries, a situation best resembling workers in the underground construction economy.62 Applying this range (32 percent to 64 percent) to the number of self-employed construction workers identified earlier results in estimates of 1,101,192 to 2,135,645 misclassified workers in the sector on a national basis in 2021.63

The authors also explored other statistical approaches. For example, instead of lumping all self-employed workers together, the authors separately added two numbers. First, they started with the entire difference between the number of construction workers who identify as “employees” and the actual number of employees reported in UI records; these workers are very likely to be misclassified. Second, they added a fraction of the construction workers who identified themselves to be “unincorporated, self-employed” based on the same income underreporting rates used above. On a national level, the results were nearly identical to those found in the approach ultimately used in this study. Nevertheless, this alternative approach was scrapped because (a) it assumes that all incorporated self-employed workers are operating entirely legally, an outcome the authors find doubtful given their discussions with industry stakeholders and (b) the use of subcategories of industry employment led to “sampling error” issues given the relatively small number of respondents in small-population states, resulting in less sensible estimates in those states.

Regardless of the approach used, the authors have reasons to believe that the estimates offered in this study are likely undercounting the extent of misclassification in the construction industry. First, University of Maryland Professor Katharine G. Abraham and colleagues offer compelling evidence that household surveys significantly undercount total employment in the United States.64 In a comparison of earnings data in the Current Population Survey (CPS)—a comparable Census survey to the American Community Survey (ACS) used in this study—to Social Security records, the authors discovered that the CPS historically undercounted 9.5 percent of wage-and-salary jobs and 45 percent of self-employment jobs. This suggests that the ACS is likely underestimating the number of construction workers in the United States; if that is the case, then the misclassification rates offered in this study would be underestimates of the true rate of illegality on construction sites. While the authors gave consideration to inflating industry employment by these rates in their calculations, they declined to do so in order to maintain a conservative approach—the research on such underreporting of employment is still developing—and because the authors could not verify that the rates for underreporting of construction jobs would be the same as more easily forgettable employment such as babysitting, driving for DoorDash, and so on.

Second, it remains an open question to the authors whether the methodology employed sufficiently accounts for the number of foreign-born workers in the construction industry. The Pew Research Center has noted that the American Community Survey undercounts the immigrant population, and the method used in this study is consistent with Pew’s suggested fix (see footnote 59 for more). However, as outlined in the main text of this report, a 2022 study published in the Journal on Migration and Human Security suggests that immigrant undercounting on the ACS may be more extensive among individuals from Central America and Cuba compared to immigrants from Europe.65 Considering the proportion of the construction workforce originally from these undercounted regions, this would imply that the correction used in this report contributes to the conservative estimated rates of misclassification identified in this paper. Researchers are encouraged to confirm and more deeply investigate the potential undercounting of foreign-born construction workers on national household surveys to inform future analyses of this issue.

Finally, this study bases its estimates of worker misclassification on documented rates of income underreporting on tax forms by construction workers. In the absence of direct evidence on misclassification, this step was taken in order to develop conservative estimates of illegality based on rock-solid numbers provided by federal and state tax agencies. But worker misclassification is not defined by tax evasion as there are plenty of workers miscast as independent contractors who pay their taxes in full. As a result, the authors underline that the estimates provided in this report likely represent lower-bound projections of the extent of the problem. Providing certainty to this conclusion is complicated by other potential concerns. For example, it is possible—if not proven— that industry miscoding on household surveys (which have been found to suffer from occupational miscoding) may work in the opposite direction. Regardless, most signs point towards misclassification in the industry being higher than the numbers offered in this study.

Appendix B: Estimating Economic Costs of Worker Misclassification

The method to estimate the economic costs of worker misclassification is a variant of the approach originally advanced in a 2019 report for the Office of the Attorney General for the District of Columbia and sharpened in a 2020 ICERES report.66 The approach used in this study is necessary given that it is practically impossible to accurately evaluate how much money is changing hands in the vast underground construction economy. This method therefore projects costs by evaluating the per-worker cost of misclassification and off-the-books employment based on what one could have expected such workers to earn if hired as a legal employee, and then multiplying that per-worker number by the number of workers affected.

The generation of state-specific estimates of the per-worker costs of misclassification in this study required a number of resources and some assumptions. First, estimates of lost overtime pay were generated by industry-level data from the Employer Costs for Employee Compensation (ECEC) series published by the Bureau of Labor Statistics (BLS).67 Second, unemployment insurance costs by state were estimated from state average costs to employers provided by the Department of Labor.68 However, since the UI costs are experience rated and are typically higher in the construction industry—where turnover is high—this report follows the direction of the 2019 report in the District of Columbia that assumed the rates in a state’s construction industry is double the state average.

Finally, estimating workers’ compensation insurance premium costs in this study presented a unique challenge. While the ECEC series of the BLS presents an industry-wide cost per $100 in payroll and was the basis of previous studies, this was deemed an ineffective measure considering that misclassification is significantly more prevalent in the most dangerous (and thus more costly) occupations. But workers’ compensation insurance rates are not uniform within the construction industry and vary wildly across different occupational codes. Efforts to find industry-wide averages from insurance stakeholders were unsuccessful, so the authors instead used the mathematical average of workers’ compensation insurance premium rates from eleven construction occupations by every state and the District of Columbia as provided in 2020 and 2022 reports published by the State of Oregon (a 2021 report was unavailable).69 While these reports seemed to oversample carpenter trades among the many available construction occupational codes, this was deemed acceptable given that the previously cited 2021 ICERES study identified carpenters as among the trades most plagued by misclassification.


  1. Misclassification and off-the-books payments are both strategies that employers use to avoid taxes and to deny their workers the rights, income, and benefits that are guaranteed under employment law. While these two types of payroll fraud are often discussed together—paying workers off the books is, in some ways, misclassification taken to its extreme—they are distinct malfeasances that often can occur under different circumstances and for different reasons.
  2. Mark Erlich, “Misclassification in Construction: The Original Gig Economy,” Industrial and Labor Relations Review 74, no. 5 (2021): 1202–230.
  3. “Contract to Cheat,” McClatchy DC,  September 2014,
  4. Tom Juravich, Russell Ormiston, and Dale Belman, “The Social and Economic Costs of Illegal Misclassification, Wage Theft and Tax Fraud in Residential Construction in Massachusetts,” UMass-Amherst Labor Center Working Paper, 2021,
  5. Nicole Prchal Svajlenka, “Undocumented Immigrants in Construction,” Center for American Progress, February 2, 2021,
  6. Jacquelyn Pavilon, and Vicky Virgin, “Climbing the Ladder: Roadblocks Faced by Immigrants in the New York City Construction Industry,” Center for Migration Studies, 2022,
  7. Tom Juravich, Essie Ablavsky, and Jake Williams, “The Epidemic of Wage Theft in Residential Construction in Massachusetts,” UMass-Amherst Labor Center Working Paper, 2015,
  8. Ken Jacobs, Kuochih Huang, Jenifer MacGillvary, and Enrique Lopezlira, “ The Public Cost of Low-Wage Jobs in the US Construction Industry,” UC Berkeley Center for Labor Research and Education, January 2022,
  9. Jacquelyn Pavilon and Vicky Virgin, “Climbing the Ladder: Roadblocks Faced by Immigrants in the New York City Construction Industry,” Center for Migration Studies, 2022,
  10. Tom Juravich, “Wage Theft at the North Square Apartments in Amherst, Massachusetts,” UMass-Amherst Labor Center Worker Paper, 2022,
  11. For an overview of how contractors use shell companies to avoid detection and commit insurance fraud, see the recent summary by U.S. Treasury’s Financial Crimes Enforcement Network published on August 15, 2023, and accessed here
  12. Russell Ormiston and Tom Juravich, “Worker Misclassification and Wage Theft in Rhode Island,” UMass–Amherst Labor Center Working Paper, 2022,; Tom Juravich, Russell Ormiston, and Dale Belman, “The Social and Economic Costs of Illegal Misclassification, Wage Theft and Tax Fraud in Residential Construction in Massachusetts,” UMass-Amherst Labor Center Working Paper, 2021,
  13. For a complete review of UI audit studies released prior to 2020, see: Russell Ormiston, Dale Belman and Mark Erlich, “An Empirical Methodology to Estimate the Incidence and Costs of Payroll Fraud in the Construction Industry,” Institute for Construction Employment Research Working Paper, 2020,
  14. “Press Release: FinCEN Highlights Concerning Increase in Payroll Tax Evasion, Workers’ Compensation Fraud in the Construction Sector,” U.S. Treasury Financial Crimes Enforcement Network, 2023,
  15. “Act 85 of 2020: Joint Task Force on Misclassification of Workers, Final Report,” Pennsylvania Department of Labor and Industry, 2022,
  16. Clayton Sinai and Ernesto Galeas, “The Underground Economy and Wage Theft in Washington DC’s Commercial Construction Sector,” Catholic Labor Network, 2021,
  17. Building Injustice,” Workers Defense Project, “Building Austin, 2009,; “Building a Better Texas: Construction Conditions in the Lone Star State,” Workers Defense Project, 2013,; Nik Theodore, Bethany Boggess, Jackie Cornejo, and Emily Timm, “Build a Better South: Construction Working Conditions in the Southern U.S.,” Workers Defense Project, 2017,
  18. Marc Roemer, “Using Administrative Earnings Records to Assess Wage Data Quality in the March Current Population Survey and the Survey of Income and Program Participation,” U.S. Census Bureau Staff Paper, 2002,
  19. Katharine G. Abraham, Brad Hershbein, Susan N. Houseman, and Beth Truedale, “The Independent Contractor Workforce: New Evidence On Its Size and Composition and Ways to Improve Its Measurement in Household Surveys,” National Bureau of Economic Research Working Paper #30997, 2023,; Katharine G. Abraham, Brad Hershbein, and Susan Houseman, “Independent Contract and Informal Work: Preliminary Evidence from Developing Better Measures in Household Surveys,” Working paper presented at 2019 Allied Social Science Association annual meeting, 2018,
  20. Russell Ormiston, Dale Belman, and Mark Erlich, “An Empirical Methodology to Estimate the Incidence and Costs of Payroll Fraud in the Construction Industry,” Institute for Construction Employment Research, 2020
  21. Tom Juravich, Russell Ormiston and Dale Belman, “The Social and Economic Costs of Illegal Misclassification, Wage Theft and Tax Fraud in Residential Construction in Massachusetts: Appendices,” University of Massachusetts–Amherst Labor Center Working Paper, 2021,
  22. Three state-specific studies applied slightly different quantitative methods using the same data to estimate a 16 percent misclassification rate in New Jersey and California, and an 11 percent to 21 percent rate in Tennessee. For more, see Oliver Cooke, Deborah Figart, and John Froonjian, “The Underground Construction Economy in New Jersey,” Stockton University, William J. Hughes Center for Public Policy, 2016,; Yvonne Yen Liu, Daniel Flaming, and Patrick Burns, “Sinking Underground: The Growing Informal Economy in California Construction,” Economic Roundtable, 2014,; William Canak, and Randall Adams, “Misclassified Construction Employees in Tennessee,” 2010, available at
  23. See Appendix A for discussion.
  24. James Parrott and L.K. Moe, “For One in 10 New York Workers, ‘Independent Contractor’ Means Underpaid and Unprotected,” Center for New York City Affairs, 2022, Note that the current study did not use the same approach given concerns that the construction industry—unlike many other sectors analyzed in their report—have a history of legitimate independent contractors who may not necessarily be incorporated (for example, the traditional “handyman”).
  25. “Table B24070 – Industry By Class Of Worker For The Civilian Employed Population 16 Years And Over,” U.S. Census Bureau, American Community Survey,
  26. Tom Juravich, Russell Ormiston, and Dale Belman, “The Social and Economic Costs of Illegal Misclassification, Wage Theft and Tax Fraud in Residential Construction in Massachusetts,” UMass–Amherst Labor Center Working Paper, 2021,
  27. Frank Neuhauser and Colleen Donovan, “Fraud in Workers’ Compensation Payroll Reporting: How Much Employer Fraud Exists? What is the Impact on Honest Employers?” The California Commission on Health and Safety and Workers’ Compensation, 2007,
  28. As explained in more detail in Appendix A, this study does incorporate a conservative correction for potential immigrant undercounting in the ACS consistent with a suggestion from the Pew Research Center. However, this most recent study in the Journal on Migration and Human Security implies that this correction is likely insufficient to fully project the number of foreign-born workers in the construction industry.
  29. Jacquelyn Pavilon and Vicky Virgin, “Climbing the Ladder: Roadblocks Faced by Immigrants in the New York City Construction Industry,” Center for Migration Studies, 2022,
  30. This cost analysis focuses on public costs that are both quantifiable and common to every state. For instance, misclassified workers do not qualify for certain types of OSHA protections, but it may be nearly impossible to untangle the causal effects of the loss of these protections on workplace injuries and the social and personal economic costs that accompany them. Further, there are other costs and benefits that occur at the state level that are not accounted for in this analysis, such as state paid sick and paid family leave programs, due to the complexity engendered by state variation in programs.
  31. Data from Bureau of Labor Statistics Occupational Employment and Wage Statistics series. This identifies the twenty-fifth percentile of annual earnings for workers in Standard Occupational Code 47-0000 within industry code 23 of the North American Industrial Classification System. The May 2022 estimates are used given that this series considered surveys from the previous three years (meaning data from 2019–22), available at
  32. See Appendix B for data sources and discussion.
  33. “Workers’ Compensation Fraud in America,” Coalition Against Insurance Fraud, 2022,
  34. Between 2017 and 2020, the construction industry accounted for approximately 47 percent of all approved cases of the Massachusetts Workers’ Compensation Trust Fund, which pays workers who are injured on the job but whose employer does not have valid workers’ compensation insurance coverage. Tom Juravich, Russell Ormiston, and Dale Belman, “The Social and Economic Costs of Illegal Misclassification, Wage Theft and Tax Fraud in Residential Construction in Massachusetts,” UMass–Amherst Labor Center Working Paper, 2021,
  35. As the data does not allow researchers to know the most relevant tax factors of individual misclassified workers—such as spousal income, number of dependents, income from second jobs, earned income tax credit eligibility—the researchers simply apply state and federal tax rates to the assumed worker income independent of these outside influences (for example, zero children, no spousal income, no additional jobs). Since assuming every worker is unmarried can single-handedly inflate tax estimates—potentially undermining the desire for conservative estimates—the authors assume that misclassified construction workers are married at the same rate as their legally employed counterparts (52.95 percent), thereby taking the standard deduction applicable to joint filers. While this approach would seem to considerably undercount workers’ taxable earnings, it is somewhat balanced by the fact that it simultaneously overlooks the earned income tax credit, child tax credit, and other programs that would reduce the tax burden of lower-income workers.
  36. Two studies by the Workers Defense Project in Texas suggest that wage theft may equal about 1 percent of workers’ earnings, but there are no corroborating studies of wage theft outside of the state to conclude whether this result is representative of the national situation. For more, see “Building a Better Texas: Construction Conditions in the Lone Star State,” Workers Defense Project, 2013; “Building Austin, Building Injustice,” Workers Defense Project, 2009.
  37. A 2022 study by researchers at the University of California-Berkeley Center revealed that 39 percent of families of construction workers are enrolled in at least one public safety net program at a cost of nearly $28 billion per year. However, it is unclear how many of these workers are misclassified and how many represent legally employed but lowly paid tradespeople as both issues affect the industry’s workforce. For more, see Ken Jacobs, Kuochih Huang, Jenifer MacGillvary, and Enrique Lopezlira, “The Public Cost of Low-Wage Jobs in the US Construction Industry,” UC Berkeley Center for Labor Research and Education, 2022,
  38. Testimony of Matt Townsend, “Hearing on Misclassification of Employees: Examining the Costs to Workers, Businesses, and the Economy,” Workforce Protections Subcommittee of the House Education and Labor Committee, September 26, 2019.
  39. “State Unemployment Insurance Trust Fund Solvency Report,” Office of Unemployment Insurance, U.S. Department of Labor, March 2023,
  40. Andrew Stettner and Laura Valle Gutierrez, “A Pandemic Lifeline, Nation’s Unemployment Insurance Programs is Inadequate and Unprepared for Next Downturn,” The Century Foundation, November 2022,
  41. Tom Juravich, Russell Ormiston, and Dale Belman, “The Social and Economic Costs of Illegal Misclassification, Wage Theft and Tax Fraud in Residential Construction in Massachusetts,” UMass–Amherst Labor Center Working Paper, 2021,
  42. “Federal Tax Compliance Research: Tax Gap Estimates for Tax Years 2008-2010,” Internal Revenue Service, IRS Publication 1415, 2016,
  43. David Weil, “The Application of the Fair Labor Standards Act’s “Suffer or Permit” Standard in the Identification of Employees Who Are Misclassified as Independent Contractors,” U.S. Department of Labor, Wage and Hour Division, July 2015, .
  44. Wage and Hour Division, U.S. Department of Labor, “Employee or Independent Contractor Classification Under the Fair Labor Standards Act: Notice of Proposed Rulemaking,” Federal Register 87, no. 197 (October 13, 2022),
  45. Wage and Hour Division, U.S. Department of Labor, “Independent Contractor Status Under the Fair Labor Standards Act (FLSA): Withdrawal,” Federal Register 86, no. 86 (May 6, 2021),
  46. For more information on the status of the rule and its regulatory history see “Misclassification of Employees as Independent Contractors,” U.S. Department of Labor, Wage and Hour Division,
  47. Moshe Marvit, “How the PRO Act Would Fix Labor Law’s Worker Classification Problem,” The Century Foundation, May 27, 2021,
  48. Lynn Rhinehart, et al. “Misclassification, the ABC test, and employee status: The California experience and its relevance to current policy debates.” Economic Policy Institute, June 16, 2021,
  49. Jon Shimabukuro, “Worker Classification: Employment Status Under the NLRA, the FLSA, and the ABC Test [R46765],” Congressional Research Service, April 20, 2021,
  50. Lynn Rhinehart, et al., “Misclassification, the ABC test, and employee status: The California experience and its relevance to current policy debates.” Economic Policy Institute, June 16, 2021,
  51. “Worker Reclassification – Section 530 Relief,” Internal Revenue Service, .
  52. Deco Logistics, Inc. d/b/a Container Connection; Universal Intermodal Services; Southern Counties Express; Universal Logistics Holdings, Inc.: and Universal Management Services, as a single employer, Case No. 21-CA-272323 (complaint filed March 17, 2022).
  53. John Schmitt, Heidi Shierholz, Margaret Poydock, and Samantha Sanders, “The economic costs of worker misclassification,” Economic Policy Institute, January 25, 2023,
  54. Mark Erlich and Terri Gerstein, “Confronting Misclassification and Payroll Fraud: A Survey of State Labor Standards Enforcement Agencies,” Harvard Law School Labor and Worklife Program, June 2019,
  55. National Labor Relations Board, “Ensuring Safe and Dignified Access for Immigrant Workers to NLRB Processes,” General Counsel Memorandum OM 22-09, May 2, 2022,
  56. “DHS Announces Process Enhancements for Supporting Labor Enforcement Investigations,” U.S. Department of Homeland Security, press release, January 13, 2023,
  57. “Process for Requesting Department of Labor Support for Requests to the Department of Homeland Security for Immigration-Related Prosecutorial Discretion During Labor Disputes: Frequently Asked Questions For Workers and Their Representatives,” U.S. Department of Labor,  July 6, 2022,
  58. For a thorough documentation of this methodology, see Russell Ormiston, Dale Belman and Mark Erlich, “An Empirical Methodology to Estimate the Incidence and Costs of Payroll Fraud in the Construction Industry,” Institute for Construction Employment Research Working Paper, 2020, The current report, however, uses an updated version of that method that is described at length at Tom Juravich, Russell Ormiston, and Dale Belman, “Appendices: The Social and Economic Costs of Illegal Misclassification, Wage Theft and Tax Fraud in Residential Construction in Massachusetts,” UMass–Amherst Labor Center Working Paper, 2021,
  59.  In the method outlined in Appendix Table A1 of Juravich, Ormiston, and Belman (2001, ibid.), total industry employment is calculated by starting with the estimated number of workers in the construction industry in a given year according to the American Community Survey; state calculations involve an adjustment to account for the primary address of their employer and not their home address. As the ACS only queries workers’ primary job, an additional 1.99 percent is added to the totals, reflecting the proportion of workers nationally who identify as having a second job in the construction industry according to the Current Population Survey. Finally, as the Pew Research Center has identified that large national household surveys undercount the immigrant population by at least 2 percent—who comprise an important portion of the nation’s construction workforce—the totals add a state-specific amount of likely undercounting of immigrant construction workers. Meanwhile, total wage-and-salary employment starts with the construction-industry total for this metric from the Bureau of Economic Analysis. This is then supplemented by adding in the estimated number of construction-occupation workers who are technically employed in the industry of “temp agencies” since these workers are operating above board and should be counted as legal employment. The difference between total industry employment and total wage-and-salary employment therefore represents total self-employment which includes the legally self-employed, misclassified independent contractors, and workers operating off-the-books.
  60. For a review of relevant studies regarding income underreporting by self-employed workers and further justification for the range provided, see: Russell Ormiston, Dale Belman and Mark Erlich, “An Empirical Methodology to Estimate the Incidence and Costs of Payroll Fraud in the Construction Industry,” Institute for Construction Employment Research Working Paper, 2020,
  61. Tom Juravich, Russell Ormiston, and Dale Belman, “The Social and Economic Costs of Illegal Misclassification, Wage Theft and Tax Fraud in Residential Construction in Massachusetts: Appendix B,” UMass–Amherst Labor Center Working Paper, 2021,
  62. “Federal Tax Compliance Research: Tax Gap Estimates for Tax Years 2008-2010,” Internal Revenue Service, IRS Publication 1415, 2016,
  63. An alternative method using this data—effectively treating incorporated and unincorporated self-employed workers differently in the data—led to nearly identical estimates on a national basis, albeit with more extreme variations between states.
  64. Katharine G. Abraham, John Haltiwanger, Claire Hou, Kristin Sandusky, & James R. Spletzer, “Reconciling Survey and Administrative Measures of Self-Employment.” Journal of Labor Economics 39, no. 4 (2020).
  65. Robert Warren, “2020 American Community Survey: Use with Caution, An Analysis of the Undercount in the 2020 ACS Data Used to Derive Estimates of the Undocumented Population,” Journal on Migration and Human Security 10, no. 2 (2022): 134–45, 2022,
  66. Dale Belman and Aaron Sojourner, “Economic Analysis of Incentives to Fraudulently Misclassify Employees in District of Columbia Construction,” Office of the Attorney General for the District of Columbia, 2019,; Russell Ormiston, Dale Belman and Mark Erlich, “An Empirical Methodology to Estimate the Incidence and Costs of Payroll Fraud in the Construction Industry,” Institute for Construction Employment Research Working Paper, 2020,
  67. For details, see “Employer Costs for Employee Compensation News Release,” U.S. Department of Labor, Bureau of Labor Statistics, December 16, 2021,
  68. For details, see “Employer Contribution Rates, Calendar Year 2021,” U.S. Department of Labor, Employment and Training Administration, 2021,
  69. See Donald Gallogy and Amelia Vargas, “Oregon Workers’ Compensation Premium Rate Ranking: Calendar Year 2022,” Oregon Department of Consumer and Business Services, 2022, Also see Chris Day and Jay Dotter, “Oregon Workers’ Compensation Premium Rate Ranking: Calendar Year 2020,” Oregon Department of Consumer and Business Services, 2020,