Executive summary

Beyond its role as a regulator, the New York state government has a critical function in the state’s construction industry: one of its largest customers. With a 2018 fiscal year capital budget of $14.5 billion, the policies and practices of the state government—as a consumer—significantly influence local construction labor markets. This presumably puts state lawmakers in the difficult position of having to balance two ethical considerations that, on the surface, appear to be mutually exclusive: the need to minimize taxpayer costs against the responsibility of ensuring fair wages, benefits, and safe working conditions for its residents on public construction projects.

To ensure that New York state fulfills the latter consideration, contractors working on state-funded government construction projects must adhere to the tenets of New York’s prevailing wage law. Enacted in 1897, this state policy requires that contractors pay their workers no less than the “prevailing” wage and benefit levels within the local construction market. In addition to its ethical underpinnings, the law also has an economic justification: it protects New York construction workers from being undercut by low-wage, often out-of-state contractors that may covet a large government construction contract and whose presence would take away jobs and erode working conditions for local residents.

State prevailing wage laws across the country have increasingly been assailed by those who appeal to lawmakers’ other responsibility—minimizing taxpayer costs—in an attempt to weaken or repeal these policies. These nationwide campaigns are built almost entirely upon a single argument: higher wages must equate to higher taxpayer costs. This narrative is simple, and its simplicity has made it a powerful political position that has been successfully exploited to overturn state laws in Indiana (2015) and West Virginia (2016). And with a recent publication by the Empire Center (McMahon and Gardner 2017), it has become apparent that some in New York will attempt to pitch the same narrative to state lawmakers.

There’s one problem. According to the most advanced economic research on state prevailing wage laws, the simple narrative largely isn’t true.

To separate fact from fiction as it relates to New York’s prevailing wage law, this report provides a thorough cost-benefit analysis of state policy relying extensively on independent, peer-reviewed research. As summarized in this report, academic economists from around the country have made prevailing wage laws a research priority over the last 15 years. In study after study, economists have found no evidence that these laws have had any significant cost effects on the biggest drivers of New York’s capital budget: highways and institutional buildings (e.g., schools).

How could the simple narrative be so wrong?

It turns out that the narrative relies on a subtle, yet egregious, logical fallacy that assumes that the only way that a contractor can minimize labor costs is by reducing the cost per hour of each worker. This is false. In any industry, an employer can also minimize labor costs by reducing turnover and/or the number of labor hours required. In construction, this would mean using higher wages to attract and hire the industry’s most productive workers and providing them with the most advanced equipment and technology. This high-wage, high-skill approach to minimizing cost is referred to as paying “efficiency wages,” a well-established strategy that is discussed in any introductory labor economics course. In the retail industry, this approach is what allows high-wage Costco and Wegmans to compete with low-wage Wal-Mart. Both paths can minimize costs, and it has been demonstrated convincingly in the academic research that high-wage contractors are able to exploit these cost offsets to place bids on public construction projects that are competitive with—if not better than—those of low-wage contractors.

In short, opponents of prevailing wage laws have shaped the political narrative surrounding these laws by, in part, successfully creating and defeating an implicit straw man argument that prevailing wage laws increase the hourly cost per worker. That may be true. But the fundamental policy question is whether state prevailing wage laws increase total construction costs. From this perspective, the academic literature is clear and unequivocal: state prevailing wage laws have no effect on the biggest drivers of a state’s construction budget. Suggestions to the contrary, as found in reports authored by the laws’ opponents, consistently (and conveniently) ignore the academic research and instead rely on an empirical methodology that has been widely discredited among economists.

Opponents of state prevailing wage laws across the country have also shaped the narrative on these laws by focusing almost entirely on the presumed costs of the policy; recent analyses of New York’s policy by the Center for Urban Real Estate (Vitullo-Martin 2012) and the Empire Center (McMahon and Gardner 2017) follow a similar approach. This single-minded focus on construction costs has altered political discussion about prevailing wage laws in a powerful way: it has obscured all attention away from the benefits that these laws offer a state and its residents. That is unfortunate, as state prevailing wage laws have been demonstrated to improve the lives of workers, communities, contractors and construction consumers.

First and foremost, New York’s prevailing wage law strengthens and protects the state’s blue-collar middle class. By ensuring that New York state residents working on public construction projects receive fair pay and benefits, state lawmakers are directly responsible for creating and promoting the types of blue-collar middle-class jobs that have long represented the backbone of communities throughout New York state. This not only brings economic and personal security to New York families and communities, but it also adds hundreds of millions of dollars to the state’s economy and substantially increases New York’s state and local tax revenues.

New York’s prevailing wage law is also one of the few effective policy levers that promote a clear pathway to the middle class for non–college educated workers. As described in this report, New York state residents without a college degree have encountered dwindling economic opportunities in recent decades, and most jobs expected to become available to them in the near future offer poverty-level wages or below. A career in construction is an exception, as this is one of the few remaining industries where workers can earn a sufficient paycheck and receive health and pension benefits without a college degree. New York’s prevailing wage law promotes these middle-class career opportunities by incentivizing contractors to hire apprentices enrolled in state-sanctioned programs, many of which only require a high school diploma for admission. In doing so, state lawmakers are directly providing opportunities for inexperienced workers to acquire the on-the-job training necessary to develop into skilled tradespeople and the next generation of New York’s blue-collar middle-class.

The benefits of lawmakers’ support for apprenticeship programs through New York’s prevailing wage law also extend to the state’s contractors and construction consumers. Contractors have bemoaned skill shortages in construction for decades. While bona fide regional skill shortages can typically be resolved by increasing wages, contractors’ oft-used solution to a perceived skill shortage—recruiting and employing available out-of-region workers—represents a missed opportunity for policymakers interested in generating new jobs for local residents. Prevailing wage laws are one of the few effective policy levers available to promote on-the-job training and apprenticeships, thereby developing a skilled, in-state labor force capable of powering New York’s economic growth.

New York’s prevailing wage law is also consistent with the obligation of state and local governments to prioritize the bids of honest, fair-dealing contractors whose workplace practices adhere to labor and employment law. Under the policy, contractors working on state-funded projects must submit a project-wide certified payroll report that includes employees’ names, Social Security numbers, pay rates, and number of hours. The prospect of more stringent government oversight discourages bids from unscrupulous contractors whose competitiveness depends on illegal, but-cost saving labor strategies. By rooting out unethical actors, prevailing wage laws also advantage contractors with a greater commitment to workplace safety, as the academic research demonstrates that fatal and nonfatal injury rates in construction are substantially lower in states with these laws.

In sum, state lawmakers overseeing New York’s construction industry must weigh two ethical responsibilities that, on the surface, appear to be mutually exclusive: the need to minimize taxpayer costs and the responsibility of ensuring fair wages, benefits, and safe working conditions for its residents on public construction projects. A critical review of the current academic research in this report indicates that New York’s prevailing wage law may offer policymakers the opportunity to fulfill both responsibilities. On the biggest components of the state’s construction budget, the consensus of academic studies is that the policy has little to no effect on taxpayer costs. There is also considerable evidence that the policy raises the standard of living of a state’s residents, improves workplace safety, and offers a clear path to the middle class for New York state residents without a college degree. Taken together, research by independent, academic economists indicates that New York’s prevailing wage law is a uniquely valuable component of state policy that simultaneously uplifts residents and communities while imposing minimal, if any, cost on taxpayers.

Introduction

Beyond its role as a regulator, the New York state government has a critical function in the state’s construction industry: one of its largest customers. With a 2018 fiscal year capital budget of $14.5 billion, the policies and practices of the state government—as a consumer—significantly influence local construction labor markets. This presumably puts state lawmakers in the difficult position of having to balance two ethical considerations that, on the surface, appear to be mutually exclusive: the need to minimize taxpayer costs against the responsibility of ensuring fair wages, benefits and working conditions for its residents on public construction projects.

To ensure that New York state fulfills the latter consideration, contractors working on state-funded government construction projects must adhere to the tenets of New York’s prevailing wage law. Enacted in 1897, this state policy requires that contractors pay their workers no less than the “prevailing” wage and benefit levels within the local construction market. In addition to its ethical underpinnings, the law also has an economic justification: it protects New York construction workers from being undercut by low-wage, often out-of-state contractors who may covet a large government construction contract and whose presence would take away jobs and erode working conditions for local residents.

As states have continued to struggle to balance their budgets following the Great Recession, opponents have increasingly appealed to lawmakers’ other responsibility—minimizing taxpayer costs—in an attempt to weaken or repeal prevailing wage laws. These nationwide campaigns are built almost entirely upon a single argument: higher wages must equate to higher taxpayer costs. This narrative is simple, and its simplicity makes it a powerful political position.

The argument is indeed simple. But is it true?

It turns out that the simple narrative is too simple. An emerging consensus from the academic research reveals that, for many public construction projects, prevailing wage laws do not increase taxpayer costs (Duncan and Ormiston 2018). If this seems counterintuitive, consider that the simple wage-cost narrative ignores basic dynamics within construction markets: high-wage contractors attract and employ the highest skilled and most productive workers while using the industry’s most advanced equipment and technology. These labor cost offsets are substantial and allow high-wage contractors to place bids for public construction projects that are competitive with—if not better than—those of low-wage, low-skill contractors (Atalah 2013a).

Whether prevailing wage policies allow state lawmakers to simultaneously fulfill both of their ethical obligations—to taxpayers and workers—represents a complicated public policy question. While the academic research cited above suggests that it does, any definitive conclusion about the answer for New York requires a more thorough examination of the state’s prevailing wage law than has been provided elsewhere. This report offers a detailed analysis of the broad array of costs and benefits of state prevailing wage laws in the context of New York’s state capital budget, emphasizing the findings of academic, peer-reviewed research.

Costs

Nationwide political campaigns by opponents of state prevailing wage laws are typically accompanied by partisan reports featuring projections about how much a state will save by repealing the policy. These estimates are typically enormous. Political efforts to weaken the law in New York state are no different, as a 2017 report by the Empire Center proclaims that the state’s prevailing wage law is costing New York taxpayers an extra $400 million annually (McMahon and Gardner 2017). However, when viewed through the lens of the most advanced academic research, it becomes clear that these reports are methodologically defective and offer estimated cost savings that are wildly optimistic.

Academic economists from around the country have made the potential cost impact of prevailing wage laws a research priority over the last 15 years. Numerous studies have analyzed the policy’s effect on public construction costs in three areas of significant government expenditure: highways, schools, and affordable housing. To demonstrate the applicability of the academic research to New York state, this report will provide a brief overview of these studies through the lens of the state’s fiscal year 2018 capital budget.1

Highways

New York State is projected to spend $5.6 billion on highways, bridges, and other transportation projects in fiscal year 2018, accounting for nearly 40 percent of the state’s capital budget. The Department of Transportation will oversee construction projects whose value will be more than four times that of any other state agency. As a result, any argument that prevailing wage laws substantially increase the state’s construction budget is severely weakened if it is shown that the policy has no impact on transportation spending.

This is exactly the conclusion reached by Duncan (2015a), the definitive academic study on the relationship between prevailing wage laws and transportation spending. In an analysis of highway maintenance costs in Colorado between 2000 and 2011, the author discovered that resurfacing projects on state-funded intrastate highways and federally funded interstate highways were required to be built to identical standards. Contractors bidding on interstate projects, however, encountered one additional requirement: federal prevailing wage law (i.e., the Davis-Bacon Act).2 In contrast, intrastate projects were exempt from prevailing wage policy since Colorado had never passed such a law to govern state-funded construction. This created a perfect “natural experiment” for the author to examine the cost impact of prevailing wage.

Analyzing data on 132 highway resurfacing projects, Duncan (2015a) applied a standard regression model to account for factors that may have affected a project’s complexity and cost (e.g., location, type of terrain). The results demonstrated that there was not any statistically significant cost differential between highway projects covered by prevailing wage and those without such requirements. In a follow-up study, Duncan (2015b) expanded the analysis to examine whether bids on 91 resurfacing projects were more aggressive when contractors switched from federal projects to less-regulated state projects; the results again failed to find any evidence suggesting that prevailing wage policy had any statistically significant impact on contractor bids.3

While not an extensive literature, the current research on prevailing wage laws offers compelling evidence that contractors on highway resurfacing projects are able to offset the higher hourly labor costs by employing the most skilled workers and using the most advanced technologies. More broadly, it would be unreasonable to expect that prevailing wage laws would increase costs on more complicated transportation projects (e.g., bridges, tunnels) given that high-wage, high-skill contractors would presumably have even more of an advantage on those projects compared to highway resurfacing. Considering that New York state is spending nearly 40 percent of its capital budget on transportation projects, the absence of any cost effect attributable to prevailing wage in this construction area substantially weakens claims that the state’s prevailing wage law has a significant effect on taxpayer costs.4

Schools

Economists have learned the most about state prevailing wage laws over the last 15 years by studying their effects on public school construction costs. School construction offers researchers an ideal environment to study the cost impact of prevailing wage, as detailed data can be collected on a large number of projects that are quite comparable in their design and construction. Using regression analysis, economists are able to control for unique aspects of each school (e.g., square feet, number of stories, high school vs. elementary school) and thus conduct an “apples to apples” comparison of the cost differential between schools based on whether such work was covered by prevailing wage policy. Regression unambiguously represents “best practices” among researchers in this area, and it has been the chosen methodology in nine studies—all five academic articles and four additional non-academic papers—published on the subject over the past two decades.5

The results of these studies offer a clear consensus: state prevailing wage laws do not affect public school construction costs. In eight of the nine studies, the researchers failed to find a statistically significant link between prevailing wage policy and increased costs.6 This includes the largest studies on the topic, as Azari-Rad, Philips, and Prus (2002, 2003) examined over 4,600 schools built nationwide between 1991 and 1999 and failed to find any evidence that the presence or relative strength of a state’s prevailing wage law had an effect on school construction costs.7

The ability of high-wage contractors to entirely offset the hourly labor costs attributable to prevailing wage laws on school construction projects is clearly demonstrated by Atalah (2013a). After Ohio exempted schools from its prevailing wage law in 1997, the author examined 8,093 bids on public school construction projects in the state between 2000 and 2007. Comparing the bid cost per square foot between union contractors—who presumably pay the highest local wages—and nonunion contractors, the author found no statistically significant difference between bids statewide. The only significant difference occurred in an examination of southern Ohio, where the author found that bids from nonunion contractors were considerably higher than their union counterparts. Given that contractors were not operating under the requirements of a prevailing wage law, it is revealing that high-wage contractors were just as competitive—if not more competitive—on cost than low-wage contractors. The results provide convincing evidence that contractors are able to offset the wage differentials attributable to prevailing wage by employing the most skilled and most productive workers and using the most advanced equipment and technology.8

The evidence offered by the academic research on school construction suggests that state prevailing wage laws do not affect taxpayer costs in this area. For New York, the implications of this conclusion are far-reaching. In addition to the $589 million that the state is projected to spend on K-12 public school construction projects in fiscal year 2018, it would be reasonable to expect that these findings would also be applicable for construction work on comparable institutional projects; this would most specifically include the $1.5 billion that the state has set aside for higher education, which accounts for over 10 percent of its capital budget. Finally, this research is likely to be of considerable interest to local governments that devote a considerable portion of their capital budget to K-12 construction spending and require contractors to pay their workers a prevailing wage on city-funded projects. This certainly extends to New York City, whose fiscal year 2018 capital budget includes $3.4 billion for K-12 school construction projects.

Affordable housing

In November 2016, New York Governor Andrew Cuomo announced a deal between real estate developers and union construction officials that revived the state’s 421a tax exemptions for new construction of affordable housing in New York City. The controversial, long-awaited Affordable New York Housing Program provides developers with a 35-year tax abatement on housing projects. In return, developers must set aside a certain proportion of affordable apartments for 40 years and agree to pay construction workers an average compensation package of $45 or $60 per hour depending on the project’s location in NYC. Developers are not required to employ union labor.

A growing consensus among researchers suggests that prevailing wage laws are likely to increase construction costs on affordable housing projects. In an academic study of 205 affordable housing projects in California between 1997 and 2002, Dunn, Quigley, and Rosenthal (2005) found that prevailing wage laws at any level (federal, state, or local) increased costs between 9 percent and 11 percent using a standard regression model.9,10 In a more recent academic paper, Littlehale (2017) examined 286 affordable housing projects in California from 2001 to 2011. Applying a more extensive regression model than previously employed in the literature, the author estimated that prevailing wage laws increased construction costs on affordable housing between 5 percent and 7 percent. While prevailing wage laws were not the primary focus of Palm and Niemeier (2017), their study of housing projects built between 2008 and 2016 in the state’s four largest metropolitan areas suggested that the policy’s cost effects were between 15 and 16 percent per unit. Finally, in a non-academic paper, the New York Independent Budget Office (NYC IBO 2016) examined 211 projects in New York City and concluded that federal Davis-Bacon regulations added 23 percent to the cost of construction. While the estimate from the New York report is consistent in direction with the two academic papers, its deviation in magnitude from the existing research—both academic and non-academic—is likely attributable to a combination of factors.11 This includes a singular focus on federal policy and a relatively sparse regression model (when compared with the academic studies) that appears to inadequately isolate the prevailing wage effect from other differences in projects that receive and do not receive federal funding (e.g., additional oversight by the U.S. Department of Housing and Urban Development), thereby inflating the estimated cost effect of labor policy.12,13

In sum, the current research on prevailing wage laws suggests that the policy likely increases construction costs on affordable housing. But the application of this research to make projections about the Affordable New York Housing Program is complicated by the fact that its required compensation—$45 or $60 per hour depending on the location—is substantially less than the wage and benefit requirements of federal and state prevailing wage laws for those areas. This renders direct comparisons to the effect of the federal Davis-Bacon Act—such as those made in the NYC IBO study—to be ineffective as cost estimates of Governor Cuomo’s program.

It should not come as a surprise that prevailing wage laws likely have a different cost effect for affordable housing when compared with other areas of construction. There are likely two critical reasons for this. First, Dunn, Quigley, and Rosenthal (2005) noted that affordable housing construction requires less skill, has lower costs of materials, and features a larger share of labor in total costs when compared with other publicly funded projects. This position is consistent with the behavior of high-wage contractors, who have traditionally avoided residential construction projects; it is likely that these contractors have found that high-wage workers are less cost-effective on residential properties when compared to more complicated industrial or institutional projects. As a result, the assertions of Dunn, Quigley, and Rosenthal (2005)—combined with the near unanimous research on highways and institutional buildings (i.e., schools)—suggest that the positive cost effect of prevailing wage may be limited to affordable housing and related areas of construction.

But there is another reason why prevailing wage laws might increase costs on affordable housing projects: it significantly disadvantages contractors who rely on illegal, but cost-saving, employment practices to remain competitive. These are hardly isolated practices in the construction industry. Audits by the New York State Department of Labor Unemployment Insurance Division between 2002 and 2005 revealed that 14.9 percent of construction employers misclassified employees—nearly 50,000 of them annually—as independent contractors (Donahue, Lamare, and Kotler 2007). Unscrupulous contractors often use this strategy as a means of evading legally required Social Security taxes and payments to the state unemployment insurance fund, robbing workers of their rightfully earned benefits and costing New York state tens of millions of dollars annually in lost UI funding.

In addition to employee misclassification, numerous studies have shown that the construction industry in New York City is rife with contractors who engage in wage theft, hire undocumented labor, and ignore unsafe working conditions (Bernhardt, Spiller, and Polson 2013; Milkman, Gonzalez, and Ikeler 2012; Theodore, Valenzuela, and Melendez 2006). These issues are especially relevant when discussing the effect of prevailing wage laws on the construction costs of affordable housing; the Brennan Center for Justice (2007) noted that these illegal labor practices are especially concentrated in the residential sector of the construction industry.

Prevailing wage laws help minimize these illegal labor practices. Consistent with similar laws across the country, the Affordable New York Housing Program requires contractors working on state-funded projects to submit a project-wide certified payroll report; this includes employees’ names, Social Security numbers, pay rates, and number of hours. The prospect of stringent government oversight of a project’s payroll provides a substantial disincentive for contractors whose low-cost bids for public works rely on cutting corners in their employment practices. At this point, there is not a definitive research study that indicates whether the Affordable New York Housing Program will increase construction costs on state-funded affordable housing. But if it does, would it be worth it—economically and ethically—for the state government to repeal the program and instead save money by rewarding contractors whose lowest bids are predicated on illegal but cost-saving employment practices?

Response to other studies

The most advanced, sophisticated research on prevailing wage laws in the United States yields no evidence that state policy increases construction costs on two of New York’s largest public expenditures: highways and schools. This consensus among academic economists, however, has not deterred non–peer reviewed reports from claiming that New York’s prevailing wage law substantially increases public construction costs in the state. The Citizens Housing and Planning Council (Roistacher, Perine, and Shultz 2008) estimated that the state law increased construction costs by 25 percent. The Center for Governmental Research (Gardner and Ruffer 2008) suggested that the law increased construction costs between 19 percent and 55 percent depending on the region. Most recently, the Empire Center (McMahon and Gardner 2017) estimated that these regional effects were anywhere between 13 percent and 25 percent. In the context of the academic research, these cost estimates do not make much sense. If the most advanced research suggests that prevailing wage laws do not affect construction costs on the key drivers of the state’s construction budget, how are these reports generating such enormous cost estimates?

A review of these three studies—and similar anti-prevailing wage articles published around the country—reveals that these reports universally rely on variants of the same empirical methodology. This approach involves a simple two-step process that can best be explained by example. Consider the 25 percent estimate offered by the Citizens Housing and Planning Council (Roistacher, Perine, and Shultz 2008). The first step in the process is to compare the per-hour compensation required by prevailing wage law to some arbitrary, lower wage and benefit level. The CHPC, for instance, estimated that the median union construction worker in New York City earned 74 percent more in wages and fringe benefits than the median nonunion construction worker, an inflated and misleading number given that union workers operate almost entirely in the higher-wage nonresidential sector of the industry. In the second step, this per-hour compensation differential is multiplied by the proportion of construction costs attributable to labor expenses; the resulting product is touted as the “cost” of a state’s prevailing wage law. Given that the CHPC estimated that one-third of public projects’ construction cost was attributable to labor, the authors multiplied 74 percent by 33 percent to produce their final estimate: a 25 percent cost increase.

This approach is mathematically simple. But it also violates the laws of economics.

In any industry, an employer can minimize labor costs by either lowering the cost per hour or by reducing turnover and/or the number of labor hours. Construction contractors are no different. Some contractors minimize costs by following the first path, offering substandard wages and employing teams of relatively unskilled laborers. Other contractors pursue the latter route, hiring and retaining the industry’s most productive workers at higher wages and providing them with the most advanced equipment and technology. This is a well-known approach in labor economics called “efficiency wages”; in the retail industry, this strategy is what allows high-wage Costco and Wegmans to compete with low-wage Wal-Mart. Both paths can minimize costs, and it has been demonstrated convincingly that high-wage contractors in the construction industry are able to exploit these cost offsets to place bids on public construction projects that are competitive with—if not better than—those of low-wage contractors (Atalah 2013a, 2013b).

With this background in mind, consider that the empirical methodology used by anti-prevailing wage studies relies entirely on a comparison between the prevailing wage and some arbitrary lower wage. This demonstrates that, all else equal, prevailing wage laws are associated with a higher hourly cost per worker. This may be true, but beware of the straw man argument at play: the policy-relevant question is whether prevailing wage laws increase overall construction costs, not the hourly cost per worker. Studies using this approach ignore, or conveniently dismiss, high-wage contractors’ ability to minimize labor costs in any other way besides reducing per-hour compensation; as an example, see the report from the Empire Center (McMahon and Gardner 2017, 9). This represents a significant methodological oversight that artificially inflates the cost estimates associated with the policy and violates the basic tenets of labor economics.

Given evidence that high-wage contractors in New York state employ more experienced, more educated and better trained workers (see Appendix B), it is unfortunate that this flawed approach has already influenced public debate over prevailing wage laws in the state. Part of this, however, may be due to a lack of clarity about the shortcomings of this methodology. Given its prevalence among anti-prevailing wage studies, the Citizens Housing and Planning Council (Roistacher, Perine, and Shultz 2008) gave equal standing to the simple, two-step approach—which they deemed the “hypothetical” cost model—when compared to the econometric, regression-based model employed in the academic literature; the New York City Independent Budget Office (2016) later incorporated the CHPC’s perspective into its report.

These two approaches, however, are not on equal standing among empirically rigorous economists and researchers. As outlined in Duncan and Ormiston (2018), the “hypothetical” approach—sometimes called the “wage differential” method—has been widely discredited in academic circles. As evidence, consider that there has not been a single known research paper accepted for academic publication in the last 16 years that has employed this simple approach in an evaluation of prevailing wage laws; every published study has instead relied on regression modeling. This latter approach is preferred among researchers given that it evaluates overall construction costs in a manner that is agnostic about a contractor’s method of minimizing costs, thereby allowing an unbiased evaluation of the cost impact of state prevailing wage laws.

Conclusion: Costs

Those opposed to prevailing wage laws have armed themselves with a simple, yet powerful, political narrative that higher wages must mean higher construction costs. They have supported their position with an equally simple empirical methodology that relies entirely on the assumption that prevailing wage laws must increase taxpayer costs. Following this well-worn playbook, opponents in New York have used this simple ideology to promise taxpayers hundreds of millions of dollars of savings if the state repealed this long-established policy.

Simple arguments can produce politically powerful narratives. But cost estimates of state prevailing wage laws based on opponents’ simple methodology is the economics equivalent of trying to build a new hospital using only a hammer and a hand saw. As any construction contractor or worker can appreciate, more complicated problems often require more complex tools. Fortunately, academic economists—using more advanced empirical tools—have made estimating the cost impact of state prevailing wage laws a research priority in recent years.

When viewed through the lens of New York state’s capital budget, the most advanced research offered by these economists indicate that opponents’ estimated cost impact of the state’s prevailing wage law is substantially overblown. Peer-reviewed studies have offered clear evidence that these laws have no significant cost effect on the biggest drivers of New York’s capital budget: highways and institutional buildings (e.g., schools). There is evidence that prevailing wage laws may increase construction costs on affordable housing, but the effect may be limited to residential construction and the unique characteristics of the Affordable New York Housing Program clouds the applicability of existing research to estimate the policy’s cost. In sum, when research on prevailing wage laws is considered in the context of New York state’s capital budget, the results suggest that the policy has minimal effect, if any, on public construction costs.

Benefits

Recent analyses of New York’s state prevailing wage law by the Center for Urban Real Estate (Vitullo-Martin 2012) and the Empire Center (McMahon and Gardner 2017) have focused almost entirely on the presumed costs of the policy. This single-minded focus on construction costs—consistent with political campaigns to repeal state policies around the country—has altered the narrative about prevailing wage laws in a powerful way: it has obscured all attention away from the benefits that these laws offer a state and its residents. That is unfortunate, as prevailing wage laws are one of the few, effective policy levers available to lawmakers to improve the standard of living of blue-collar workers, promote a skilled workforce, and provide an established pathway to the middle class for New York residents without a college degree.

To expand the public narrative on prevailing wage laws in New York, this section will explore how the policy improves the lives of the state’s residents. Given that there is limited academic research exploring the benefits of prevailing wage laws, this study will supplement the analysis using government data and other available sources. To provide a framework to guide policy discussions, this study classifies benefits as belonging to one of two broad categories: “workers, families, and communities” and “Contractors and consumers.”

Workers, families, and communities

Prevailing wage laws strengthen wages for blue-collar workers

Political efforts to weaken or repeal state prevailing wage laws—in New York and across the country—are implicitly rooted in the perspective that a state government should save money by paying blue-collar construction workers less. While the academic research offers evidence that prevailing wage laws generally do not increase public construction costs, economists have clearly demonstrated that the law is effective in improving worker wages. In the definitive study of the policy’s effect on construction labor markets, Kessler and Katz (2001) estimated that the repeal of a state’s prevailing wage law results in a 2 percent to 4 percent decline in the average hourly wage for its blue-collar construction workers on all projects, both public and private.14,15

Given these results, the ethical and financial implications of repealing or weakening New York’s prevailing wage law are enormous; for the average worker in the sector, a 3 percent reduction in the hourly wage would result in lost earnings approaching $2000 annually.16 By itself, this would devastate some New York workers and their families. But the pursuit of a policy initiative that intentionally weakens the earning power of blue-collar workers at this time seems unconscionable. As presented in Table 1, New York state residents working in blue-collar occupations across all industries have experienced stagnant paychecks or scarce employment opportunities since 2000. This trend has been especially acute within the construction industry, as Figure A reflects that blue-collar workers in this sector had their inflation-adjusted weekly earnings devastated by the run-up and aftermath of the Great Recession. While the New York economy grew by 28 percent from 2000 to 2016 after accounting for inflation, these trends demonstrate that the state’s blue-collar workers have not shared in its prosperity. The prospect of a policy initiative—repealing or weakening the state’s prevailing wage law—that further undercuts the position of blue-collar New York state residents is economically and ethically disturbing.17

Table 1

Figure A

Prevailing wage laws stimulate the economy

One of the first lessons in any introductory macroeconomics course is that when workers earn higher wages, a region’s economy will grow. Not only does the worker (and their family) improve their standard of living, but their subsequent spending of these gains ripples through the economy to create economic growth through the “multiplier effect.” This basic set of economic principles has been overlooked in previously published analyses of New York’s prevailing wage laws.

This oversight is unfortunate, as the repeal or weakening of New York’s prevailing wage policy—and the subsequent reduction in worker wages in construction—would damage the economic outlook for thousands of state residents, their families, and their communities. As an estimate, Table 2 analyzes New York’s blue-collar construction labor force and the statewide earnings losses that would be attributable to the repeal of the state’s policy. Given the findings of Kessler and Katz (2001), Table 2 examines how 2 percent and 4 percent reductions in the average hourly wage of New York’s blue-collar construction workers would impact the state’s economy.

Table 2

As presented in Table 2, it is estimated that blue-collar construction workers in New York state earned a total of $18.975 billion in earnings in 2016. If New York’s prevailing wage policy was repealed and the average industry wage fell between 2 percent and 4 percent, this would equate to earnings losses between $379.5 million and $759.0 million. These earnings losses would only minimally be offset by increased hiring attributable to lower industry wages, as Maiti and Indra (2016) estimate the wage elasticity of labor demand in construction is just -0.14. This would indicate that the earnings of newly hired employees would be between $53.1 million and $106.3 million.18 On net after repeal, aggregate earnings by blue-collar construction workers in New York state would decline by between $326.4 million and $652.7 million. This decline would not only distress New York state residents employed in these jobs, but their subsequent reductions in spending would ripple through the state’s economy.19 Short of repeal, the earnings impact of policy weakening would depend on the particulars of the proposal; it is nevertheless obvious that any policy movement in that direction would have a significant and deleterious effect on New York’s economy.

Prevailing wage laws increase tax revenues

Any statewide loss of income attributable to the repeal or weakening of the New York’s prevailing wage law will also result in lower tax revenues for state and local governments. Using the figures estimated in the previous section, Table 3 projects the additional state sales and income tax revenues that are produced by the state’s prevailing wage policy directly from the increase in net earnings among blue-collar construction workers in the state. Given that a significant portion of New York state residents’ incomes are spent on items exempt from the state sales tax (4 percent), this study estimates the increase in state revenues from the sales tax by comparing the state’s annual sales tax revenues to its aggregate level of personal income (via the Bureau of Economic Analysis) annually between 2011 and 2016. That ratio—1.06 percent—has been stable year over year, indicating that the increase in the incomes for blue-collar construction workers in New York attributable to prevailing wage laws results in an additional $3.5 million to $6.9 million in sales tax revenue.20 Given that the median local sales tax in New York is also 4 percent, it would be expected that prevailing wage laws would have a comparable impact on local tax revenue.

Table 3

The progressive nature of New York state’s income tax system complicates efforts to project the estimated tax effect generated by the prevailing wage law; there is not sufficient data to precisely identify the proportion of blue-collar construction workers in each tax bracket. As a result, this study calculates the average ratio of state tax liability (after credits) to family income for New York’s blue-collar construction workforce from 2000 to 2016 as provided by the Annual Social and Economic Supplement of the Current Population Survey. The resulting ratio (1.70 percent) is small, but it reflects that a sizeable proportion of the state’s construction workers earn poverty-level wages and subsequently have a 0 percent effective tax rate. While there are many reasons to consider this a conservative estimate, this result suggests that New York’s prevailing wage law added between $5.6 million and $11.1 million in income tax to the state’s revenues in 2016 from blue-collar construction workers.21 Taken together, the results of Table 3 indicate that the state’s prevailing wage policy added between $9.0 million and $18.0 million to New York’s tax revenues last year.22

Prevailing wage laws reduce poverty

Blue-collar construction workers account for nearly 7 percent of New York state’s “working poor,” as there are an estimated 25,500 workers in the state employed in construction whose standard of living is below the poverty line.23 Substandard wages in some sectors of the industry are exacerbated by regular unemployment stints due to insufficient construction demand because of the weather or the health of the economy. Repealing or weakening the state’s prevailing wage law will likely make this problem worse. Table 4 shows that poverty among construction workers is far worse in states with weakened or nonexistent prevailing wage laws. In New York, the results indicate that 8.36 percent of blue-collar construction workers were living below the poverty line. This is comparable to the 8.14 percent among all states—like New York—with a “strong” or “average” prevailing wage law.24 However, in states with a “weak” or nonexistent prevailing wage law, 10.68 percent of blue-collar construction workers live in poverty.

Table 4

As would be expected, increased poverty rates by blue-collar construction workers in these states lead to an expanded reliance on government subsidies. As demonstrated in Table 4, receipt of the earned income tax credit (EITC) is substantially higher by blue-collar construction workers in states with a weak or nonexistent prevailing wage law (17.52 percent) compared with states with a stronger state policy (13.97 percent). To be clear, there may be other state-specific reasons to explain the differences in poverty rates beyond prevailing wage laws. But when paired with the findings of Kessler and Katz (2001), the results in Table 4 would logically suggest that the repeal or weakening of New York’s law would exacerbate poverty concerns for those employed in blue-collar construction occupations.25

Prevailing wage laws promote employment-based health insurance

In a time of instability in health insurance markets, state prevailing wage laws are one of the few public policies available to directly promote employment-based health insurance. Contractors working on state-funded projects are mandated under the law to compensate workers with the prevailing rate of fringe benefits in the area. As demonstrated in Table 5, blue-collar construction workers in states with a strong or average prevailing wage law—including New York—have substantially higher rates of employment-based health insurance (41.88 percent) than comparable workers in states with weak or nonexistent policies (30.41 percent). Although there may be a number of contributing factors to this outcome, the compensation mandates included in state prevailing wage laws are directly responsible for at least part of this differential. Any weakening of New York’s prevailing wage law would therefore impose considerable social and economic costs on the workers and their families who would lose their health insurance as a result of this change in state policy.

Table 5

Prevailing wage laws improve workers’ financial security in retirement

In addition to health insurance, the fringe benefit package required of contractors working on state-funded projects typically includes payments into an employer-sponsored pension fund. As presented in Table 5, the mandates included in prevailing wage laws directly contribute to higher rates of workplace pensions among blue-collar construction workers in states with a strong or average prevailing wage law (35.69 percent) when compared with states with a weak or nonexistent policy (24.86 percent). Increased wages attributable to prevailing wage laws also help workers accrue larger Social Security accounts, further strengthening their retirement security. Finally, requirements that contractors working on state-funded construction projects must submit a certified payroll to regulators help ensure that employees will not be misclassified as independent contractors. Being correctly classified as an employee requires the firm to pay the employer portion of Social Security taxes. Increased financial security at retirement is of critical concern given that blue-collar construction workers often retire with more physical limitations than those who were employed in other sectors of the economy, and prevailing wage laws are one of the most effective policy levers that lawmakers can utilize to improve their economic situation upon retirement.

Prevailing wage laws offer a pathway to the middle class

Perhaps the most important characteristic of a thriving state economy is its ability to create “good jobs” for New York residents, or employment that features sufficient pay, health insurance, and a retirement plan. These middle-class jobs have long represented the backbone of communities throughout New York state. But as workplaces have changed in recent decades, New York state residents have increasingly discovered that the pathways to the middle class that were taken by their parents’ generation are no longer viable. This is particularly true for those without a college degree and, as Figure B demonstrates, has been particularly acute since the turn of the century. Among New York state residents with a high school diploma but less than an associate degree, the proportion with a “good” job—defined as one with inflation-adjusted earnings of $40,000 or more, employer-sponsored health insurance, and a workplace retirement plan—declined sharply from 20.8 percent in 2000 to just 13.9 percent in 2016.26

Figure A

The economic plight of non–college educated workers in New York is only projected to get worse, as most of the jobs that will become available to these individuals through 2024 are likely to feature poverty-level wages or below. To demonstrate this, Table 6 documents the six fastest-growing occupations on a national basis for which at least 50 percent of job incumbents in New York do not have a post-secondary degree. The fastest-growing occupational category—health care support—is expected to feature brisk job growth through 2024, however New York state residents in these positions earn an average income of $30,540. This isn’t enough to support a family, as the average salary in these positions is so low that it would qualify a family of four for food stamps in New York state.27 Residents employed in the second-fastest growing occupation for the non–college educated—personal care and services—earn an even lower income ($29,870). At these rates of pay, these are not the kinds of jobs that provide the personal and economic stability required for healthy New York residents, families and communities.

Table 6

There are, however, occupations—such as construction—that offer non–college educated workers a clear pathway to the middle class. As outlined in Table 6, demand for construction workers is expected to grow by 10.1 percent between 2014 and 2024, with the average New Yorker employed in the construction trades earning $62,960 as of last year. These kinds of jobs are what can rebuild—literally—New York’s middle class.

New York state’s prevailing wage law is an important and effective policy lever that opens a pathway to the middle class for non–college educated residents. In addition to ensuring that experienced construction workers earn a middle-class income and receive benefits on state-funded projects, state prevailing wage laws also incentivize contractors to hire apprentices enrolled in a state-sanctioned program, many of which only require a high school diploma for admission. Under the law, contractors are able to compensate apprentices at a rate below the required prevailing wage. This motivates contractors to provide opportunities for inexperienced workers to acquire the on-the-job training necessary to develop into skilled tradespeople and the next generation of New York’s blue-collar middle-class. As highlighted by Bilginsoy (2005), Philips (1998) and Philips et al. (1995), repealing or weakening a state’s prevailing wage law will reduce apprenticeship opportunities. In a time of limited middle-class job opportunities for non–college educated New York residents, this would be an egregious mistake in this economy that would further exacerbate the decline of the state’s middle class.

Contractors and consumers

Prevailing wage laws level the playing field for law-abiding contractors

The construction industry—both nationally and in New York—is populated by a widely divergent set of contractors. On one hand, there are honest contractors who operate with transparency, in accordance with the law, and with the goal of doing right by their employees and customers. On the other hand, there are also unscrupulous contractors who engage in wage theft, exploit undocumented laborers, misclassify workers, evade tax payments (e.g., Social Security), and ignore unsafe working conditions. Construction firms whose profits depend on unethical actions are prevalent in New York (Bernhardt, Spiller, and Polson 2013; Donahue, Lamare, and Kotler 2007; Milkman, Gonzalez, and Ikeler 2012; Theodore, Valenzuela, and Melendez 2006), with many operating in sectors where there are numerous opportunities to bid on publicly-funded projects (Brennan Center for Justice 2007).

The illegal, cost-saving actions of these unethical actors put New York’s honest contractors at a competitive disadvantage. The state’s prevailing wage law, however, helps level the playing field. Under the law, contractors working on state-funded projects must submit a project-wide certified payroll report that includes employees’ names, Social Security numbers, pay rates, and number of hours. The prospect of more stringent government oversight substantially discourages bids from contractors whose competitiveness is dependent on unlawful employment strategies. By rooting out unscrupulous contractors, prevailing wage laws help effectuate the moral obligation of state and local governments to prioritize the bids of honest, fair-dealing contractors whose workplace practices adhere to labor and employment law.

Prevailing wage laws improve workplace safety

Debate over public policy in the construction industry is complicated by its high rates of workplace injuries and jobsite fatalities; the industry was responsible for one in five job-related deaths in 2014.28 As lawmakers and construction consumers, government officials in New York have an ethical responsibility to ensure that state residents are not unduly imperiled while working on public projects. While prevailing wage laws typically do not include safety requirements, the elimination of low-road contractors from bidding on public projects results in an increased reliance on high-road contractors with a greater commitment to workplace safety and worker training.

The relationship between state prevailing wage laws and increased worker safety has been demonstrated in a series of research studies. In the most advanced, peer-reviewed article to date, Azari-Rad (2005) used data from the Survey of Occupational Injury and Illness between 1976 and 1999 and found that nonfatal injury rates were 7 percent to 10 percent lower in states with prevailing wage laws. These results are supported by a pair of unpublished studies that showed that states with prevailing wage laws exhibit lower fatality rates in construction (Dickson Quesada et al. 2013) and decreased rates of disability among construction workers (Philips 2014). While the safety impact may be indirect, the research offers clear evidence that New York’s prevailing wage law effectively advantages high-road contractors whose commitment to safety helps minimize workplace injuries to residents working on state-funded construction projects.29

Prevailing wage laws advantage local contractors

Those opposing prevailing wage laws have long relied on an unsubstantiated claim that the policy reduces bid competition on public projects (e.g., Leef 2010). Two recent academic studies have suggested that this claim is inaccurate. In an analysis of 140 municipal projects in California in 2006 and 2007, Kim, Kuo-Liang, and Philips (2012) found no evidence that the absence of prevailing wage laws increased the number of bidders, nor that contractors changed their bidding strategy based on whether a project required prevailing wages to be paid.30 Duncan (2015a) found similar results in a study of 132 highway resurfacing projects in Colorado, as the presence or absence of prevailing wage requirements had no effect on the number of bidders.31

The results of these two studies demonstrate that, on net, prevailing wage laws do not affect the number of contractors bidding on public projects. But additional evidence suggests that prevailing wage policies may substantially advantage local, in-state contractors in the bidding process. In a non-academic review of 110 Ohio school construction projects open to bid between 2013 and 2016, Onsarigo et al. (2017) discovered that out-of-state contractors submitted the lowest bid on 21 percent (16 of 77) of projects that were not covered by prevailing wage law. In comparison, out-of-state contractors won just 3 percent (1 of 33) of the bids for projects that required prevailing wages to be paid. The effect of prevailing wage laws to advantage local contractors has yet to be examined in the academic press, however the results of Onsarigo et al. (2017) are consistent with the reasons why some prevailing wage laws were originally enacted. In addition to ensuring minimum labor standards, prevailing wage laws—including the federal Davis-Bacon Act of 1931—were designed to protect local contractors from being undercut by out-of-state firms relying on low-cost, low-road employment strategies that erode local labor market conditions (Gujarati 1967).

Prevailing wage laws develop a skilled workforce

Contractors have bemoaned skill shortages in the construction trades for decades (e.g., Weinberg 1969). The basic laws of supply and demand suggest that bona fide skill shortages in a particular region can typically be resolved by increasing wages. In construction, however, contractors are more often inclined to employ an alternative strategy when dealing with a perceived short-term skill shortage: recruiting and employing available out-of-region workers. This represents a missed opportunity for policymakers interested in generating new jobs for local residents, and the repeal of the state’s prevailing wage law—and the corresponding decline in industry wages—would only exacerbate the problem.

Perceived skill shortages in the industry may exist because of the cyclical and seasonal nature of construction work and the resulting instability of the employment relationship. For many contractors, workers hired during times of high construction demand are quickly jettisoned once demand slows. This loose attachment between worker and employer suppresses the incentives for many contractors to incur the costs of training. Employers cannot ensure that they will retain employees that have benefited from potential training investments, and are discomforted by the likelihood that these workers will eventually be employed by competing contractors. Given frequent spells of unemployment (combined with relatively low wages in entry-level positions), individual workers often have insufficient resources to incur the costs of training themselves. In sum, the cyclical and seasonal nature of construction may lead to a less than optimal level of training in the industry.

State prevailing wage laws are one of the few effective policy levers available to lawmakers to increase the number of skilled workers in construction. The policy’s compensation exemption for apprentices incentivizes contractors—both union and nonunion—to hire a fixed number of relatively inexperienced workers admitted to a state-sanctioned apprenticeship program to work alongside more skilled, experienced workers on a job site; this on-the-job training is critical for workforce development. The promotion of apprenticeship programs is amplified by state prevailing wage laws in other ways. For union contractors, a portion of most unions’ hourly compensation package sets aside a per-hour portion to be contributed to a joint union-management apprenticeship training fund that supports classroom instruction that trains the next generation of tradespeople available to all union contractors. Nonunion contractors seeking to bid on prevailing wage projects require a sufficiently trained workforce whose productivity and skill make higher wages economically viable and allow them to be competitive with union firms that traditionally have employed better-trained workforces. To incentivize this caliber of firm-based training and development, nonunion contractors require access to a sufficient number of projects requiring these skills unavailable to low-road, rival contractors; this ensures they will maintain enough labor demand to continue the employment of their skilled workforce and reap the benefits of training investments.

Prevailing wage laws improve quality and on-time completion

The singular focus of public debate on construction costs ignores two other critical construction outcomes: quality craftsmanship and on-time completion. Poorly constructed bridges and long-delayed school openings can have disastrous consequences for New York families and communities. To date, most experts posit that prevailing wage laws—through its advantaging of high-road contractors employing the industry’s most skilled workers—lead to better construction quality and greater on-time completion (e.g., Philips 2014; Kelsay 2016). While these conclusions may follow logically from the other demonstrated benefits of prevailing wage laws, it should be noted that a lack of available data on quality and timeliness have limited academic researchers from empirically evaluating these hypotheses.

Conclusion: Benefits

The single-minded focus on costs by the Center for Urban Real Estate (Vitullo-Martin 2012) and the Empire Center (McMahon and Gardner 2017) has altered the public narrative about New York’s state prevailing wage law by obscuring the benefits that these laws provide to the state’s residents and communities. That is misguided, as there is considerable evidence that prevailing wage laws improve the standard of living for blue-collar workers, improve workplace safety, advantage in-state contractors, minimize illegal employment practices, develop worker skills, and provide a pathway to the middle class unavailable in other sectors for state residents without a college degree. Those who ignore the benefits of prevailing wage laws in framing public debate are, at best, offering a one-sided, incomplete depiction of the policy that does a disservice to New York’s blue-collar workers and the communities in which they reside.

Concluding discussion

Those seeking to repeal or weaken state prevailing wage laws have been incredibly effective in recent years in shaping public and political opinion. Their approach has relied on (a) a singular focus on the costs of the policy and (b) an argument—higher wages mean higher taxpayer costs—whose simplicity and intuitiveness seems to powerfully resonate with lawmakers and residents alike. Opponents of prevailing wages have exploited this narrative in campaigns across the country, successfully convincing lawmakers to repeal state prevailing wage laws in Indiana (2015) and West Virginia (2016) and politically or effectively weaken these laws in many other states across the country.

But there is a fundamental problem with the simple narrative that prevailing wage laws increase costs: the statistical evidence says otherwise. The cost of state prevailing wage laws has been a considerable focus of independent, academic economists over the last 15 years. In study after study, the results demonstrate a clear consensus: state prevailing wage laws have not been shown to increase taxpayer costs on the biggest components of state construction budgets (roads and schools). If this seems counterintuitive, consider that high-wage contractors employ the most skilled and most productive workers and use the industry’s most advanced technology and equipment; this allows them to place bids on public construction projects that are competitive with—if not better than—those of low-wage, low-skill contractors. Essentially, state lawmakers “get what they pay for” when it comes to hiring contractors and workers to build public construction projects.

There is another fundamental problem with the current narrative on state prevailing wage laws: it entirely ignores the many benefits that the law provides a state’s residents and communities. In a time when economic opportunities for blue-collar workers are slipping away—devastating families and communities—prevailing wage laws are one of the few effective policies available to state lawmakers that increase the standard of living for these workers, incentivize employers to provide opportunities for training and skill development, and offer a clear pathway to the middle class for non–college educated state residents. Prevailing wage laws also advantage in-state and law-abiding contractors, reduce illegal employment practices, and improve workplace safety for a state’s residents. Any public discussion about state prevailing wage laws that ignores the benefits of the policy does an incredible disservice to a state’s workers, families, and communities.

It is acknowledged that, in discussing New York’s prevailing wage policy, state lawmakers are faced with two ethical obligations that, on the surface, appear to be mutually exclusive: the need to minimize taxpayer costs against the responsibility of ensuring fair wages, benefits, and working conditions for its residents on public construction projects. From a review of the current academic research, it would appear that New York’s prevailing wage law potentially offers policymakers the opportunity to fulfill both responsibilities. On the biggest components of the state’s construction budget, the consensus of academic studies would be that the policy has little to no effect on taxpayer costs. There is also considerable evidence that the policy raises the standard of living of a state’s residents, improves workplace safety, and offers a clear path to the middle class for New York state residents without a college degree. In sum, research by independent, academic economists suggests that New York’s prevailing wage law is a uniquely valuable component of state policy that simultaneously uplifts residents and communities while imposing little, if any, cost on taxpayers.

Response to Empire Center

This report would be remiss without directly addressing the points raised by the Empire Center, which published a 2017 report (McMahon and Gardner 2017) critical of the state’s prevailing wage law and proposing a series of amendments to substantially weaken the policy. First, the authors generate cost estimates of New York’s policy using a variant of the widely discredited “hypothetical” approach discussed earlier in this paper. By comparing the level of prevailing wage with some arbitrary, lower wage level, the Empire Center does demonstrate that prevailing wages may increase labor cost per hour on some projects, if it assumes that the state government would have used a low-wage contractor instead of a high-wage contractor. But the authors commit a significant logical fallacy in blindly suggesting that this equates to showing that prevailing wages would increase total construction costs. “Efficiency wages” are a basic concept in labor economics, as it explains how firms using a high-wage, high-skill workforce can effectively compete with firms that employ low-wage, low-skill workers; in essence, this is how Costco and Wegmans are able to compete against Wal-Mart in the grocery industry. Within the construction industry, the productivity differential between high-skill and low-skill construction workers is substantial (Atalah 2013a, 2013b), which makes the authors’ explicit dismissal of productivity differences (McMahon and Gardner 2017, 9) to be, at best, puzzling.

Lawmakers and residents should also be clear-eyed that many of the Empire Center’s basic arguments against the state’s prevailing wage law—beyond a cost estimate that relies on a discredited methodology—have little to do with the actual law itself. Instead, it seems that the authors’ strategy to undermine state policy is to explicitly connect the law to New York’s construction unions—evidenced, in part, by its cartoonish cover—and then, by proxy, discredit the policy by disparaging the unions. Besides the use of simple misdirection, this approach conceals two more logical fallacies. First, it is not clear that many of these critiques have any relationship with prevailing wage or construction costs. The Empire Center criticizes unions’ work rules and underfunded pensions, but increased job security stipulations and pension contributions are typically offset in collective bargaining through lower wage requirements.32 Second, and perhaps more importantly, the authors’ arguments rely on their implicit narrative that union contractors—and only union contractors—bid on prevailing wage projects (pg. 7). This is simply untrue. In fact, in the study of prevailing wage laws in California, Kim, Kuo-Liang, and Philips (2012) found that 26 percent of bidders on prevailing wage projects were nonunion contractors. Duncan (2015a) found similar results in Colorado, and there is nothing to suspect that things would be markedly different in New York state. Much of the Empire Center’s argumentation relied on the assumption that prevailing wage projects were the explicit domain of construction unions; evidence to the contrary calls into question their approach and, as a result, their conclusions.

About the authors

Russell Ormiston is associate professor of economics at Allegheny College and research scholar for the Institute for Construction Economic Research. Dale Belman is professor of labor relations and human resources at Michigan State University and president of the Institute for Construction Economic Research. Matt Hinkel is a Ph.D. Student, Labor Relations and Human Resources, Michigan State University.

Appendix A

Appendix B

Endnotes

1. The New York State Fiscal Year 2018 Enacted Capital Program and Financing Plan can be accessed at: https://www.budget.ny.gov/budgetFP/enactedfy18.html.

2. Bids for federally funded interstate projects also had to adhere to the requirements of the Disadvantaged Business Enterprise Program. However, given that the results of the study showed that the policies collectively had no effect on cost, it is highly unlikely that this latter policy is offsetting an undetected cost increase attributable to prevailing wage law.

3. The only other recent study of the effects of prevailing wage laws on transportation spending was Vitaliano (2002). Using data on all 50 states from 1996, the author used a sparse, six-variable regression model to link the presence of a state prevailing wage law with total expenditure by the respective state’s department of transportation. While the results suggest that prevailing wage increases state costs by 8 percent, the approach failed to include the most critical variables that predict total expenditure, such as the level of law enforcement staffing, number and scope of projects ordered, and so on. This debilitating omitted variable bias substantially weakens the paper to the point that it has questionable validity as a part of public policy discussion.

4. Concerns about the applicability of Colorado research to New York markets are attenuated by the fact that the two states seemingly have a comparable wage differential between high-wage and low-wage contractors, as presented in Appendix A. Given that data on individual contractors is not publicly available, this report instead used data from the Outgoing Rotation Group of the Current Population Survey between 2000 and 2016 to compile a reasonable sample size of blue-collar construction workers from Colorado and New York (n=498). Proxying union members as employed by high-wage contractors and nonunion members to be working for low-wage contractors, the results of a regression model on workers’ hourly wage does indicate that union members, as expected, are paid more (14.8 percent) on an hourly basis. However, the results indicate the union differential in New York is statistically indistinguishable from that of Colorado. While this approach is rife with caveats—e.g., the sample cannot separate residential from nonresidential workers, it relies on workers’ state of residence and not state of employment—it represents the best approach to compare the wage differential between high-wage and low-wage contractors across the two states. The results of the regression—and the nonsignificance of the union-state interaction term—offer preliminary evidence that contractors in New York state should be able to substitute low-wage workers with high-wage workers at a similar rate to those featured in the research on Colorado.

5. A complete summary of these studies is provided in Duncan and Ormiston (2018). The five academic articles are Bilginsoy and Philips (2000); Azari-Rad, Philips, and Prus (2002, 2003); Vincent and Monkkonen (2010); and Duncan, Philips, and Prus (2014). The four non-academic studies are: Prus (1999), the Ohio Legislative Service Commission (2002), Kelsay (2016), and Onsarigo et al. (2017).

6. The lone dissenting study—Vincent and Monkkonen (2010)—found results linking state prevailing wage laws to higher school construction costs, however their approach has been questioned given their failure to control for the cost effect of local economic conditions (Onsarigo et al., 2017).

7. The Ohio Legislative Service Commission (2002) report posited that the school construction exemption in the state’s prevailing wage law saved the state 10.7 percent; however this estimate was derived from three regression models in which none of the prevailing wage coefficients were even close to conventional levels of statistical significance. The decision to explicitly dismiss concerns over statistical significance is at odds with normal statistical practice as found in introductory econometrics textbooks, and has been criticized in multiple subsequent reports (Onsarigo et al. 2017; Duncan and Ormiston 2018).

8. In a follow-up study, Atalah (2013b) employed the same data set and approach but instead compared union and nonunion bids within each trade. While within-trade samples tended to be small, the study found no statistically significant difference between bids from union and nonunion contractors in 13 of the 18 trades studied. Of the five trades where there was a significant difference, three featured higher bids from union contractors while two demonstrated higher bids from nonunion contractors.

9. Dunn, Quigley and Rosenthal (2005) used two different types of regression methods—ordinary least squares and instrumental variables—to study the effect of prevailing wage laws on the construction costs of affordable housing. Under OLS, which was used by the NYC IBO study, the authors suggested that the cost effect was between 9 percent and 11 percent. Using instrumental variables, Dunn, Quigley and Rosenthal (2005) estimated that the cost impact was between 19 percent and 37 percent, however this approach was heavily critiqued by Mukhopadhyay, Harris and Wiseman (2013) and Littlehale (2017) for its use of weak first-stage instruments that inflated the estimates.

10. The Minnesota Office of the Legislative Auditor (2007) critiqued Dunn, Quigley and Rosenthal (2005) for their failure to consider the role of HUD requirements, thereby suggesting that their estimates may also be inflated. While the failure to properly control for other federal requirements—beyond prevailing wage—may complicate the research on affordable housing construction costs, these concerns are not as applicable when examining research on highways and schools. For the former, Duncan (2015a) specifically notes that the building specifications of state and federal highways are identical with two exceptions: federal prevailing wage and the requirements of the Disadvantaged Business Enterprise Program. In regards to schools, many studies examine the cost impact of prevailing wage by looking at costs before and after the onset or repeal of prevailing wage (e.g., the exemption of schools from Ohio’s prevailing wage law in 1997); this would minimize the influence of changes in complementary public policies.

11. While not published in academic press, a collection of California government agencies prepared a study that collected data on 400 affordable housing projects in the state from 2001 to 2011 (California Department of Housing and Community Development et al. 2014). Using regression analysis, the study estimated that prevailing wage laws increased construction costs by 11 percent, however the authors noted that the size and statistical significance of this estimate were particularly sensitive to the specification of the model.

12. Excluding year and region fixed effects, the NYC IBO study featured nine variables; in contrast, the two academic papers featured models that included nearly three times as many variables. Many of the variables included in the academic studies—but not in the NYC IBO report—were statistically significant, This includes, but is not limited to, data on profit/nonprofit status of the developer, architecture and engineering costs, parking characteristics (not just a dummy variable indicating whether parking was a part of the project), total number of units in the structure (not just the number of affordable units), and nonresidential components of the project.

13. A subsequent, non-academic study by Yildirim and Lee (2016) used the same methodology employed by the NYC IBO study to estimate how prevailing wage laws affect construction costs in affordable housing across each of New York City’s five boroughs for the same time period (2010 through 2015). The study’s reliance on the same regression model, however, exposes it to the same concerns as the NYC IBO paper.

14. While Kessler and Katz (2001) find that hourly wage declines are more concentrated among union workers following the repeal of a state’s prevailing wage law, their results demonstrate that wages for nonunion workers do not increase as a result.

15. Estimates between 2 percent and 4 percent provide conservative projections. In contrast, Kelsay (2016) cites a series of studies to use projections between 3 percent and 5 percent.

16. According to the Bureau of Labor Statistics, the average “production” worker in the construction industry earned $36.16 per hour and worked 37.9 hours per week in New York during 2016. While the BLS does not provide the average number of weeks worked—a critical issue given the on-again, off-again nature of construction employment—a 50-week work-year would result in the average blue-collar construction losing $2,056 assuming a 3 percent reduction in the hourly wage.

17. According to the Bureau of Economic Analysis, the real GDP for New York state in 2000 was $1,161.4 billion (2016 dollars). In 2016, the BEA estimated the state’s real GDP to be $1,488.0 billion, indicating a growth rate of 28.1 percent since 2000.

18. The conclusion that newly hired workers would earn between $53 million and $106 million is based on the assumption that these workers would be paid the industry average hourly wage. In all likelihood, new hires would be paid much lower, entry-level wages. Projections based on the industry average, however, are used as a means of being conservative in the estimates of the overall economic impact of the state’s prevailing wage law.

19. There are numerous reasons to believe that these estimates may be conservative. First, Onsarigo et al. (2017) discovered that bidding by out-of-state contractors substantially increased on Ohio school projects after the state excluded schools from prevailing wage. While this was based on a small sample (n=110) and was not a part of a peer-reviewed study, any increase in out-of-state contractors would take jobs—and income—away from New York workers, thus exacerbating earnings losses attributable to a potential repeal of the state policy. Second, in a non-academic study, Philips (2014) suggested that the decline in the average hourly wage of construction workers following the repeal of a state’s prevailing wage law may approach 8 percent; if that is true, that would nearly double the maximum earnings impact estimated for New York’s law in Table 2.

20. Using the aggregate state ratio of sales tax revenues to income in New York represents a conservative estimate of how earnings losses among New York’s blue-collar construction workers would affect state tax revenues. Given differences in marginal propensities to consume and save, high-income New York state residents likely devote relatively less of their aggregate income to sales tax when compared with most blue-collar construction workers.

21. Kessler and Katz (2001) outlined that high-wage employees are likely to experience more extreme income losses following the repeal of a state’s prevailing wage law when compared with low-wage employees. Given the progressive nature of New York’s income tax structure, it is likely that the income tax estimates in Table 3 underestimate the annual effect. Further, a considerable percentage of New York’s blue-collar construction workers featured in the Annual Social and Economic Supplement of the Current Population Survey were estimated to have negative effective tax rates. Since it is highly unlikely that additional workplace income would further decrease the tax burden for these workers as the negative ratio would indicate, some thought was given to replacing those negative rates with zeroes in the calculation (the aggregate ratio increases from 0.0170 to 0.0198 as a result). However, given the existence of the Earned Income Tax Credit (EITC) and in the interest of keeping the estimates conservative, this study used the lower (0.0170) rate.

22. The emphasis on state and local taxes ignores the additional monies received by the federal government as a result of New York’s prevailing wage law, including income tax and workplace payments into Social Security and Medicare.

23. Using the 2000 to 2016 Annual Social and Economics Supplement (ASEC) to the Current Population Survey, workers in the construction industry accounted for 6.84 percent of individuals who were employed and were deemed to be below the poverty line. According to the Bureau of Labor Statistics Current Employment Statistics program, there were 372,900 workers employed in construction in New York state in 2016. The resulting estimate—25,500—may be conservative given that the CPS likely undercounts low-paid, undocumented workers employed in construction.

24. The “strength” of a state’s prevailing wage law was originally introduced by Thieblot (1995). This scoring system classifies state policies on the basis of five categories—minimum contract threshold, contracts covered, enforced wage rate, breadth of work covered, and an “other” category—and awards points based on established criteria. Thieblot (1995) classifies “strong” laws as those with 12+ points, “average” laws as those with 7–11 points and “weak” laws as those with 1–6 points. Belman, Ormiston, and Petty (2017) used these guidelines to reevaluate state prevailing wage laws in 1979, 1994, and 2006. For this report, the 2006 scores are used to categorize states.

25. To date, there has not been an academic study that has attempted to isolate the effect of prevailing wage laws in predicting poverty rates among blue-collar construction workers. But in a study of Ohio’s prevailing wage law, Onsarigo et al. (2017) used a cross-sectional regression model to demonstrate that, on a national level, states with a weak or nonexistent prevailing wage law had 3.1 percent higher poverty rates than states with a strong or average law.

26. Wage and salary income is measured in 2015 US dollars.

27. See https://otda.ny.gov/programs/snap/#eligibility.

28. Fatalities in construction represented 20.5 percent of all fatalities in private industry and 18.6 percent of total fatal injuries in 2014. See “Fatal Occupational Injuries by Industry and Event or Exposure, all United States, 2014” via the Bureau of Labor Statistics at http://www.bls.gov/iif/oshwc/cfoi/cftb0286.pdf.

29. The positive safety impact of prevailing wage laws are indirectly supported by an overwhelming amount of research demonstrating that union jobsites are substantially safer than nonunion jobsites. For example, Zullo (2011) finds that fatalities in the construction industry are lower in states with greater union density. Miller et al. (2013) demonstrate that nonunion contractors commit significantly more OSHA violations than their union counterparts in Missouri. Better safety records among union contractors is the result of a number of factors, including better worker training (Bilginsoy 2005), more stringent awareness and enforcement of OSHA requirements at union worksites (Weil 1992) and a workplace culture that more greatly emphasizes safety concerns (Gillen et al. 2002).

30. While California state prevailing wage laws govern state and locally financed public projects, “charter cities” are able to choose to exempt themselves from paying prevailing wage rates (Kim, Kuo-Liang, and Philips 2012). This led the authors to be able to compare projects built under prevailing wage provisions and those projects that were exempt from the law.

31. If prevailing wage laws discourage some low-road contractors from bidding on a public project, the results of Kim, Kuo-Liang and Philips (2012) and Duncan (2015a) would indicate that either (a) the reduction in the number of bids by low-road contractors is sufficiently small so as to not be statistically significant or (b) this reduction is offset by increased bidding from other high-road contractors.

32. On page 16 of its report, the Empire Center implicitly suggests that New York’s prevailing wage law is racially discriminatory given that it advantages labor unions. The Empire Center’s implied position is contradicted by a volume of academic-caliber studies that has thoroughly debunked the hypothesis that state prevailing wage laws have any discriminatory effect (Kessler and Katz 2001; Azari-Rad, Philips, and Prus 2003; Belman 2005; Manzo, Duncan and Lantsberg 2015; Belman, Ormiston, and Petty 2017). Further, in a non-academic examination of the data on New York City construction workers from the Bureau of Labor Statistics, Mishel (2017) demonstrates that claims that New York’s unions are discriminatory are historically anachronistic; as an example, African-Americans were more represented in the union construction workforce (21.2 percent) than the nonunion sector (15.8 percent) between 2006 and 2015.

References

Atalah, Alan. 2013a. “Comparison of Union and Non-Union Bids on Ohio School Facilities Commission Construction Projects.” International Journal of Economics and Management Engineering, Vol. 3, Issue 1, pp. 29-35.

Atalah, Alan. 2013b. “Impact of Prevailing Wages on the Cost among the Various Construction Trades.” Journal of Civil Engineering and Architecture, Vol. 7, No. 6, pp. 670-676.

Azari-Rad, Hamid. 2005. “Prevailing Wage Laws and Injury Rates in Construction.” In Hamid Azari-Rad, Peter Philips, and Mark Prus (Eds.), The Economics of Prevailing Wage Laws, pp. 169–187. Aldershot, UK: Ashgate.

Azari-Rad, Hamid, and Peter Philips. 2003. “Race and Prevailing Wage Laws in the Construction Industry: Comment on Thieblot.” Journal of Labor Research, Vol. 24, 161-168.

Azari-Rad, Hamid, Peter Philips, and Mark Prus. 2002. “Making Hay When It Rains: The Effect Prevailing Wage Regulations, Scale Economies, Seasonal, Cyclical and Local Business Patterns Have On School Construction Costs.” Journal of Education Finance, Vol. 27, 997-1012.

Azari-Rad, Hamid, Peter Philips, and Mark Prus. 2003 ‘State Prevailing Wage Laws and School Construction Costs.’ Industrial Relations, Vol. 42, No. 3, pp. 445-457.

Belman, Dale. 2005. “Prevailing Wage Laws, Unions, and Minority Employment in Construction.” In Hamid Azari-Rad, Peter Philips, and Mark Prus (Eds.), The Economics of Prevailing Wage Laws, pp. 101–119. Aldershot, UK: Ashgate.

Belman, Dale, Russell Ormiston, and Ryan Petty. 2017. “The Effect of State Prevailing Wage Laws on African-American Employment in Construction.” Unpublished working paper.

Bernhardt, Annette, Michael Spiller, and Diana Polson. 2013. “All Work and No Pay: Violations of Employment and Labor Laws in Chicago, Los Angeles and New York.” Social Forces, Vol. 91, No. 3, pp. 725-746.

Bilginsoy, Cihan. 2005. “Wage Regulation and Training: The Impact of State Prevailing Wage Laws on Apprenticeship,” in Hamid Azari-Rad, Peter Philips and Mark J. Prus (eds.) The Economics of Prevailing Wage Laws, pp. 149-168. Aldershot, UK: Ashgate.

Bilginsoy, Cihan and Peter Philips. 2000. ‘Prevailing Wage Regulations and School Construction Costs: Evidence from British Columbia.’ Journal of Education Finance, Vol. 24, pp. 415-432.

Brennan Center for Justice. 2007. “Unregulated Work in the Construction Industry in New York City.” Accessed at: https://www.brennancenter.org/sites/default/files/legacy/d/download_file_49377.pdf.

Bureau of Labor Statistics, Current Employment Statistics program (BLS CES). Various years. State and Metro Area Employment, Hours and Earnings [online data tables].https://www.bls.gov/sae/

Bureau of Labor Statistics, Occupational Employment Statistics (BLS OES). Various years. OES Data [online data tables].

California Department of Housing and Community Development, California Tax Credit Allocation Committee, California Housing Finance Agency, and California Debt Limit Allocation Committee. 2014. “Affordable Housing Cost Study.” Accessed at: http://www.treasurer.ca.gov/ctcac/affordable_housing.pdf.

Dickson Quesada, Allison, Frank Manzo, Dale Belman, and Robert Bruno. 2013. “A Weakened State: The Economic and Social Impacts of Repeal of the Prevailing Law in Illinois.” Labor Education Program, School of Labor and Employment Relations, University of Illinois at Urbana-Champaign. Accessed at: https://ler.illinois.edu/wp-content/uploads/2015/01/PWL_policy-brief_spreads041.pdf.

Donahue, Linda H., James Ryan Lamare, and Fred B. Kotler. 2007. “The Cost of Worker Misclassification in New York State.” Digital Commons at Cornell University, School of Industrial and Labor Relations. Accessed at: http://digitalcommons.ilr.cornell.edu/reports/9/.

Duncan, Kevin. 2015a. “The Effect of Federal Davis-Bacon and Disadvantaged Business Enterprise Regulations on Highway Maintenance Costs.” Industrial and Labor Relations Review, Vol. 68, No. 1, pp. 212-237.

Duncan, Kevin. 2015b. “Do Federal Davis-Bacon and Disadvantaged Business Enterprise Regulations Affect Aggressive Bidding? Evidence from Highway Procurement Auction,” Journal of Public Procurement, Vol. 15, Issue 3, pp. 291-316.

Duncan, Kevin and Alex Lantsberg. 2015. “Building the Golden State: The Economic Impacts of California’s Prevailing Wage Policy.” Accessed at: http://www.smartcitiesprevail.org/2015/04/14/study-prevailing-wage-adds-17-5k-jobs-california-economy/.

Duncan, Kevin, and Russell Ormiston. 2018. “What Does the Research Tell Us About Prevailing Wage Laws?” Labor Studies Journal, forthcoming.

Duncan, Kevin, Peter Philips, and Mark Prus. 2014. “Prevailing Wage Regulations and School Construction Costs: Cumulative Evidence from British Columbia.” Industrial Relations, Vol. 53, No. 4, October, pp. 593-616.

Dunn, Sarah, John Quigley, and Larry Rosenthal. 2005. “The Effects of Prevailing Wage Regulations on the Cost of Low-Income Housing,” Industrial and Labor Relations Review, Vol. 59, No. 1, pp. 141-157.

Flood, Sarah, Miriam King, Steven Ruggles, and J. Robert Warren. 2017. Integrated Public Use Microdata Series, Current Population Survey: Version 5.0. [dataset]. Minneapolis: University of Minnesota. https://doi.org/10.18128/D030.V5.0.

Gardner, Kent, and Rochelle Ruffer. 2008. “Prevailing Wage in New York State.” Published by the Center for Governmental Research.

Gillen, Marion, David Baltz, Margy Gassel, Luz Kirsch, and Diane Vaccaro. 2002. “Perceived Safety Climate, Job Demands, and Coworker Support among Union and Nonunion Injured Construction Workers.” Journal of Safety Research, Vol. 33, No. 1, pp. 33-51. Accessed at: http://www.sciencedirect.com/science/article/pii/S0022437502000026.

Gujarati, Damodar. 1967. “The Prevailing Wage Law in the USA,” Indian Journal of Industrial Relations, Vol. 2, No. 4, pp. 514-520.

Kelsay, Michael. 2016. “The Adverse Economic Impact from Repeal of the Prevailing Wage Law in Missouri.” Accessed at: http://www.faircontracting.org/PDFs/prevailing_wages/The%20Adverse%20Economic%20Impact%20from%20Repeal%20of%20the%20Prevailing%20Wage%20Law%20in%20Missouri.pdf.

Kessler, Daniel and Lawrence Katz. 2001. “Prevailing Wage Laws and Construction Labor Markets,” Industrial and Labor Relations Review, Vol. 54, No. 2, pp. 259-274.

Kim, Jaewhan, Chang Kuo-Liang, and Peter Philips. 2012. “The Effect of Prevailing Wage Requirements on Contractor Bid Participation and Behavior: A Comparison of Palo Alto, California with Four Nearby Prevailing Wage Municipalities.” Industrial Relations, Vol. 51, No. 4, pp. 874-891.

Leef, George. 2010. “Prevailing Wage Laws: Public Interest or Special Interest Legislation?” Cato Journal, Vol. 30, No. 1, pp. 137-154.

Littlehale, Scott. 2017. “Revisiting the Costs of Developing New Subsized Housing.” Berkley Planning Journal. Forthcoming.

Maiti, Abhradeep, and Debarshi Indra. 2016. “Regional Variations in Labor Demand Elasticity: Evidence from U.S. Counties.” Journal of Regional Science, Vol. 56, No. 4, pp. 635-658.

Manzo, Frank, Alex Lantsberg, and Kevin Duncan. 2015. “The Economic, Fiscal, and Social Impacts of State Prevailing Wage Laws: Choosing Between the High Road and the Low Road in the Construction Industry.” Unpublished working paper.

McMahon, E.J., and Kent Gardner. 2017. “Prevailing Wa$te: New York’s Costly Public Works Pay Mandate.” Published by the Empire Center. Accessed at: http://www.thepartnership.org/wp-content/uploads/2017/05/Prevailing-Wage-Report-Empire-Center.pdf

Milkman, Ruth, Ana Luz Gonzalez, and Peter Ikeler. 2012. “Wage and Hour Violationsin Urban Labour Markets: A Comparison of Los Angeles, New York and Chicago.” Industrial Relations Journal, Vol. 43, No. 5, pp. 378-398.

Miller, Harry, Tara Hill, Kris Mason, and John Gaal. 2013. “An Analysis of Safety Culture & Safety Training: Comparing the Impact of Union, Non-Union, and Right-to-Work Construction Venues.” Online Journal of Workforce Education and Development, Vol. VI, No. 2.

Minnesota Office of the Legislative Auditor. 2007. “Evaluation Report: Prevailing Wages.” Accessed at: http://www.auditor.leg.state.mn.us/ped/pedrep/prevailingwages.pdf.

Mishel, Larry. 2017. “Diversity in the New York City Union and Nonunion Construction Sectors.” Economic Policy Institute. Accessed at: http://www.epi.org/publication/diversity-in-the-nyc-construction-union-and-nonunion-sectors/.

Mukhopadhyay, Sankar, Thomas Harris and Nathan Wiseman. 2013. “A Report on the Direct and Indirect Effects of Prevailing Wage Legislations on Society.” Accessed at: http://www.faircontracting.org/wp-content/uploads/2014/05/A-Report-on-the-Direct-and-Indirect-Effects-of-Prevailing-Wage-Legislations-on-Society.pdf.

New York City Independent Budget Office (NYC IBO). 2016. “Assessing the Costs: The Impact of Prevailing Wage Requirements on Affordable Housing Construction in New York City.” Accessed at: http://www.ibo.nyc.ny.us/iboreports/the-impact-of-prevailing-wage-requirement-on-affordable-housing-construction-in-new-york-city.pdf

Ohio Legislative Service Commission, “The Effects of the Exemption of School Construction Projects From Ohio’s Prevailing Wage Law, S.B. 102 Report, Staff Research Report No. 149, May 20, 2002. Accessed at: http://www.lsc.ohio.gov/research/srr149.pdf.

Onsarigo, Lameck, Alan Atalah, Frank Manzo IV, and Kevin Duncan. 2017. “The Economic, Fiscal, and Social Effects of Ohio’s Prevailing Wage Law,” Accessed at: https://midwestepi.files.wordpress.com/2016/05/bowling-green-su-kent-state-ohio-pw-study-4-10-17.pdf.

Palm, Matthew and Deb Niemeier. 2017. “Does Placing Affordable Housing Near Rail Raise Development Costs? Evidence from California’s Four Largest Metropolitan Planning Organizations.” Housing Policy Debate. Accessed online at: https://doi.org/10.1080/10511482.2017.1331367.

Philips, Peter. 1998. “Kansas and Prevailing Wage Legislation.” Accessed at: http://www.faircontracting.org/PDFs/prevailing_wages/kansas_prevailing_wage.pdf.

Philips, Peter. 2014. “Kentucky’s Prevailing Wage Law: An Economic Impact Analysis.” Accessed at: http://www.faircontracting.org/wp-content/uploads/2014/02/Kentucky-Report-2014-Philips.pdf.

Philips, Peter, Garth Mangum, Norm Waitzman, and Anne Yeagle. 1995. “Losing Ground: Lessons from the Repeal of Nine “Little Davis-Bacon” Acts.” Accessed at: http://www.faircontracting.org/PDFs/prevailing_wages/losingground.pdf.

Prus, Mark. 1999. “Prevailing Wage Laws and School Construction Costs: An Analysis of Public School Construction in Maryland and the Mid Atlantic States.” Accessed at: http://eric.ed.gov/?id=ED456630.

Roistacher, Elizabeth, Jerilyn Perine, and Harold Shultz. 2008. “Prevailing Wisdom: The Potential Impact of Prevailing Wages on Affordable Housing.” Published by the Citizens Housing and Planning Council. Accessed at: http://chpcny.org/wp-content/uploads/2011/02/Prevailing-Wisdom-web-version1.pdf

Theodore, Nik, Abel Valenzuela and Edwin Melendez. 2006. “La Esquina (the corner): Day Laborers on the Margins of New York’s Formal Economy.” Working USA, Vol. 9, pp. 407-23.

Thieblot, Armand. 1995. “State Prevailing Wage Laws: An Assessment at the Start of 1995,” Rosslyn, Va.: State Relations Department, Associated Builders and Contractors.

Vincent, Jeffery and Paavo Monkkonen. 2010. “The Impact of State Regulations on the

Cost of Public School Construction,” Journal of Education Finance, Vol. 35, No. 4, pp. 313-330.

Vitaliano, Donald. 2002. “An Econometric Assessment of the Economic Efficiency of State Departments of Transportation,” International Journal of Transportation Economics, Vol. 29, No. 2, pp. 167-180.

Vitullo-Martin, Julia. 2012. “The Complex Worlds of New York Prevailing Wage.” Published by the Center for Urban Real Estate. Accessed at: http://www.nysafah.org/cmsBuilder/uploads/The-Complex-World-of-Prevailing-Wage.pdf.

Weil, David. 1992. “Building safety: The Role of Construction Unions in the Enforcement of OSHA.” Journal of Labor Research, Vol. 13, Issue 1, pp. 121-132.

Weinberg, Edgar. 1969. “Reducing Skill Shortages in Construction.” Monthly Labor Review, Vol. 92, No, 2, pp. 3-9.

Yildirim, Yildiray, and Eunkyu Lee. 2016. “The Impact of Prevailing Wages on Construction Project Costs in NYC.” Published by the Newman Real Estate Institute. Accessed at: http://www.baruch.cuny.edu/realestate/research-publishing/documents/NREI_PrevailingWage_October2016_110316.pdf.

Zullo, Rolland. 2011. “Right-to-Work Laws and Fatalities in Construction.” Ann Arbor, MI: Institute for Research on Labor, Employment, and the Economy, University of Michigan. Accessed at: http://irlee.umich.edu/wp-content/uploads/2016/05/RightToWorkLawsAndFatalitiesInConstruction.pdf.

The expansion of collective bargaining that followed the passage of the National Labor Relations Act (NLRA) in 1935 led to decades of faster and fairer economic growth that persisted until the late 1970s. But since the 1970s, declining unionization has fueled rising inequality and stalled economic progress for the broad American middle class. These trends need to be halted and reversed.

Figure A

An idea that is gaining some traction is to address the fact that, under current law, the NLRA “preempts” any state or local law related to labor relations for private sector workers. Unlike with most federal wage, hour, and anti-discrimination laws — which establish floors above which states and localities are able to innovate in ways that provide greater worker rights and protections — the NLRA is both a floor and a ceiling. One obvious idea would be to reform the NLRA so that it remains a floor but not a ceiling. This would allow states and localities to experiment with ways that could empower new kinds of collective action without imperiling existing rights and protections.

Some have further suggested that, if it is not possible politically to do away with just the ceiling, doing away with NLRA preemption entirely should be considered. The paper investigates the question of what union coverage numbers might look like without NLRA preemption.

Replicating Freeman approach with updated data

There is only one other paper we know of that attempts to address this question quantitatively, the 2006 paper “Will Labor Fare Better Under State Labor Relations Law?” by Richard B. Freeman. In his paper, Freeman compares estimates of the potential loss of coverage in states most likely to enact state laws unfavorable to private sector collective bargaining with the potential gain in states most likely to enact laws favorable to private sector bargaining. He notes that his estimates are “back-of-the-envelope/excel computations that give orders of magnitudes only,” and we approach this exercise in a similar spirit.

Freeman’s analysis turns on the fact that while federal law governs private sector labor relations (and, except in the case of “right-to-work” (RTW) legislation, states are precluded from taking independent action), state law sets labor relations policy for state and local public sector workers. Thus every state has both private employees covered by federal law, and public employees covered by state law. Further, state laws vary enormously. He notes that this variance provides a natural experiment “in which we can contrast outcomes for the treatment group – public sector employees covered by that state’s public sector laws – against the control group of private sector employees in that state covered by the ubiquitous federal law.”

Based on a 1996 update by Kim Rueben of the NBER Valletta-Freeman state public sector labor law data set (http://www.nber.org/publaw/), Freeman places states into three broad categories of favorability of the legal environment for public sector bargaining: “Favorable” states where laws require that public employers bargain with unions; “Intermediate” states where the law requires public employers to meet and confer with unions but not to bargain with them; and “Unfavorable” states where the law either explicitly outlaws bargaining or contains no provision for bargaining. This categorization can be found in column (1) of Table A1 in the Appendix.

Freeman regresses public sector union coverage rates on dummy variables for whether the state has public sector laws that were either “Favorable” or “Unfavorable,” (the omitted dummy is “Intermediate”), controlling for private sector union coverage rates and a dummy variable for whether the state has a RTW law. Given that private sector workers in each state are covered by a common federal law, controlling for private sector union coverage and RTW is assumed to control for the state’s general attitude toward unionism. In other words, the coefficients on the “Favorability” dummies capture the effect on public sector union coverage of the state laws themselves, apart from the state’s general favorability towards unions.

Freeman used 2004 data on union coverage. We update this by using data for the 12 months ending in April 2017 (the latest data available at the time of analysis). Our main data can be found in columns (4), (5), and (6) of Table A1. Using the same state favorability rankings that Freeman used (which is based on the state of the world in 1996), we find that favorable public sector labor laws increase public sector coverage by approximately 26.1 percent, and that unfavorable public sector labor laws decrease public sector coverage by approximately 26.9 percent. The results of this regression can be found in column (2) of Table A2.

Table 1 shows the results of applying these regression results to private sector union coverage numbers, under Freeman’s (in our view heroic) assumption that, freed from NLRA preemption, all states with unfavorable public sector labor laws will enact equally unfavorable laws in the private sector, and states with favorable public sector labor laws will enact equally favorable laws in the private sector. With this assumption, we project that if states set labor relations policy for the private sector, there would be a loss of roughly 400,000 private sector union members in states likely to enact unfavorable legislation, and an increase of roughly 1.9 million in states likely to enact favorable legislation. This appears to show that unions would fare better on net at the national level in the absence of NLRA preemption.

Table 1

However, one obvious flaw in the above analysis is that Freeman’s favorability classifications are more than 20 years old at this point. During that time, state politics have shifted significantly to the right; the Republican Party controls 33 governorships and has majority representation in both chambers of most state legislatures. This results in a number of states classified in 1996 as “Favorable” labor environments that may now be deemed “Intermediate” or that had been classified as “Intermediate” and may now be deemed “Unfavorable,” based on legislation adopted in the past two decades.

We have updated the categorization, attempting to adhere as closely as possible to Freeman’s definitions of favorable, unfavorable and intermediate labor law regimes. The cohort of no-longer-favorable states is led by Wisconsin, which with the passage of Act 10 in 2011, has become one of the single bleakest legal environments for public employee unions. The complete updated categorization can be found in column (2) of Table A1. The table also includes an explanation for any changes. Critically, there are no states whose legislative history over the past decade might suggest upgrading their favorability for pro-union labor law reform.

It should be noted that while changes in public sector labor laws have the most direct application to Freeman’s model, other areas of state labor and employment law that provide relevant input for analyses later in this paper are also included in the explanation of changes. These include RTW laws; the extension or revocation of collective bargaining rights to employees outside NLRA jurisdiction; laws governing union dues and union political activity in both the public and private sectors; prevailing wage and project labor agreements in the construction industry; and laws governing the non-union labor market including minimum wage, sick leave, employment discrimination, unemployment insurance, or pension rights. All of these provide a critical measure of a given state’s center of political gravity when it comes to regulating the labor market.

Using our updated favorability rankings, Freeman’s methodology results in a finding that favorable public sector labor laws do not have a significant effect on public sector coverage, and that unfavorable public sector labor laws decrease public sector coverage by approximately 43.1 percent. The results of this regression can be found in column (4) of Table A2. Because we do not believe that there would be no positive impact on union coverage of states adopting favorable laws, we view this as a breakdown of the empirical approach in the context of updated state rankings.

For completeness, however, we apply the updated regression results to private sector union coverage numbers, again under Freeman’s assumption that, in the absence of NLRA preemption, all states with unfavorable public sector labor laws will enact equally unfavorable laws in the private sector, and states with favorable public sector labor laws will enact equally favorable laws in the private sector. Using this method, we find that if states controlled labor relations policy for private sector workers, there would be a loss of roughly 700,000 private sector union members in states likely to enact unfavorable legislation, and no change in states likely to enact favorable legislation.

Table 2

The breakdown of Freeman’s back-of-the-envelope empirical strategy in the context of updated state favorability rankings points to a need for a more robust theory regarding the contemporary politics of labor law reform in order to project the outcome of labor policy debates in current state legislatures. In particular, it can no longer be assumed that the politics governing private sector labor law reform would be identical to those that drove deliberations over public sector union rights — deliberations that in many cases took place decades ago.

There are several factors regarding the political landscape that need to be considered. First, the power of money in politics has grown ever-more extreme in recent decades – particularly in the aftermath of the Supreme Court’s 2010 Citizens United ruling. Because the political spending of employer lobbies dramatically outpaces that of unions, the political influence of anti-union forces has significantly strengthened in the years since Freeman’s original analysis. Secondly, for reasons explained below, the political influence of the corporate lobbies has been even greater at the state level than in the federal government; waiving NLRA preemption would turn labor law over to the legislative arena that is most favorable to anti-union employer lobbies. Third, corporate political activism has always been most intense around questions of private sector – rather than public – labor law. Thus we cannot assume that the politics of public sector labor policy would be reproduced for the private sector. Fourth, while public employers are fixed in place and are not in competition with other states’ services, the ability of private employers to move from one state to another, or to force unions to bargain with a multi-state employer, may undermine the bargaining power of workers even in the most politically favorable states. Fifth, as employer lobbies have been growing ever stronger, their ambition to cripple labor unions has produced much more draconian legislative initiatives than what was imaginable even a decade ago. Sixth, there are substantial barriers that limit the ability of even the most progressive state governments to adopt far-reaching pro-worker reforms (for example, even New York’s Democrats have been unable to overcome the opposition of the agriculture lobby to extend organizing rights to this industry). And finally, there are important dynamic differences between pro- and anti-union labor law reforms in that pro-union reforms in recent years typically usher in a period of significant but incremental growth, while anti-union laws often pose immediate and near-existential threats to the labor movement. Each of these factors will be described in detail below.

Employers’ political influence has increased dramatically since the 2010 Citizens United ruling

The Supreme Court’s 2010 Citizens United decision ushered in a new legislative era, shaped by the impact of unlimited corporate spending on politics. That fall’s elections were the first conducted under the new rules, and they brought dramatic change. Eleven state governments switched from Democratic or divided control to Republican control of the governorship and both houses of the legislature. Since these lawmakers took office in early 2011, the US has seen an unprecedented wave of legislation aimed at lowering labor standards and slashing public services.

At the heart of this activism are the country’s premier business lobbies – the Chamber of Commerce, the National Association of Manufacturers, and the National Federation of Independent Business – along with the AFP and industry-specific groups such as the National Grocers Association and the National Restaurant Association.

The steadily growing economic inequality of the past four decades has led, in turn, to increased political inequality. While business has always enjoyed outsized political influence, its voice has grown even more dominant in recent years. Elections for public office have become dramatically more expensive, rendering politicians all the more dependent on those with the resources to fund their campaigns.1 The increasing concentration of wealth has produced a growing class of mega-donors prepared to spend enormous sums to influence outcomes. And, finally, the progressive loosening of campaign finance regulations – culminating in the Citizens United decision – has enabled corporations and the wealthy to spend unlimited amounts on elections, and to do so secretly. The intersection of these three factors has dramatically increased the political influence of those at the top of the economy.

Political spending was heavily tilted toward corporations long before Citizens United; on average, from 2000-2010, business typically outspent labor unions by ten to one in federal elections.2 In theory, Citizens United allowed for unlimited spending by both corporations and unions. In reality, however, the resource imbalance between these groups is so extreme as to render the comparison meaningless. The annual revenue of Fortune 500 companies alone is 350 times that of the labor movement.3

The 2008 election – which put Democrats in control of both houses of Congress as well as the White House – struck fear into the hearts of energy executives worried about climate change legislation, financial titans anticipating new regulations, and insurance companies facing the demand for affordable healthcare. These fears, in turn, drove scores of new donors to join the Chamber of Commerce or the Koch network. As one Koch insider describes, “Obama’s election had sparked such vitriol on the right that [the Kochs] were almost overwhelmed by the number of wealthy donors eager to join them. Suddenly they were raising big money!”4 The first Koch donor summit, in 2003, drew only 15 participants.5 As late as June 2009, the event raised just $13 million. By 2015, however, the network attracted 450 donors and solicited nearly $900 million for the upcoming election season.6 Longtime campaign finance watchdog Fred Wertheimer commented that “we’ve had money in the past, but this is so far beyond what anyone has thought of it’s mind-boggling.”7 The Citizens United decision was critical to the growth of the Kochs’ and similar corporate advocacy groups – particularly those who wanted to shield donors’ identity.8

Corporate spending has increased dramatically since the Citizens United decision. Advocacy organizations spent just over $140 million on the presidential and congressional elections in 2008. Four years later, advocacy organizations spent $1 billion on federal elections, the great bulk of it from businesspeople and corporate organizations.9 Both the U.S. Chamber of Commerce and the Club for Growth more than doubled their spending from 2008 to 2012. In addition, a slew of new corporate-funded advocacy organizations appeared during this period. Taken together, spending by the major corporate-funded groups was more than six times higher in 2012 than in 2008.10

American Legislative Exchange Council (ALEC), the most important national organization advancing the corporate agenda
at the state level, brings together 2,000 member legislators (one-quarter of all state lawmakers, including many state senate presidents and House Speakers), and the country’s largest corporations to formulate and promote business-friendly legislation.11 According to the group’s promotional materials, it convenes bill-drafting committees – often at posh resorts – in which “both corporations and legislators have a voice and a vote in shaping policy.” Thus, state legislators with little time, staff, or expertise are able to introduce fully formed and professionally supported bills. The organization claims to introduce 800-1,000 bills each year in the 50 state legislatures, with 20 percent becoming law.12

Ultimately, the “exchange” that ALEC facilitates is between corporate donors and state legislators. The corporations pay ALEC’s expenses and contribute to legislators’ campaigns; in return, legislators carry the corporate agenda into their statehouses.13 Member corporations also fund the ALEC-affiliated State Policy Network, whose pro-corporate think tanks produce policy papers in support of model legislation.14 In the first decade of this century, ALEC’s leading corporate backers contributed more than $370 million to state elections, and over 100 laws each year based on ALEC’s model bills have been enacted.15 Through this network, corporate lobbyists have established a well-funded, highly effective operation that combines legislative drafting, electoral politics, lobbying, grassroots activism, and policy promotion.

Figure C

Employer lobbies’ political influence is greatest at the level of state legislatures

Many of the factors that strengthen corporate political influence are magnified in the states. First, far fewer people pay attention to state government, implying wider latitude for well-funded organized interests. Political scientist Martin Gilens notes that only when policy debates attract widespread public attention are politicians even modestly responsive to the bottom 90 percent of the population.16 Yet if such attention is rare at the federal level, it is rarer still in the states. Less than one-quarter of adults are able to name their state senator or representative, and less than half even know which party is in the majority.17 Thus, even the crudest form of political accountability – voting against the party in power when the economy turns bad – does not function at the state level.18

If most people can’t name their legislators, how many are likely to have a well-formed opinion on whether prevailing wages should be required on public construction projects worth more than $25,000? How many can possibly be paying attention to debates about changing the definition of employee “misconduct” – changes that affect eligibility for unemployment insurance benefits? For all practical purposes, these debates take place in a vacuum. Apart from labor unions and a handful of progressive activists, the corporate agenda on such topics encounters little public resistance at the state level, because hardly anyone knows about or understands the issues.

By contrast, an organization such as ALEC simply could not flourish at the federal level. Senators and House members cannot sit down with lobbyists three times a year to draft model bills – subject to the approval of corporate funders – that are then introduced in the Congress. There are too many interest groups, too much media attention, and too many rank-and-file voters watching what happens. ALEC’s ability to operate so effectively for so long owes much to the invisibility of state government to most citizens.

So, too, corporate lobbies’ financial advantage is magnified in the states. Citizens United marked a sea change in state as well as federal politics. As of 2010, 22 states maintained bans on independent political expenditures by corporations or labor unions; all were overturned by the Supreme Court’s decision. The first major analysis measuring the impact of the legal change on state legislatures found that the net result was to increase the odds of a Republican being elected by four percentage points, primarily as a result of increased business contributions.19

Because state legislative races are so much cheaper than federal elections, those contributions go much further. Consider North Carolina, where a network of organizations overseen by supermarket executive and corporate activist Art Pope spent $2.3 million on 27 legislative races in 2010.20 In 2008, the average North Carolina Senate candidate spent a little under $140,000, and the average House candidate spent approximately $60,000.21 Two years later, Pope’s network contributed an average of $134,000 to each of the targeted Senate races, and $59,000 to each House race – effectively doubling the previous campaign budgets.22 Republicans won nearly 80 percent of the seats Pope targeted in 2010, and the GOP gained complete control of the state legislature for the first time since Reconstruction.

The corporate lobbies’ resource advantage is manifested not only in elections, but also in growing influence over the legislative process once lawmakers are seated. Over the past four decades, state legislative districts have been repeatedly redrawn in order to maximize the number of safe seats for each party.23 As a result, incumbent Republicans have no incentive to govern as centrists, but strong reason to fear a well-funded primary opponent. Such a system, in turn, maximizes the power of the corporate purse. In competitive districts, the threat to primary an incumbent Republican from the right would be self-defeating – even if effective it would likely result in failure in November. Under current conditions, however, corporate lobbies can wield the threat of a primary challenger to enforce discipline on sitting legislators. In 2013-15, for example, Americans for Prosperity carried out an aggressive, well-funded campaign to attack Republican governors and legislators who sought to take advantage of funds available under the Affordable Care Act to expand Medicaid for their low-income citizens. In Florida, Kansas, Tennessee, and Utah, AFP put scores of staff to work, spent millions of dollars on advertising, conducted intensive phone-bank and canvassing operations in key legislators’ districts, and succeeded in reversing plans that initially enjoyed broad Republican support. In Kansas, a national AFP representative publicly warned senators that “we certainly plan to hold accountable every legislator who supports this misguided scheme.” The bill’s sponsor was shocked, noting that he had never before “heard a [witness] threaten members in testimony.”24

Such threats – whether implied or explicit – reflect the effective nationalization of state politics in the years since Citizens United. The overwhelming majority of anti-worker legislative initiatives noted in this article were crafted with the aid of corporate lobbyists and promoted through ALEC in statehouses across the country. The bills supported by ALEC, the Chamber of Commerce, AFP, NAM, and the NFIB are strikingly similar across a diverse range of states. Goals that are impossible to achieve in Congress are often feasible in the states; thus, the corporate lobbies have set out to reshape the national economy and labor market by moving a common agenda in 50 venues at once, based on the assessment that their political influence is more powerful at the state than the federal level of government.

Employer political mobilization is most concerned with the private sector

As mentioned above, we are highly skeptical of the notion that a state’s laws governing public sector labor relations is a good proxy for what that state might do in the private sector in the absence of NLRA preemption. Our skepticism is underscored by the fact that some states that have a “Favorable” orientation toward public sector labor law also have RTW laws governing the private sector labor market. How could a state be progressive toward its public employees but hostile toward private sector unions? The answer to this discrepancy lies in understanding the interests of the corporate lobbies. The money behind ALEC, the Chamber of Commerce and the other dominant corporate lobbies comes from private sector corporations. By definition, they have a much stronger interest in preventing private sector organizing than in restricting the influence of public employees. The primacy of private sector labor regulations has long been evident at the federal level: while employer lobbies are unified in seeking cutbacks to public employees, their peak moments of political mobilization over the past fifty years have all revolved around private sector reforms: proposed NLRA reform under president Carter, Clinton’s striker replacement bill, and the Employee Free Choice Act. Thus, in imagining what state legislation might emerge in response to lifting NLRA preemption, we must anticipate much more intensive employer political mobilization than what has been seen in debates over public sector regulation – including even the most contentious debates of the past decade.

The impact of employer mobility and multi-state employers

In the absence of NLRA preemption, unions in “favorable” states that operate in a national or regional (multi-state) market would be adversely affected by declining unionization in “unfavorable” states. In a national market, unions in favorable states will have a foothold in only part of the market and consumers can readily decide to avoid the somewhat higher-priced union-made goods or services. The employers can readily relocate production to avoid the union presence. In many ways, this is simply a heightened result paralleling what has happened to Northern and Midwestern-based unions as a consequence of the South remaining, for the most part, relatively unorganized. The UAW, for instance, was able to blunt the move of the Big Three into southern production but has been unable to organize the foreign transplants that have been set up almost exclusively in the South. Increasingly uneven union density across states can also undercut the bargaining power in regional, multi-state firms such as utilities or telecommunications.

Unions in “favorable” states that operate in conventionally local product markets may also find that their bargaining power is challenged. As an illustration, consider supermarkets, and in particular the fact that there are national supermarket chains. A locally-based movement with strengthened bargaining rights (scope of bargaining, enhanced ability to organize) will have limited leverage against a chain. For example, a chain can take on a local strike by relying on its profits in other locations. Even if a local union has a strike fund provided by a national union this can pose a great challenge, particularly considering that unions in many locations of that chain will be substantially weaker as a result of the change in preemption. Further, competitors – considers Wal-Mart, Publix and Wegman’s — can enter into better organized locations and undercut standards. These firms have scale to obtain low prices for the products they sell. They have substantial logistical capabilities and financial strength. Their entry will erode whatever the union standard is that is established in high-density areas, unless one believes that the improvements in labor law will facilitate the ready organization of these firms when they enter new regions. The prime example of this dynamic was the lowering of union standards in Southern California based on the fear that Wal-Mart would enter.

Potential for increasingly aggressive anti-worker legislation

One of the difficulties of projecting future labor law reform on the basis of existing statutes is that politics is not static – it is dynamic and ever-evolving. In our time and in unfavorable states, employer lobbies have been growing ever stronger, and their ambition to cripple labor unions has produced much more draconian legislative initiatives than what was imaginable even a decade ago.

For instance, so-called “paycheck protection” legislation has evolved in ever-more ambitious forms over the past decade. Both Arizona and Missouri move beyond traditional “paycheck protection” legislation (i.e., requiring that unions obtain prior approval from employees to spend dues money on behalf of political parties, political candidates, or other political advocacy) to create added reporting requirements that pose significant new financial threats to unions engaged in political activity. Under current state and federal law, unions must file annual reports showing what percentage of their expenses was devoted to political activities. Each year’s report determines the percentage rebate that must be awarded in the following year to those who opt out of supporting their union’s political agenda. By contrast, the new Arizona and Missouri laws require unions to determine in advance, and announce at the start of each year, how much they will spend on political activities in the coming year. This declaration in turn determines what share of dues is waived for non-supporters. It is easy to imagine how political events unforeseen at the start of a year might end up requiring greater than anticipated resources, leaving unions facing untenable financial choices. The law thus puts unions in a no-win bind. If a union underestimates its political needs, it risks being unable to respond to potentially existential legislative threats. Yet to the extent a union cautiously overestimates its anticipated political budget, it forgoes that much greater a share of dues revenue.

Together with such high risk reporting requirements, labor opponents have also sought to broaden the definition of “political” activities subject to funding restrictions. Where the first generation of such bills aimed at a traditional understanding of union political activity – financial, advertising and field support for candidate or initiative campaigns – recent laws passed in Alabama, Arizona, Kansas and Missouri all aim at a broader scope of activity. Alabama’s Act 761, for example, defines “political activity” to include “public opinion polling,” “any form of political communication,” “any type of political advertising,” “phone calling for any political purpose,” or “distributing political literature of any type.”25 For Alabama union members to make voluntary dues contributions through the payroll system, their unions would have to disavow any of these activities, conducted with any part of the organization’s budget. Kansas’ law similarly makes it illegal – even for employees who voluntarily choose to pay dues – to use electronic payroll deductions to process dues payments earmarked for “political purposes,” broadly defined to include any “act done … in a way to influence or tend to influence, directly, or indirectly, any person … to vote for or against any candidate for public office.”26 The state Chamber of Commerce testified in support of the bill, with its director of legislative affairs explaining to lawmakers that “I need this bill passed so we can get rid of public sector unions.”27 Paul Kersey, longtime “labor expert” for the Heritage Foundation, ALEC, and multiple states’ corporate-funded think tanks, declared the 2013 Kansas law a “good step” but rued that “it could have been stronger.”28

Finally, Arizona’s 2011 law — designated a “priority bill” by the state Chamber of Commerce — requires that both public and private sector unions get annual written authorization from each member before spending any amount of dues money for “political purposes,” defined as “supporting or opposing any candidate … political party, referendum, initiative, [or] political issue advocacy.” Thus at the same time that corporations are being freed to engage in ever more expensive political activity, unions are being restricted from engaging in activities that were legal even before Citizens United.

So too, the type of restrictions on bargaining created by Wisconsin’s Act 10 were unprecedented when they were adopted in 2011. But elements of this legislation have now been adopted as corporate goals for private sector labor law reform. This evolution is most immediately evident at the federal level. During debate over the Employee Free Choice Act, the Republican and corporate-backed alternative was Rep. Charlie Norwood’s Secret Ballot Protection Act, which mandated that unions could only be certified through an NLRB election.29 In 2017, the primary labor law reform proposal supported by House Republicans is dramatically more ambitious. The Employee Rights Act, introduced in May 2017 with 81 co-sponsors, starts at the same place as the 2010 Norwood bill – requiring NLRB elections and making card-check recognition illegal – but adds a series of new items, including:30

• Union certification requires a “yes” vote by a majority of all employees in the bargaining unit. In other words, non-voters automatically become “no” voters.
• Requires that elections be held only after all legal questions and challenges have been heard and settled.
• Union must be re-certified by majority of the bargaining unit whenever there’s been more than 50 percent turnover in the workforce since the previous certification election. These elections cannot be delayed because of employer ULP’s.
• Employees may notify employer that they don’t want their name or contact info provided to union, and will then be excluded from any Excelsior list.
• Provides liquidated damages for union-side ULP’s. Unions found to have coerced employees filing decertification petitions are barred from filing objections in the decertification election.
• Non-members must be allowed to vote on contract ratification, strikes or other job actions.
• No dues money can be spent on non-representation activities (whether political, new organizing, charity, or anything else) without the annual written authorization of each member whose dues are being used.
• No strikes are legal except if approved by majority of all bargaining unit employees through secret ballot vote overseen by a private independent party chosen by agreement of employer and union and paid for entirely by the union.
• Removes NLRA protection from comprehensive campaigns and subjects unions to charges of “extortion.” Proponents explain that “This would effectively criminalize many of the more aggressive union tactics that organizers use to unethically pressure employees into union membership against their will.”31

Official corporate supporters of this bill include a virtual Who’s Who of corporate lobbies: Americans for Prosperity, Americans for Tax Reform, Heritage Action, Club for Growth, FreedomWorks, National Taxpayers Union, American Conservative Union, ALEC, Freedom Partners Chamber of Commerce, Mackinac Center, Center for Worker Freedom, North Carolina Retail Merchants Association, and Wisconsin Manufacturers and Commerce. These same organizations have been active in state legislative fights across the country. We must expect, then, that this represents the agenda that these organizations would seek to implement in states where their legislative allies dominate the lawmaking process.

Finally, lifting NLRA preemption would create new and unique vulnerabilities for the construction industry. Construction is a key industry that is large, growing, cannot be offshored, and where skilled workers without college degrees can make a decent living. In 2016, there were 323,000 construction workers who were union members in RTW work states. Construction unions remain viable in RTW states due to the unique nature of their industry. While RTW mandates that non-members receive the same contractual benefits as dues-paying members, apprenticeship training programs and union hiring halls are not affected by this requirement. Union apprenticeship program typically provides the highest quality training and the most reliable path to becoming a journey-level worker. And when union signatory contractors need to hire workers for an upcoming project, they typically turn to the union’s hiring hall. Because both apprenticeship programs and hiring halls are reserved for union members, workers have strong reason to value the union and to remain dues-paying members. For this reason, construction unions have been uniquely positioned to maintain membership even in states that adopt RTW laws. In the absence of NLRA preemption, however, hostile state legislators would be free to dismantle these structures, thus undermining the viability of one of the few union sectors that remains relatively protected in RTW states.

The upside potential may be limited

The state categorizations used in the Freeman-based regression analysis reflect the favorability of the legal environment for public sector bargaining, since Freeman’s empirical strategy rested on the assumption that states with unfavorable (favorable) public sector labor laws will enact equally unfavorable (favorable) laws in the private sector. We develop a third categorization (found in column (3) of Table A1), that includes information on changes in other areas of state labor and employment law that provide relevant information regarding the likelihood of favorable or unfavorable state-level private sector labor law reform in the absence of federal preemption.

We believe there are substantial barriers that limit the ability of even the most progressive state governments to adopt far-reaching pro-worker reforms. To assess the potential for progressive reform, we add a fourth category to Freeman’s analysis, focusing on states that not only have “favorable” public sector labor regimes, but are also “actionable,” i.e. states that unions could look to in hopes of progressive reforms in the absence of NLRA preemption. We determined the “Favorable and Actionable” set of states by first selecting those states where Democrats have controlled all three branches of government for at least the past four years (unfortunately, there are only six: California, Delaware, Hawaii, New York, Oregon, and Rhode Island) and adding to that list both Washington state and Massachusetts. We believe our list of “Favorable and Actionable” states is not too restrictive, particularly given that even these states have been limited in their ability to regulate the labor market in ways unions would like. For instance, in only two of these states – California and Hawaii – do farmworkers have collective bargaining rights equal to those of other private sector workers. Even New York’s Democrats, for instance, have been unable to overcome the opposition of the agriculture lobby to extend organizing rights to this industry.32 While most of these states have raised their minimum wage substantially above the federal minimum, tipped workers still make less than the normal minimum wage in Delaware, Hawaii, Massachusetts, New York, and Rhode Island – these most favorable of states remain unable to do away with the “tip credit” laws beloved by the restaurant industry. Of these eight states, only California, Massachusetts, Oregon, and Washington have established a right to paid sick leave, and only California and New York have established an effective mechanism for recovering stolen wages (though Massachusetts is close to passing one). In all these ways, then, we note that even in what appear to be the most favorable states for pro-union labor law reform, the actual potential for legislators to enact reforms is quite limited in the face of what we assume will be unprecedented opposition by private sector employers (more on this below).

Table 3 provides the breakdown of private sector union workers by this new state categorization. We find that 38 percent of private sector union workers are in states that are “Favorable and Actionable,” i.e. states where there would be a likelihood of progressive reform, whereas nearly half (49 percent) are in “Unfavorable” states, where the likelihood of anti-union reform is high.

Table 3

Table A3 in the appendix provides a further breakdown of Table 3 by race/ethnicity and gender. Figure B below shows the share of workers in “Favorable and Actionable” states and in “Unfavorable” states, by race/ethnicity and gender. Black non-Hispanic workers are less likely to be concentrated in “Favorable and Actionable” states and more likely to be concentrated in “Unfavorable” states than private sector union workers overall. This means the upsides of getting rid of NLRA preemption would be smaller, and the downsides would be greater, for black private sector union workers. That is also true of white non-Hispanic workers, who are even less concentrated in “Favorable and Actionable” states and more concentrated in “Unfavorable” states than black workers. The opposite is true of Hispanic and Asian/other workers, who are substantially more concentrated in “Favorable and Actionable” states and less concentrated in “Unfavorable” states.

By gender, we find that female private sector union workers are somewhat more concentrated in “Favorable and Actionable” and somewhat less concentrated in “Unfavorable” states than their male counterparts.

Figure B

Incremental Growth and Existential Threats

Finally, it is critical to account for the dynamic differences between pro- and anti-union labor law reforms. Essentially, pro-union reforms in recent years typically usher in a period of significant but incremental growth, while anti-union laws often pose immediate and near-existential threats to the labor movement. Thus, in states that have provided for card-check recognition, or in NLRB elections conducted under the most recent Obama-era administrative reforms, workers’ ability to organize was improved to a modest degree. Over a sustained number of years, such changes would effect very significant improvements – but progress is incremental. By comparison, in the six years since Wisconsin’s Act 10, one of that state’s largest public employee unions saw its membership fall from 70,000 to 10,000, forcing a cutback from 130 staff to just 15.33 Such institutional setbacks not only hinder the union’s ability to service its current members, but also its ability to grow. Even if Wisconsin’s labor law were restored to its 2010 status, it would likely take many years before unions could organize their way back to previous membership levels. Thus, if anti-union officials control a state government even for only a few legislative cycles, they may undermine the labor movement’s organizational capacity in ways that take decades to repair. Such dangers must be accounted for in our weighing the costs and benefits of waiving NLRA preemption.

Conclusion

This analysis points to the need for a huge amount of realism in thinking about opportunities and risks in waiving NLRA preemption. We estimate that around 4.1 million private sector union workers live in states where the political environment is ripe for the adoption of anti-union legislation (“Unfavorable” states), whereas 3.2 million private sector union workers live in states where the environment is ripe for the adoption of progress labor reforms (“Favorable and Actionable” states). The following examples provide a broad sense of how these groups might be affected. In the absence of NLRA preemption, if the anti-labor reforms in “Unfavorable” states were to reduce union coverage in those states by 50 percent (in our view a highly conservative assumption based on the above discussion), that would imply that the pro-union reforms in “Favorable and Actionable” states would have to increase coverage in those states by roughly two-thirds in order to have union coverage at the national level break even. If the anti-labor reforms in “Unfavorable” states were to reduce union coverage in those states by 75 percent, that would imply that the pro-union reforms in “Favorable and Actionable” states would have to nearly double coverage in those states in order to have union coverage at the national level break even. Finally, if private sector unions in “Unfavorable” states were completely eviscerated, pro-union reforms in “Favorable and Actionable” states would have to increase coverage in those states by roughly 130 percent in order to break even. The below table shows what these “required” increases would mean in terms of private sector union coverage in “Favorable and Actionable” states.

Table 4

In light of these numbers, a key question becomes whether or not progressive state governments would be able to adopt pro-union reforms that are far-reaching enough to achieve increases at that scale. Above, we point out some cause for concern in that even very progressive states have been limited in their ability to regulate the labor market in pro-worker ways (for example, even New York’s Democrats have been unable to overcome the opposition of the agriculture lobby to extend organizing rights to this industry).

Another question that is beyond the scope of this paper is what the economic, political, and human implications would be of a dramatic increase in the existing polarization of the U.S. into a few states where workers have strong rights and protections and a wide swath of the country where they do not.

Authors

Heidi Shierholz
Economic Policy Institute
hshierholz@epi.org

Gordon Lafer
University of Oregon Labor Education & Research Center
glafer@uoregon.edu

Appendix

Table A1

Table A2

Table A3

Endnotes

1. Modern campaign finance rules were first adopted in 1974. Since then, the real cost of Congressional campaigns has increased four-fold, fueled by the rise first of television and later of consultants, polling, digital media and ever-sophisticated data analysis. The cost of presidential election campaigns doubled between 2000 and 2008, and then doubled again in 2012. Center on Responsive Politics, https://www.opensecrets.org/pres12. The dominant role of money is further visible in how candidates spend their time: during the course of his reelection campaign, president Obama’s schedule included twice as many fundraising as public speaking events. Nicholas Confessore, “Result Won’t Limit Campaign Money Any More Than Ruling Did,” New York Times, November 11, 2012, www.nytimes.com/2012/11/12/us/politics/a-vote-for-unilmited-campaign-financing.html?_r=o. Once elected, Congresspeople commonly report that they spend one-third of their hours fundraising for their next campaign.

2. Federal elections from 2000-2014. (Center for Responsive Politics 2015)

3. Liz Kennedy and Sean McElwee, Do Corporations & Unions Face the Same Rules on Political Spending? Demos, 2014, http://www.demos.org/sites/default/files/publications/CorpExplainer.pdf.

4. Quoted in Jane Mayer, Dark Money: The Hidden History of the Billionaires Behind the Rise of the Radical Right, Doubleday, New York, 2016.

5. Mayer 2016, p. 7.

6. Kenneth Vogel, “A Koch love-fest in California,” Politico, August 3, 2015. http://www.politico.com/story/2015/08/koch-love-fest-in-california-120928; Nicholas Confessore, “Koch Brothers’ Budget of $889 Million is On Par With Both Parties’ Spending,” New York Times, January 26, 2015. http://www.nytimes.com/2015/01/27/us/politics/kochs-plan-to-spend-900-million-on-2016-campaign.html.

7. Quoted in Mayer 2016, p. 378.

8. In 2006, just 2 percent of outside political spending came from undisclosed donors; by 2010 this figure jumped to 40 percent, representing hundreds of millions of dollars in secret spending. Mayer, Dark Money, 248.

9. “Advocacy” organizations here denotes independent expenditures by groups other than candidate campaigns or political parties.

10. Data is from FEC, compiled by Center on Responsive Politics, https://www.opensecrets.org/outsidespending/index.php?type=Y.

11. Over 100 corporations and nearly 20 nonprofit groups have resigned their membership in ALEC since 2011. Following the murder of Trayvon Martin—broadly perceived as, in part, the result of ALEC-promoted “Stand Your Ground” laws—public outcry led to a rash of corporate resignations, including both Wal-Mart and Coca-Cola (see Center for Media and Democracy, “Corporations That Have Cut Ties to ALEC,” http://www.sourcewatch.org/index.php/Corporations_that_Have_Cut_Ties_to_ALEC). These corporations are noted as ALEC supporters in this book for several reasons. First, they were active ALEC supporters during the period that most of the bills discussed in this report were formulated and initially promoted. Second, although these companies distanced themselves from ALEC due to the controversy surrounding the Martin killing or ALEC’s position on immigration or climate change, they in no way distanced themselves from ALEC’s economic or labor agenda, and it is possible that these companies will either renew ties with ALEC in the future, or find parallel channels through which to promote the same policy goals. In some cases, these interests may already be supporting ALEC’s activities through other channels. For instance, while Wal-Mart resigned from ALEC, the Walton Family Foundation remains an active member. So too, many of the companies that resigned ALEC membership are members of the U.S. Chamber of Commerce, which in turn is an active supporter of ALEC. It is possible that some corporations may shield themselves from public criticism by resigning direct ALEC membership, but continue to support the organization’s activities with funds channeled through the Chamber of Commerce or the many other trade associations that remain active ALEC members.

12. American Legislative Exchange Council, “ALEC 101,” . Undated, http://alecexposed.org/w/images/5/5b/ALEC_101_Exposed_1.pdf.

13. For a 2011 database of ALEC affiliated corporations that have donated to the campaigns of ALEC-affiliated legislators, see ProPublica, ALEC-Related Contributions, August 2011. http://projects.propublica.org/alec-contributions.

14. Fang, The Machine, 202.

15. Common Cause, Legislating Under the Influence: Money, Power and the American Legislative Exchange Council, June 24, 2011, http://cldc.org/wp-content/uploads/2011/12/MONEYPOWERANDALEC.pdf. The discrepancy between this number and ALEC’s claim of 20 percent success may reflect the difficulty that outsiders face in tracing ALEC’s full impact, as ALEC-affiliated lawmakers may put forth bills that accomplish the organization’s aims without mirroring its exact model language.

16. Martin Gilens, Affluence and Influence: Economic Inequality and Political Power in America (Princeton: Princeton University Press, 2010), 173.

17. Center for the Study of Democratic Institutions, Vanderbilt Poll, May 2012, http://www.vanderbilt.edu/csdi/tl2012.pdf; American National Election Studies, Evaluations of Government and Society Study, 2010, http://www.electionstudies.org/studypages/2010_2012EGSS/2010_2012EGSScriteria.htm.

18. Steven Rogers, “Accountability in State Legislatures: How Parties Perform in Office and State Legislative Elections,” Center for the Study of Democratic Institutions, Vanderbilt University, October 17, 2013, http://www.stevenmrogers.com/Dissertation/ChapterDrafts/CollectiveAccountability/Rogers-CollectiveAccountability.pdf.

19. Tilman Klumpp, Hugo Mialon and Michael Williams, “The Business of American Democracy: Citizens United, Spending, and Elections,” July. 2014. The authors culled data from over 38,000 state legislative races over seven election cycles – five preceding Citizens United and two following the decision. The analysis compared the impact of the Supreme Court decision in states that had previously allowed corporate independent expenditures with those that had banned them before 2010. This differences-in-differences analysis provides the first rigorous statistical measure of the law’s impact on state legislative elections. The data show that independent expenditures increased the likelihood of Republican incumbents seeking reelection, decreased the number of Democrats choosing to stand as candidates, and increased the odds of the Republican candidate winning.

20. Chris Kromm, “How much did Art Pope’s network really spend on North Carolina’s 2010 legislative election?” October 7, 2011, http://www.artpopeexposed.com/explainer_how_much_did_art_pope_really_spend.

For a detailed account of Pope’s political activities, see Mayer 2011.

21. Author’s calculations based on Kromm 2011 and National Institute on Money in State Politics, overview of North Carolina 2008 election. https://www.followthemoney.org/election-overview?s=NC&y=2008.

22. Author’s calculations based on National Institute on Money in State Politics, overview of North Carolina 2008 election. https://www.followthemoney.org/election-overview?s=NC&y=2008.

23. Competitiveness in state legislative races – measured by the share of all races decided by 5 percent or less – has been declining since the 1970s. 2014 saw the fewest closely contested races of any state election on record. Reid Wilson, “Study: elections becoming less competitive,” Washington Post, May 7, 2015. https://www.washingtonpost.com/blogs/govbeat/wp/2015/05/07/study-state-elections-becoming-less-competitive.

24. Peter Hancock, “Pro-business groups wielding big influence in Legislature, Democrats say,” Lawrence Journal-World, April 5, 2015, http://www2.ljworld.com/news/2015/apr/05/pro-business-lobby-groups-wielding-big-influence-l.

25. Alabama Secretary of State, Act 2010-761, December 20, 2010, http://www.alabamaschoolboards.org/PDFs/Act percent202010-761.pdf.

26. Kansas State Legislature, House Bill No. 2023, 2013a, http://kslegislature.org/li_2014/b2013_14/measures/documents/hb2023_00_0000.pdf.

27. John Celock, “Kansas Chamber of Commerce Lobbyist: Bill Is Needed to End Public Sector Unions,” Huffington Post, January 23, 2013, www.huffingtonpost.com/2013/01/23/kansas-chamber-of-commerce_n_2536360.html.

28. In 2013, Paul Kersey was Labor Policy Director for the Illinois Policy Institute, a member of the ALEC-affiliated State Policy Network. Kersey previously served in a similar role for the Mackinac Center in Michigan, and before that at the Heritage Foundation. Kersey presented several pieces of model anti-union legislation at ALEC’s Spring 2012 Task Force Summit.

29. www.congress.gov/bill/110th-congress/house-bill/866.

30. HR 2723, The Employee Rights Act, May 25, 2017. https://www.congress.gov/bill/115th-congress/house-bill/2723.

31. http://employeerightsact.com.

32. www.usnews.com/topics/locations/new_york.

33. Personal communication from Wisconsin union officials.

This paper presents the methodology and data sources used in the 2015 update of the Economic Policy Institute’s Family Budget Calculator. The budget calculator draws upon the most recent available data, which in many instances is data for 2014. In cases where 2014 data are not available, data from the latest available year are inflated to 2014 dollars. As such, the budgets should be considered as applying to 2014, even though they were published in 2015.

Use the EPI Family Budget Calculator

Read the overview

Definition of families

The size of a family dramatically affects the budget needed to maintain a secure yet modest standard of living. We have constructed budgets for 10 different types of families in each area. These families include a single person with no children; a married couple with no children; single-parent families with one, two, three, or four children; and a married couple with one, two, three, or four children.

Our definition of a single person with no children assumes that he or she is employed and is the head of household for federal income tax purposes. Our definition of a married couple with no children assumes both are employed, live together, and jointly file federal income taxes. Our definition of single-parent families assumes that the head of household is employed, lives with his or her children, and files as the head of household for federal income tax purposes. Our definition of two-parent families assumes that both partners are employed, live together with their children, and jointly file federal income taxes.

Families with one child are assumed to have a 4-year-old. Families with two children are assumed to have a 4-year-old and an 8-year-old. Families with three children are assumed to have a 4-year-old, an 8-year-old, and a 12-year-old. Families with four children are assumed to have a 4-year-old, an 8-year-old, a 12-year-old, and a 16-year-old.

Definition of areas

The 2015 EPI Family Budget Calculator presents data for 618 areas. Of these, 48 are statewide averages of rural areas; Rhode Island, New Jersey, and the District of Columbia do not have rural areas. Among the remaining 570 family budget areas, 485 are non-overlapping metropolitan statistical areas. Thirty-nine metropolitan areas cross state lines and generate 85 family budget areas to account for state-based variance within metropolitan areas.

A metropolitan statistical area (MSA) is defined by the Office of Management and Budget (2009) as having at least one urbanized area of 50,000 or more people, plus adjacent territory that has a high degree of social and economic integration with the core, as measured by commuting ties. Some of our data (those pertaining to housing) require us to use fair market rent (FMR) areas. FMR areas are published by the U.S. Department of Housing and Urban Development (HUD 2014a). They are divided into metropolitan FMR areas and nonmetropolitan FMR areas.

Since most metropolitan FMR areas correspond to an MSA, where possible we replaced metropolitan FMR areas with the corresponding MSAs in the list of family budget areas. Those without corresponding MSAs remain in the list of family budget areas as metropolitan FMR areas. The remaining nonmetropolitan FMR areas were labeled as non-MSA. The rural areas were also labeled as non-MSA. When regional breakdowns were used for budget calculations, they were based on the Census Bureau regions, per data availability (U.S. Census Bureau 2013).

Several of our components (child care and out-of-pocket medical costs) depend on the MSA categorization (i.e., whether or not an area is designated as an MSA) or on the population size of the MSA (transportation).

To make the 2015 EPI family budgets more user-friendly, we organized the budgets by United States Postal Service (USPS) Zone Improvement Plan (ZIP) codes. Using the U.S. Department of Housing and Urban Development’s USPS ZIP Code Crosswalk Files (HUD 2015) and data from the Missouri Data Center (2012), we crossed our 618 unique MSA-based family budgets into their respective USPS ZIP codes.

Components of the 2015 EPI family budgets

The 2015 EPI family budgets consist of seven individual components: rent, food, transportation, child care, health care, taxes, and other items of necessity. The following sections describe the methodology used to construct a monthly cost for each of these seven components across the 618 areas.

Rent

Data for rental costs come from the U.S. Department of Housing and Urban Development (2014a). HUD estimates FMRs in order to establish cost information for the federal government’s Section 8 housing assistance programs (HUD 2014b). FMRs are used to ensure a sufficient supply of housing for these programs. HUD calculates FMRs using five-year data from the American Community Survey (ACS) and relies on the Office of Management and Budget for definitions of metropolitan areas. All counties that are not classified as metropolitan areas are classified as rural. To establish a family budget cost for housing in a rural area, rental costs for rural counties are averaged into one price to be applied as “rural rent” for the entire state. Data extracts of these cost estimates are made publicly available, and EPI made use of these data to construct our family budget measure.

Fair market rent estimates are provided at the 40th percentile of rent cost—the dollar amount below which 40 percent of standard quality rental units are rented. HUD also provides the 50th percentile of rental cost for each MSA. In fiscal 2014, 505 of the 524 MSAs had data for the 40th percentile, and 19 metro areas had data available only at the 50th percentile. For these 19 areas, the 40th percentile rental cost is derived by applying a ratio of the average 40th and 50th percentile rental costs for the other MSAs in the state.

HUD makes rental rates available for studio apartments and one-bedroom through four-bedroom apartments. The EPI family budgets assume that a one-person household uses a studio and a two-person household uses a one-bedroom apartment. Families with one or two children use the two-bedroom rate. Families with three or four children use the rate for a three-bedroom unit. Rental costs include shelter plus all tenant-paid utilities, excluding telephone service, cable or satellite service, and Internet service.

Food

Data for food costs are taken from the Center for Nutrition Policy and Promotion (CNPP) publication Official USDA Food Plans: Cost of Food at Home at Four Levels (USDA 2014). Presented there are the official USDA costs for four types of food plans that serve as national standards for nutritious diets: the “Thrifty Plan,” “Low-Cost Plan,” “Moderate-Cost Plan,” and “Liberal Food Plan.” We use the USDA “Low-Cost Plan,” which assumes that almost all food is bought at a grocer and then prepared at home. We use June 2014 data, which represents the annual average monthly cost (Carlson, Lino, and Fungwe 2007). The data are only available at the national level, and are thus the same for all family budget areas (except Alaska and Hawaii, as discussed below).

Family food costs are constructed from data for the following age categories: child age 4–5; child age 6–8; and averages are created from data for males and females age 12–13, age 14–18, and age 19–50.

• For single-parent households, we use an average of the male age 19–50 data and the female age 19–50 data to represent the adult in the household. For married-parent households, we assume one male age 19–50 and one female age 19–50 are the two adults in the household. All costs in the USDA food plans table are for individuals in four-person families; for individuals in families of other sizes, USDA suggests making the following adjustments to account for differences in returns to scale:
  • One-person family: add 20 percent
  • Two-person family: add 10 percent
  • Three-person family: add 5 percent
  • Five-person family: subtract 5 percent
  • Six-person family: subtract 5 percent
• To calculate overall household food costs, we first adjust food costs for each person in the household and then sum the adjusted food costs.
• Example: For a one-parent, two-child household:

Food cost = [(average (female (age 19–50), male (age 19–50)))*1.05] + [child (age 4–5)*1.05] + [child (age 6–8)*1.05]

Note that for Alaska and Hawaii, separate food cost data are available in half-year increments. We use data for the second half of 2013 to compute household food costs for the four Alaska areas and the two Hawaii areas because it most closely represents the annual national data used for the other states. Note that only the “Thrifty Plan” is available for these states; there is no “Low-Cost Plan.” In addition, Alaska and Hawaii only have data available for children ages 6–8 and 9–11, so we imputed values for 4-year-old children, 12-year-old children, and 16-year-old children by applying the ratio of costs for relative age groups using the national data for the age groups available for Alaska and Hawaii.

The USDA food plans represent a nutritious diet at four different cost levels. The nutritional foundation of the plans consists of the 1997–2005 Dietary Reference Intakes, 2005 Dietary Guidelines for Americans, and 2005 MyPyramid food intake recommendations. In addition to cost, plans vary according to specific foods and quantities of foods. Another assumption underlying the food plans is that all meals and snacks are prepared at home. All four food plans are based on 2001–2002 data and updated to current dollars by using the Consumer Price Index for specific food items.

Child care

We utilize the Child Care Aware of America (2014) publication Parents and the High Cost of Child Care, which relies on data from the January 2013 State Child Care Resource and Referral Network survey. For the purposes of this study, we use Appendix Table 1, “2013 Average Annual Cost of Full-Time Care by State.” Several states in the survey report data on a delay, including Alabama, California, Delaware, Idaho, Iowa, Kentucky, Louisiana, Missouri, Pennsylvania, South Carolina, South Dakota, Texas, Vermont, and Wyoming, which report 2012 data, and Nevada, New Hampshire, and New York, which report 2011 data. If an MSA is in multiple states, the dominant state containing the primary city, as defined by the Office of Management and Budget, is used.

For available years, we inflate all data to reflect real 2014 dollars using the Consumer Price Index of child care and nursery school for all urban consumers from the Bureau of Labor Statistics (BLS 2014b).

We calculate our child care costs for our family types based on the following assumptions:

• One child = cost of 4-year-old care
• Two children = cost of 4-year-old care + cost of one school-aged child
• Three children = cost of 4-year-old care + cost of one school-aged child + one-sixth the cost of one school-aged child
• Four children = cost of 4-year-old care + cost of one school-aged child + one-sixth the cost of one school-aged child

The following subsections explain these assumptions in greater detail.

Center care

We use cost estimates for center-based child care in the 570 MSAs. We chose center-based care because it is more regulated than family care, and because the costs of center care do not fluctuate as much as the costs of family care.

Family care

We use cost estimates for family-based care for the remaining 48 rural areas, operating under the assumption that they are simply more accessible to those located in rural areas.

Infant care

The family budgets do not include infant care in their child care costs because we do not have an infant as part of any family type. It should be noted, however, that infant center care is significantly more expensive than 4-year-old center care, so the child care component for some families may be underestimated.

Four-year-old care

Four-year-old care is full-time care. To approximate MSA and non-MSA care costs, we use center and family-based care estimates for all 4-year-olds, taken from Appendix 1 in CCAA (2014).

School-age child care

The survey for school-age care specifically represents the average annual cost of before- and after-school care, and therefore it does not include full-time, weekend, or full-day summer care. Because of the need for 8-year-olds to be in care during the summer, the cost of school-aged child care is somewhat underestimated.

We estimate that 12-year-olds need full-day care during the summer months only; thus, one-sixth the cost of care for one school-aged child is added to families with three and four children. For families with four children, we assume child care is not necessary for the fourth child, who is assumed to be 16 years old.

State-level estimates for school-age child care are not available for Minnesota and North Dakota. Regional averages, based on the Census Bureau regions and divisions, are taken for these states. Minnesota and North Dakota fall into the West North Central Division; for these states we thus use regional averages constructed from the states in this division (Iowa, Kansas, Minnesota, Missouri, Nebraska, North Dakota, and South Dakota).

Transportation

Data on costs of transportation are derived from the Federal Highway Administration’s 2009 National Highway Transportation Survey (FHA 2009) and the Internal Revenue Service Announcement 2013-95 (IRS 2013). We use annualized vehicle miles traveled (VMT) for calculating both the total annual miles driven and to determine the trip purpose. While it is possible to use other metrics, such as person miles traveled, we judge that in many MSAs, the use of a vehicle may be necessary to get to and from major destinations, such as work, medical appointments, a grocery store, etc. In areas in which public transportation is accessible for traveling to and from major destinations, this cost may be overstated.

Our equations for calculating total transportation costs are as follows:

One-adult family transportation costs =

[(% work & non-social trips)/100*(average miles per month by MSA size)*(cost/mile)]

Two-adult family transportation costs =

[(% work & non-social trips)/100*(average miles per month by MSA size)* (cost/mile)]+ [(% work trips)/100*(average miles per month by MSA size)*(cost/mile)]

Equation components

• The share of work and non-social trips is calculated using the 2009 National Household Transportation Survey (NHTS), as 2009 is the most recent year for which data are available. The variable WHYTRP1S, or trip purpose summary, is used in conjunction with the variable Travel Day VMT to categorize each vehicle trip into the following purposes: home; work; school, day care, religious activity; medical, dental services; shopping, errands; social, recreational; family, personal business, obligations; transport someone; meals; and other purposes. We chose to make non-social trips the share of trips to home; school, day care, religious activity; medical, dental services; shopping, errands; family, personal business, obligations; and to transport someone.
• We decompose the trip purpose and average vehicle miles traveled by MSA size using the variable MSASIZE from the NHTS. MSASIZE uses the same definition of an MSA as does the Office of Management and Budget. The NHTS reports MSASIZE in six sizes, based on household population within a given area:
  • Not in an MSA (0–49,999 inhabitants)
  • In an MSA of 50,000–249,999 inhabitants
  • 250,000–499,999 inhabitants
  • 500,000–999,999 inhabitants
  • 1,000,000–2,999,999 inhabitants
  • 3,000,000+ inhabitants
• We group our MSAs by these six population categories using population data from the Office of Management and Budget.
• The IRS reports the standard mileage rates used to calculate the costs of operating an automobile for businesses, charitable, medical, or moving purposes. For 2014, the revised and most accurate standard mileage rates for the use of car, van, pickup, or panel truck is 56 cents per mile for business miles driven. The mileage rate includes fixed costs such as depreciation, lease payments, insurance, registration and license fees, and personal property taxes, and variable costs such as gasoline, oil, tires, and routine maintenance and repairs.

Example

Single parent in a rural area:

= (% work & non-social trips)/100*(average annual miles by MSA size)/12*(cost/mile) = (84.4%/100)* $14,607/12*($0.56) = .844*$1217.2 *$0.56

= $570.18

Thus, $570.18 is the monthly transportation cost for a single parent who lives in a rural area.

Health care

Health care expenses have two components: Affordable Care Act (ACA) insurance premiums and out-of-pocket expenditures.

Premiums

Premiums were obtained through a review of insurer rate filings to state regulators, as well as through data published by the U.S. Department of Health and Human Services and The Henry J. Kaiser Family Foundation’s Health Insurance Marketplace Calculator (Kaiser 2014).

Premiums are based on the lowest-cost bronze plan in the rating area, adjusted for family size, age of user, and tobacco surcharge (Kaiser 2014). The family budgets assume all adults are 40-year-old nonsmokers. EPI’s 2015 family budgets do not take into consideration the two types of health insurance subsidies available through the state and federal health insurance exchanges: the premium tax credit and the cost-sharing subsidy. Therefore, the health budget may be overestimated and can be reduced by the size of the subsidy.

Out-of-pocket costs

Out-of-pocket costs are from the MEPS Household Component (Full-Year Consolidated File) for 2012, in 2012 dollars (HHS 2014).

We assume that everyone has private health insurance (defined by the variable PRIV12). Out-of-pocket medical expenditures are calculated for adults and children separately by region and are differentiated between MSAs and non-MSAs for those covered by private insurance (HHS 2014). Costs are estimated as follows:

• We use the regional breakdown of costs for both adults and children (using the variable REGION12, with the regions defined as Northeast, Midwest, South, and West).
• The data are divided into MSA data and non-MSA data (using the variable MSA12). For out-of-pocket costs, we use MSA data for those areas that are strictly MSAs, and we use non-MSA data for nonmetropolitan FMR areas and rural areas (see the above section titled “Definition of areas” for more detail on the distinction).
• We classify a child (regardless of family size) as age 17 and under, and an adult as age 18–64 (using the variable AGE12X). We did not break down data for children into smaller age groups or by gender because the resulting sample sizes were too small.
• For each family budget area, adult out-of-pocket costs are the mean costs (variable TOTSLF12) for adults age 18–64 with private insurance in one of four regions and the metropolitan classification in that region.
• Child out-of-pocket costs are the mean costs for children age 0–17 with private insurance in one of four regions and the metropolitan classification in that region in 2012.
• We compute total out-of-pocket costs (OOP) in the following way:

[(number of parents) * (adult OOP)] + [(number of children) * (child OOP)]

• Since out-of-pocket costs are annual numbers, we divided by 12 to get the total monthly out-of-pocket costs.
• The total out-of-pocket costs were adjusted for inflation to 2014 dollars using the regional breakdowns of the Consumer Price Index-All Urban Consumers for Medical Care (CPI-U-MC) from the Bureau of Labor Statistics (2013c).
• When computing the mean, we used a population weight (variable PERWT12F).

Total health care costs

We compute total health care costs in the following way:

[Total premium] + [Total out-of-pocket cost]

Change in methodology

The health care methodology of the 2015 Family Budget Calculator differs from that of previous editions because it reflects changes to the private market for health insurance and the creation of state-based health insurance exchanges. These changes are one of the main reasons why the 2015 family budgets are not comparable to earlier family budgets.

In the 2013 edition of the Family Budget Calculator, we assumed that everyone had employer-sponsored health insurance. We used the total premium cost (what both the employee and employer contribute) to better reflect a measure of total compensation. Therefore, if the employer didn’t provide it, the family budget still allowed for the full amount. This affected estimates of premiums and out-of-pocket costs because employer-sponsored health insurance was assumed in the compilation of both data sets.

In the current edition, we calculate premiums based on the lowest-cost bronze plan and calculate out-of-pocket expenditures based on estimates for all types of private insurance, both employer-sponsored and non-group insurance.

Other necessities

Our calculation of “other necessities” is derived from the Bureau of Labor Statistics (BLS) Consumer Expenditures Survey (BLS 2014a). We consider other items of necessity as items that do not fall into the aforementioned categories, but are nevertheless necessary for a reasonably secure yet modest standard of living. These items include apparel, entertainment, personal care expenses, household supplies (including furnishings and equipment, household operations, housekeeping supplies, and telephone services), reading materials, school supplies, and other miscellaneous items of necessity.

We use the Consumer Expenditure Survey (CES) data for families in the second fifth of the overall income distribution. In the 2013 CES expenditure table “Quintiles of income before taxes,” “other necessities” is the proportion of costs for these items in relation to the costs of food and housing. In 2013, the proportion was 48.3 percent. Therefore, we devise our estimate of other necessities by applying this percent to each respective family budget’s food and housing costs.

Change in methodology

In the previous editions of the Family Budget Calculator, other necessities did not take into account a number of items, including household supplies, furnishings and equipment, household operations, housekeeping supplies, and telephone services. Therefore, the costs of other necessities in the 2015 Family Budget Calculator are significantly higher than those in the previous edition.

Taxes

The family budget components, without taxes, sum to the family’s post-tax income. To calculate the family budget tax component, a pre-tax income level had to be estimated using a tax rate and the post-tax income.

We utilized the National Bureau of Economic Research’s Internet TAXSIM Version 9.3 with ATRA to calculate these tax rates (NBER 2013). The TAXSIM model accepts 22 input variables, including state, marital status, dependent exemptions, wage income, other incomes, rent paid, child care expenses, and capital gains and losses (Feenberg et al. 1993). We ran the TAXSIM model for each family type across all 618 areas.

Our input variables were (variables not listed were input as zero):

• State
• Marital status (single for one-parent families, married for two-parent families)
• Dependent exemptions (one for each child)
• Wage and salary income of taxpayer (entire post-tax family budget for one-parent families)
• Wage and salary income of spouse (for two-parent families, the post-tax family budget was split evenly between the two parents)
• Rent paid (the annual cost of rent for each family budget, which is used to calculate state property tax rebates in certain states)
• Child care expenses (the annual cost of child care for each family budget)
• Number of dependents under age 17 (one for each child)

The TAXSIM model takes these inputs and calculates three outputs: federal tax liability, state tax liability, and FICA tax liability. All of these liabilities are for year 2013 tax law. Additionally, the TAXSIM model calculates FICA liability as the full 15.3 percent tax from both the employer and employee side; we cut this in half to more accurately represent the typical taxpayer. Local taxes, such as county- or city-level income taxes, are not included in this model. Sales taxes are also not included.

As aforementioned, it is not accurate to simply input the post-tax family budgets as the wage incomes and use the TAXSIM output as the tax rates. To correct for this, we input the post-tax family budgets and obtained the tax rates and established these as a lower floor for tax rates (because the pre-tax incomes will almost always be higher than these post-tax incomes, these tax rates must be lower given our assumptions about sources of income and the income ranges we are considering). We then established an upper bound of tax rates by taking the post-tax family budget and multiplying it by 1.25 and inputting these budgets into the TAXSIM model.

Once we had the lower and upper bounds of tax rates, we calculated an accurate average of these tax rates using a weighting procedure, described below:

1. Multiply the lower bound (post-tax family budget) and upper bound (post-tax family budget * 1.25) inputs by (1 – calculated tax rate)
2. Calculate the difference between the actual post-tax family budget and the lower bound calculated in step one: [post-tax family budget – lower bound]
3. Calculate the difference between the upper bound and the actual post-tax family budget calculated in step one: [upper bound – post-tax family budget]
4. Calculate the difference between the upper bound and the lower bound calculated in step one: [upper bound – lower bound]
5. Calculate the weight for the lower bound, which is equal to the upper–post-tax budget difference divided by the upper–lower difference:
6. Calculate the weight for the upper bound, which is equal to [1  – lower weight (calculated in step five)]
7. Multiply the lower bound tax rate from TAXSIM by the lower bound weight from step five: lower bound tax rate * 
8. Multiply the upper bound tax rate from TAXSIM by the upper bound weight from step six: upper bound tax rate * [1 – lower weight (calculated in step five)]
9. Add these two weights to get the final, weighted tax rate: [step seven + step eight]

The final tax rate calculated in step nine is then applied to the post-tax family incomes [post-tax family budget * (1 + final weighted tax rate)], to obtain a pre-tax income. The difference between the pre- and post-tax incomes is the annual tax bill for the family budget unit. This annual tax bill is then adjusted for inflation to 2014 dollars using the regional breakdowns of the Consumer Price Index-Research Series Using Current Methods (CPI-U-RS) from the Bureau of Labor Statistics (BLS 2014c).

In cases where the post-tax budget exceeds the bounds, we increase the budget multiplier by increments of .05 (1.3, 1.35, 1.4, 1.45, 1.5) until the post-tax budget no longer exceeds the upper bound.

About the authors

Elise Gould, senior economist, joined EPI in 2003 and is the institute’s director of health policy research. Her research areas include wages, poverty, economic mobility, and health care. She is a co-author of The State of Working America, 12th Edition. In the past, she has authored a chapter on health in The State of Working America 2008/09; co-authored a book on health insurance coverage in retirement; published in venues such as The Chronicle of Higher Education, Challenge Magazine, and Tax Notes; and written for academic journals including Health Economics, Health Affairs, Journal of Aging and Social Policy, Risk Management & Insurance Review, Environmental Health Perspectives, and International Journal of Health Services. She holds a master’s in public affairs from the University of Texas at Austin and a Ph.D. in economics from the University of Wisconsin at Madison.

Tanyell Cooke joined EPI in 2014. As a research assistant, she supports the research of EPI’s economists on topics such as wages, labor markets, inequality, education, race and ethnicity, and immigration. She holds a B.A. in economics and statistics from The George Washington University.

Alyssa Davis joined EPI in 2013 as the Bernard and Audre Rapoport Fellow. She assists EPI’s researchers in their ongoing analysis of the labor force, labor standards, and other aspects of the economy. Alyssa aids in the design and execution of research projects in areas such as poverty, education, health care, and immigration. She also works with the Economic Analysis and Research Network (EARN) to provide research support to various state advocacy organizations. Alyssa has previously worked in the Texas House of Representatives and the U.S. Senate. She holds a B.A. from the University of Texas at Austin.

Will Kimball joined EPI in 2013. As a research assistant, he supports the research of EPI’s economists on topics such as wages, labor markets, macroeconomics, international trade, and health insurance. Prior to joining EPI, Will worked at the Center on Budget and Policy Priorities and the Center for Economic and Policy Research. He holds a B.A. in economics and political science from the University of Connecticut.

References

Bureau of Labor Statistics (BLS). 2014a. Consumer Expenditure Survey. Current Expenditure Tables.

Bureau of Labor Statistics (BLS). 2014b. Consumer Price Index Program. All Urban Consumers (Current Series)–Child Care and Nursery School [database].

Bureau of Labor Statistics (BLS). 2014c. Consumer Price Index Program. All Urban Consumers (Current Series)–Medical Care [database].

Carlson, A., Lino, M., and Fungwe, T. 2007. The Low-Cost, Moderate-Cost, and Liberal Food Plans, 2007. U.S. Department of Agriculture, Center for Nutrition Policy and Promotion.

Child Care Aware of America (CCAA). 2014. Parents and the High Cost of Child Care: 2014 Report. 

Federal Highway Administration (FHA). 2009. National Household Travel Survey (NHTS) [tabulation created on the NHTS website]

Feenberg, Daniel Richard, and Elizabeth Coutts. 1993. “An Introduction to the TAXSIM Model.” Journal of Policy Analysis and Management, vol. 12 no. 1, 189–194.

Henry J. Kaiser Family Foundation (Kaiser). 2014. 2014 Health Insurance Marketplace Calculator.

Internal Revenue Service (IRS). 2013. “2014 Standard Mileage Rates.” IRS Announcement 2013-95.

Missouri Census Data Center. 2012. “MABLE/ Geocorr12 Version 1.2: Geographic Correspondence Engine.”

National Bureau of Economic Research (NBER). 2013. “TAXSIM Model Version 9.3 with ATRA.”

Office of Management and Budget (OMB). 2009. “Update of Statistical Areas Definitions and Guidance on Their Uses.” OMB Bulletin No. 10-02.

U.S. Census Bureau. 2013. Census Regions and Divisions of the United States.

U.S. Department of Agriculture Center for Nutrition Policy and Promotion (USDA). 2014. Cost of Food at Home: U.S. Average at Four Cost Levels (June 2014 annual average).

U.S. Department of Health and Human Services (HHS). 2014. Medical Expenditure Panel Survey HC-155 2012 Full Year Consolidated Data File [microdata].

U.S. Department of Housing and Urban Development (HUD). 2014a. “Fair Market Rents dataset” [county- level data file].

U.S. Department of Housing and Urban Development (HUD). 2014b. “Proposed Fair Market Rents for the Housing Choice Voucher Program, Moderate Rehabilitation Single Room Occupancy Program and Other Programs Fiscal Year 2015.” Federal Register, vol. 79, no. 158

U.S. Department of Housing and Urban Development (HUD). 2015. “HUD USPS ZIP Code Crosswalk Files.”  

This paper is part of the Full Employment Project of the Center on Budget and Policy Priorities. It was presented by Valerie Wilson on Monday, March 30, 2015, as part of a forum entitled “Full Employment: How Can We Get There and Stay There? Why Does It Matter?”.

By the end of 2014, the U.S. economy had experienced 58 consecutive months of job growth, and the unemployment rate had fallen to 5.6 percent from a high of 10 percent in October 2009. In fact, 2014 was by far the strongest year of the recovery, with job growth averaging over 246,000 per month, the highest monthly rate since before the recession. Economic growth also picked up, with gross domestic product rising at annual rates of 4.6 percent and 5 percent during the second and third quarters, respectively, following a first-quarter decline of 2.1 percent.

Last year’s solid job growth proved to be especially beneficial to communities of color, whose unemployment rates rose well above 10 percent during the worst years of the recession. In particular, after reaching a high of 16.8 percent in March 2010, the African American unemployment rate fell to 10.4 percent in December 2014. Between December 2013 and December 2014, African Americans had the largest increase in the share of adults with a job and the largest increase in their labor force participation rate,1 translating to nearly 200,000 fewer unemployed black workers.

Despite this progress, the labor market is nowhere near a full recovery, much less full employment. For African American workers in particular, much is at stake in whether the economy is allowed to reach a full recovery and full employment, as evidence from the last four recoveries strongly suggests:

• On average, the black unemployment rate is more volatile with respect to aggregate labor market changes than the white rate. Between 1979 and 2014, the average annual black unemployment rate changed by 1.7 percentage points for every 1 percentage-point change in the national unemployment rate, compared to a change of 0.91 percentage points for whites.
• Wages of black workers are more responsive to aggregate labor market changes. Doubling the national unemployment rate is estimated to reduce real hourly wages by at least 8 percent for the median black worker compared to 3 percent for the median white worker.

In the five-year period between 1995 and 2000, during which the annual unemployment rate dropped to 4 percent:

• The black unemployment rate fell to 7.6 percent, the lowest rate on record and the closest it has ever been to the white rate (within 4.1 percentage points) during a period of economic expansion.
• Real wage growth for African Americans narrowly exceeded that of whites, as median hourly wages of black workers grew by 2 percent per year compared to 1.7 percent per year for whites.
• The African American middle class expanded more than in periods of economic recovery when the economy was further from full employment. The share of African American households in the middle 60 percent of the income distribution increased 3 percentage points between 1995 and 2000, while it declined during the recoveries of the 1980s and the 2000s as well as during the current one.
• The fact that all of these positive developments occurred without setting off an inflationary spiral suggests that policymakers should be willing to experiment aggressively with low rates of unemployment in order to bring the benefits of a full recovery to African American households.

As the Recovery Builds Steam, Critical Labor Market Weaknesses Persist

The good news about the fall in the unemployment rate since its 2009 peak is tempered by the fact that labor force participation continued to decline sharply even after the recession officially ended. A better measure of labor market strength is the prime-age employment-to-population ratio (EPOP), which measures the share of adults between the ages of 25 and 54 who are working. The prime-age EPOP declined 4.8 percentage points between 2007 and 2010 (peak to trough), and the average for 2014 remained 3.2 percentage points below the pre-recession rate. The employment gap for prime-age African Americans is even wider: Having declined 8.1 percentage points between 2007 and 2011 (peak to trough), it remained 4.2 percentage points below the pre-recession rate.

Another key indicator of remaining slack in the labor market has been slow wage growth. Real wages have been essentially flat since 2009, and even with the recent drop in inflation wages in 2014 grew at a slightly slower pace than wages in 2013.2 Wage growth is as an important signal of inflationary pressure that the Federal Reserve considers as it sets interest rate policy. But as Bivens and Gould point out, as long as nominal wages grow at or below the rate of productivity growth plus the Fed’s price inflation target, there is no risk that excess labor market tightness will push inflation above that target.3 If this is the case, then with productivity growth averaging 1.5 to 2 percent and nominal wage growth sitting at 2 percent over the last year, wage growth could essentially double without putting any pressure on the Fed’s 2 percent price inflation target.

In spite of the gains in employment, for most workers slow and stagnant wage growth continues to challenge any sense of confidence about the economic recovery. History strongly suggests that full employment provides the best chance for American workers to experience wage and income growth. Baker and Bernstein show that incomes have grown faster and more equally when the economy was at or near full employment, notwithstanding other factors that influence wage growth.4 They also show that lower-income and African American families suffer greater income losses during periods of slack labor markets and faster income growth when labor markets are tighter.

Federal Reserve Policy Looms Large Over Prospects for Full Employment and Wage Growth

The Federal Reserve’s monetary policy decisions have tremendous influence over economic growth and employment in the United States and across the world. The Great Recession tested the limits of this power, as the Federal Reserve took extraordinary actions to stabilize the financial sector, end the recession, and bring about economic recovery. At the beginning of 2015, near-zero interest rates are the last policy tool that the Fed continues to use to maximum effect to promote a full economic recovery. Still, the Fed is under pressure from monetary hawks to raise short-term interest rates as a guard against price and wage inflation.

The timing of the Fed’s decision to raise interest rates will be a deciding factor in how close the economy gets to full employment and, thus, how fast the wages of American workers will grow. Given the amplified relationship between labor market strength and African American income growth estimated by Baker and Bernstein,5 the Fed’s decision also has major implications for how complete the recovery will be for African American workers. Evidence on today’s labor market suggests that it would be premature to slow the recovery in the name of staying ahead of any possible inflationary pressure. And the potential consequences of premature tightening are huge — years of foregone income and wage growth for precisely those American families that have disproportionately suffered over the past decade.

African Americans Experience Stronger Employment Gains When Labor Markets Are Tight

This paper examines how full employment, or the lack thereof, has affected African American employment and wages over the last four economic cycles. This examination argues that genuinely full employment has been much rarer than one might expect taking at face value estimates of the natural rate of unemployment (i.e., the NAIRU, the non-accelerating inflation rate of unemployment). The implication for policy is that we should be willing to experiment aggressively with low rates of unemployment.

The analysis begins by contrasting the economic recoveries following the 1981, 1991, 2001, and 2007 recessions with the five-year period between 1995 and 2000, when the unemployment rate dropped to its lowest level in generations—well below existing estimates of the NAIRU—and wages rose strongly across the wage distribution. I then estimate the statistical relationship between the unemployment rate and real wages for black and white workers since 1979.

Tables 1 and 2 present two basic indicators of labor market strength — the unemployment rate and the employment-population ratio — along with the changes in those indicators from the trough to the peak of the last four economic recoveries as well as from 1995 to 2000. These numbers are reported for all workers and for white and black workers separately. The first panel of each table summarizes conditions for the entire working-age population (age 16 and older), and the second panel includes only prime-age workers (age 25 to 54).

Table 1

Table 2

Looking across the four periods of recovery, it is clear that the 1990s recovery produced the strongest labor market in recent history. The overall unemployment rate reached a low of 4 percent on an annual basis, and the lowest rates for whites and blacks were 3.5 percent and 7.6 percent, respectively. (See Figure 1.) Although the roughly 2-to-1 black-white unemployment-rate ratio held across each of the economic recoveries, the black unemployment rate was closest to the white rate (within 4.1 percentage points) during the 1990s. Moreover, the lowest annual black unemployment rate of the 1990s was nearly 4 percentage points lower than the low of the 1980s and the low of the current recovery. For white workers, the difference between the 1990s and other periods was less dramatic: The lowest annual white unemployment rate of the 1990s was only 1 percentage point below the low of the 1980s and 1.8 percentage points below the most recent low. Though unemployment rates were relatively low during the post-2001 recovery, their level was due more to an extension of the gains of the late 1990s, and a relatively mild recession, than to a particularly strong labor market recovery. In fact, in terms of simple rates of job growth, the recovery following the 2001 recession remains the worst post-World War II recovery on record; unemployment rates barely changed between the trough and peak. Further, wage-growth trends seemed to retain their momentum from the late 1990s recovery into the 2001 recession. Wages for most workers rose until 2002 or even 2003 before reversing as the recovery matured (an odd historical pattern).

Figure A

For prime-age workers, labor market outcomes were even better during the 1990s than they were for workers overall. Prime-age workers typically have the most consistent labor market attachment because they are less likely to voluntarily leave the labor market to enroll in school or retire. In the right-hand panel of Table 2 we see that, during the 1990s, up to 81.5 percent of all people between the ages of 25 and 54 were employed. The peak rate for whites, at 82.4 percent, was 4.3 percentage points higher than in the current recovery, and the peak rate for blacks, 77.2 percent, was 6.6 points higher.

The Bureau of Labor Statistics (BLS) does not report prime-age EPOP ratios by race for years prior to 1994. However, we can glean the story by comparing prime-age unemployment rates of blacks and whites and get an idea of how well this core group of African American workers fared during the late 1990s. While the unemployment rate for African American workers age 16 or older reached a low of 7.6 percent during the 1990s recovery, the rate for prime-age African American workers fell to 5.6 percent (see the left and right panels of Table 1). The latter was 3.4 percentage points below the low of the 1980s recovery and 3.9 points below the low of the current recovery. By focusing on the segment of the workforce with the strongest labor market attachment, we get a sense of just how important full employment is to bringing down the black unemployment rate. Since BLS began reporting black unemployment rates in 1972, 2000 was the only year the annual rate for African American workers age 16 or older was less than 8 percent. The rate for prime-age African American workers has been less than 6 percent in only three years, two of which occurred during the 1990s recovery — 1973 (5.7 percent), 1999 (5.9 percent), and 2000 (5.6 percent).

In addition to historically high black employment during the 1995–2000 period of full employment, another important point to draw from Table 1 is the fact that blacks experienced relatively larger labor market gains than whites during the 1990s recovery. While the black unemployment rate declined by twice as much as the white rate during both the 1980s and 1990s recoveries, the larger absolute change in unemployment rates during the 1980s should be evaluated within the context of the depth of the preceding recession, a downturn that, prior to the 2007 recession, held the distinction as the worst on record since the Great Depression. However, relative changes in employment, as measured by the EPOP, were much higher for blacks than whites in the 1990s recovery than in the 1980s. During the 1980s recovery, employment of black working-age adults increased 1.5 times more than it did for whites (7.3 versus 4.9 percentage points). During the 1990s, the increase for blacks was 2.7 times greater (4.3 versus 1.6 percentage points).

The discussion so far has focused on labor market outcomes for prime-age workers who typically have stronger labor force attachments. The relative strength of the labor market over different periods of recovery can be further underscored by gains among teens and young adults, groups that traditionally have weaker labor market attachments or face greater barriers to employment. Beginning in 1994, BLS began reporting alternative measures of labor utilization that are useful for assessing the strength of the labor market. One of those measures, commonly referred to as underemployment, accounts for people officially classified as unemployed but also for marginally attached workers and people working part time for economic reasons. Marginally attached workers are people who searched for work in the past 12 months (though not the most recent four weeks) and are willing and available to take a job. Those working part time for economic reasons are people who would prefer full-time employment if it were available.

Table 3 presents changes in unemployment, underemployment, and EPOP ratios for the last three recoveries and for 1995–2000 specifically for workers age 16-24. Underemployment for all young workers fell to a much lower rate (14.4 percent) during the 1990s than it has during the current recovery (22.8 percent). For young African American workers, the underemployment rate reached a low of 24.9 percent during the 1990s, compared to 32.3 percent in the current recovery. Comparing the change in the unemployment rate over the entire 1990s recovery to the change from 1995 to 2000, we see that all the improvement in this group’s unemployment rate took place during the 1995–2000 part of that recovery, during which the overall unemployment rate was below 5 percent for four of the five years. Similarly, the share of all 16-24-year-olds who were employed peaked at 59.7 percent in 2000 but has declined steadily since then. While school enrollment has been trending upward over the last two decades, it does not explain the decline in employment for this group: Enrollment rates haven’t increased any faster since 2000 than they did between 1995 and 2000, and in fact they have plummeted since 2012.6 Rather, tighter labor markets in the mid- to late-1990s served to effectively expand job opportunities for young job seekers who traditionally face greater barriers to employment.

Table 3

This examination carries an important lesson: Premature interest-rate tightening in the mid-1990s would have robbed millions of the most disadvantaged workers of a half-decade of historically fast job growth. And to be clear —many in the mid-1990s argued precisely that unemployment should not be allowed to get much below the 5.6 percent rate that prevailed at the beginning of 1996, as pushing it below this level, they believed, would surely set off accelerating inflation. Four years later, overall unemployment averaged 4 percent for two solid years, and accelerating price inflation never arrived.

The evidence presented in this section shows that, on average, the black unemployment rate is more volatile with respect to aggregate labor market changes. Between 1979 and 2014, the average annual black unemployment rate changed an average of 1.7 percentage points for every 1 percentage-point change in the national unemployment rate (see Figure 2); the corresponding figure for whites was 0.91. In short, the stakes of effective macroeconomic stabilization policy are extraordinarily high for African American employment. African American families suffer disproportionately from labor market downturns and reap disproportionate gains during recoveries. Given the vast suffering inflicted by the Great Recession, and given the slow job growth of the early phase of recovery, it would be a tragedy to deny the gains that could be achieved should the economy be allowed to reach a full recovery, and even full employment.

Figure B

The distinction between a full recovery and full employment is an important one. A full recovery would simply mean a return to pre-Great Recession labor market conditions without any regard to how good or bad those conditions were. Full employment, on the other hand, raises the bar to a point at which all who are willing to work at the prevailing wage rate are employed. Another way of thinking about full employment is as the point at which all unemployment is “frictional,” i.e., consists of people searching for the best job fit.

African Americans Experience Stronger Growth in Income and Hours Worked When Labor Markets Are Tight

Thus far, our analysis has focused primarily on differences in employment outcomes across each of the last four economic recoveries. Since most Americans derive the large majority of their income from working, increased employment is also associated with increased income. Figure 3 shows how income increased more for African American households7 than for white households during the recoveries that followed the 1981 and 1991 recessions, but also declined more during the recoveries from the 2001 and 2007 recessions. Household labor income can increase because household members work more hours, because hourly wages increase, or both. As Table 4 shows, in all periods when median household income for African Americans increased more than that for whites (1982-1990, 1991-2001, and 1995-2000), African American households also experienced a greater increase in hours worked than did white households, especially if the households had lower incomes.

Figure C

Table 4

Trends in annual median wage growth show that inflation-adjusted median wages for all workers grew by less than 0.5 percent per year during the 1982–1990 recovery but grew by about 1 percent per year during the recovery of 1991–2001. (See Figure 4.) Real median hourly wage growth for African Americans narrowly exceeded that of whites between 1995 and 2000; median wages for black workers grew by 2 percent per year compared to 1.7 percent per year for whites. The combination of increased hours of work and higher annual wage growth for low-income families during the late 1990s recovery had the added benefit of shifting more African American households into the broad middle class, defined here as the middle 60 percent of the income distribution. Between 1995 and 2000 the share of African American households in the middle 60 percent of the income distribution grew by nearly 3 percentage points, while the share of white middle-income households declined by just under 1 percentage point. (See Figure 5.) This expansion of the black middle class was unique to this period of economic growth. During each of the other recoveries, the share of black households in the middle 60 percent actually declined.

Figure D

Figure E

The Role of Full Employment in Driving Wage Growth of Black and White Workers

The preceding sections of this report presented evidence on broad trends in employment, income, hours worked, and wages over the last four economic recoveries in order to demonstrate the importance of full employment in promoting greater employment and wage growth for all workers and for African Americans in particular. Next, I use regression analysis to formalize the relationship between the unemployment rate and real hourly wages of white and black workers. This exercise is an extension of the original Phillips curve, which posited a negative relationship between the unemployment rate and nominal wage growth.8 The acceleration of inflation in the 1970s led many macroeconomists to assert that low rates of unemployment did not just lead to higher rates of wage growth but also to accelerating rates of wage growth that could lead to spiraling price inflation. The lowest level of unemployment consistent with stable wage (and hence price) inflation was labeled the NAIRU, and an entire industry of economists sprung up trying to provide policy-relevant ex ante estimates of it.

The experience of the late 1990s showed that these ex ante NAIRU estimates were too conservative, but it is the only period since 1979 in which policymakers pushed actual unemployment rates down anywhere near these too-conservative targets. As others have shown, trends in real wage growth, productivity growth, and full employment since 1979 suggest that sustained periods of unemployment above official estimates of the NAIRU have been at least one of the factors working to suppress real wages. Bivens et al. show that wages generally kept pace with productivity during most of the post-World War II era, but starting in 1979 this relationship started to diverge.9 One of several factors contributing to the weakening of this relationship for most earners has been the failure to pursue full-employment policies. According to Baker and Bernstein, the unemployment rate was an accumulated 31 percentage points above official estimates of the NAIRU for the years 1980 to 2012, compared to an accumulated 15 percentage points below this measured NAIRU from 1949 to 1979.10

At the median, wage growth of black workers has been more responsive to changes in labor market slack than wage growth of white workers. (See Figure 6.) This relationship held for the entire period between 1979 and 2014 as well as between 1979 and 2007, a period that excludes the effects of the Great Recession and subsequent recovery. These results are based on annual state-level real median hourly wages for whites and blacks and state unemployment rates. (Under this specification, individual state and year “dummy variables” are included to control for any effects on wages, other than unemployment, that might be unique to a particular state or year.)

Figure F

Row 1 of Table 5 shows in statistical terms that a doubling of the state unemployment rate reduced median real hourly wages by 7.9 percent for blacks and 3.0 percent for whites between 1979 and 2014. In the period before the Great Recession, a doubling of the state unemployment rate reduced real median hourly wages by 10.1 percent for blacks and 3.4 percent for whites. The lower estimates for the period including the Great Recession are consistent with the concept of downward nominal wage rigidities.11 With inflation relatively low and employers reluctant to cut the dollar value of workers’ pay during tough economic times, real wages have been less responsive to changes in the unemployment rate than in the years preceding the Great Recession. Estimates of wage responsiveness to changes in the EPOP and the underemployment rate reveal similar patterns through 2007, with the median real hourly wage of black workers changing more than the wage of white workers in response to fluctuations in the labor market.

Table 5

Because population size restricts the number of states and years for which reliable median wage estimates are available for African Americans, the above analysis includes fewer observations for black workers than for whites. The estimates in row 5 of Table 5 are based on the subsample of eight states for which a full panel of median wage data for African Americans is available.12 Again, the effect of changes in the unemployment rate on median wage growth is roughly twice as large for African American workers as it is for whites.

Conclusion

This paper presents a case for the role of full employment in helping to reduce racial disparities in unemployment and wages. On average, African American families suffer disproportionately from labor market downturns, but they also reap disproportionate gains during recoveries. These labor market fluctuations affect employment levels as well as wage growth, the primary means by which the majority of families maintain and improve their standard of living.

Based on evidence from the economic recoveries following the 1981, 1991, 2001, and 2007 recessions, this examination argues that genuinely full employment has been much rarer than one might expect taking estimates of the NAIRU at face value. But in the five-year period between 1995 and 2000, during which the annual unemployment rate dropped to 4 percent, the difference between the black and white unemployment rates was smaller than it’s ever been — within 4.1 percentage points — during an economic expansion; real median hourly wage growth for African Americans narrowly exceeded that of whites; and the African American middle class expanded. Further, all of this happened without causing price inflation to accelerate or even increase.

Given the vast suffering caused by the Great Recession and the slow job growth of the early phase of recovery, much is at stake for African American families in whether the economy is allowed to reach a full recovery and full employment. Based on evidence over the past year as well as the estimates presented in this report, a 1 percentage-point decline in the national unemployment rate means that roughly 200,000 fewer African American workers will be unemployed.13 From a policy perspective, this likely outcome strongly suggests that we should experiment aggressively with low rates of unemployment.

Endnotes

1. Valerie Wilson, “Keep the Jobs Coming! People of Color Have Actually Benefited More From Job Growth This Year,” Working Economics (Economic Policy Institute blog), November 7, 2014 ; Valerie Wilson, “Single Digit Black Unemployment May Not Be so Far Away,” Working Economics (Economic Policy Institute blog), January 9, 2015.

2. Elise Gould, “Average Real Hourly Wage Growth in 2014 Was No Better Than 2013,” Working Economics (Economic Policy Institute blog), January 16, 2015.

3. Josh Bivens and Elise Gould, “What to Watch on Jobs Day: It’s No Longer a Jobless Recovery but It’s Undoubtedly a Wage-Growth-Less Recovery,” Working Economics (Economic Policy Institute blog), September 4, 2014.

4. Dean Baker and Jared Bernstein, Getting Back to Full Employment: A Better Bargain for Working People (Washington, DC: Center for Economic Policy Research, 2013).

5. Ibid.

6. Heidi Shierholz, Alyssa Davis, and Will Kimball, “Class of 2014: The Weak Economy Is Idling Too Many Young Graduates,” Economic Policy Institute, Briefing Paper #377.

7. The race of the household is based on the race of the household head.

8. Alban Phillips, “The Relation Between Unemployment and the Rate of Change of Money Wages in the United Kingdom, 1861-1957,” Economica (25, 1958), pp. 283-99.

9. Josh Bivens, Elise Gould, Lawrence Mishel, and Heidi Shierholz, “Raising America’s Pay: Why It’s Our Central Economic Policy Challenge,” Economic Policy Institute, Briefing Paper #378.

10. Baker and Bernstein 2013.

11. Mary C. Daly and Bart Hobijn, “Downward Nominal Wage Rigidities Bend the Phillips Curve,” Federal Reserve Bank of San Francisco, Working Paper 2013-08.

12. African American wage rates are missing in 1980 for Arizona and New Mexico, but since excluding these states did not change the results, there were included to maximize the number of observations.

13. This estimate assumes a continuation of trends in black labor force participation and unemployment over the last year (from December 2013 to December 2014). The black labor force grew by 3.4 percent over the year.

Acknowledgments

We thank Hilary Wething for outstanding research assistance. We are grateful to David Autor for generously making his data and programs available, and for an ongoing lively and helpful discussion. We thank Dean Baker, Annette Bernhardt, David Card, Michael Handel, David Howell, Frank Levy, Jesse Rothstein, Ben Sand, and participants at the “Inequality in America: Contending Theories” panel at the 2013 ASSA annual meeting, the University of California-Berkeley labor economics seminar, seminar participants at the Council of Economic Advisers and the Brookings Institution, and participants at the Institute for Work and Employment Research seminar series at the MIT Sloan School of Management. We appreciate the Institute for New Economic Thinking (INET) support of this work.

Executive summary

Many economists contend that technology is the primary driver of the increase in wage inequality since the late 1970s, as technology-induced job skill requirements have outpaced the growing education levels of the workforce. The influential “skill-biased technological change” (SBTC) explanation claims that technology raises demand for educated workers, thus allowing them to command higher wages—which in turn increases wage inequality. A more recent SBTC explanation focuses on computerization’s role in increasing employment in both higher-wage and lower-wage occupations, resulting in “job polarization.” This paper contends that current SBTC models—such as the education-focused “canonical model” and the more recent “tasks framework” or “job polarization” approach mentioned above—do not adequately account for key wage patterns (namely, rising wage inequality) over the last three decades. Principal findings include:

1. Technological and skill deficiency explanations of wage inequality have failed to explain key wage patterns over the last three decades, including the 2000s.

The early version of the “skill-biased technological change” (SBTC) explanation of wage inequality posited a race between technology and education where education levels failed to keep up with technology-driven increases in skill requirements, resulting in relatively higher wages for more educated groups, which in turn fueled wage inequality (Katz and Murphy 1992; Autor, Katz, and Krueger 1998; and Goldin and Katz 2010). However, the scholars associated with this early, and still widely discussed, explanation highlight that it has failed to explain wage trends in the 1990s and 2000s, particularly the stability of the 50/10 wage gap (the wage gap between low- and middle-wage earners) and the deceleration of the growth of the college wage premium since the early 1990s (Autor, Katz, and Kearney 2006; Acemoglu and Autor 2012). This motivated a new technology-based explanation (formally called the “tasks framework”) focused on computerization’s impact on occupational employment trends and the resulting “job polarization”: the claim that occupational employment grew relatively strongly at the top and bottom of the wage scale but eroded in the middle (Autor, Levy, and Murnane 2003; Autor, Katz, and Kearney 2006; Acemoglu and Autor 2012; Autor 2010). We demonstrate that this newer version—the task framework, or job polarization analysis—fails to explain the key wage patterns in the 1990s it intended to explain, and provides no insights into wage patterns in the 2000s. We conclude that there is no currently available technology-based story that can adequately explain the wage trends of the last three decades.

2. History shows that middle-wage occupations have shrunk and higher-wage occupations have expanded since the 1950s. This has not driven any changed pattern of wage trends.

We demonstrate that key aspects of “job polarization” have been taking place since at least 1950. We label this “occupational upgrading” since it primarily consists of shrinkage in relative employment in middle-wage occupations and a corresponding expansion of employment in higher-wage occupations. Lower-wage occupations have remained a small (less than 15 percent) and relatively stable share of total employment since the 1950s, though they have grown in importance in the 2000s. Occupational upgrading has occurred in decades with both rising and falling wage inequality and in decades with both rising and falling median wages, indicating that occupational employment patterns, by themselves, cannot explain the salient wage trends.

3. Evidence for job polarization is weak.

We use the Current Population Survey to replicate existing findings on job polarization, which are all based on decennial census data. Job polarization is said to exist when there is a U-shaped plot in changes in occupational employment against the initial occupational wage level, indicating employment expansion among high- and low-wage occupations relative to middle-wage occupations. As shown in Figure E (explained later in the paper but introduced here), in important cases, these plots do not take the posited U-shape. More importantly, in all cases the lines traced out fit the data very poorly, obscuring large variations in employment growth across occupational wage levels.

4. There was no occupational job polarization in the 2000s.

In the 2000s, relative employment expanded in lower-wage occupations, but was flat at both the middle and the top of the occupational wage distribution. The lack of overall job polarization in the 2000s is a phenomenon visible in both the analyses of decennial census/American Community Survey data provided by proponents of the tasks framework/job polarization perspective (Autor 2010; Acemoglu and Autor 2012) and in our analysis of the Current Population Survey. Thus, the standard techniques applied to the data for the 2000s do not establish even a prima facie case for the existence of overall job polarization in the most recent decade. This leaves the job polarization story, at best, as an account of wage inequality in the 1990s. It certainly calls into question whether it should be a description of current labor market trends and the basis of current policy decisions.

5. Occupational employment trends do not drive wage patterns or wage inequality.

We demonstrate that the evidence does not support the key causal links between technology-driven changes in tasks and occupational employment patterns and wage inequality that are at the core of the tasks framework and job polarization story. Proponents of job polarization as a determinant of wage polarization have, for the most part, only provided circumstantial evidence: both trends occurred at the same time. The causal story of the tasks framework is that technology (i.e., computerization) drives changes in the demand for tasks (increasing demand at the top and bottom relative to the middle), producing corresponding changes in occupational employment (increasing relative employment in high- and low-wage occupations relative to middle-wage occupations). These changes in occupational employment patterns are said to drive changes in overall wage patterns, raising wages at the top and bottom relative to the middle. However, the intermediate step in this story must be that occupational employment trends change the occupational wage structure, raising relative wages for occupations with expanding employment shares and vice-versa. We demonstrate that there is little or no connection between decadal changes in occupational employment shares and occupational wage growth, and little or no connection between decadal changes in occupational wages and overall wages. Changes within occupations greatly dominate changes across occupations so that the much-focused-on occupational trends, by themselves, provide few insights.

6. Occupations have become less, not more, important determinants of wage patterns.

The tasks framework suggests that differences in returns to occupations are an increasingly important determinant of wage dispersion. Using the CPS, we do not find this to be the case.  We find that a large and increasing share of the rise in wage inequality in recent decades (as measured by the increase in the variance of wages) occurred within detailed occupations.  Furthermore, using DiNardo, Fortin, and Lemieux’s reweighting procedure, we do not find that occupations consistently explain a rising share of the change in upper tail and lower tail inequality for either men or women.

7. An expanded demand for low-wage service occupations is not a key driver of wage trends.

We are skeptical of the recent efforts of Autor and Dorn (2013) that ask the low-wage “service occupations” to carry much or all of the weight of the tasks framework. First, the small size and the slow, relatively steady growth of the service occupations suggest significant limitations of a technology-driven expansion of service occupations to be able to explain the large and contradictory changes in wage growth at the bottom of the distribution (i.e., between middle and low wages, the 50/10 wage differential), let alone movements at the middle or higher up the wage distribution. The service occupations remain a relatively small share of total employment; in 2007, they accounted for less than 13 percent of total employment, and just over half of employment in the bottom quintile of occupations ranked by wages. Moreover, these occupations have expanded only modestly in recent decades, increasing their employment share by 2.1 percentage points between 1979 and 2007, with most of the gain in the 2000s. Relative employment in all low-wage occupations, taken together, has been stable for the last three decades, representing a 21.1 percent share of total employment in 1979, 19.7 percent in 1999, and 20.0 percent in 2007.

Second, the expansion of service occupation employment has not driven their wage levels and therefore has not driven overall wage patterns. The timing of the most important changes in employment shares and wage levels in the service occupations is not compatible with conventional interpretations of the tasks framework. Essentially all of the wage growth in the service occupations over the last few decades occurred in the second half of the 1990s, when the employment share in these occupations was flat. The observed wage increases preceded almost all of the total growth in service occupations over the 1979–2007 period, which took place in the 2000s, when service occupation wages were falling (another trend that contradicts the overall claim of the explanatory power of service occupation employment trends).

8. Occupational employment trends provide only limited insights into the main dynamics of the labor market, particularly wage trends.

A more general point can and should be drawn from our findings: Occupational employment trends do not, by themselves, provide much of a read into key labor market trends because changes within occupations are dominant. Recent research and journalistic treatment of the labor market has highlighted the pattern of occupational employment growth to assess the extent of structural unemployment, the disproportionate increase in low-wage jobs, and the “coming of robots”—changes in workplace technology and the consequent impact on wage inequality. The recent academic literature on wage inequality has highlighted the role of changes in the occupational distribution of employment as the key factor. In particular, occupational employment trends have become increasingly used as indicators of job skill requirement changes, reflecting the outcome of changes in the nature of jobs and the way we produce goods and services. Our findings indicate, however, that occupational employment trends give only limited insight and leave little imprint on the evolution of the occupational wage structure, and certainly do not drive changes in the overall wage structure. We therefore urge extreme caution in drawing strong conclusions about overall labor market trends based on occupational employment trends by themselves.

I. Introduction

Skill-biased technological change (SBTC) has been a leading explanation for the rise in wage inequality almost since economists first noticed the increase in wage inequality that began at the end of the 1970s. A recent wave of research, however, has questioned important aspects of the standard version of SBTC models of wage inequality (Autor, Levy, and Murnane 2003; Autor, Katz, and Kearney 2006, 2008; Acemoglu and Autor 2011, 2012; and others), frequently invoking arguments made in an earlier round of criticism of SBTC-based explanations (Mishel and Bernstein 1994, 1998; Howell 1994, 1999; Mishel, Bernstein, and Schmitt 1997; Galbraith 1998; Howell and Wieler 1998; Card and DiNardo 2002, 2006). The new research rejects key features of the long-standing SBTC models, but is itself closely tied to an alternative, technology-based explanation of rising wage inequality. This new “tasks framework” grows out of important insights about the role of technology in production that were first discussed by Autor, Levy, and Murnane (2003) and has had its most formal presentation in a model developed by Acemoglu and Autor (2011, 2012).

This paper seeks to assess the usefulness of the “tasks framework” as implemented in a growing body of empirically oriented papers. Elsewhere, we have offered an alternative explanation of widening wage inequality since the late 1970s, which does not appeal to technology as an important explanatory factor.1 Here, however, we make a narrower argument that current models of SBTC —either what Acemoglu and Autor call the “canonical model” or the more recent tasks framework offered to replace it— do not adequately account for key wage trends over the last three decades. We largely concur with Acemoglu and Autor’s critique of the “canonical model” (we have been making similar arguments since at least the mid-1990s). As a theoretical exercise, we also find Acemoglu and Autor’s formal modeling of the tasks framework elegant and much richer than the “canonical model” it seeks to supplant. But, we argue here that, its insights and elegance notwithstanding, the tasks framework fails to explain the most important developments in wage trends observed since the end of the 1970s.

A central empirical feature of the tasks framework is the concept of “job polarization,” usually defined as stronger employment growth in jobs at the top and bottom of the wage distribution than in the middle. Job polarization is closely linked to the argument that the last several decades have seen a “hollowing out” of the middle of the wage distribution. To be clear from the outset, we have no strong views about whether or not computerization caused job polarization over the last three decades. In what follows, we comment extensively on the timing, direction, and magnitude of changes in occupational employment patterns, but our purpose is not to suggest that job polarization did or did not take place. Instead, our interest lies in whether any employment polarization that did take place is consistent with predictions made by the tasks framework and, more importantly, whether and to what degree any observed changes in occupational employment can contribute to our understanding of changes in the overall wage distribution, which is the focus of the canonical SBTC model, the tasks framework, and other competing models.

To preview our main findings, we believe that the tasks framework fails as an explanation of rising wage inequality. Technological forces may be behind observed changes in the wage distribution, but, if so, current versions of the tasks framework do not adequately represent those forces.

Our critique of the tasks framework has several elements. We begin by demonstrating that key aspects of job polarization have been taking place since at least 1950. We label this “occupational upgrading” since it primarily consists of a shrinkage of middle-wage occupations and a corresponding expansion of higher-wage occupations. This enduring decline in middle-wage jobs—most typically in manufacturing and administrative and clerical work—makes job polarization a weak candidate for explaining the rise in wage inequality that only began at the end of the 1970s. Focusing on just the period since the late 1970s, the continued smooth decline in employment in middle-wage occupations, combined with the continued smooth rise in employment in higher-wage occupations and the relative stability of lower-wage occupations (until the 2000s) makes these employment changes an unlikely driver of the sharp rise (from about 1979) and the subsequent fall (beginning in about 1986–1987) in the size of the earnings differential between workers at the 50th percentile and the 10th percentile of the wage distribution. The inability of the tasks framework to provide an empirical accounting of the decline in the 50/10 differential after 1986–87 is particularly problematic because the need to explain differences in the divergent paths of the 90/50 and 50/10 differentials after the late 1980s was an important motivation for the theoretical innovations first proposed by Autor, Levy, Murnane (2003) and Autor, Katz, and Kearney (2006, 2008), and later formalized in Acemoglu and Autor (2011, 2012).

We also demonstrate significant problems with the evidence that is most commonly used to establish the existence of job polarization—the U-shaped plots of changes in occupational employment against the initial occupational wage level. In important cases, these plots do not take the posited U-shape and in all cases the lines traced out fit the data poorly, obscuring large variations in employment growth across occupational wage levels.2

When, as is often done, these same plots lump together the experience of the last two or three decades, they also mask substantial differences in occupational employment patterns for the 1980s, 1990s, and 2000s. Most importantly, the now-standard plots show no signs of overall job polarization in the 2000s. In the 2000s, relative employment expanded at the bottom, but was flat at both the middle and the top of the occupational wage distribution. The lack of job polarization in the 2000s is a phenomenon visible in both the decennial census / American Community Survey data and in our analysis here of data from the Current Population Survey (CPS). Thus, the standard techniques applied to the data for the 2000s do not establish even a prima facie case for the existence of job overall polarization in the most recent decade. This leaves the job polarization story, at best, as an account of wage inequality in the 1990s and clearly not a description of current economic trends on which one should base policy.

We also document that data from the CPS—unexpectedly—reveal occupational employment polarization the 1980s. In earlier empirical work using the decennial census, occupational employment rises monotonically with initial occupational wages, which is consistent with the monotonic rise in overall wages by initial position in the wage distribution. But, using the CPS data, occupational employment shows almost as much of a tendency toward job polarization in the 1980s as it did in the 1990s. As a result, and contrary to the conventional interpretation of the tasks framework, we have two cases (the 1980s and the 2000s) in which rising occupational employment shares at the bottom were associated with falling wages at the bottom; and only one case—the 1990s—when rising occupational employment shares at the bottom were associated with rising relative wages.

More generally, we find little evidence consistent with the causal story of the tasks framework. In the most commonly told version of the story, technology drives changes in the demand for tasks (increasing demand at the top and bottom relative to the middle) and this change in demand for tasks drives changes in occupational employment (increasing employment at the top and the bottom relative to the middle). The change in occupational employment is then supposed to drive changes in overall wages by operating through occupational wages. The literature, however, has neglected to carefully examine the intervening dynamic of changing the occupational wage structure (raising wages more at the top and the bottom than at the middle).We demonstrate that there is little or no connection between decadal changes in occupational employment shares and occupational wage growth, and little or no connection between decadal changes in occupational wages and overall wages. The only evidence presented by proponents of the tasks framework for the shifting pattern of the 50/10 wage gap between the 1980s and 1990s, for instance, has been circumstantial evidence: Job polarization and wage polarization both occurred in the 1990s.

We also question the view that differences in returns to occupations are an increasingly important determinant of wage dispersion, possibly even exceeding in statistical importance differences in the returns to education. Informally, we show that the goodness-of-fit of the standard plots of occupational-employment changes against initial occupational-wage-levels declines steadily and sharply between the 1980s, 1990s, and 2000s, suggesting that occupations have substantially less explanatory power in recent years. A more formal regression analysis of the contribution of occupation categories using the CPS finds, contrary to the findings in Acemoglu and Autor (2011), that occupations do not explain an increasing share of wage variation over time. We find a large and increasing share of the rise in wage inequality in recent decades (as measured by the increase in the variance of wages) occurred within detailed occupations.  Furthermore, using DiNardo, Fortin, and Lemieux’s reweighting procedure, we do not find occupations consistently explain a rising share of the change in upper tail and lower tail inequality for either men or women.

We are also skeptical of the recent research of Autor and Dorn (2013) that asks the typically low-wage “service occupations” to carry much or all of the weight of the tasks framework. First, the magnitude of employment changes in the service occupations (excluding public safety occupations, which generally pay much better than other service occupations) make them poor candidates for explaining overall wage trends. The service occupations remain a relatively small share of total employment. In 2007, for example, these occupations (excluding workers in public safety occupations, which generally pay much better) accounted for less than 13 percent of total employment, and just over half of employment in the bottom quintile of the occupations ranked by wages. Moreover, these occupations have expanded only modestly in recent decades, increasing their employment share by 2.1 percentage points between 1979 and 2007, with most of the gain in the 2000s. The small size and the slow, relatively steady, growth of the service occupations suggest significant limitations on the ability of a technology-driven expansion of service occupations to explain the large and contradictory changes in wage growth at the bottom of the distribution (i.e., the 50/10 wage differential) let alone movements at the middle or higher up the wage distribution.

Second, the timing of the most important changes in employment shares and wage levels in the service occupations is not compatible with conventional interpretations of the tasks framework. Seen over a long period—say, 1980 to 2005, as in Autor and Dorn (2012)—both employment and wages in the service occupations increased, which Autor and Dorn interpret as suggesting that increasing demand for service occupations drove up, first, employment and, then, wages in these occupations. Essentially all of the wage growth in the service occupations over this period, however, occurred in the second half of the 1990s, when the employment share in these occupations was flat. The observed wage increases preceded almost all of the total growth in service occupations over the period, which took place in the 2000s, when wages in service occupations were falling, in clear contradiction to the usual interpretation of the tasks framework.

A much more general point can and should be drawn from our findings: occupational employment trends, by themselves, provide only limited insights into the main dynamics of the labor market, particularly wage trends. Recent research has used occupational employment trends to discern the extent of structural unemployment and changes in workplace technology and the consequent impact on wage inequality. The recent wage inequality literature has highlighted the role of changes in occupational distribution of employment as drivers, and key indicators, of labor market trends. In particular, occupational employment trends have become increasingly used as indicators of changes in job skill requirements, reflecting the outcome of changes in the nature of jobs and the way we produce goods and services. Our findings indicate, however, that occupational employment trends give only limited insight and leave little imprint on the evolution of the occupational wage structure, let alone the overall wage structure. We therefore urge caution in drawing strong conclusions about the labor market based on occupational employment trends alone.

Structure of the paper

The next section of the paper reviews key elements of the tasks framework, including the basis for its compelling critique of the “canonical” SBTC model. Section 3 provides an overview of long-term trends in occupational employment and wage inequality, showing that job polarization, broadly defined, is a long-standing feature of the U.S. labor market, while rising wage inequality is strictly a phenomenon of the last three decades or so. Section 4 takes a closer look at changing patterns of occupational employment, occupational wages, and overall wages. We use the CPS data both to provide an independent test of earlier results based primarily on the decennial census and the American Community Survey and to take advantage of the annual nature of the CPS in order to examine more closely the timing and magnitude of relative occupational employment changes across occupations ranked by their 1979 level of wages. Section 5 continues the analysis in the preceding section, focusing on the chain of causality that runs from changes in occupational employment, through changes in occupational wages, and finally to changes in the overall wage distribution. Section 6 focuses on the role of service occupations and low-wage occupations in job polarization and wage trends. Section 7 concludes.

II. Theory, claims, and data

In two recent important papers, Acemoglu and Autor (2011, 2012) raise significant concerns about the ability of the “canonical model” of skill-biased technical change (SBTC) to explain rising wage inequality. The “canonical model” argues that the main cause of the increase in wage dispersion since the late 1970s is the rise in skill-biased production technologies that place a premium on the skills of more-educated and better-paid workers at the expense of less-educated and lower-paid workers. Acemoglu and Autor propose a “tasks framework” that subsumes the “canonical model” as a special case and claims to explain key wage patterns and rising wage inequality over the last three decades. This section reviews their critique of the “canonical model,” and lays out the basics and key empirical claims of their tasks framework.

A. The ‘canonical model’

For the last two decades, the standard explanation for rising wage inequality has relied heavily on a fairly simple model of the interplay of supply and demand. Daron Acemoglu and David Autor (2012) have succinctly summarized the main workings of the “canonical model” as follows: “In this model, technological progress raises the demand for skill and human capital investments slake that demand. When demand moves outward faster than does the supply of human capital, inequality rises, and vice versa when supply outpaces demand” (p. 428).

Claudia Goldin and Lawrence Katz build on the prior literature and offer a comprehensive theoretical and empirical description of the “canonical model” in The Race between Education and Technology (2010). In the “canonical model,” the primary driver of labor demand is skill-biased technological change (SBTC) in the production of goods and services, which leads employers to hire more high-skilled workers whose skills complement new production technologies (including, especially, computers). In most empirical implementations of the model, the key indicator of the supply of skills is the share of college-educated workers in the labor force. As Acemoglu and Autor (2011, 2012) stress, the “canonical model” features two types of workers: college graduates—or, more precisely “college equivalents,” which includes all those with a college degree or further degree and half of those with “some college”—and non-college graduates (not-college equivalents). The level of computerization is the most common empirical measure on the demand side, but SBTC is often simply inferred from a time trend or the pattern of employment and wages (a key figure is in Acemoglu and Autor (2011), reproduced here as Figure A, see also Katz and Murphy, 1992).3

Almost since its inception, the “canonical model” has been subject to two distinct critiques. The first, in chronological terms, questioned the connection between technological change and inequality. As Lawrence Mishel and Jared Bernstein (1994, 1998) argued, technological change has been a constant feature of the economy throughout the entire 20th century, with no obvious associated increase in wage or income inequality for much of that period. David Howell (1994) noted that the timing of the microcomputer revolution, which was widely believed to be a key source of the skill bias, was also off: the largest increase in wage inequality took place in the few years between 1979 and 1982, well before personal computers, let alone the Internet, had transformed workplaces. Consistent with this critique, the pace of growth in inequality declined even as computerization spread steadily in the late 1980s and 1990s (Mishel, Bernstein, and Schmitt 1997; Card and DiNardo 2002). Card and DiNardo (2002) expanded the critique, noting  “puzzles and problems for the [SBTC] theory in nearly every dimension of the wage structure” including the failure to explain important dimensions of wage inequality by gender, race, and age.

One common feature of the Mishel, Bernstein and Schmitt (1997) and the Card and DiNardo (2002) critiques was the failure of the 50/10 wage differential to expand after the late 1980s, and the subsequent fall of the 50/10 differential in the 1990s. These movements in the bottom half of the wage distribution were inconsistent with the canonical claim that the rising price of skills (reflected in wage levels) in all its dimensions was driving wage inequality, since it was clear that low-wage workers were seeing wage gains as large or larger than middle-wage workers.

The second, and later, critique of the “canonical model” (Autor, Levy, and Murnane 2003; Autor, Katz, and Kearney 2006; Acemoglu and Autor 2011, 2012) did not seek to replace the “canonical model” so much as to use it as the foundation for a more general skill-biased technological change framework, which essentially subsumed the canonical approach.4 Acemoglu and Autor (2011, 2012), who provide the most articulate and comprehensive discussion, identify three key shortcomings of the standard account and offer a formal, task-based, rather than skills-based, model that is explicitly an extension of the “canonical model.”5

The first shortcoming identified by Autor, Katz, and Kearney (2006),6 and later by Acemoglu and Autor, is that even though the simple supply-and-demand framework can account well, in their view, for the rising wage differential between college-educated and high school–educated workers from the 1970s through the early 1990s, the same model substantially overstates the rise in the college premium thereafter (as shown by Figure B, reproduced from Acemoglu and Autor 2012).The deceleration in the college premium, they note, would suggest that the relative demand for high-skilled workers decelerated after about 1992 which, in their judgment “does not accord with common intuitions regarding the nature or pace of technological changes occurring in this era” (p. 437).

The second shortcoming noted by Acemoglu and Autor is the “canonical model’s” inability to explain the divergence in the pattern of wage behavior of inequality between the top and bottom halves of the wage distribution in the early 1990s relative to the pattern of the 1980s.7 In the 1980s, wage growth was a smoothly increasing function of the initial wage level. From the early 1990s (or slightly earlier) onward, however, inequality continued to grow in the top half of the distribution (the 90/50 differential widened), but inequality was flat or falling in the bottom of the distribution (the 50/10 differential declined somewhat).8 Since the “canonical model” is usually framed around two kinds of workers—more- and less-skilled workers, often operationalized as college- and non-college-educated workers—these non-monotonic movements in the wage distribution in the 1990s are difficult to explain in that framework.

The final shortcoming noted by Acemoglu and Autor is the “canonical model’s” inability to explain how technological progress could produce stagnant and, especially, declining real wages for middle- and less-skilled workers over long periods, even as technology was increasing average productivity.9

All three of these critiques were made by Mishel, Bernstein, and Schmitt (1997) 10 years earlier:

…the experience since the mid- to late- 1980s does not accord with a technology explanation, whose imagery is of computer-driven technology bidding up the wages of “more-skilled” and “more-educated” workers, leaving behind a small group of “unskilled” workers with inadequate skills. The facts are hard to reconcile with the notion that technological change grew as fast or faster in the 1990s than in earlier periods. If technology were adverse for “unskilled” or “less-educated” workers, then we would expect a continued expansion of the wage differential between middle-wage and low-wage workers (the 50/10 differential).

Yet, the 50/10 differential has been stable or declining among both men and women since 1986 or 1987. Instead, we are seeing the top earners pulling away from nearly all other earners. Therefore, there seem to be factors driving a wedge between the top 10 percent and everyone else, rather than a single factor aiding the vast majority but leaving a small group of unskilled workers behind. Further confirmation of the breadth of those left behind is that wages have been stable or in decline for the bottom 80 percent of men and the bottom 70 percent of women over the 1989–95 period, with wages falling for the entire non-college-educated workforce (roughly 75 percent of the workforce). Of course, even high-wage, white-collar, or college-graduate men have failed to see real wage growth in 10 years.

The flattening of the growth of education differentials in the late 1980s and 1990s among men also does not easily fit a technology story. Since the wages of college-graduate men are not being “bid up” relative to others at the same pace as in the early and mid-1980s, one can only conclude that there has been a deceleration of the relative demand for education (given that the supply of college workers did not accelerate).

We also note that Autor and Acemoglu do not consider several additional critiques of the “canonical model,” particularly: (1) the failure of education wage differentials to capture much, if not the majority, of the growth of wage inequality, which happened among workers with similar education and experience (so called “within-group” wage inequality); (2) the failure to explain the extraordinary rise of wages among the top 1 percent of earners; and (3) the possibility that observed education wage gaps could be driven by factors other than changes in the price of skills, including changes in the minimum wage and unionization, industry deregulation, and globalization, all of which could affect relative wages by education but have nothing to do with technological change.

B. The tasks framework

To address these three shortcomings,10 Acemoglu and Autor (2012, Section 4) construct a formal model, which builds on the framework originally developed by Autor, Levy, and Murnane (2003), where the fundamental units of the production process are job “tasks,” rather than workers’ skills. Probably the most important feature of the model is that it defines three kinds of tasks: non-routine cognitive tasks (high-skilled), non-routine manual tasks (low-skilled), and routine tasks (middle-skilled, some of which are cognitive and some of which are manual). The second important feature of their model is that it separates tasks from skills. Workers of different skill levels (and different mixes of workers of different skill levels) can perform any of the tasks.11

In this framework the driving force is computerization. Computers have the capacity to compete directly with workers who perform routine tasks, such as those performed by clerical and administrative workers or production workers in manufacturing. Computers, however, are poor substitutes for workers performing non-routine cognitive jobs, such as managers, lawyers, or doctors. In fact, in practice, computers are likely to be strong complements to such workers. Nor are computers effective substitutes for workers performing non-routine manual tasks, including many personal services, such as food preparation and cleaning.

The introduction of tasks and of three kinds of labor (low-, middle-, and high-skilled), however, allows the new model to overcome the main shortcomings of the standard approach identified earlier. Most prominently, by allowing for three types of tasks, their model can potentially explain the divergent inequality trends after the 1980s between the top and middle, on the one hand, and the middle and the bottom, on the other hand—something that the “canonical model,” with only two types of labor, cannot.

1. Key empirical claims

Economists working in this framework have made several empirical claims, which are the focus of the empirical work in the remainder of this paper.

Claim 1: In the 1990s, employment growth was polarized, with the employment share of high-skilled and low-skilled occupations expanding and the employment share of middle-skilled occupations contracting. This pattern marked a stark change relative to the 1980s when, across occupational skill levels, employment grew least at the bottom, more in the middle, and most at the top.

Autor and various co-authors argue that sometime after the late 1980s, employment growth became polarized, with a simultaneous increase in the employment shares of high-skilled and less-skilled occupations coinciding with declining employments shares for middle-skill occupations.12 This pattern was in stark contrast to the 1980s, when employment fell at the bottom of the skills distribution, whether skill was measured by education level or using the occupational wage rankings used in the tasks model. Simply put, the first claim is that job polarization occurred in the 1990s but not in the 1980s, when occupational employment at the bottom contracted, rather than expanded.

For the 1990s, Acemoglu and Autor (2012, Figure 5), Autor, Katz, and Kearney (2008, Figure 11) covering 1990–2000, Autor (2010, Figure 1) covering 1989–1999, and Acemoglu and Autor (2011, Figure 10) covering 1989-1999 find an increase in the employment share for low-skill occupations up to roughly the 10th percentile of the occupational skill distribution.13 These same estimates show contemporaneous increases in the employment share of high-skilled occupations from about the 75th percentile of the occupational skill distribution (though, perhaps about as low as the 65th percentile in the case of Acemoglu and Autor (2011)). Since employment shares must add to 100 percent, increases in employment shares at the bottom and the top imply declining employment shares for occupations in the middle.

Most discussions of job polarization have emphasized the change in employment patterns between the 1980s and the 1990s. Researchers, however, have paid almost no attention to a similarly stark break in employment patterns between the 1990s and the 2000s. For the period 2000–2007, Autor (2010, Figure 1) and Acemoglu and Autor (2011, Figure 10) show employment gains for the bottom 30 percent or so of the occupational skill distribution, but no increase in relative employment for workers above that level in the distribution. These results suggest that by the 2000s, job polarization had ceased to be a factor in the U.S. labor market. This is acknowledged only indirectly, unfortunately, in the literature, but this finding undercuts the claim that occupational employment patterns correspond to key wage patterns, a topic explored below.

Claim 2: This switch in occupational employment patterns in the 1990s caused corresponding shifts in wage patterns. Specifically, occupational employment polarization in the 1990s can explain the shift in the 50/10 wage differential that the “canonical model” fails to explain: after wage inequality at the top and the bottom of the distribution grew symmetrically in the 1980s, wage inequality in the 1990s was distinctly asymmetric, with growing inequality in the top half (90/50 wage gap) of the distribution and declining inequality in the bottom half (50/10 wage gap).14

If the first claim is essentially “job polarization occurred in the 1990s but not in the 1980s” then the second claim is that “the arrival of job polarization in the 1990s explains changes in wage inequality at the bottom of the wage scale.”

The most recent tasks framework theoretical work by Autor and Dorn (2012) indicates that job polarization’s impact on wages is ambiguous though the empirical claim is that job polarization generates wage inequality, in this instance at the bottom of the wage distribution.15

Claim 3: Autor and Dorn (2012) argue that a “key fact” is that “rising employment and wages in service occupations account for a substantial share of aggregate polarization and growth of the lower tail of the U.S. employment and earnings distributions between 1980 and 2005.”

The most recent research in the tasks framework has focused strongly on the role that low-wage “service occupations” have played in the process of both employment polarization and wage polarization and ignores the wage and employment patterns in the top half of the occupational and wage structure. The service occupations (excluding public safety workers, as Autor and Dorn do) are usually low-wage, non-routine manual jobs that accounted for less than 13 percent of total employment in 2007. Typical occupations in this category include food preparation, security guards, and janitorial services, but unless explicitly excluded from the analysis, the formal “service occupations” category also includes police and firefighters, who earn substantially more and are on average much better educated than other service occupations. The emphasis on the service occupations represents a methodological break with much of the earlier research in this area, which focused on much finer occupational categories (typically three-digit occupations) where the skill ordering was determined by the initial average wage in the narrow occupation group.

III. An introductory look at wage inequality and occupational employment trends

The “tasks framework” attempts to explain the patterns of wage inequality over the last 30 years by examining changes in the occupational employment mix. This section provides an introductory look at occupational employment and wage trends with two key goals. The first is to put the post-1979 occupational trends in a longer historical context. The second is to examine the correspondence of occupational shifts since 1979 to the evolution of key wage gaps—particularly the 50/10 and 90/50 wage gaps using the annual data available from the CPS.

A. The long view of occupational employment trends

We begin with a review of occupational employment trends over the postwar period in order to provide historical context. The focus is on the expansion and contraction of high-, middle-, and low-wage occupations (categories defined by Acemoglu and Autor (2011) and used in other papers). Acemoglu and Autor (2011) provide a long look back at the occupational composition of 10 occupations since 1959, which is reproduced in Table 1 and supplemented with: (1) changes in the aggregate categories of low-, middle-, and high- wage occupations; and (2) the percentage-point change in occupational employment shares for each decade and the most recent time period (for the 2000s, we show the period 2000–2007, adjusted to a 10-year rate of change). The Acemoglu and Autor data are derived from the decennial census data except for the latest year, which is based on the American Community Survey. The trend in these aggregate occupational shares are shown in Figures C-A and C-B. These figures also include comparable annual trends for the 1979–2007 period based on our tabulations of the Current Population Survey.16

Acemoglu (2010), in a newspaper column, summarized his and his co-author’s view of the historical record:

U.S. employment and demand for labour have been undergoing profound changes over the last 30 years. While the demand for high skill workers, who can perform complex, often non-production tasks, has increased, manufacturing jobs and other “middling occupations” have been in decline. Also noteworthy is that over the last 10-15 years, many relatively low-skill, low-pay service occupations have been expanding rapidly.

Table 1 and Figure C-A show that middle-wage occupations declined at a comparable pace in each of the last three periods dating back to 1979, eroding by roughly 4.5 percentage points per decade. A critically important observation, however, is that such “middling occupations” have been declining throughout the entire period covered by their data, 1959–2007, which includes periods when wage inequality was stable as well as ones when wage inequality was growing. The decline in the “middle” that is highlighted in the discussion of job polarization is not unique at all to the 1990s. Moreover, the blue-collar occupations, “Production, craft and repair” and “Operators, fabricators and laborers,” eroded more quickly in the 1960s, 1970s, and 1980s than in the 1990s or 2000s, so any notion that job polarization represents some new technological turn against blue-collar middle-wage jobs is not accurate.

We also note that discussions of occupational employment trends in the job polarization literature consistently assume that these observed occupational shifts are solely a reflection of technological change. In fact, this assumption is clearly false since globalization trends and, in particular, the rise of trade deficits over the last three decades, has also changed the occupational composition of employment. We flag the likely important contribution of trade to the occupational employment mix. We hope to address the impact of trade on these occupational employment trends in future work.

Table 1 and Figure C-B also show that high- wage occupations have expanded over the entire 1959–2007 period, though faster in the 1980s and 1990s. An important development is that the growth of high-wage occupational employment in the 2000s was slower than it had been in the 1990s and grew comparably to the slow trend in the 1960s. Analysis of occupational employment growth by wage percentile, presented by Autor (2010), shows that employment in high-wage occupations grew no faster than employment in middle-wage occupations in the 2000s. This dramatic slowdown in demand for workers in high-wage occupations in the 2000s relative to demand in the 1980s and 1990s has important implications for the analysis of wage trends. The slowdown in the growth in top occupations in the 2000s was due to the failure of “technician” employment to expand (which also occurred in the 1990s), a remarkable slowdown in management occupations (up only 0.4 percentage points, about a fifth as fast as in the 1980s and 1990s), and a deceleration in the growth of professional employment. This overall slowdown in high-wage occupations in the 2000s is consistent with the slower growth of the 90/50 wage gap. The slower expansion of high-wage employment is consistent with recent technological change generating more modest growth in the demand for college graduates and could help explain the flattening of the growth of college wages and the college wage premium.

Given the continuing trends in both the middle- and high-wage occupations, an explanation of a changed pattern in the 50/10 wage gap rests heavily on the expansion of low-wage occupations in the 1990s, relative to the 1980s: Such an expansion can explain the differing pattern of low-wage (10th percentile) and 50/10 trends in each decade as demand for low-wage workers expanded in the 1990s relative to the 1980s, with the result that low-wage workers fared better and the 50/10 wage gap stopped expanding and even fell in the 1990s. As noted, low-wage occupations are equated to the aggregate of the three service occupations. However, service occupations represent only about half of the occupational employment of the bottom fifth (in terms of the lowest-paying occupations) of occupational employment. Section 6 provides a more in depth analysis of employment trends in service occupations and in low-wage occupations. For our purposes here our analysis focuses only on the aggregate of service occupations.

As shown in Table 1 and Figure C-B, service occupations were relatively stable in size over the 1970s and 1980s, then expanded modestly in the 1990s (up 1.0 percentage point over the full decade), and then grew far faster in the 2000s (up 3.0 percentage points at a 10-year rate).

Acemoglu and Autor (2011), citing Autor and Dorn (2010), suggest a sharper increase in the 1990s than seen in Table 1, indicating “the employment share of service occupations was essentially flat between 1959 and 1979. Thus, their rapid growth since 1980, marks a sharp trend reversal [emphasis added].” More importantly, they note:

Indeed, Autor and Dorn (2010) show that rising service occupation employment accounts almost entirely for the upward twist of the lower tail … during the 1990s and 2000s. All three broad categories of service occupations, protective service, food preparation and cleaning services, and personal care, expanded by double digits in the both the 1990s and the pre-recession years of the past decade (1999-2007). Protective service and food preparation and cleaning occupations expanded even more rapidly during the 1980s.

Our reading of these initial data suggests that the scale and timing of the changes in service occupations does not correspond to Acemoglu and Autor’s interpretation. The key empirical issue is whether trends in service occupations are consistent with the view that there was a substantial expansion of demand for low-wage workers in the 1990s relative to the 1980s as indicated by trends in service occupations. Such an expansion would need to have occurred in the late 1980s (the specific timing depends on whether the analysis is of all workers, men, or women).

The appropriate metric, in our view, is the change in employment shares rather than (percent) growth in employment because the issue is change in relative demand. In our reading of Acemoglu and Autor’s data, displayed in Table 1, there was growth in service occupations in the 1990s (up 1.0 percentage point) that exceeded that of the 1980s (up just 0.3 percentage points). On first glance, however, this 1990s expansion does not seem sufficient to explain why a major increase of the 50/10 wage gap in the 1980s failed to continue in the 1990s. This is especially the case since measurement issues—the occupation coding change that occurs between the 1990 and 2000 decennial censuses— artificially expands service occupation employment in the 1990s in the Acemoglu and Autor series. This is examined in greater detail in a later section and in the data appendix.

The largest change in employment in service occupations appears to have taken place in the 2000s, when the service-occupation employment share increased at a 3.0-percentage-point 10-year rate. These observed employment (share) trends in the service occupations raise some important questions. One is whether a modest expansion of the low-wage occupation expansion in the 1990s relative to the 1980s is sufficient to explain a sharp change in wage patterns at the bottom. A second question is whether the timing of the change in service occupation employment occurred coincident with the change in wage patterns, i.e., in the late 1980s or early 1990s. An even bigger question: If one believes the modest 1990s change in service occupation employment was sufficient to dramatically alter wage patterns in the 1990s, then why didn’t a doubling of that rate of expansion in low-wage occupations in the 2000s lead to an acceleration of wage growth at the bottom and a sharp contraction in the 50/10 wage gap from 2000 to 2007? This is the first of many instances where the story which is said to explain the 1990s is clearly inconsistent with the trends in the 2000s (which is not adequately acknowledged, in our view, in the tasks framework/job polarization literature).

We have also examined the historical pattern of occupational employment shifts presented in two other papers. Autor and Dorn (2012, Table 1) present data for 1950, 1970, 1980, 1990, 2000, and 2005 based on shares of total hours worked and including farming (excluded from the Acemoglu and Autor (2011) data). The Autor and Dorn data use the same grouping of occupations into high-, middle- and low-wage as Acemoglu and Autor and confirm that the shrinkage of middle-wage occupational employment and the expansion of high-wage occupational employment also occurred in the 1950-70 period, though at a slower pace than in later periods.

Katz and Margo (2013) provide an even longer series, going back to 1920, and include agricultural occupations since they were very significant up through 1960. Katz and Margo, however, categorize occupations differently, expanding the “low” category beyond service occupations to also include blue-collar operatives and laborers. This more than doubles the size of the low-wage group even near the end of the period: In 2000, low-wage occupations comprised 27.5 percent of total non-agriculture employment while service occupations alone (the Acemoglu and Autor grouping for low-wage) were just 13.0 percent. Katz and Margo do not explain the basis for their categorization and it does seem odd to us to put more highly paid blue-collar workers (operatives and laborers) into the low-skill category while leaving lower-paid clerical/administrative workers in the middle-skill group. Katz and Margo’s share of non-agriculture employment in the low-skill group declined in the 1950s, 1960s, 1970s, 1980s, and 1990s and increased in the 2000s. Middle-skill occupational employment declined and high-skill occupational employment expanded in every decade starting with the 1950s. Thus, the Katz and Margo data show that occupational upgrading was a constant feature of the postwar economy up through 2000, with low- and middle-skill occupational employment shrinking and high-skill occupational employment expanding. Moreover, Katz and Margo’s data show no job polarization in the 1990s as low-skill occupational employment shrank absolutely and more than middle-skill occupational employment shrank.17

This ongoing shrinking of middle-wage occupations and expansion of high-wage occupations over the last 60 years represents the occupational upgrading associated with technological change (but also the impact of trade impact in the last few decades). This upgrading process should not be surprising to labor economists and has long been noted (Spenner 1988; Mishel and Bernstein 1994, Mishel, Bernstein, and Schmitt 1997; Handel 2005; and Howell and Wolff 1991).

Having now analyzed the postwar historical trajectory of occupational employment changes we can now characterize ways in which technology has affected the labor market (shaping the occupational employment structure) and ways it has not affected the labor market (causing wage inequality). Some analysts have mistakenly asserted that if an analysis suggests that technological change has not greatly affected wage inequality that this is equivalent to saying that technology has had no impact on the labor market. We have just documented a longstanding trend of occupational upgrading—more white-collar and less blue- and pink-collar work—for many decades. These changes in the occupational structure are primarily technology-driven and have increased the skills and education employers seek in the labor market which, in turn, necessitates an educational upgrading of the workforce. This is what Claudia Goldin and Lawrence Katz refer to as the “race between technology and skills.” We believe there has been such a race, that technology has had a major effect, but also that the education and skills have greatly improved and satisfied that increased demand.

There has also been an increase in so-called “skill premiums,” such as the college wage premium. We view that increase as reflecting other factors such as deregulation of industries, globalization, an eroded minimum wage, excessive unemployment, and declining unionization rather than the product of technology-driven skill shortages. That is, in the race between skills and technology since 1979 there has been roughly a tie. This has especially been the case since 1995, after which the college premium has barely grown. Moreover, as shown in Mishel et al. (2012), over the last 10 years, real wages have been flat or falling for a majority of college graduates, including those in nearly every occupational group (e.g., business occupations). In these circumstances, where technology and skills have run neck and neck, technology has had a large impact on the labor market but it has not generated wage inequality. We have faced a “wage deficit” rather than a “skills deficit,” meaning that jobs at every education and skill level have not seen appropriate wage growth. This is evident in the failure of wages of both high school– and college-educated workers to keep pace with productivity, and in the extraordinary share of profits in the economy, especially in the 2000s.

B. Occupational employment shifts and wage gap shifts

This section continues the analysis of occupational employment shifts and focuses on their correspondence over time with key wage gap trends. The starkest shift in wage patterns was for the 50/10 wage gap, which substantially expanded in the 1980s, stopped expanding in the late 1980s and then stabilized and contracted somewhat in the 1990s. This pattern indicates that in the 1990s, wage growth for low-wage workers was as good or better than it was for middle-wage workers in the same period—exactly the reverse of what occurred in the 1980s. The shift in wage patterns was most stark among women: The 50/10 wage gap grew from 1.47 in 1979 to 1.91 in 1988, a gap generated by a 17 percent decline at the 10th percentile and an 8 percent gain at the median. The 50/10 wage gap grew far less among men (from 2.04 in 1979 to 2.28 in 1986) because wages fell less at the 10th percentile and wages also fell at the median. The second wage gap, the 90/50 gap, continued to grow throughout the 1980s and the 1990s but rose more slowly among men after 1993 and among women after 1994.

Tables 2-A through 2-C and Figures D-A through D-C employ annual data from the CPS to portray the timing of occupational employment and wage-gap shifts. The 1973–2010 annual trend in the key occupational employment shares (hours-weighted) and the corresponding wage gap are presented in Figure D-A for all workers, and separately for men and women (Figures D-B and D-C). These figures allow us to see whether the shifts in occupational employment (e.g., changes in the size of service occupations) correspond to the shifting pattern of wages (e.g., the flattening in the 50/10 wage gap after the late 1980s). The 50/10 wage gap is displayed along with the employment shares in middle- and low- wage occupations and, likewise, the 90/50 wage gap is displayed alongside the employment shares in high- and middle- wage occupations. Table 2A through 2C use the same data to present the trends in particular time periods: business cycle periods such as 1979–1989, 1989–2000 and 2000–2007 plus the two distinct subperiods of the 1990s; 1989–1995 and 1995–2000.

Two metrics for assessing the occupational shifts are used. The first is the simplest, the annual percentage-point change in an occupation’s employment share in each period. The percentage-point change, however, is not scaled at all to the size of the occupation group and since the occupation groups vary tremendously in size—the middle group represented roughly 60 percent of employment in 1979 while service occupation represented only about 14 percent in the same year. Given differing sizes, the same percentage-point change means a larger expansion or contraction for service occupations than it would for middle-wage occupations. The second metric, therefore, divides the percentage-point change in a period by the starting share and thereby reflects the degree to which that occupation expanded or contracted.

Using the CPS-ORG for both the occupational employment trends and the wage trends has several advantages. One is that the CPS occupational employment trends have not yet been used in an analysis of job polarization, so they provide new information and an additional test of the robustness of the finding of job polarization.18 Second, using the CPS allows us to examine annual trends and to test whether shifts in employment patterns are coincident with shifts in the 50/10 and 90/50 wage gaps. A third reason to use annual CPS data is that we can make an adjustment for the coding changes that occurred in 1983 and 2003. The annual data series we employ eliminates the change in occupational shares in the year of the coding changes, 1982–1983 and 2002–2003 by substituting the average of the change in share in the preceding and subsequent two years.19 Last, the CPS-ORG is the acknowledged best data source for examining hourly wages.

Figures D-A through D-C also display the trends in low-, middle-, and high- wage occupational employment shares. Employment shares for all three groups move smoothly over the entire 1973–2010 period. Meanwhile, wage patterns show abrupt changes, particularly the sharp growth in the 1980s in the 50/10 wage gap and the subsequent shrinkage (among all and men) or flattening (among women) thereafter.

The prima facie evidence, therefore, does not support the claim that occupational employment polarization emerged in the late 1980s and early 1990s and caused a change in wage patterns. For instance, the trend in the employment share of service occupations, which Autor and Dorn (2010) and Acemoglu and Autor (2011) claim drives the 50/10 wage gap, has been smoothly changing over nearly 40 years. Similarly, the shrinkage of middle-wage occupations has been steady as well. Steady trends in occupational employment composition cannot cause discontinuous shifts in key wage gaps.

Tables 2A through 2C provide the detailed data that confirm the lack of correspondence between low-wage (service) occupation share trends and trends in the 50/10 wage gap. Our reading of the tasks framework literature is that it postulates that the expansion of service occupations should lead to a falling 50/10 wage gap as the demand for low-wage workers expands relative to workers in the middle. In the 1980s, service occupations expanded employment overall and among men, but, contrary to the expectations built into the tasks framework literature, the 50/10 wage gap expanded. Nor was there any shift to a faster expansion of service occupations overall or among men in the early 1990s (1989–1995), when the 50/10 wage gap flattened among men and declined overall. Thus, shifts in occupational employment shares for service occupations do not appear to have any influence on the development of the 50/10 wage differential, raising doubts about a key claim of the task s framework literature. Our analysis of the broader low-wage occupational group presented below affirms this finding.

The corresponding data for women are also inconsistent with the standard tasks framework. In the 1980s, service occupation employment shares for women fell modestly (-.03 percentage points per year), which does not tightly correspond to a stark widening of the 50/10 wage gap. Service occupation employment shares fell at a faster pace in the first half of the 1990s (-0.09 percentage points per year) but the 50/10 gap declined (0.56 log points per year). The abrupt change in the trend of the 50/10 differential for women was the largest change in wage patterns between the 1980s and 1990s. Yet, low-wage service occupation employment trends among women cannot explain this key shift in wage patterns among women, a major failure of the tasks framework.

The trends in the wage gap at the top, the 90/50 wage gap, are not tightly linked to shifts in occupational employment shares either. Among men, the 90/50 wage gap grew more strongly in the early 1990s than in the 1980s even though the high-wage occupation growth slowed. The 90/50 wage gap grew far more slowly in the late 1990s (0.67 log points per year) though the growth of high-wage occupations did not slow at all. In the 2000s the growth of high-wage occupations among men was minimal (0.05 percentage points per year) but the 90/50 wage gap continued to grow (0.51 log points per year) though at a reduced pace compared with the 1980s  or 1990s.

Among women, there was a clear deceleration in the growth of the 90/50 wage gap in the late 1990s and 2000s, but it is difficult to see how the trajectory of high-wage occupations could explain this pattern. The growth of high-wage occupations was slowed somewhat between the first and second halves of the 1990s (falling from 0.68 to 0.59 percentage points per year) while the 90/50 wage gap grew only a third as fast in the second half of the 1990s as it had in the first half of the decade. In the 2000s, the 90/50 wage gap grew a bit faster than in the second half of the 1990s, but high-wage occupations grew more slowly. The decline in middle-wage occupations among women was similar in the 1980s and the entire 1990s, which does not help resolve this puzzle. The faster shrinkage of middle-wage occupations in the 2000s, however, can help explain the pickup in 90/50 wag gap growth.

IV. What does the CPS tell us about occupational employment trends?

In this section, we first use data from the Current Population Survey Outgoing Rotations Group (CPS-ORG) to replicate a key figure in the job polarization literature—a figure that was created using data from the decennial census and the American Community Survey— Figure 10 in Acemoglu and Autor (2011) (which is also Figure 1 in Autor (2010)). Though the CPS-ORG has smaller sample sizes, its key advantages are that it has a better hourly wage measure, and data are available for each year. This exercise is also useful simply to see how robust this key finding in the job polarization literature is to changing the data source. Our treatment of the CPS-ORG data is described in the Appendix.

In this discussion, we find it helpful to distinguish between what we call “absolute” job polarization and “relative” job polarization. Absolute polarization is both what the job polarization literature conveys and what is most commonly referred to in popular discussions of this topic: employment share growth at both the top and bottom of the occupational distribution, with losses in the middle. Relative polarization refers to growth across the occupational distribution that has a U-shape, whether or not both ends see absolute growth in shares. So, for example, employment share growth across the occupational distribution where the middle loses substantial employment share, the bottom loses employment share but not as much as the middle, and the top gains employment share would represent relative but not absolute polarization. Of course, relative polarization of employment growth, even without absolute polarization of employment growth, would be expected to generate a polarization of relative wage trends.

Another important issue is the impact of changes in occupation coding over time. As described in the Appendix, we employ the same occupation crosswalk used in, among other papers on this topic, Acemoglu and Autor (2011), to attempt to obtain a consistent series across several coding changes in the CPS over the 1979–2007 period. As shown in Appendix Figure A, using this crosswalk, we are able to almost exactly replicate the underlying major occupational employment shares in each year used in Figure 12 of Acemoglu and Autor (2011).20 There are 10 occupations and this graph shows trends for each of these 10 occupations from both our tabulations and Acemoglu and Autor’s. The fact that only 10 lines are observable means that the lines based on our tabulations almost exactly duplicate their tabulations. An examination of Figure A reveals that the major changes in occupation coding between 1982 and 1983 and between 2002 and 2003 are difficult to bridge. Even with the careful crosswalk used in Acemoglu and Autor (2011), there are large, visible discontinuities in employment shares between 1982 and 1983 and between 2002 and 2003.21 The impact of the coding breaks is nontrivial. We find that the coding break between 1982 and 1983 masks the decline in middle-wage jobs in the 1980s, which is a key reason that, as discussed below, we find job polarization in the 1980s while Acemoglu and Autor (2011) do not (since we remove the effect of the coding break and they do not). We also find that the coding break between 2002 and 2003 leads to a substantial overstatement of low-wage job growth in the 2000–2007 period, and that when we remove the effect of the 2002/2003 coding break there is much more modest growth in low-wage jobs (when occupations are ranked according to their 1979 mean wage) in the 2000–2007 period. Our method for removing the effect of the coding breaks is to simply replace the change in employment share over the break years (1982–1983 or 2002–2003) with the average change of the two years on either side of each break. For more detailed information about the impact of the coding break, see the Appendix.

Figure E shows our replication of Figure 10 in Acemoglu and Autor (2011) (which is also Figure 1 in Autor (2010)), using data that are adjusted for the coding breaks using the simple adjustment procedure described above and in the Appendix. Figure E fairly closely replicates the Acemoglu and Autor (2011) figure, with some notable differences. While Acemoglu and Autor (2011) find monotonic increases in employment across occupational wage percentiles in the 1980s, we find relative (though not absolute) job polarization, with less job loss for the bottom of the occupational ranking than the middle during this period. The finding of relative job polarization in the 1980s is inconsistent with the Acemoglu and Autor claim that there was a stark shift in occupational employment patterns between the 1980s and the 1990s that can explain the shift in the 50/10 wage gap at the end of the 1980s (specifically, the observation that the 50/10 wage differential substantially expanded in the 1980s, stopped expanding in the late 1980s, and then contracted in the 1990s). For the 1990s, we replicate the Acemoglu and Autor (2011) job polarization finding, though here too we find no absolute job polarization, only relative polarization, with less job loss for the bottom of the occupational ranking than the middle during this period.22 Between 2000 and 2007, Acemoglu and Autor (2011) find much more employment growth at the bottom of the occupational employment distribution than we do using CPS data; of the bottom half of the distribution we find that only the first five percentiles saw any employment share growth.

Most discussions of job polarization have emphasized the change in employment patterns between the 1980s and the 1990s. Researchers, however, have paid almost no attention to the stark break in employment patterns between the 1990s and the 2000s. For the period 2000–2007, Autor (2010, Figure 1) and Acemoglu and Autor (2011, Figure 10) show employment gains for the bottom 30 percent or so of the occupational skill distribution, but no increase in relative employment for workers above that level in the distribution. Our CPS results confirm that there was little or no employment expansion of occupations in the upper half of the wage scale. These results suggest that by the 2000s, job polarization had ceased to be a factor in the U.S. labor market. This is acknowledged only indirectly, unfortunately, in the literature.23

These descriptions fail to acknowledge how different the trends in the 2000s are from those of the 1990s and how those differences are at variance with the tasks framework interpretation of wage inequality trends. First, the failure to find a sizeable difference in the employment share growth of middle- and high- wage occupations in the 2000s means that one key dimension of job polarization that was present in the 1990s (and 1980s) is absent in the 2000s. This has the important implication that occupational employment trends in the 2000s cannot explain continued growth in the 90/50 wage gap. Second, the faster (than in the 1990s) expansion of low-wage occupational employment share in the 2000s generates an additional puzzle for the job polarization interpretation of wage inequality: a smaller expansion of low-wage occupational employment in the early 1990s is said to have narrowed the 50/10 wage gap over that decade, yet, a larger expansion of low-wage occupations in the 2000s generates no contraction in the 50/10 differential in the 2000s.

An important and unspoken implication is that “job polarization,” which features in many popular and policy discussions of the contemporary economy, is in fact, at best, only a theory of developments through the 1990s and does not contribute to our understanding of more recent trends.

A. The smoothing typically used in this context masks substantial variation

The key data presented to demonstrate occupational employment trends across wage percentiles (i.e., such as job polarization in the 1990s) rely on a locally weighted smoothing regression. Such analyses are useful, but unfortunately are not typically presented with an assessment of goodness of fit. In fact, the smoothed lines in the standard presentations of employment polarization mask substantial variation of occupational employment growth across the entire distribution of occupations. What is presented as a reliable picture of employment patterns obscures substantial variation in the underlying data (Lefter and Sand, 2011).

The smoothed lines in Figures F-A through F-C are the exact lines from Figure E, now displayed on a y-axis with a scale wide enough to incorporate the unsmoothed log employment share changes at each occupational percentile. We also report the “lowess” R-squared of the smoothed line to provide a measure of goodness-of-fit.24 In every decade, the smoothed lines mask a great deal of underlying variability in the data. The lowess R-squared from the 1979–1989 period is the largest of the three periods, at 0.213. The lowess R-squares declines with each subsequent period, to 0.174 in the 1989–2000 period, and to 0.039 in the 2000–2007 period. In other words, for any of the three time periods, but most pronounced in the 2000s, the variation of employment shifts across detailed occupations is not well captured by the smoothed lines presented in the job polarization literature.25 Knowing an occupation’s initial wage level does not provide much information about employment growth in that occupation or about employment changes in nearby occupations. Much of the literature focuses on how the shapes of these smoothed lines shift from period to period, most prominently from “monotonic” in the 1980s to “polarized” in the 1990s. Such analyses are implicitly differencing lines which are not well estimated, suggesting that any claims about differences are subject to a substantial margin of error.

As mentioned, our preferred use of the data in this context is to adjust for the major occupational coding breaks in 1982/1983 and 2002/2003. However, since the empirical work in this literature typically does not make these adjustments, here we also present the lowess R-squared for the unadjusted data. The lowess R-squared for the unadjusted 1979–1989 data, at .043, is substantially lower than for the adjusted data (which was 0.213). The lowess R-squared for the unadjusted 2000–2007 data, also at .043, is slightly higher than for the adjusted data (which was 0.039). There was no major coding break (and therefore no adjustments) in the 1990s; the lowess R-squared for the 1989–2000 period is 0.174. We also note that whether we use the adjusted or the unadjusted data, the ability of initial occupational wage levels to predict employment changes declines substantially from the 1990s to the 2000s. This appears to contradict the claim in Acemoglu and Autor (2011) that the explanatory power of occupations has increased since the 1980s.26

V. Occupational employment shifts and wage determination

We now leave behind issues regarding the shape of occupational employment patterns and ask how occupational employment trends have affected wages, assessing Claim #2 above. As noted earlier, the literature has not offered direct evidence of how occupational shifts shape wage patterns: rather, the only evidence presented has been circumstantial, merely showing that when job polarization occurred in the 1990s there was also a polarization of wages (with the 90/50 gap expanding and the 50/10 gap closing).

We examine the relationship between occupational employment shifts and wages by first examining the main channel through which occupational shifts can be expected to affect wages: occupational employment shifts affect the wage patterns across occupations which, in turn, drive overall wage patterns. We find only weak empirical links between occupational employment changes, occupational wage changes, and changes in the overall wage distribution. We then move from observing the weak underlying relationship between wage changes at various parts of the wage distribution and changes in employment in occupations at different “skill” percentiles to an assessment in a regression framework of how much the variance of wages can be explained by occupations. We find that the importance of occupations in explaining wage variance slowed in the 1990s and reversed somewhat in the 2000s, and that a large and increasing share of the increase in wage inequality in recent decades is occurring within occupations, a phenomenon for which the tasks framework has no explanation.

A. Occupational employment shifts, occupational wage differentials and overall wage differentials and wage determination

In the tasks framework, first, technology changes occupational employment shares by changing demand for workers by occupation, which in turn drives changes in occupational wages, and finally, changes in occupational wages drive changes in the overall wage distribution. The poor fit of the smoothed occupational employment lines already seen in Figures F-A through F-C suggest a weak link between technology and the occupational employment structure. Figures G-A through G-C, however, take those occupational employment changes as real and examine their possible impact on occupational wages and then take the changes in occupational wages as given, and examine their possible impact on overall wages.

The poor links between changes in occupational employment, occupational wages, and overall wages are evident in Figure G-B, which presents data for the 1990s, the period when the occupational employment patterns best fit the tasks framework predictions. The blue, roughly U-shaped, line is the same as the fitted occupational employment line in Figure E and Figure F-B. The black line shows the corresponding (smoothed) change in occupational wages—using the same occupations in each percentile as the employment share line but portraying the percent change in mean log real hourly wages for those occupations. Occupational wages grew fastest at the bottom (through about the 30th percentile) and grew at the same—slower—rate through the rest of the occupational wage distribution. Rising occupational employment at the bottom could plausibly have driven rising occupational wages at the bottom, but rapidly rising occupational employment shares did not lead to higher occupational wage growth among higher-skilled occupations. The red line in the figure traces the (smoothed) wage change at the percentiles of the overall wage distribution. Again, the rise in occupational wages at the bottom of the occupational skill distribution could plausibly explain the more rapid rise at the bottom of the overall wage distribution. But, for the top half, growth in the occupational wage distribution was inconsistent with growth in the overall wage distribution. Wage growth rose steadily from about the median of the overall wage distribution, but was basically flat over the corresponding range of the occupational wage distribution. In fact, the biggest increases in overall wages were at the very top of the distribution, while the very top of the occupational wage distribution experienced among the smallest wage increases across all occupations.27 Based on the data for the 1990s, then, occupational employment changes in the upper half appear to be poor predictors of the corresponding occupational wage changes, which are, in turn, poor predictors of the corresponding changes in the upper half of the overall wage distribution. (It should be noted that one way stronger employment growth in high-wage occupations could be contributing to stronger wage growth at the top of the overall wage distribution is through composition changes. However, in the tasks framework, shifts in the overall wage distribution are due to changing demand for workers in different occupations, which would have to operate through occupational wages.)

The empirical links are also weak for the 1980s and the 2000s. Figure G-A shows the same set of lines for the 1979–1989 period. As we saw in Figure E and Figure F-A, the occupational employment growth shows a slight U-shaped relationship for the 1980s (in contrast with Autor’s work, using the decennial census, which shows a monotonic rise in employment growth by occupational skill level). Occupational wage changes in the CPS, however, increase monotonically. The increase in the occupational employment shares of lower-waged occupations did not translate into more rapid wage growth for these occupations. The overall wage distribution also increases monotonically in the initial wage percentile, though much more steeply than the relationship followed by occupational wages.28 For the 1980s then, we find that the corresponding changes in occupational employment, occupational wages and overall wages in the bottom half to be poorly aligned.

Figure G-C repeats the analysis using CPS data covering 2000–2007. Again, the occupational employment line is the same as the one in Figure E and Figure F-C. In the bottom occupational quintile, occupational employment and occupational wages move in opposite directions, with occupational employment growth declining through the bottom quintile, while occupational wage growth increases through the bottom quintile. As noted in earlier sections, the expansion of employment among low-wage occupations was greater in the 2000s than in prior decades but this does not appear to have translated into faster wage growth at the bottom. Occupational wages grew at essentially the same rate over the rest of the distribution. The overall wage distribution increased monotonically, with somewhat larger increases in the top quintile, inconsistent with the occupational wage patterns. As with the prior two plots, we find here that occupational employment changes, occupational wage changes, and changes in the overall wage distribution do not generally follow similar trends.

The tasks framework attempts to move from changes in production technologies to changes in occupational employment to changes in the overall wage distribution without pausing to examine the intervening changes in occupational wages. The initial, and still most common, empirical implementation of this framework operationalizes tasks as (roughly three-digit) occupations and then groups these tasks into “skill” percentiles based on the mean occupational wage. A review of the published evidence and our own analysis of the CPS data, however, demonstrate that the relationship between tasks and employment changes is fragile (as suggested by the poor fit between employment changes and occupational skill levels), inconsistent over time, and does not appear to hold at all in the 2000s. More importantly, occupational employment changes are not tightly linked to changes in the wage structure and the channel through which occupational employment shifts affect the wage structure—changes in the occupational wage structure—doesn’t match up. Occupational employment changes are poor predictors of occupational wage changes and occupational wage changes are poor predictors of changes in overall wages.

Figures H-A through H-C highlight just how little occupational employment changes have to do with occupational wage changes in the tasks framework.29 Figure H-A is a scatterplot of the log wage changes versus the log employment share changes at each occupational percentile for the 1979–1989 period. The slope of the least-squares regression line through the data is positive but insignificant, with an R-squared of 0.0102. In other words, roughly 1 percent of the variation in occupational wage changes is explained by occupational employment changes in the 1980s. Figures H-B and H-C present the same exercise for the 1990s and 2000. Between 1989 and 2000, the relationship between occupational wage changes and occupational employment changes is positive but insignificant, with less than 1 percent of the variation in occupational wage changes explained by occupational employment changes. Between 2000 and 2007, the relationship between occupational wage changes and occupational employment changes is actually negative, though insignificant. In other words, changes in occupational employment shares explain virtually none of the variation in occupational wage changes in any of the last three decades. If occupational employment changes do not drive occupational wage structure changes, as these figures show, then they surely do not drive overall wage structure changes.

As mentioned, in the tasks framework, first, technology moves occupational employment shares, then, changes in occupational employment drive changes in occupational wages, and finally, changes in occupational wages drive changes in the overall wage distribution. However, these links turn out to be extremely weak empirically. Table 3 summarizes the empirical links between occupational employment, occupational wages, and the overall wage structure. The first column of Table 3 gives the R-squares from the simple linear regressions of the change in the log occupational wage on the change in log employment share discussed in the previous paragraph. As mentioned, almost none of the variation of log occupational wages can be explained by changes in occupational employment share. The second column looks at the second link, the link between occupational wages and the overall wage distribution, and also finds it to be very weak, particularly in the 1990s and 2000s. In the 1980s, changes in wages by occupation percentile explained 11.7 percent of the changes in the overall wage distribution, and that figure drops below 3 percent for the 1990s and 2000s. In other words, very little of the variation of the changes in the overall wage structure can be explained by changes in wages by occupation. These two weak links means that occupational employment shifts explain virtually none of the shifts in the overall wage distribution.

The last column shows that even if we do what is generally done in this literature and ignore the channel through which occupational employment shifts affect the wage structure (i.e., through changes in the occupational wage structure), the relationship between occupational employment shifts and the overall wage structure is still very weak. In the 1980s and 1990s, the R-squared of the OLS regression of log wage changes by percentile on log employment share change by occupational percentile is less than 0.1, and it drops to less than 0.03 in the 2000s. These results tell us that analyses of occupational employment trends do not provide much leverage for understanding shifts in the wage distribution, and make us skeptical that occupational employment trends by themselves are useful in explaining key labor market dynamics.

B. Role of occupations in explaining wage variation

In the tasks framework, differences in changes in demand for different tasks are posited to be driving relative wage trends. Empirically, detailed occupations are used as a proxy for tasks. We have shown above that changes in the relative wages across occupations (as ranked by their 1979 mean wage) depart from changes in the overall wage distribution in critical ways. This casts doubt on whether occupational employment shifts are driving changes in the wage distribution since this would happen through changes in the wage distribution across occupations, rather than changes within occupations. Further, we have shown that there is in fact almost no relationship between occupational employment shifts and changes in the occupational wage structure. Here, we look at the direct link between occupations and wages using a regression analysis. A key result in this section is that we do not find that occupations explain an increasing share of wage variation over time, contradicting the claim in Acemoglu and Autor (2011) that occupations are becoming more important in wage determination. A related result is that a large and increasing amount of the rise in wage inequality in recent decades (as measured by the increase in the variance of wages) occurred within detailed occupations, even after accounting for the fact that a rising share of the workforce is in demographic groups—e.g., older workers—with higher residual wage inequality.

Tables 4A and 4B present, by gender, R-squares from cross-sectional OLS regressions of log hourly wages on the set of 250+ detailed occupation dummies along with a quartic in age, and dummies for region of the country (using the four major census regions), marital status, and race/ethnicity (using mutually exclusive categories of white non-Hispanic, black non-Hispanic, Hispanic any race, and other). We also calculate the partial R-squares of the set of detailed occupation dummies net of the remaining controls (i.e., net of the quartic in age and the dummies for region, marital status, and race/ethnicity).30 The partial R-squares are equivalent to the explanatory power of the set of detailed occupation dummies in the log hourly wage regressions. The partial R-squares are also plotted in Figures I-A and I-B.31

Columns (2) and (4) of Table 4, along with Figures I-A and I-B, show how the importance of occupations in wage determination has evolved over time for women and men. Column (4) shows that for men, the increase in the share of wage variation explained by detailed occupations slowed dramatically from the 1980s to the 1990s and then declined in the 2000s. Between 1979 and 1989, the share of wage variation explained by detailed occupations increased by 6.3 percentage points, whereas it increased by just 0.5 percentage points between 1989 and 2000. Between 2000 and 2007, the share of wage variation explained by detailed occupations dropped by 0.8 percentage points. Column (2) shows that for women, the importance of occupations in wage determination dropped in both the 1990s and the 2000s; the share of wage variation explained by detailed occupations dropped by 2.8 percentage points between 1989 and 2000, and dropped 2.7 percentage points further between 2000 and 2007. Table 4 and Figures I-A and I-B show that, contrary to findings in Acemoglu and Autor (2011), we find that differences across occupations accounted for a smaller share of rising wage inequality in the 1990s and 2000s than they did in the 1980s.32

One implication of the results in Figures I-A and I-B and Table 4 is that a small share of the total increase in wage inequality between 1989 and 2007 can be explained by even detailed occupations. Figures I-C and I-D and Table 4B investigate how much of rising wage variance can be explained by detailed occupations. They show the variance of log wages over time by gender, plus the variance explained by detailed occupations, which is simply the wage variance multiplied by the partial R-squared values given in Figures I-A and I-B and Tables 4A and 4B. It is striking how little of the increase in wage inequality can be accounted for by occupations, particularly in the 1990s and 2000s. Between 1989 and 2000, just 32.7 percent of the increase in male wage variance can be explained by detailed occupation dummies, and that drops to 17.0 percent between 2000 and 2007. For women, just 6.0 percent and 9.9 percent of the increase in wage variance can be explained by detailed occupation dummies in the 1990s and 2000s, respectively. In other words, the vast majority of the increase in wage inequality is occurring within occupations. By themselves, occupations provide very little leverage for understanding shifts in the wage distribution.

One drawback of the analysis presented in Figures I-A through I-D and Tables 4A and 4B is that using wage variance as a measure of wage inequality does not allow for an investigation of the impact of occupations on trends in inequality at different parts of the wage distribution, which is such a crucial component of the job polarization explanation of rising wage inequality. We turn now to the reweighting approach developed in Dinardo, Fortin, and Lemieux (1996) to examine how much of the changes in upper-tail and lower-tail inequality—as measured by the 50/10 and the 90/50 wage gaps—can be explained by detailed occupations.

Table 4C presents the results. Column (1) shows the 50/10 wage gap, column (2) shows the 50/10 wage gap holding the demographic composition constant across years, and column (3) shows the 50/10 wage gap holding the demographic and occupational composition constant across years. Column (4) gives the difference between (3) and (2), providing a measure of the contribution of the composition of occupations to changes in the 50/10 gap over various periods. Columns (5)–(8) are analogous to columns (1)–(4), but for the 90/50 wage gap. The demographic characteristics held constant in this table are the same ones used in Table 4A, (a quartic in age, and region, marital status, and race/ethnicity dummies), and the occupations are the 250+ detailed occupations described above.33

The only decade when the 50/10 wage gap for men changed substantially was in the 1980s, when it increased by 4.4 log points. This increase was not due to occupational shifts—occupations decreased the 50/10 wage gap for men by 3.1 log points in the 1980s.

The major changes in the 50/10 wage gap were among women. In the 1980s, the female 50/10 wage gap increased by 25.5 log points, but occupations were responsible for only 2.0 log points of this increase. In the 1990s, the female 50/10 wage-gap decreased by 5.5 log points, while occupations slightly increased the 50/10 wage gap for women over this period (by 0.7 log points). In the 2000s, occupations mattered for women’s wage shifts at the low-end of the wage distribution; the female 50/10 wage gap increased by 4.9 log points and 4.4 log points of that was due to occupations.

Occupational composition played a relatively small role in changes in the 90/50 wage gap. For men, occupations explained only 0.8 log points of a 10.7 log point increase in the 90/50 wage gap in the 1980s, and for women over this period occupations explained only 0.2 log points of a 10.4 log point increase in the 90/50 wage gap. For men in the 1990s, occupational composition explained only 1.0 log points of an 8.4 log point increase in the 90/50 wage gap. However, for women occupational composition did in fact explain a substantial share of the increase in the 90/50 wage gap in the 1990s, 3.4 log points out of an 8.0 log point increase.

In the 2000s, given the lack of job growth in high-wage occupations relative to middle-wage occupations (see Figure E), the job polarization explanation of wage inequality would not predict occupations having much of an effect on the 90/50 wage gap in the 2000s. This was largely true for women; occupational composition decreased the female 90/50 wage gap by a small amount, 0.5 log points. For men, however, the increase in the 90/50 wage gap due to occupations was larger in the 2000s than in the 1990s (1.2 log points versus 1.0 log points), which runs counter to the idea that job polarization was driving wage trends in the 2000s.

In short, for both men and women, occupational composition explains little of the changes in upper-tail and lower-tail inequality in most time periods examined.

To further investigate rising within-occupation wage inequality, we turn to an analysis of residual wage variance.34 Tables 5A and 5B and Figures J-A and J-B show the variance of log hourly wages and the residual wage variance from cross-sectional OLS regressions of log hourly wages on the 250+ detailed occupation dummies along with a quartic in age and dummies for region of the country (using the four major census regions), marital status, and race/ethnicity (using mutually exclusive categories of white non-Hispanic, black non-Hispanic, Hispanic any race, and other). (This is the same model used to generate the R-squares in Table 4.) In order to account for changes in residual wage inequality that are due to composition effects linked to the secular increase in groups with higher within-group wage variance—for example the aging of the workforce over this period—we also calculate the composition-adjusted residual wage variance according to the reweighting approach developed in Dinardo, Fortin, and Lemieux (1996) and described in the previous section. 35

The last two columns of Tables 5-A and 5-B show how much of the increase in the wage variance is due to an increase in residual (within-group) variance, both unadjusted and adjusted for composition changes. A key result from this table is that even after accounting for composition changes, a large and increasing share of the wage variance since 1989 is occurring within detailed occupations and demographic controls. For men, 25.5 percent of the increase in wage variance occurred within groups between 1989 and 2000, and that rose to 55.9 percent between 2000 and 2007. For women, 55.2 percent of the increase in wage variance occurred within groups between 1989 and 2000, and that rose slightly to 56.1 percent between 2000 and 2007. It is notable that less than half of the increase in wage variance for both men and women can be accounted for by detailed occupation and demographic controls between 2000 and 2007. These results reinforce the view that analysis of occupational changes provides only a limited assessment of changes in wage inequality, particularly in the 2000s.

VI. Service and low-wage occupations

In the latest iteration of the polarization literature, Autor and Dorn (2012) focus on the role of service occupation employment, offering “an integrated explanation and empirical analysis of the polarization of U.S. employment and wages between 1980 and 2005, and the concurrent growth of low-skill service occupations.” Autor and Dorn claim that the “rapid rise of employment and wages in service occupations” (page 4) is tightly linked to the overall polarization of employment and wages from 1980 to 2005. Specifically, they conclude that:

Between 1980 and 2005, the share of hours worked in service occupations among non-college workers rose by more than 50 percent. Simultaneously, real hourly wages of non-college workers in service occupations increased by 11 log points, considerably exceeding wage growth in other low-skill occupations. This phenomenon is broadly important because it offers insight into the polarization of employment and earnings in the U.S. and, potentially, other industrialized countries. Indeed, a key fact documented by this paper is that rising employment and wages in service occupations account for a substantial share of aggregate polarization and growth of the lower tail of the U.S. employment and earnings distributions between 1980 and 2005.

And, their bottom line is: “Our results suggest a critical role for changes in labor specialization, spurred by automation of routine task activities, as a driver of rising employment and wage polarization in the U.S. and potentially in other countries.”

The key claim then is that workplace technological change eroded middle-class routinized jobs and simultaneously increased demand for tasks done in service occupations, and this higher relative demand for service occupation work boosted wage growth in these occupations. This increase in service occupation employment and wages, in turn, boosted low-wage occupational employment and wages at the bottom, and largely explains wage and employment polarization at the bottom end.

This new line of research suffers from many of the weaknesses of the earlier research. First, and most fundamentally, over the period Autor and Dorn study, there was no wage polarization between service occupations and the middle of the overall wage distribution. As shown in Table 9, between 1979 and 2007, the median hourly wage grew faster (7.8 percent in real terms) than wages in service occupations (3.9 percent). The median family wage also grew faster than the 10th-percentile wage in the overall wage distribution (down 0.9 percent).36 These wage trends make it difficult to use the pattern of growth in service occupations to explain a compression between the bottom and the middle of the overall wage distribution that did not occur. Second, all of the data and the regression analyses presented focus on decadal periods (1980s, 1990s, and sometimes the 2000s) and the key analysis of the wage impact of employment changes is examined for changes over the entire 1979–2005 period, thus ignoring annual patterns and failing to examine the issue of the precise timing of trends.37 The timing is important because the increase in real wages of service occupations (primarily a late 1990s trend) preceded the expanded employment share of service occupations (primarily a 2000s trend). Third, by focusing on changes over the long period 1980–2005, Autor and Dorn paper over the earlier finding in the literature that wage and employment polarization did not occur in either the 1980s or the 2000s. That is, the discussion presumes that the research is explaining job polarization over the entire 1980–2005 period, a phenomenon that did not occur throughout that time period and was notably absent in the most recent period (2000–2005 or 2000–2007). Fourth, job polarization over the 1980–2005 period is not well rooted in the data. When, following Autor and Dorn (their Figure 1), we estimate the smoothed changes in employment shares across occupations ranked by wages we find evidence of polarization but the Lowess R-squared for our fitted line is just 0.111.38 Using the smoothed or unsmoothed data for the entire period, there is only a small absolute employment share expansion for low-wage occupations and only for those in the bottom decile—occupations in the second decile of the 1979 occupational wage distribution actually saw their employment share decline. Finally, although Autor and Dorn attempt to provide insights into the growth of wage inequality due to technological change, they never present any metric for assessing technology’s impact on wages nor compare any such estimates to changes in the overall growth of wage inequality. Therefore, it is not possible to assess the quantitative significance of the findings.

The Autor and Dorn research on the role of service occupations shifts the focus of attention in three important ways. First, the emphasis is now almost exclusively on the bottom of the occupational ladder, the low-wage service occupations. It therefore avoids the topic of why high-wage occupations expanded relative to middle-wage occupations in the 1980s and 1990s but not in the 2000s. Second, the focus is no longer on explaining the change in the trend of the 50/10 wage gap between the 1980s (when it grew) and the 1990s (when it stabilized and fell). In fact, nothing in their analysis helps to explain trends in the 50/10 wage gap, even though doing so was the key motivator for the tasks framework (Autor, Katz, and Kearney 2006, 2008; Acemoglu and Autor 2011, 2012). Last, research focuses on the share of non-college educated workers (specifically, those with a high school degree or less) in service occupations rather than on the share of all workers in service occupations. This appears to muddy the water in our view because, as shown below, the share of workers with a high school degree or less education working in service occupations has not been primarily driven by technological (and other) trends affecting the share of total employment in these occupations, but rather by the educational upgrading that has depleted the share of workers with such education (from 58.4 percent in 1979 to just 38.9 percent in 2007).

The service occupation intensity among workers with a “high school or less education” is:

where N refers to employment, subscripts refers to occupation and superscripts refers to the education category of the workforce: HS indicates workers with a high school degree or less; T stands for the all workers, and S stands for service occupations.

Thus, this ratio, on which Autor and Dorn focus, is determined by three distinguishable factors:

1. the share of “high school or less” workers among all workers;
2. service occupation employment’s share of total employment; and
3. the share of “high school or less” workers among those working in service occupations.

Although Autor and Dorn treat the share of those with a “high school or less” education in service occupations as being driven by technological change, this ratio clearly depends upon the relative importance of these three factors above. The first factor reflects labor supply and has declined in recent decades because of the “educational upgrading” of the workforce as more workers attend at least some college. The second factor, the share of employment in service occupations, is most clearly the result of technology.39 The third factor is ambiguous, potentially reflecting the decreased use of “less educated” workers in the production process (a result of technology), but also possibly reflecting underemployment of those with more education who are forced to work in lower-paying service occupations.

We present a decomposition of the role of these three factors in driving changes in the share of those with a “high school or less” education in service occupations below. Before doing so, we explore trends in the first two factors.

We have shown earlier that the main technological driver in the Autor and Dorn framing, the service occupation share of employment, has remained low and relatively stable over the last three decades, with some expansion occurring in the 2000s. As such it would seem that technological change has played an inconsequential role in concentrating “high school or less” in service employment. Table 6 gathers an array of measures of the service occupation employment to further assess trends in service occupations’ employment share. We draw on several sources of information: our tabulations of the CPS-ORG data; published tabulations of decennial census and American Community Survey data from Acemoglu and Autor (2011), Autor and Dorn (2012), and Katz and Margo (2013); and published data from the BLS Occupational Employment Statistics (OES) program. Some of the data reflect shares of employment while other data reflect share of total hours worked. Last, Autor and Dorn present trends for service occupation employment that exclude public safety employment (i.e., police and firefighters) and the table presents comparable results from our tabulations of the CPS.

All of the CPS tabulations—service occupations’ employment and work hours shares and the hours share of service occupations excluding public safety occupations—show that the importance of service occupations remained relatively low (12 to 14 percent) and stable over the 1980s and 1990s, with some expansion in the 1980s but stability in the 1990s. A faster growth in the 1980s than the 1990s is inconsistent with a story of expanded service occupation employment shifting the 50/10 wage gap pattern in the 1990s.

In contrast, the decennial census data from Acemoglu and Autor (2011) and Autor and Dorn (2012) indicate a modest expansion of service occupations in the 1980s and a somewhat larger expansion in the 1990s. There is a measurement problem, however, in the decennial census comparisons for the 1990s. The coding change that occurs in the CPS data between 2002 and 2003 occurs between 1990 and 2000 in the decennial census data. The CPS annual data for 2002–2003 indicate that this coding change, using Autor and Dorn’s occupational coding crosswalk, generates an artificial 1.0 percentage-point increase, equivalent to the entire expansion of service occupations in the 1990s.40 If the effect of the coding changes between the censuses in 1990 and 2000 is similar to the impact of the same change in the CPS between 2002 and 2003, then it is possible that the entire service occupations expansion in the 1990s shown by census data reflects the coding change implemented between the two censuses.41 Katz and Margo (2013), whose analysis relies on a different census crosswalk (one that puts all occupations into the 1950 occupation scheme), shows a slight decline in service occupation employment in the 1990s. The Katz and Margo data therefore lend support to our sense that the differences between the analysis by Acemoglu and Autor (2011) and Autor and Dorn (2012) and our own analysis of CPS data—we find stable service occupation employment while they find an expansion in the 1990s—is based on differing occupational coding schemes (with the one employed by Acemoglu and Autor overstating the expansion of service occupations in the 1990s by about 1.0 percentage point).

Data from BLS employment projections (Alpert and Auyer 2003) provide another estimate of the growth of service occupations in the 1990s, showing an expansion of 0.5 percentage points from 1988 to 2000. Other BLS employment projections provide reads on earlier periods, with expansion over 1983–1994, 1986–1996, and 1988–1998 of, respectively, 0.7, 0.5 and 0.5 percentage points.42 In these various BLS series there appears to be little, if any, expansion of service occupation employment between 1994 and 2000 and the fastest expansion shown is from 1983 to 1994.

None of the available data series available show any sizeable expansion of service occupation employment or hours, including or excluding public safety workers, in the 1980s and 1990s, and no evidence of any sizeable acceleration of service occupation growth in the 1990s. This is true even for the census data, which show the greatest expansion in the 1990s (Autor and Dorn 2012; Acemoglu and Autor 2011). This relative stability of the share of service occupation employment goes back to 1959 and of service occupation hours share goes back to 1949. Given this stability, it is difficult to explain how any technology-driven change in service occupations could have materially affected wage patterns and wage inequality in the 1980s or 1990s. By contrast, all of these same data series (including data from the establishment survey based Occupational Employment Statistics) show a sizeable expansion in service occupations in the short period from 1999 to 2005, or during the 2000s more generally.

The last line in Table 6 presents Autor and Dorn’s tabulation of the service employment share among those with a “high school or less” education, rising from 12.9 percent in 1979 to 19.8 percent in 2005. This is the only indicator to show a substantial expansion of service occupations. Given the relative stability of the more inclusive measures, it follows that the expansion highlighted in the table reflects the focus on a limited subset of the workforce. This previews our results below, which show that it is educational upgrading that is driving this trend, not a technology-driven change in the importance of service occupations in the economy. Last, we would note that Autor and Dorn characterize the roughly 7 percentage-point expansion (12.9 to 19.8) over the 26 years from 1979 to 2005 as a “50 percent” increase. That description, in our view, exaggerates the roughly 2.8 percentage-points per-decade expansion and is not the best description of these trends.

In the literature, service occupations are frequently equated with “low-wage occupations.” However, service occupations comprise about half the employment in the bottom fifth of occupational employment ranked by wage level. Autor and Dorn (2012, footnote 7) report that service occupations’ share of employment in the bottom fifth rose from 47 to 55 percent between 1979 and 2005, suggesting this indicates their rising importance. In fact, these same data indicate that service occupations (in the bottom fifth) rose from 9.4 to 11.0 percent of total employment between 1979 and 2005, another confirmation of the low and relatively stable role of low-wage service occupation employment—a rise of only 1.6 percentage points over 26 years.

We use the CPS data to examine the Autor and Dorn claim that service occupations are responsible for the expansion of employment in low-wage occupations. The CPS data suggest the opposite conclusion, showing that low-wage occupational employment actually contracted both in the 1980s and the 1990s before expanding modestly in the 2000s: the modest expansion of service occupation employment was offset by a larger contraction among other low-wage occupations.

Table 7 provides CPS tabulations of the breakdown of the hours-weighted employment shares of the lowest-wage occupations, separating service (excluding public safety occupations) and other occupations, in key years over the 1979–2007 period. This analysis uses the Autor and Dorn occupational coding crosswalk but adjusts for the coding breaks in 1983 and 2003. These “decile” breakdowns are “lumpy”—not being exact deciles—because the analysis relies on the actual cumulative shares of detailed occupations.43 The data in Table 7 are, in other terms, displaying growth in the unsmoothed occupational employment share for the bottom two deciles of occupations (defined by wage level); the numbers in Table 7 directly correspond to the growth in the smoothed occupation shares portrayed in the graphical analysis used by Autor and Dorn to examine job polarization trends.44

These data indicate that service occupations comprise the vast majority of the bottom decile of occupations and roughly half of the bottom fifth of occupational employment. Service occupation employment in each of the lowest two deciles was relatively stable in the 1980s and 1990s and expanded in the 2000s in the lowest decile but remained stable in the second decile. The lowest-wage occupations in the first decile (as determined by 1979 wage levels) expanded modestly over the entire 1979–2007 period in the first decile (rising to 10.8 from 10.2 percent). In the second decile, however, the lowest-wage occupations shrank over the same period (from 10.9 to 9.3 percent). As a consequence, on net, among the lowest-paying fifth of occupations there was no employment expansion over the last three decades. A direct implication of this result is that there was no employment polarization at the bottom to speak of over this period. The service occupations at the low end did expand modestly (from 9.6 to 11.0 percent) but this was offset by employment shrinkage in other (meaning “non-service”) low-paying occupations. This analysis of CPS employment data, therefore, contradicts the Autor and Dorn claim that service occupation trends have generated employment polarization at the bottom end of the occupation wage scale since 1979.

As noted earlier, Autor and Dorn focus much of their discussion on service occupation employment among a subset of the workforce, those with no more than a high school education. This metric is partially driven by the eroded share of the workforce in this education category, what we refer to as educational upgrading. Figure K displays the educational upgrading of the workforce over the 1973 to 2007 period, showing the share of employment represented by those with less than a high school degree, those with a high school degree (or GED) and those with some college but not a four-year college degree (including those with associate degrees). In contrast to the stability of the service occupations’ share of employment and hours, Figure K shows a rapidly declining share of those in each of these educational categories.

Autor and Dorn do not explain their selection of “high school or less” as their focus rather than those lacking a four-year college degree when analyzing “non-college” employment. They label their category as “non-college” which invites confusion because it excludes those who have no degree past a high school credential, the fifth of the workforce that has “some college” but no further degree (including not having an associate’s degree). The “some college” group’s wage levels and wage behavior mirror those of high school graduates much more than it does those with college degrees. We choose to label the Autor and Dorn category as “high school or less” rather than “non-college” to avoid any confusion about the category. The share of the non-college when defined as those lacking a four-year college degree comprised 69 percent of all workers in 2007 (down from 85.4 percent in 1973), a large share of the workforce. In contrast, the share of workers that have “high school or less” educations in 2007 was just 39.4 percent (down from 66.9 percent in 1973).

Table 8 presents a formal decomposition of the changes in the share of “high school or less” workers employed in service occupations. This is done by using the equation shown above, taking logs in each relevant year and differencing across time periods. Our focus is on the role of the increased share of employment in service occupations since that is the factor clearly identifiable as being driven by technological change. The decomposition shows that only about 30 percent (29.8) of the growth of the service occupation employment intensity among workers with a “high school or less education” from 1979 to 1999 was due to a growth of service occupation employment in the economy. In contrast, the growing importance of service occupation employment among “high school or less” workers in the 1980s and 1990s was heavily driven by educational upgrading. This group comprised 58.4 percent of all workers in 1979 but just 42.4 percent 20 years later. Consequently, movement in the share of “high school or less” workers employed in service occupations, at least over the 1980s and 1990s, hardly seem to reflect technological developments regarding routinization of middle-wage occupations or the expansion of low-wage service occupation employment. The 2000s, on the other hand, is the only period when service occupation employment’s expansion drove its importance among “high school or less” workers.

What role has service occupation employment played in wage patterns? We have already seen that service occupations did not expand at a notably different pace in the 1990s than in the 1980s (in the CPS there was no expansion). Even among workers with a “high school or less” education there was not a greater expansion of service employment in the 1990s relative to the 1980s.45 In our view, the similar expansion of service occupation employment in both the 1980s and 1990s should have produced similar wage trends for workers at the bottom of the wage distribution in the two decades. That the same employment patterns were associated with workers at the bottom of the wage distribution losing relative to the middle of the 1980s, but gaining relative to the middle in the 1990s, suggests that forces other than occupational employment patterns are the main drivers of shifting wage inequality.

What of the Autor and Dorn claim that “real hourly wages of non-college workers in service occupations increased by 11 log points, considerably exceeding wage growth in other low- skill occupations”? This phenomenon is broadly important because it offers insight into the polarization of employment and earnings in the United States. First, the wage polarization between 1979 and 2005 evident in their Figure 1 shows that wage growth for each of the percentiles in the bottom third of occupations were greater than that of each percentile in the middle third of occupations, something that probably cannot be explained by service occupation employment which comprised less than 13 percent of employment in 2005. Second, Autor and Dorn’s empirical analysis examines differences between 1979 and 2005 across local labor markets in order to identify the role of service occupation employment on wages. Their key finding is that, “7 percentage point higher routine share in 1980, equal to the gap between the 80th and 20th percentile commuting zones, predicts approximately 3 log points greater wage growth in service occupations between 1980 and 2005.” So, the technological process on which Autor and Dorn focus yields at most a modest change—3 log points—in wages over 26 years, which does not get us that far in explaining wage polarization.

Last, and most importantly, an examination of the timing of trends within the 1979–2005 period indicates that rising service occupation employment occurred after the notable real wage growth. The key data are presented in Table 9 and in Figure L for the 1979–2007 period.46 The median hourly wage grew faster than wages in service occupations between 1979 and 2007, 7.8 versus 3.9 percent, indicating that there was not any polarization between the middle and bottom wages over the period of interest to Autor and Dorn. The premise of that research is incorrect since there isn’t any wage polarization over the longer 1979–2007 period to explain. Most damaging to this framework, however, is that wage growth in service occupations only occurred in the late 1990s (up 15.2 percent), a period when service occupation employment did not expand relative to other occupations.

Moreover, the during the period from 2000 to 2007, when service occupation employment did expand, real wages for service occupations fell 6.2 percent. This pattern of stable employment shares with rising wages followed by rising employment shares with falling wages runs completely counter to the idea that the demand for service occupation work has been lifting wages at the bottom of the wage distribution relative to the middle and that such a dynamic provides insight into the evolution of overall wage inequality. Put in other terms, service occupation wages rose relative to the median wage in the late 1990s (see the column on the ratio) when service occupation employment wasn’t expanding and fell relative to the median wage over 2000–2007 when service occupation employment expanded more in seven years than it had over the preceding 21 years. The timing just does not work for the service occupation variation of the tasks framework analysis. (See Figure L, which shows the annual data.)

The trend in service occupation wages is certainly disappointing, rising only 3.9 percent over 28 years. Productivity in the economy as a whole rose 59 percent over that same period, so service occupation workers clearly have not gained correspondingly. Note also that Table 9 demonstrated that workers in service occupations had substantially more education in 2007 than in 1979 (the share with more than a high school education grew from 30 to 43 percent from 1979 to 2007). It seems at least as plausible to attribute the wage gains for service occupations in the second half of the 1990s to persistently low unemployment and 20 percent boost in the minimum wage enacted in 1996 and 1997.

VII. Conclusion

As Acemoglu and Autor (2011, 2012) and others before them have argued, the “canonical model” of SBTC fails to explain important developments in the U.S. wage distribution. The most notable of these gaps are the inability to explain the following: the reversal of the 50/10 differential after 1986–1987; the deceleration in the college premium after the mid-1990s; and the stagnation or decline over long periods in the real wage for important segments of the wage distribution. The tasks framework that Acemoglu and Autor developed to replace the “canonical model” is a distinct improvement over the “canonical model,” not simply because it can, in principle, account for these three key developments, but also because it provides a richer description of the labor market and the production process.

The tasks framework, however, suffers from its own empirical failings. Central among these are: the relatively smooth, long-standing nature of job polarization, which appears poorly suited to explaining the abrupt rise in inequality at the end of the 1970s or, more importantly, the sharp change in the path of the 50/10 wage differential after 1986–1987; the failure of the most conventional measure of job polarization (plots of occupational employment change against occupational wage level) to show any signs of occupational employment polarization in the 2000s, even as wage inequality continued to grow; and, the consistent lack of correspondence across the 1980s, 1990s, and 2000s between changes in occupational employment, occupational wages, and the overall wage distribution.

Recent research that has focused on the role of typically low-wage service occupations has not rescued the tasks framework from these shortcomings. Over the last three decades, service occupations comprised only about one-tenth of the workforce and increased their total employment share by less than two percentage points. A group of occupations that is small to begin with and that grows only slightly over the course of three decades seems unlikely to be a major driver of the overall wage distribution and almost by definition can tell us nothing about developments at the middle and the top of the occupational or overall wage distributions. More importantly, the timing of employment and wage changes in this sector is the exact opposite of what is suggested by the tasks framework. Service occupation wages grew first (in the boom of the late 1990s), at a time when employment in the sector was basically flat; when service occupation employment finally grew (after the 1990s boom went bust), wages in the sector actually fell.

The tasks framework offers a theoretical model that can, in principle, account for wage and employment trends that the “canonical model” cannot. In practice, however, the tasks framework suffers from its own empirical failings. Technology may be a factor in widening wage inequality, but, if so, the tasks framework is not the model that captures those dynamics.

About the authors

Lawrence Mishel has been president of the Economic Policy Institute since 2002. Prior to that he was EPI’s first research director (starting in 1987) and later became vice president. He is the co-author of all 12 editions of The State of Working America. He holds a Ph.D. in economics from the University of Wisconsin-Madison, and his articles have appeared in a variety of academic and non-academic journals. His areas of research are labor economics, wage and income distribution, industrial relations, productivity growth, and the economics of education.

John Schmitt is a senior economist at the Center for Economic and Policy Research and a research associate of the Economic Policy Institute. He has written for popular and academic publications on wage inequality, the minimum wage, education, work-life balance and comparative economic performance. He has a Ph.D. in economics from the London School of Economics.

Heidi Shierholz joined the Economic Policy Institute as an economist in 2007. She conducts research on employment, unemployment, and labor force participation; the wage, income, and wealth distributions; the labor market outcomes of young workers; unemployment insurance; the minimum wage; and the effect of immigration on wages in the U.S. labor market. She previously worked as an assistant professor of economics at the University of Toronto, and she holds a Ph.D. in economics from the University of Michigan-Ann Arbor.

Data appendix

By Heidi Shierholz and Hilary Wething

Unless otherwise noted, data used throughout this analysis are from the Outgoing Rotation Groups of the Current Population Survey (CPS-ORG) and the May CPS files 1973–1978. Our sample includes all public or private sector (unincorporated self-employed excluded) wage and salary workers who are age 18–64 and have valid wage and hour data. A detailed description of our treatment of the data, including calculation of hourly wages and treatment of top-coded earnings, is available in Appendix B of Mishel et al. (2012) (http://stateofworkingamerica.org/files/book/Appendices.pdf). Throughout this paper, unless otherwise noted, data are weighted by hours, in particular using CPS person weights multiplied by usual hours worked. It should be noted that we do not exclude from any analysis in this paper observations where wages are imputed, (unless the observations is being excluded for other reasons, for example if unincorporated self-employed). Future work will test whether our findings are affected by removing observations with imputed wages.

To obtain a consistent series across several occupation coding changes in the CPS over the 1973-2007 period, we employ the same occupation crosswalk as was used to create Figure 12 in Acemoglu and Autor (2011) (which is also Figure 3 in Autor, 2010). In particular, we use the programs generously available for download on David Autor’s MIT website under Acemoglu and Autor (2010), May/ORG Wage Prep, http://economics.mit.edu/~dautor/hole-vol4/morg/morg-wage-prep.zip. We document that we are able to implement this occupation crosswalk successfully by replicating the underlying occupation employment shares over time used to make Figure 12 in Acemoglu and Autor (2011), the programs and data for which are also available for download under Acemoglu and Autor (2011), Figure 12 – Percent Change in Employment by Occupation, 1979-2009, http://economics.mit.edu/~dautor/hole-vol4/figs/fig-12-rev.zip.

Appendix Figure A shows these occupation employment shares over time. There are 10 major occupations and this graph shows trends for each of these 10 occupations from both our tabulations (in light grey) and Acemoglu and Autor’s tabulations (in dashed dark blue). The fact that only 10 lines are observable means that the lines based on our tabulations exactly duplicate their tabulations. It should be noted that in order to replicate the Acemoglu and Autor employment shares, the sample we used is slightly different that the sample we use for the rest of this paper. In particular, it includes workers age 16 and 17, includes all non-agriculture, non-military workers who report having worked and report an occupation, not just those with valid wage and hour data, and the data are person-weighted instead of hours-weighted.

One finding from looking at the year-by-year data is that there are discontinuities in 1983 and 2003, when major occupation coding changes were implemented. That is, despite “consistent” occupation coding schemes used by Acemoglu and Autor (2011) and Autor and Dorn (2012) there are substantial inconsistencies remaining. These discontinuities do not surface in the job polarization literature because all of the analysis covers changes across multiple years and no data are presented on an annual basis. We now turn to assessing the quantitative impact of the discontinuities in the occupation employment data due to coding changes. In Appendix Table 1 we compare the changes in employment shares for each of the 10 major occupation groups using both our CPS data with the occupation crosswalk described above, and an “adjusted series” (which substitutes the average of the trends in 1980–1982 and 1983–1985 for the 1982–1983 change and the average of the trends in 2000–2002 and 2003–2005 for the 2002–2003 change). Appendix Table 2 compares the changes in the unadjusted and adjusted series for 1982–1983 and 2002–2003 and presents the changes in the adjusted series over the business cycles, 1979–1989 or 2000–2007. We consider the difference between the two series a measure of the error introduced by the coding changes. The last column measures the scale of that error relative to the longer period trend—the periods for which occupation employment trends are presented in the literature. It turns out that the distortions due to the coding changes are large relative to the trends presented in the literature. Most prominent is that the increase in the employment share for low-wage service occupations between 2000 and 2007 in the unadjusted series was 2.3 percentage points, substantially higher than the 1.3 percentage-point growth in the adjusted series. This implies that the coding change was responsible for 42 percent (1.0 percentage point of the observed 2.3 percentage-point change) of the observed expansion of low-wage service occupations in the unadjusted CPS series employed in the job polarization literature. Also, the unadjusted data series seriously understates the erosion of middle-wage occupational employment in the 1980s. The unadjusted series shows a decline of just 3.9 percentage points, whereas the adjusted series shows a decline of 5.6 percentage points. This implies the coding change was responsible for a 45 percent understatement (1.7 percentage points on the observed 3.9 percentage-point drop) of the erosion of middle-wage occupations. Correspondingly, the unadjusted series understates the expansion of high-wage occupations in the 1980s. (It should be noted that the employment shares in these tables are calculated from the subsample described in the first paragraph of this data appendix and used throughout this paper, not the somewhat expanded subsample used in Appendix Figure A that matches the data used in Acemoglu and Autor (2011, Figure 12.)

Appendix Table 3 explores the extent to which the effect of the coding change for key annual changes (1982–1983 and 2002–2003) varies depending on whether the data are person-weighted or hours-weighted. We find that the 1982/83 coding change leads to an understatement of the erosion of middle-wage occupations (and an overstatement of the growth of high-wage occupations) using both employment and hours weights, but the under/overstatements are more pronounced using employment weights. We find that there is little difference on the impact of the 2002/2003 coding change using employment-weighted versus hours-weighted occupational shares.

An important issue with the 1982/1983 coding break is that there are many occupations in 1983 and later that do not exist in the earlier data. In particular, of the 318 occupations that have workers in them with valid CPS wage data in 1983, 63 do not exist in 1982. They are relatively small occupations on average, but altogether 9.1 percent of workers in 1983 are in occupations that do not exist in the 1982 coding. This is a concern because in Acemoglu and Autor (2010), occupations are ranked by 1979 wages prior to this coding break, and any occupations that don’t exist in the year occupations are ranked, are, for obvious reasons, dropped from the analysis. This means that a nontrivial chunk of the post-1982 data are dropped. Appendix Table 4 shows the distribution across the major occupations of employment in 1983 by whether or not the worker is classified into an occupation that existed in 1982. Workers that are in occupations that did not exist in prior to 1983 are necessarily dropped from any analyses in which occupations are ranked prior to 1983. Dropping these occupations meaningfully increases the share of personal care and personal service workers, operators, fabricators, and laborers, and professionals, and meaningfully decreases the share of workers in sales, management, and production, craft and repair.

Finally, Appendix Tables 5A through 5C provide our complete annual data of our aggregate occupations shares, with the adjustments made for the 1982/1983 and 2002/2003 coding breaks, between the years 1973 and 2010 for all workers and by gender.

Means vs. Medians: In most of the job polarization research, occupations are ranked by their mean wage in 1979. However, since the idea of the ranking is to capture wages paid for the typical set of tasks associated with each occupation, arguably one would want to rank occupations by their median wage to ensure that the occupational rankings are not distorted by atypical wages within occupations. In any event, ideally the results would be robust to ranking by either means or medians. We find that the broad contours of Figure E—which, consistent to what is typically done in this literature, is generated with occupations ranked using means—are indeed quite similar when ranking occupations by either their means or median wage.

Endnotes

1. See Mishel et al. (2012).

2. Our findings using the Current Population Survey (CPS) corroborate similar results first described by Lefter and Sand (2011) using decennial census and ACS data.

3. The “canonical model’s” supply-and-demand framework has sometimes been augmented by a consideration of labor-market institutions (most typically, the minimum wage and unions), but as Goldin and Katz (2009) have argued: “The rise and decline of unions plays a supporting role in the story [of wage inequality], as do immigration and outsourcing. But not much of a role. Stripped to essentials, the ebb and flow of wage inequality is all about education and technology.” (p. 28)

4. The tasks model “…nests the canonical model as one parametric case; thus, this model builds upon rather than dispenses with the many strengths of the canonical model.” (Acemoglu and Autor 2012, 445)

5. As we note below, all three of these shortcomings were identified by Mishel, Bernstein, and Schmitt (1997). Acemoglu and Autor (2011, 2012) build on task-based models proposed by Autor, Levy, and Murnane (2003).

6. The abstract of the article (Autor, Katz, and Kearney 2008) that grew from the initial job polarization paper (Autor, Katz, and Kearney 2006) presented at the 2006 AEA meetings says, “The slowing of the growth of overall wage inequality in the 1990s hides a divergence in the paths of upper-tail (90/50) inequality—which has increased steadily since 1980, even adjusting for changes in labor force composition—and lower-tail (50/10) inequality, which rose sharply in the first half of the 1980s and plateaued or contracted thereafter…. Models emphasizing rapid secular growth in the relative demand for skills—attributable to skill-biased technical change—and a sharp deceleration in the relative supply of college workers in the 1980s do an excellent job of capturing the evolution of the college/high school wage premium over four decades. But these models also imply a puzzling deceleration in relative demand growth for college workers in the early 1990s…”

7. Our reading of the data is that the key inflection point occurred somewhat earlier in 1987 or 1988.

8. Acemoglu and Autor, characterizing the difference in wage patterns between the 1980s and 1990s, focus on the shift in the 50/10 wage gap differential: “During the initial period of 1974 through 1988, the monotonicity of wage changes by percentile is evident. Equally visible is the U-shaped (or ‘polarized’) growth of wages by percentile in the 1988 through 2008 period. Interestingly, the steep gradient of wage changes above the median is nearly parallel for these two time intervals. Thus, the key difference between these periods turns on the evolution of the lower tail, which fell steeply in the 1980s and then regained ground relative to the median thereafter” (Acemoglu and Autor 2012, 440-1, emphasis added).

9. As they observe, “The canonical model of factor-augmenting technical change robustly predicts that demand shifts favoring skilled workers will raise the skill premium and boost the real earnings of all skill groups (e.g., college and high school workers)... This prediction appears strikingly at odds with the data…” (2012, 439, emphasis added).

10. Acemoglu and Autor (2012, 444) are very explicit that the motivation for developing the tasks framework is to overcome the deficiencies of the “canonical model”: “We believe that these discrepancies between the data and the predictions of the canonical model— specifically, the heterogeneous behavior of the top, middle and bottom of the earnings distribution, falling real wages of some skill groups, polarization of earnings growth, and polarization of occupational growth are sufficiently important to warrant enriching the canonical model to gain leverage on these trends.” And they note: “We do not wish to suggest that this model resolves the puzzles posed above; to some degree, it was purpose-built to interpret them. Our claim — or at least our hope — is that this framework is a productive conceptual tool for confronting key facts that currently lie beyond the canonical model’s scope” (448).

11. Acemoglu and Autor (2012) write: “Many of the shortcomings of the canonical model can, we believe, be overcome by relaxing the implicit equivalence between workers’ skills and their job tasks in the model. In our terminology, a task is a unit of work activity that produces output. A skill is a worker’s stock of capabilities for performing various tasks. Workers apply their skills to tasks in exchange for wages. Thus, the task-based approaches emphasize that skills are applied to tasks to produce output—skills do not directly produce output. The distinction between skills and tasks is irrelevant if workers of a given skill always perform the same set of tasks. The distinction becomes important, however, when the assignment of skills to tasks is evolving with time, either because shifts in market prices mandate reallocation of skills to tasks or because the set of tasks demanded in the economy is altered by technological developments, trade, or offshoring” (444-45).

12. See, most recently, Acemoglu and Autor (2012, Figure 5) covering 1990–2007, as well as Autor, Katz, and Kearney (2008, Figure 11) covering 1990–2000; Autor (2010, Figure 1) covering 1989–1999; and Acemoglu and Autor (2011, Figure 10) covering 1989–1999. In all cases, occupational skill level is measured by the mean occupational wage in 1980, except Autor, Katz, and Kearney, who use the median occupational wage in 1980.

13. There is relative polarization (a term further elaborated below) further up the distribution to the 20th percentile, meaning employment shares among low-wage occupations were expanding greater (falling less) than those in middle-wage occupations.

14. The tasks framework also is said to address two other shifts in wage patterns: (1) from the early 1990s, education wage differentials stabilized among the non-college educated (that is, those with some college, a high school degree, or no high school degree); and (2) from the mid-1990s, the college premium decelerated sharply and grew at a rate well below what would have been predicted by the “canonical model.” This paper does not examine these empirical claims. Our research, not included here, does lend support to the second claim—occupational employment-driven changes in educational requirements do correspond to the deceleration of the college wage premium. Thus, the tasks framework and the data contradict “common intuitions regarding the nature or pace of technological changes occurring in this era,” meaning that SBTC actually did slow in the late 1990s, just like the “canonical model” suggests.

15. The authors note, “This result is of signal importance to our analysis because it underscores that despite ongoing, skilled labor augmenting technological progress and a fixed skill endowment, wage inequality need not rise indefinitely. If goods and services are at least weakly complementary, inequality between high- and low-skill labor either asymptotes to a constant or reverses course. Thus, consumer preferences determine whether the rising marginal physical product of high-skill workers translates into a corresponding rise in their marginal value product” (1566). Moreover, the tasks framework focuses on the demand-side and not the supply-side of the labor market.

16. These are hours-weighted employment trends “adjusted” to remove the inconsistencies generated by occupation coding changes in 1983 and 2003. We employ the consistent occupation coding developed by Autor and Dorn. The Appendix provides methodological detail.

17. The ratio of the employment shares of middle-skill and low-skill occupations grew from 1.233 in 1990 to 1.255 in 2000.

18. Acemoglu and Autor (2011) use CPS-ORG data for computing wage trends for the 10 broad occupational groups in Tables 2A through 2C based on their consistent occupation coding crosswalk, but do not use these data for annual employment trends.

19. That is, the change in 1982 to 1983 is the average of the changes from 1980 to 1982 and from 1983 to 1985. Comparably, the change in 2002–2003 is the average of the change from 2000–2002 and from 2003–2005. The changes from 1983 to 2002 are the 1983 occupation share obtained this way plus the change in each year’s occupation share change in the CPS. Similarly, the changes from 2003 to 2010 are based on the actual percentage-point change in occupation shares in the CPS.

20. Acemoglu and Autor (2011) do not show annual data but do have them in their publicly available program.

21. The consistency problems associated with the Autor and Dorn occupational classification system are also discussed in Foote and Ryan (2012).

22. This difference may reflect the fact that the occupational coding change that occurs between 2002 and 2003 in the CPS occurs between 1990 and 2000 in the census data, used by Acemoglu and Autor (2011). We find, for instance, that the 2002–2003 coding change artificially inflated service occupation (excluding public safety) growth by 1.0 percentage point. This measurement issue is addressed in more detail in the section on service occupations.

23. Acemoglu and Autor (2011, 17):

“In contrast, during the most recent decade for which census/ACS data are available, 1999-2007, employment growth was heavily concentrated among the lowest three deciles of occupations. In deciles four through nine, the change in employment shares was negative, while in the highest decile, almost no change is evident. Thus, the disproportionate growth of low-education, low-wage occupations became evident in the 1990s and accelerated thereafter.”

And Autor (2010, 3) describes the 1997–2007 trends in the upper 60 percent as “flat,” which better accords with our read of the facts:

“Fast forward to the period 1999 to 2007. In this interval, the growth of low-skill jobs comes to dominate the figure. Employment growth in this period was heavily concentrated among the lowest three deciles of occupations. In deciles four through nine, growth in employment shares was negative. In the highest decile of occupations, employment shares were flat. Thus, the disproportionate growth of low-education, low-wage occupations becomes evident in the 1990s and accelerates thereafter.”

24. The lowess R-squared is calculated in the same way an R-squared is calculated: the sum of the squared deviations of the lowess fitted values from the mean of the lowess fitted values divided by the total sum of squares of the dependent variable. Jacoby (2000) notes that a summary fit statistic of this type “cannot, strictly speaking, be interpreted as variance explained because the lowess fitting procedure does not partition the total sum of squares in Y neatly into additive components representing the sums of squares in the fitted values and the residuals, respectively. … Users should simply give the lowess R-squared value a more limited interpretation. It conveys the size of the fitted value variance, expressed as a ratio of the total variance in Y. While the latter is not really variance explained in the traditional sense, it does provide an effective summary of the degree to which the lowess fitted values track the empirical data points in the scatterplot.”

25. We also fit a quadratic curve to these data in each period. For the 1979–1989 period, the R-squared associated with a quadratic curve is 0.232. The t-statistic on the linear term is -2.58, and the t-statistic on the quadratic term is 3.68. Thus, there is a statistically significant U-shape in the 1980s. For the 1989–2000 period, the R-squared is 0.1830. The t-statistic on the linear term is -1.35, and the t-statistic on the quadratic term is 2.42, so there is also a statistically significant U-shape in the 1990s. For the 2000–2007 period, the R-squared is 0.0306. The t-statistic on the linear term is 0.10, and the t-statistic on the quadratic term is 0.33. The two terms are not individually or jointly significant (F-statistic of 1.53).

26. Acemoglu and Autor (2011, 13) note: “The explanatory power of occupation reaches a nadir in 1979 and then, like the education measures, rises over the subsequent three decades. Distinct from the education measures, however, the explanatory power of the occupation variables rises less rapidly than education in the 1980s and more rapidly than education thereafter – overtaking education by 2007. Thus, as hypothesized, occupation appears to gain in importance over time. This is most pronounced starting in the 1990s, when the monotone growth of employment and earnings gives way to polarization.”

27. An obvious implication of the divergence between the occupational wage distribution and the overall wage distribution is that an important part of the changes in the overall wage distribution is taking place within occupations, not between them.

28. Again, this suggests that much of what is happening to wages is happening within not between occupations.

29. Figures H-A through H-C are similar in spirit to figures 2.7 and 2.8 in Howell, Houston, and Milberg (2001), though we do not subdivide occupations by industry sector.

30. In particular, the log hourly wage and each occupation dummy is separately regressed on the remaining controls (the quartic in age and the dummies for region, marital status, and race/ethnicity), and using the residuals from each of these regressions, residual log hourly wages are regressed on the residuals from the occupation dummy regressions. The partial R-squares are the R-squares from these regressions.

31. Though the specifications differ slightly, Figures IA-B here are similar in spirit to Figures 17a and 17b in Acemoglu and Autor (2011). One key difference is that we calculate the partial R-squares using the 250+ detailed occupations. The Acemoglu and Autor figure shows rising partial R-squares of occupation dummies since 1989, while we do not.

32. It should be noted in Figures I-A and I-B that there is a visible drop between 1993 and 1994 in the partial R-squared values. This drop coincides with major changes in the CPS survey, however it does not coincide with occupational coding breaks (as can be seen in Appendix Figure A). This merits further investigation. There is also a noticeable rise in the partial R-squared values, particularly for women, between 1982 and 1983. This is unsurprising since, as noted in the Appendix, there are 63 more occupation dummies for 1983 than there are for 1982. To test whether these breaks may affect the finding in Table 4 of a slowdown in the importance of occupations between the 1980s and the 1990s, we simply subtract the changes associated with these two “break” years (1982/1983 and 1993/1994). We find the following: For men, the increase in the share of wage variation explained by detailed occupations drops from 5.6 percentage points between 1979 and 1989 to 3.0 percentage points between 1989 and 2000 (instead of dropping from 6.3 percentage points to 0.5 percentage points). For women, the increase in the share of wage variation explained by detailed occupations drops from 5.2 percentage points between 1979 and 1989 to 1.5 percentage points between 1989 and 2000 (instead of dropping from 7.5 percentage points to -2.8 percentage points). Thus, after removing the effect of these breaks, the slowdown in the importance of occupations between the 1980s and the 1990s is somewhat smaller, but it remains pronounced. Of course, the finding also remains that for both men and women the partial R-squared values associated with detailed occupations declines between 2000 and 2007.

33. It should be noted that in this reweighting approach any year can be selected as the “base” year, the composition of which is held constant; typically the base year is either the first or last year of the series. We select 2007 rather than 1979 as the base year because of the major expansion of occupations that occurs between 1982 and 1983; since we are composition-adjusting for occupations, if we were to composition-adjust using the 1979 occupations, the data from the 60+ occupations that existed in 1983 and later but did not exist in 1979 would be dropped from the composition-adjusted series.

34. We abandon our investigation of 50/10 and 90/50 trends in this context since the residual wage distribution does not map on to the overall wage distribution and therefore the 90/50 in the residual wage distribution does not provide information about inequality in the upper-tail of the wage distribution and the 50/10 in the residual wage distribution does not provide information about inequality in the lower-tail of the wage distribution.

35. It should be noted that the break in the composition-adjusted residual wage variance between 1982 and 1983 that is visible in Figures J-A and J-B is likely due to the occupational coding break that happened at that time. As in Figures I-A and I-B, there is also a visible break between 1993 and 1994 in Figures J-A and J-B with major changes in the CPS survey (not occupational coding breaks). This merits further investigation.

36. Mishel et al. (2012, Table 4.4)

37. Though Autor and Dorn refer to the decennial census data in terms of the year the data were collected (1980, 1990, or 2000) we refer to the data for the year the data capture (1979, 1989, and 1999) since those are the years we use in our CPS analysis.

38. This follows the same procedures as the estimates explained in an earlier section. We did not adjust the employment shares to account for the inconsistencies resulting from coding changes in 1983 and 2003 since we are seeking to replicate Autor and Dorn’s finding. Our analysis uses hours-weighted employment data from the CPS-ORG data for 1979 and 2005, the same years as the 1980–2005 period presented by Autor and Dorn.

39. Of course, globalization and consumer preferences also matter here.

40. The Appendix presents this analysis.

41. David Autor, in correspondence, has provided a more detailed breakdown of the 1.1 percentage-point growth in service occupations in the Autor and Dorn data analysis: “food occupations” (0.4 ppt.); health aides (0.4 ppt.); child care (0.3 ppt.); janitors (-0.2 ppt.) and recreation (0.2 ppt.). Our CPS analysis shows a 0.1 ppt. decline, a gap of 1.2 ppt. with the Autor and Dorn analysis. To gauge the correspondence of our CPS analysis and the impact of the coding change (judged by the 2002–2003 impact) we computed the detailed change of the sum of the CPS change plus the coding impact and find: “food occupations” (0.5 ppt.); health aides (0.4 ppt.); child care (0.1 ppt.); janitors (-0.3 ppt.) and recreation (0.3 ppt.) which sum to a 0.8 ppt. change. The relatively close correspondence by detailed service occupation subgroup indicates that the impact of the coding change represents most of the difference between the Autor and Dorn and our CPS computations of the growth of service occupations in the 1990s.

42. See Silvestri (1995, 1997) and Braddock (1999).

43. There are 18 occupations in the first decile with 1979 average wages of $4.04 or less and another 26 occupations in the second decile with 1979 average wages of greater than $4.04 but less than or equal to $4.84. The cutoffs for the first and second decile occupational employment groups in 2012 dollars are $11.91 and $14.27. These cutoffs are dramatically different than the first and second decile wage cutoffs for the overall wage distribution in 1979—$8.71 and $9.93, another indication of how different the occupational and actual wage structures are.

44. The data in Table 7 show changes in employment shares while the “polarization graphs” show log changes in employment shares.

45. This is true in the census data presented by Autor and Dorn and the CPS data show a much smaller expansion in the 1990s than in the 1980s.

46. Extending the analysis to 2007 from 2005 does not change any conclusions.

References

Acemoglu, Daron. 2010. “Yes the Labour Force Hasn’t Responded to Shifting Demand for Skills.” A response to, “Is America Facing an Increase in Structural Unemployment?” The Economist: Economics, by invitation. July 25, 2010.

Acemoglu, Daren, and David Autor. 2012. “What Does Human Capital Do? A Review of Goldin and Katz’s The Race between Education and Technology.” Journal of Economic Literature 50(2), 426–463.

Acemoglu, Daron, and David Autor. 2011 “Skill, Tasks and Technologies: Implications for Employment Earnings.” In O. Ashenfelter and D. Card (eds.), The Handbook of Labor Economics vol. 4b. Amsterdam:  Elsevier.

Alpert, Andrew, and Jill Auyer. 2003. “Evaluating the BLS 1988–2000 employment projections.” Monthly Labor Review 126(10),13–37.

Autor, David. 2010. “The Polarization of Job Opportunities in the U.S. Labor Market, Implications for Employment and Earnings.” Center for American Progress and Brookings Institution: The Hamilton Project.

Autor, David, and David Dorn. 2009. “This Job is Getting Old: Measuring Changes in Job Opportunities using Occupational Age Structure.” American Economic Review: Papers and Proceedings 99(2), 45–51.

Autor, David, and David Dorn. 2010. “Inequality and Specialization: the Growth of Low-skilled Service Employment in the United States.” MIT Working Paper.

Autor, David, and David Dorn. 2013. “The Growth of Low-Skill Service Jobs and the Polarization of the U.S. Labor Market.” American Economic Review 103(5), 1553–1597.

Autor, David, David Dorn, and Gordon H. Hanson. 2013.“The China Syndrome: Local Labor Market Impacts of Import Competition in the United States.” American Economic Review, forthcoming.

Autor, David, Lawrence F. Katz, and Alan Krueger. 1998. “Computing Inequality: Have Computers Changed the Labor Market?” Quarterly Journal of Economics 113(4), 1169–1214.

Autor, David, Lawrence F. Katz, and Melissa S. Kearney. 2006. “The Polarization of the U.S. Labor Market.” NBER Working Paper 11986.

Autor, David, Lawrence F. Katz, and Melissa Kearney. 2008. “Trends in US Wage Inequality: Re-Assessing the Revisionists.” Review of Economics and Statistics 90(2), 300–323.

Autor, David, Frank Levy,  and Richard J. Murnane. 2002. “Upstairs Downstairs: Computers and Skills on Two Floors of a Large Bank.” Industrial and Labor Relations Review 55 (3), 432–447.

Autor, David, Frank Levy,  and Richard J. Murnane. 2003. “The Skill Content of Recent Technological Change: an Empirical Exploration.” Quarterly Journal of Economics 116 (4), 1279-1333.

Autor, David, Alan Manning, and Christopher Smith. 2010. “The Contribution of the Minimum Wage to U.S. Wage Inequality over Three Decades: A Reassessment.” NBER Working Paper No. 16533.

Baker, Dean. 2007. The United States since 1980. New York: Cambridge University Press.

Bakija, Jon, Adam Cole, and Bradley T. Heim. 2012. “Jobs and Income Growth of Top Earners and the Causes of Changing Income Inequality: Evidence from U.S. Tax Return Data.” Working paper: http://web.williams.edu/Economics/wp/BakijaColeHeimJobsIncomeGrowthTopEarners.pdf

Berman, Eli, John Bound, and Stephen Machin. 1998 “Implications of Skill Biased Technological Change: International Evidence.” Quarterly Journal of Economics 113(4), 1245–1279.

Berman, Eli, John Bound, and Zvi Griliches. 1994. “Changes in the Demand for Skilled Labor within U.S. Manufacturing: Evidence from the Annual Survey of Manufactures.” The Quarterly Journal of Economics 109(2), 367–397.

Bivens, Josh. 2008. Everybody Wins, Except for Most of Us: What Economics Teaches about Globalization. Washington, D.C.: Economic Policy Institute.

Bivens, Josh. 2011. Failure By Design: A Story behind America’s Broken Economy. An Economic Policy Institute book. Ithaca, N.Y: Cornell University Press.

Blinder, Alan S, and Alan B. Krueger. 2009. “Alternative Measures of Offshorability: A Survey Approach.” NBER Working Paper No. 15287.

Borjas, Goerge J., Richard B. Freeman, and Lawrence F. Katz. 1992. “On the Labor Market Effects of Immigration and Trade.” In George J. Borjas and Richard B. Freeman (eds.), Immigration and the Work Force: Economic Consequences for the United States and Source Areas. University of Chicago Press.

Braddock, Douglas. 1999. “Occupational Employment Projections to 2008,” Monthly Labor Review: Employment outlook: 1998–2008. Office of Employment Projections, Bureau of Labor Statistics.

Card, David, and John DiNardo. 2002. “Skill Biased Technological Change and Rising Wage Inequality: Some Problems and Puzzles.” Journal of Labor Economics 20(4), 733–783.

Card, David, and John DiNardo. 2006. “The Impact of Technological Change on Low Wage Workers: A Review“ in Rebecca M. Blank, Sheldon Danziger, and Robert F. Schoeni (eds.), Working and Poor: How Economic and Policy Changes are Affecting Low-Wage Workers. New York: Russell Sage Foundation.

DiNardo, John, Nicole M Fortin, and Thomas Lemieux. 1996. “Labor Market Institutions and the Distribution of Wages, 1973-1992: A Semiparametric Approach.” Econometrica 64(5), 1001–1044.

DiNardo, John, and Jorn-Steffen Pischke. 1997. The Returns to Computer Use Revisited: Have Pencils Changed the Wage Structure too? The Quarterly Journal of Economics 112(1), 291–303.

Fernandez-Macias, Enrique. 2012. “Job Polarization in Europe? Changes in the Employment Structure and Job Quality, 1995-2007.” Work and Occupations. 20(10), 1–26.

Fernandez-Macias, Enrique, John Hurley, and Donald Storrie. 2012. “Transformation of the Employment Structure in the EU and USA, 1995-2007.” New York: Palgrave Macmillan.

Firpo, Sergio, Nicole M. Fortin, and Thomas Lemieux. 2011. “Occupational Tasks and Changes in the Wage Structure.” IZA Discussion Paper No. 5542.

Foote, Christopher L., and Richard W. Ryan. 2012. “Labor-Market Polarization Over the Business Cycle.” Federal Reserve Bank of Boston. Public Policy Discussion Papers No. 12-8.

Fortin, Nicole M., and Thomas Lemieux. 1997. “Institutional Changes and Rising Wage Inequality: Is there a Linkage?” The Journal of Economic Perspectives. 11(2), 75–96.

Freeman, Richard B. 1992. “How Much Has De-unionization Contributed to the Rise in Male Earnings Inequality?” In Sheldon Danziger and Peter Gottschalk (eds.), Uneven Tides: Rising Inequality in America. New York: Sage Press.

Galbraith, James K. 1998. Created Unequal: The Crisis in American Pay. New York: The Free Press, Simon and Shuster Inc.

Goldin, Claudia, and Lawrence Katz. 2009. The Future of Inequality: The Other Reason Education Matters So Much. Washington, D.C.: The Milken Institute.

Goldin, Claudia, and Lawrence Katz. 2010. The Race between Education and Technology. Cambridge, Mass.: Belknap Press of Harvard University Press.

Handel, Michael. 1999. “Computers and the Wage Structure.” Levy Institute Working Paper No. 285.

Handel, Michael. 2005. Worker Skills and Job Requirements: Is There a Mismatch? Washington, D.C.: Economic Policy Institute.

Howell, David. 1994. “The Collapse of Low-skill Male Earnings in the 1980s: Skill Mismatch or Shifting Wage Norms?” Levy Institute Working Paper. No 105.

Howell, David. 1999. “Theory Driven Facts and the Growth of Earnings Inequality.” Review of Radical Political Economics 31(1), 54–86.

Howell, David. 2002. “Increasing Earnings Inequality and Unemployment in Developed Countries: Markets, Institutions, and the “Unified Theory.” Politics and Society 30(2), 193-243.

Howell, David R., Ellen Houston, and William Milberg. 2001. “Skill Mismatch, Bureaucratic Burden, and Rising Earnings Inequality in the U.S.: What Do Hours and Earnings Trends by Occupation Show?” In  Jim Stanford, Lance Taylor, and Ellen Houston (eds.), Power, Employment, and Accumulation: Social Structures in Economic Theory and Practice. M.E. Sharpe, Inc.

Howell, David R., and Susan S. Wieler. 1998. “Skill-biased Demand Shifts and Wage Collapse in the United States, A Critical Perspective.” Eastern Economic Journal 24(3), 343–366.

Howell, David R., and Edward N. Wolff. 1991. “Skills Bargaining Power and Rising Interindustry Wage Inequality Since 1970.” Review of Radical Political Economics 23(1 & 2), 30–37.

Jacoby, William. 2000. “Loess: A Nonparametric, Graphical Tool for Depicting Relationships Between Variables.” Electoral Studies 19(4), 577–613.

Juhn, Chinhui, Kevin M. Murphy, and Brooks Pierce. 1993. “Wage Inequality and the Rise in Returns to Skill.” Journal of Political Economy 101(3), 410–442.

Kaplan, Greg. 2012. “Moving Back Home: Insurance against Labor Market Risk.” Journal of Political Economy 120(3), 446–512.

Katz Lawrence F, and Alan B. Krueger. 1992. “The Effect of the Minimum Wage on the Fast Food Industry.” Industrial and Labor Relations Review 46(1)6–21.

Katz, Lawrence F., and Alan B. Krueger. 1999. “The High Pressure U.S. Labor Market of the 1990s.” Brookings Papers on Economic Activity 1999(1), 1–87.

Katz, Lawrence F., and Robert A. Margo. 2013. “Technical Change and the Relative Demand for Skilled Labor: the United States in Historical Perspective.” National Bureau of Economic Research. Working Paper No. 18752.

Katz, Lawrence F., and Kevin M. Murphy. 1992. “Changes in Relative Wages 1963-1987: Supply and Demand Factors.” Quarterly Journal of Economics 107(1), 35–78.

Kopczuk, Wojciech, Emmanuel Saez, and Jae Song. 2010. “Earnings Inequality and Mobility in the United States: Evidence from Social Security Data since 1937.” Quarterly Journal of Economics 125(1): 91–128.

Krugman, Paul. 1995. “Growing World Trade: Causes and Consequences.” Brookings Papers on Economic Activity 26(1), 327–377.

Lee, David. 1999. “Wage Inequality in the United States during the 1980s: Rising Dispersion or Falling Minimum Wage?” Quarterly Journal of Economics. 114(3), 977–1023.

Lefter, Alexandru, and Benjamin M. Sand. 2011. “Job Polarization in the U.S.: A Reassessment of the Evidence from the 1980s and 1990s.” University of St. Gallen Discussion Paper No. 2011–03.

Lemieux, Thomas. 2006. “Post-Secondary Education and Increasing Wage Inequality,” American Economic Review 96(2), 195–99.

Levy, F., and R.J. Murnane. 1994. “Skills, Demography and the Economy: Is There a Mismatch?” In Lewis C. Solomon and Alec R. Levenson (eds.),  Labor Markets, Employment Policy and Job Creation. Milkin Institute Series in Economics and Education. Boulder: Westview Press.

Levy, Frank, and Peter Temin. 2007. “Inequality and Institutions in 20th Century America.” NBER Working Paper No. 13106.

Maarten, Goos, and Alan Manning, 2007. “Lousy and Lovely Jobs: The Rising Polarization of Work in Britain,” The Review of Economics and Statistics 89(1), 118–133.

Meyer, Peter, and Anasastiya Osborne. 2005. “Proposed Category System for 1960-2000 census Occupations.” U.S. Bureau of Labor Statistics Working Paper No. 383

Mishel, Lawrence, and Jared Bernstein. 1994. “Is the Technology Black Box Empty?: An Empirical Examination of the Impact of Technology on Wage Inequality and the Employment Structure.” Economic Policy Institute Technical Paper.

Mishel, Lawrence, and Jared Bernstein. 1998. “Technology and the Wage Structure: Has Technology’s Impact Accelerated since the 1970s?” Research in Labor Economics, 17: 305–356.

Mishel, Lawrence, Jared Bernstein, and John Schmitt. 1997. The State of Working America 1996–1997. Economic Policy Institute. M.E. Sharpe. 

Mishel, Lawrence, Josh Bivens, Elise Gould, and Heidi Shierholz. 2012. The State of Working America, 12th Edition. An Economic Policy Institute book. Ithaca, N.Y.: Cornell University Press.

Mishel, Lawrence, and R.A. Teixeira. 1991. “The Myth of the Coming Labor Shortage: Jobs, Skills and Incomes of America’s Workforce 2000.” Economic Policy Institute.

Murphy, Kevin M., and Finis Welch. 1992. “The Structure of Wages.” The Quarterly Journal of Economics 107(1), 285–326.

Schmitt, John. 2002. “Supply, Demand, and the Wages of Recent College Graduates in the 1990s.” http://www.noapparentmotive.org/papers/RCG_III.pdf

Schmitt, John. 2009. “Inequality as Policy.” Center for Economic and Policy Research. Washington, D.C

Silvestri, George. 1995. “Occupational Employment to 2005.” Monthly Labor Review: Employment Outlook: 1994–2005. Office of Employment Projections, Bureau of Labor Statistics.

Silvestri, George. 1997. “Occupational Employment Projections to 2006.” Monthly Labor Review: Employment Outlook: 1996–2006. Office of Employment Projections, Bureau of Labor Statistics.

Spenner, Kenneth. I. 1988. “Technological Change, Skill Requirements and Educations: The Case for Uncertainty.” In R.M. Cyert and D.C. Mowery (eds.), The Impact of Technological Change on Employment and Economic Growth. Cambridge, Mass.: Ballinger Books.

Western, Bruce, and Jake Rosenfeld. 2011. “Unions, Norms, and the Rise in the U.S. Wage Inequality,” American Sociological Review. 76(4), 513–537.

Tables and figures

Table 1

Table 2a

Table 2b

Table 2c

Table 3

Table 4a

Table 4b

Table 4c

Table 5a

Table 5b

Table 6

Table 7

Table 8

Table 9

Figure A

Figure B

Figure Ca

Figure Cb

Figure Da

Figure Db

Figure Dc

Figure E

Figure Fa

Figure Fb

Figure Fc

Figure Ga

Figure Gb

Figure Gc

Figure Ha

Figure Hb

Figure Hc

Figure Ia

Figure Ib

Figure Ic

Figure Id

Figure Ja

Figure Jb

Figure K

Figure L

Appendix Table 1

Appendix Table 2

Appendix Table 3

Appendix Table 4

Appendix Table 5a

Appendix Table 5b

Appendix Table 5c

Appendix Figure A

This working paper presents the methodology and data sources used to compute the Economic Policy Institute’s 2013 Family Budget Calculator.

Definition of family

The size of a family dramatically affects the budget needed to maintain a safe and comfortable, but modest, standard of living. We have constructed budgets for six different types of families in each area. These families include single-parent families with one, two, or three children, and two-parent families with one, two, or three children.

Use the EPI Family Budget Calculator

Read the overview

Download data [Excel]

[/pullquote]

Our definition of single-parent families assumes that the head of household is employed, lives with his or her children, and is considered the head of household for federal income tax purposes. Our definition of two-parent families assumes that both partners are employed, live together with their children, and file federal income taxes jointly. An employment assumption needs to be made in order to calculate child-care costs, transportation needs, and tax liabilities.

Families with one child are assumed to have a 4-year-old. Families with two children are assumed to have a 4-year-old and an 8-year-old. Families with three children have a 4-year-old, an 8-year-old, and a 12-year-old.

Definition of areas

There are 615 areas in the 2013 EPI Family Budget Calculator. Forty-seven of these 615 areas are statewide averages of rural areas; Rhode Island, New Jersey, and Massachusetts do not have rural areas. The remaining 568 areas are defined using the following technique:

A Metropolitan Statistical Area (MSA) is defined by the Office of Management and Budget as having at least one urbanized area of 50,000 or more people, plus adjacent territory that has a high degree of social and economic integration with the core as measured by commuting ties (OMB 2009). Some of our data (housing section) require us to use Fair Market Rent areas (FMR areas). FMR areas are published by the U.S. Department of Housing and Urban Development (2013). They are divided into metropolitan FMR areas and nonmetropolitan FMR areas. Several of our components depend on the MSA categorization (child care and out-of-pocket medical costs in the health care section) and on population size of the MSA (transportation).

Since most metropolitan FMR areas match up with an MSA, we replaced all metropolitan FMR areas with the MSA in the list of “Family Budget” areas. The remaining nonmetropolitan FMR areas were labeled as non-MSA. The rural areas were also labeled as non-MSA.

When regional breakdowns were used for budget calculations, they are based on the Census Bureau Regions, per data availability (U.S. Census Bureau 2013).

Components of the 2013 EPI Family Budget Calculator

The 2013 EPI Family Budget Calculator consists of seven individual components: rent, food, child care, transportation, health care, other items of necessity, and taxes. The following sections will describe the methodology used to construct a monthly cost for each of these seven components across the 615 areas for which data have been made available.

Rent

Annually, HUD estimates Fair Market Rents (FMRs) in order to establish cost information for the federal government’s Section 8 housing assistance programs (HUD 2013). FMRs are used to assess the sufficiency of housing supply in MSAs for housing assistance programs. HUD obtains the data using 5-year data from the American Communities Survey (ACS). All counties that do not fall into a “family budget area” (see documentation on family budget areas) are counted as rural for that state. To establish a “rural” FMR, and thus a family budget cost for housing in a rural area, rental costs for rural counties are averaged into one price to be applied as “rural” for the entire state. Data extracts of these cost estimates are made publicly available, and EPI made use of these data to construct our family budget measure.

HUD FMR estimates are at the 40th percentile of rent cost—the dollar amount below which 40 percent of standard quality rental units are rented. HUD makes rental rates available for studio apartments and one-bedroom through four-bedroom apartments. For the EPI family budget, we assumed that families with one or two children use the two-bedroom rate. Families with three children use the rate for a three-bedroom unit. Rental costs include shelter plus all tenant-paid utilities, excluding telephone service, cable or satellite service, and Internet service.

Food

Data for food costs are taken from the Center for Nutrition Policy and Promotion (CNPP) publication “Official USDA Food Plans: Cost of Food at Home at Four Levels” (USDA 2013). Presented there are the official USDA costs for four types of food plans that serve as national standards for nutritious diets: the “Thrifty Plan,” “Low-Cost Plan,” “Moderate-Cost Plan,” and “Liberal Food Plan.” We use the USDA “Low-Cost Plan,” which assumes that almost all food is bought at a grocer and then prepared at home. We use June 2012 data that represent the annual average monthly cost (Carlson, Lino, and Fungwe 2007). The data are only available at the national level, and are thus the same for all family budget areas (except Alaska and Hawaii, which are discussed later). Since the costs of raw, unprepared foods vary relatively little over geographic areas and the data present only nationally representative costs of raw, unprepared foods, our calculations for costs of food vary only by the size of the family and not by geographical area.

Families are constructed from data for the following age categories: male 19–50, female 19–50, child 4–5, child 6–8, average of male and female for individual age 12–13.

• All costs in the USDA food plans table are for individuals in four-person families. For individuals in other size families, the following adjustments are suggested to account for differences in returns to scale for different family sizes (USDA 2013):
  • Two-person – add 10 percent
  • Three-person – add 5 percent
  • Five-person – subtract 5 percent
• To calculate overall household food costs, we first adjust food costs for each person in the household and then sum the adjusted food costs.
• Example: For a one-parent, two-child household:

Food cost = [female(age19–50)*1.05]+[child(age 4–5)*1.05]+[child(age 6–8)*1.05]

Note that for Alaska and Hawaii, separate food cost data are available in half-year increments. We use data for the first half of the year to compute household food costs for the four Alaska areas and the two Hawaii areas because it most closely represents the annual national data used for the other states. (Note that only the “Thrifty Plan” is available for these states; there is no “Low-Cost Plan.”)

The USDA food plans represent a nutritious diet at four different cost levels. The nutritional bases of the plans are the 1997–2005 Dietary Reference Intakes, 2005 Dietary Guidelines for Americans, and 2005 MyPyramid food intake recommendations. In addition to cost, differences among plans are in specific foods and quantities of foods. Another basis of the food plans is that all meals and snacks are prepared at home. All four food plans are based on 2001–2002 data and updated to current dollars by using the Consumer Price Index for specific food items.

Child care

We utilize the Child Care Aware of America (2012) (formerly NACCRRA) publication Parents and the High Cost of Child Care, which relies on information provided by the January 2012 State Child Care Resource and Referral Network survey. For the purposes of this study, we use Appendix Table 1, “2011 Average Annual Cost of Full-Time Care by State,” and Appendix Table 10a, “2011 Urban-Rural Cost Difference for Center Care, by State.” Several states in the survey report data on a delay, including the District of Columbia, Iowa, Louisiana, North Carolina, Oregon, Pennsylvania, South Dakota, and Vermont, which report 2010 data, and California and New Hampshire, which report data for 2009. If an MSA is in multiple states, the child care cost is assigned separately for each state within the MSA. Therefore, for example, child care costs would vary between St. Louis, Mo., and St. Louis, Ill.

From available years, we inflate all data to reflect real 2012 dollars using the Consumer Price Index of “Child care and nursery school” for all urban consumers from the Bureau of Labor Statistics, or BLS (BLS 2013b).

We calculate our child care costs for our family types based on the following assumptions:

• One parent, one child = cost of 4-year-old care
• One parent, two children = cost of 4-year-old care + cost of one school-age child
• One parent, three children = cost of 4-year-old care + cost of two school-age children
• Two parents, one child = cost of 4-year-old care
• Two parents, two children = cost of 4-year-old care + cost of one school-age child
• Two parents, three children = cost of 4-year-old care + cost of two school-age children

Center care

• We use center care estimates instead of in-home care for our child care costs because the costs of center care do not fluctuate as much as the costs of in-home care. We assume that other family members are not available to provide care.

Infant care

• The family budget does not include infant care in its child care costs because we do not have an infant as part of any “Family Budget” family. It should be noted, however, that infant center care is significantly more expensive than 4-year-old care, so the child care component for some families may be underestimated.

Four-year-old care

• Four-year-old care is full-time care. To approximate MSA and non-MSA care, we use urban and rural estimates for all 4-year-old center care costs, taken from Appendix 10a in the Child Care Aware of America (2012) publication. Urban and rural care cost estimates are not available for Alabama, Arizona, Connecticut, Idaho, Mississippi, New Jersey, North Carolina, Rhode Island, Vermont, and West Virginia, so state averages for 4-year-old care are used instead.

School-age child care

• The survey for school-age care specifically represents the average annual cost of before- and after-school care, and therefore it does not include full-time, weekend, or full-day summer care. Because of the need for 4-year-olds (and some 12-year-olds) to be in care during the summer, the value of school-aged child care is significantly underestimated.
• Urban and rural data do not exist for school-age children, so we use state averages for all the MSAs with the exception of the District of Columbia, Idaho, Minnesota, North Carolina, North Dakota, Texas, and West Virginia, as state-level data are not reported for these states. Regional averages, based on the Census Bureau Regions and Divisions, are taken for these states. The following indicate the regions the states with missing data fall into. We construct regional averages for these states to be used instead:

• The District of Columbia falls into the South Atlantic Division, which is composed of Delaware, Florida, the District of Columbia, Georgia, Maryland, North Carolina, South Carolina, Virginia, and West Virginia.
• Minnesota falls into the West North Central Division, which is composed of Iowa, Kansas, Minnesota, Missouri, Nebraska, North Dakota, and South Dakota.
• North Dakota falls into the West North Central division, which is composed of Iowa, Kansas, Minnesota, Missouri, Nebraska, North Dakota, and South Dakota.
• North Carolina falls into the South Atlantic Division, which is composed of Delaware, Florida, the District of Columbia, Georgia, Maryland, North Carolina, South Carolina, Virginia, and West Virginia.
• West Virginia falls in the South Atlantic Division, which is composed of Delaware, Florida, the District of Columbia, Georgia, Maryland, North Carolina, South Carolina, Virginia, and West Virginia.
• Idaho falls into the Western division, which is composed of Alaska, Arizona, California, Hawaii, Idaho, Nevada, Oregon, and Washington.
• Texas falls into the Southwest division, which is composed of Arkansas, Louisiana, New Mexico, Oklahoma, and Texas.

Transportation

Data on costs of transportation are produced by the Federal Highway Administration’s 2009 National Household Travel Survey, or NHTS (FHA 2009) and IRS Announcement 2011-116 (IRS 2012). We choose to use annualized vehicle miles traveled (VMT) for calculating both the total annual miles driven and to determine the trip purpose. While it is possible to use other metrics, such as person miles traveled, we judge that in many MSAs the use of a vehicle is necessary to get to and from nearly all major destinations, such as work, medical appointments, a grocery store, etc. In areas in which public transportation is accessible and available for traveling to and from major destinations, this cost may be overstated (though obviously even public transportation carries significant costs if used every day).

Our equations for calculating total transportation costs are as follows:

Single-parent families’ transportation costs =

[(share of work & non-social trips)*(average miles per month by MSA size)*(cost/mile)]

Two-parent families’ transportation costs =

[(share of work & non-social trips)*(average miles per month by MSA size)*(cost/mile)] + [(share of work trips)*(average miles per month by MSA size)*(cost/mile)]

Equation components

• The share of work and non-social trips is calculated using the 2009 National Household Travel Survey (FHA 2009), as 2009 is the most recent year for which data are available. The variable WHYTRP1S, trip purpose summary, is used in conjunction with the variable Travel Day VMT to categorize each vehicle trip into the following purposes: home; work; school, day care, religious activity; medical, dental services; shopping, errands; social, recreational; family, personal business, obligations; transport someone; meals; and other purposes. We chose to make non-social trips the share of trips to home; school, day care, religious activity; medical, dental services; shopping, errands; family, personal business, obligations; and to transport someone.
• We decompose the trip purpose and average vehicle miles traveled by MSA size using the variable MSASIZE, from the National Household Travel Survey. The MSASIZE uses the same definition of a Metropolitan Statistical Area as the Office of Management and Budget. The NHTS reports MSASIZE in six sizes, based on household population within a given area:
  • 0–49,999 inhabitants
  • 50,000–249,999 inhabitants
  • 250,000–499,999 inhabitants
  • 500,000–999,999 inhabitants
  • 1,000,000–2,999,999 inhabitants
  • 3,000,000+ inhabitants
• We group our MSAs by these six population categories using population data from the Office of Management and Budget.
• The IRS reports the standard mileage rates used to calculate the costs of operating an automobile for businesses, charitable, medical, or moving purposes. 2012 data are available at the beginning of January and are revised semi-annually for a more accurate estimate. For 2012, the revised and most accurate standard mileage rate for the use of car, van, pick-up, or panel truck is 55.5 cents per mile for business miles driven. The mileage rate includes fixed costs such as depreciation, lease payments, insurance, registration and license fees, and personal property taxes, and variable costs such as gasoline, oil, tires, and routine maintenance and repairs.

Example

Single parent in a rural area:

= (share of work & non-social trips)*(average annual miles by MSA size)/12*(cost/mile)
= (84.4/100)*$14,607/12*($0.555)
= .844*$1217.2 *$0.555
= $570.18

So $570.18 is the monthly transportation cost for a single parent who lives in a rural area.

Health care

There are two components to the health care data: total insurance premiums and total out-of-pocket costs. Data for premiums are average total premiums (in 2012 dollars) for private-sector establishments for areas within states for 2011. These data are from Table IX.A.2 from the insurance component of the Medical Expenditure Panel Study (MEPS) published by the Agency for Healthcare Research and Quality (AHRQ) at the U.S. Department of Health and Human Services (HHS 2013b). Out-of-pocket costs are from the MEPS Household Component (Full-Year Consolidated File) for 2010 (HHS 2013a). To calculate data for the 40th percentile, we use Table 14 from the BLS Employee Benefits Survey (BLS 2013d).

Premiums

Benchmarking decisions

• Our benchmark insurance plan cost in the Family Budget Calculator is the cost of an employer-sponsored insurance plan. We use this benchmark for a number of reasons. First, there is wide variation in the cost of individual plans, but a good portion of this variation is due to individual characteristics (i.e., young, healthy people get cheaper insurance on the individual market). To have a benchmark influenced less by individual characteristics, we use the employer-provided plan, which is pooled across many people. Further, the cost of individual plans can vary significantly depending on the actuarial generosity of each particular plan (i.e., what share of expected medical costs will be covered by insurance versus covered out of the patient’s pocket). The actuarial generosity of employer-based plans is more tightly distributed, which also makes an employer-based plan a better single benchmark. We obviously recognize that a significant share of families do not receive health care coverage through employer-sponsored plans, but we think this is a decent benchmark and defensible minimum standard for our Family Budget Calculator.
  • We use insurance from private-sector employers because data on employer-sponsored insurance from public-sector employers were only available by region and not by state. Additionally, public-sector premiums were only available for state and local government employees (there were no data on federal government employee premiums). To control for this, we would have to weight the regional state and local government premiums by public-sector employment in a state and assume that federal employees get the same premium. This would also only add a level change. Since we were already approximating the 40th percentile premiums from a proxy ratio, we did not want to create any potential inaccuracy. Premiums in the public sector tend to be more expensive, so we are more likely to understate, not overstate, the costs by using only private-sector and not public-sector premiums.
• We use total premium costs rather than focusing only on employee contributions. We do this because while we think the cost of an employer-sponsored plan is a good benchmark for what a given standard of health insurance costs, we cannot assume that people have access to any particular level of employer-provided benefit to help pay for it. No other part of the family budget construction requires assuming any particular behavior by employers, so we did not want to introduce one here. In future research we hope to look at the resources side of the equation to see what share of American families can meet the basic family budget thresholds. In constructing these shares, we will take into account the receipt of employer contributions to health insurance premiums (as well as the potential assistance received through public insurance such as Medicare and Medicaid).

Data source and selection

• We use “employee-plus-one premium” and “family premium” data. “Employee-plus-one premium” data are used for one-parent, one-child families from MEPS Table IX.A.2 (HHS 2013b). Premiums for “employee-plus-one” are generally less expensive than “family” premiums, and we assume that when one-parent, one-child families are faced with two different plans, they will opt for the less expensive alternative when it exists. “Family premium” data are used for all other family sizes.
• Some areas from MEPS Table IX.A.2 (HHS 2013b) do not match the areas chosen for the Family Budget Calculator (“Family Budget” areas). However, there usually was a very close match, so the MEPS areas were matched with the closest Family Budget area. For Family Budget areas that did not have a corresponding MEPS area, the “Remainder of state” value was used. “Remainder of state” was also used for rural Family Budget areas.

Data compilation

• Since the average total premiums from MEPS Table IX.A.2 (HHS 2013b) are annual figures, we divided by 12 to get the monthly premium.
• To calculate the premium for the 40th percentile, we used data from BLS Employee Benefits Survey Table 14, “Medical care benefits, family coverage: Employer and employee premiums by employee contribution requirement, private industry workers from the National Compensation Survey, March 2011” (BLS 2013d).
• We used a ratio of total premiums for the lowest 25th percentile to the average total premium:

  total average flat monthly premium for lowest 25th percentile

  —————————————————————————–

  total average flat monthly premium for all workers

  where the numerator is a weighted premium from the “All workers” row of data:

  [(percent of participating employees in employee contribution not required) * (average flat monthly employer premium if employee contribution not required)] + [(percent of participating employees in employee contribution required) * (average flat monthly employer + average flat monthly employee contribution in employee contribution required)]

  and the denominator is a weighted premium (using the same formula from above) from the “Lowest 25 percent” row of data under the heading “Average wage within the following categories.”

• We use the lowest 25th percentile because the data in this table have an income distribution based only on people with employer-sponsored insurance. The distribution is very tight and skewed upwards. We assumed that an income earner in the 40th percentile on the total income distribution would fall into the lowest 25th percentile of the income distribution of those with insurance.
• The 40th percentile premium is adjusted for inflation to 2012 dollars using the regional breakdowns of the Consumer Price Index-All Urban Consumers for Medical Care (CPI-U-MC) from the Bureau of Labor Statistics (BLS 2013c).

Out-of-pocket costs

Benchmarking decisions

• We assume that everyone has the equivalent of an employer-sponsored health insurance plan (defined by the variable PRIEU10).

Data source and selection

• Out-of-pocket costs are from the MEPS Household Component (Full-Year Consolidated File) for 2010 (HHS 2013a).
• We use the regional breakdown of costs for both the adult and child numbers (we used the variable REGION10, with the regions defined as Northeast, Midwest, South, and West).
• The data are divided into MSA data and non-MSA data (we used the variable MSA10). For out-of-pocket costs, an area only gets MSA data if it is strictly an MSA, and nonmetropolitan Fair Market Rent areas and rural areas get non-MSA data (see “Definition of areas” documentation).
• We classify a child (regardless of family size) as age 17 and under and an adult as age 18–64 (we used the variable AGE10X). We did not break down children into smaller age groups or by gender because the resulting sample sizes were too small.
• For each Family Budget area, adult out-of-pocket costs are the mean costs (variable TOTSLF10) for adults age 18–64 with private employer-sponsored insurance in one of four regions and the metropolitan classification in that region.
• Child out-of-pocket costs are the mean costs for children age 0–17 with private employer-sponsored insurance in one of four regions and the metropolitan classification in that region in 2009.
• We compute total out-of-pocket costs (OOP) in the following way:

  [(number of parents) * (adult OOP)] + [(number of children) * (child OOP)]

• Since out-of-pocket costs are annual numbers, we divided by 12 to get the total monthly out-of-pocket costs.
• The total out-of-pocket costs were adjusted for inflation to 2012 dollars using the regional breakdowns of the Consumer Price Index-All Urban Consumers for Medical Care (CPI-U-MC) from the Bureau of Labor Statistics (BLS 2013c).
• When computing the mean, a population weight needs to be used (variable PERWT10F).

Other notes

• The health care costs for the 2013 Family Budget Calculator are significantly higher than the health care costs for the previous edition of the Family Budget Calculator (2007 and 2008 update), and these higher health care costs are the main reason why overall 2013 Family Budget Calculator budgets are higher than 2007 and 2008 budgets. A significant portion of the increase in health costs in the 2013 family budgets relative to the 2007 and 2008 budgets is due to a change in methodology.
  • The previous editions did not solely use a benchmark based on employer-sponsored health insurance plans. They also used non-group market premiums as an input. The “sticker prices” of such non-group plans are often significantly lower than the post-underwriting premiums—and these post-underwriting premiums are not available from public data sources. We did not want to allow our health costs to become biased down by reporting less-accurate, pre-underwriting costs. The employer-sponsored health insurance premium benchmark we use instead is fully post-underwriting, reflecting the actual price.
  • Additionally, previous editions made strong assumptions about resources available to households. These previous editions calculated total cost of health care as a weighted average of employer-provided plans including an average employer contribution, as well as the (substantially subsidized) costs of public plans. All of these decisions required making strong assumptions about resources available to families rather than simply costs of a benchmarked necessity.

Other necessities

Our calculation of “other necessities” is derived from the Bureau of Labor Statistics Consumer Expenditures Survey (BLS 2013a). We consider other items of necessity as items that do not fall into the aforementioned categories but are nevertheless necessary for a reasonably safe and comfortable, but modest, standard of living. These items include apparel, entertainment, personal care products and services, reading, education, and miscellaneous items.

Using 2011 data (the latest available data) on families in the bottom 40 percent of the overall income distribution from the 2011 Consumer Expenditures Survey expenditure table “Quintiles of income before taxes,” “other necessities” is the proportion of costs for these items in relation to the costs of food and housing (also from the Consumer Expenditures Survey). In 2011, this was 25.6 percent. We use this figure to calculate “other necessities” in our Family Budget Calculator by taking 25.6 percent of the family budget’s food and housing cost for each family budget as the cost for “other necessities.”

Taxes

The Family Budget Calculator’s components, without taxes, sum to the family’s post-tax income. To calculate the family budget tax component, a pre-tax income level had to be estimated using a tax rate and the post-tax income.

We utilized the National Bureau of Economic Research’s Internet TAXSIM Version 9.2 with ATRA to calculate these tax rates (NBER 2013; Feenberg, Richard, and Coutts 1993). The TAXSIM model accepts 22 input variables, including state, marital status, dependent exemptions, wage income, other incomes, rent paid, child care expenses, and capital gains and losses. We ran the TAXSIM model for each family type across all 615 areas.

Our input variables were (variables not listed were input as zero):

• State
• Marital status (single for one-parent families, married for two-parent)
• Dependent exemptions (one for each child)
• Wage and salary income of taxpayer (entire post-tax family budget for one-parent families)
• Wage and salary income of spouse (for two-parent families, the post-tax family budget was split evenly between the two parents)
• Rent paid (the annual cost of rent for each family budget, which is used to calculate state property tax rebates in certain states)
• Child care expenses (the annual cost of child care for each family budget)
• Number of dependents under age 17 (one for each child)

The TAXSIM model takes these inputs and calculates three outputs: federal tax liability, state tax liability, and FICA tax liability. All of these liabilities are for year 2011 tax law. Note that in 2011 the 2.0 percent payroll tax holiday was in effect; we eliminate this payroll tax cut to move it to its 2013 level. Additionally, the TAXSIM model calculates FICA liability as the full 15.3 percent tax from both the employer and employee side; we cut this in half to more accurately represent the typical taxpayer. Local taxes, such as county- or city-level income taxes, are not included in this model. Sales taxes are also not included.

We made the conservative assumption that the full cost of health care premiums would be excluded from taxation (as there exist substantial tax advantages to employer-sponsored premiums). As aforementioned, it is not accurate to simply input the post-tax family budgets as the wage incomes and use the TAXSIM output as the tax rates. To correct for this, we input the post-tax family budgets and obtained the tax rates and established these as a lower floor for tax rates (because the pre-tax incomes will almost always be higher than these post-tax incomes, these tax rates must be lower given our assumptions about sources of income and the income ranges we are considering). We then established an upper bound of tax rates by taking the post-tax family budget and multiplying it by 1.25 and inputting these budgets into the TAXSIM model.   Of the 3,690 family budgets, 158 budgets fall outside our initial upper and lower bounds. For these “out of bounds” observations, we progressively widen the upper and lower bounds until all observations fall within them. Of the 19 that fall below our lower bound (which is possible because refundable tax credits can result in negative effective tax rates), we decrease the bound to be 90 percent of a family budget, and this adjustment captures all “below-bound” observations. Of the 139 that fall above our upper bound, we increase the bound to 135 percent of the post- tax family budget for 126 of the budgets, 140 percent for 9 family budgets and 145% for 4 of the family budgets.

Once we had the lower and upper bounds of tax rates, we calculated an accurate average of these tax rates using the following weighting procedure:

1. Multiply the lower- and upper-bound inputs to TAXSIM by (1 – tax rate calculated by TAXSIM).
2. Calculate the difference between the actual post-tax family budget and the lower bound calculated in step one.
3. Calculate the difference between the upper bound and the actual post-tax family budget calculated in step one.
4. Calculate the difference between the upper bound and the lower bound calculated in step one.
5. Calculate the weight for the lower bound, which is equal to the upper post-tax budget difference divided by the upper–lower difference.
6. Calculate the weight for the upper bound, which is equal to (1 – lower-bound weight from step five).
7. Multiply the lower-bound tax rate from TAXSIM by the lower-bound weight from step five.
8. Multiply the upper-bound tax rate from TAXSIM by the upper-bound weight from step six.
9. Add these two weights together to get the final, weighted tax rate.

The final tax rate calculated in step nine is then applied to the post-tax family incomes (multiply that income by 1 + the tax rate) to obtain a pre-tax income. The difference between the pre- and post-tax incomes is the annual tax bill for the family budget unit.

About the authors

Elise Gould joined the Economic Policy Institute in 2003. Her research areas include employer-sponsored health insurance, inequality and health, poverty, mobility, and the employer tax exclusion. She has published her research in a range of venues from academic journals to general audience periodicals, been quoted by various news sources, and testified before the U.S. Congress. Also, she teaches health economics and econometrics to graduate students at Johns Hopkins University and The George Washington University, respectively. She holds a master’s in public affairs from the University of Texas-Austin and a Ph.D. in economics from the University of Wisconsin-Madison.

Nicholas Finio is a research assistant at the Economic Policy Institute. He provides research assistance to EPI’s economists on a variety of topics, including health insurance, wages and living standards, macroeconomics, and the labor market. His areas of interest and research include wage inequality and the labor market. He has a B.A. in economics from Gettysburg College.

Natalie Sabadish is a research assistant at the Economic Policy Institute, providing support to EPI’s economists in a variety of policy areas. She also works with the Economic Analysis and Research Network (EARN), answering data-related inquiries and coordinating a nationwide internship program. She has previously interned at the Delaware Department of Labor and the Keystone Research Center. She has a B.S. in economics from the University of Delaware.

Hilary Wething is a senior research assistant at the Economic Policy Institute. She provides support to EPI’s economists and policy analysts for a variety of topics, including health insurance, labor markets, inequality, living standards, and macroeconomics. She has a B.S.B.A. in mathematics and economics from Creighton University.

References

Bureau of Labor Statistics (BLS). Consumer Expenditure Survey. 2013a. Annual Calendar Year Tables Current Expenditure Tables. http://www.bls.gov/cex/#tables

Bureau of Labor Statistics (BLS). Consumer Price Index Program. 2013b. All Urban Consumers (Current Series) –Child Care and Nursery School [database]. http://www.bls.gov/cpi/home.htm#data

Bureau of Labor Statistics (BLS). Consumer Price Index Program. 2013c. All Urban Consumers (Current Series) – Medical Care [database]. http://www.bls.gov/cpi/home.htm#data

Bureau of Labor Statistics (BLS). Employee Benefits Survey. 2013d. “Table 14. Medical Care Benefits, Family Coverage: Employer and Employee Premiums by Employee Contribution Requirement, Private Industry Workers, National Compensation Survey, March 2012.” http://www.bls.gov/ncs/ebs/benefits/2012/ownership/private/table09a.htm

Carlson, A., Lino, M., and Fungwe, T. 2007. The Low-Cost, Moderate-Cost, and Liberal Food Plans, 2007. U.S. Department of Agriculture, Center for Nutrition Policy and Promotion. http://www.cnpp.usda.gov/Publications/FoodPlans/MiscPubs/FoodPlans2007AdminReport.pdf

Child Care Aware of America. 2012. Parents and the High Cost of Child Care: 2012 Report. http://www.naccrra.org/sites/default/files/default_site_pages/2012/cost_report_2012_final_081012_0.pdf

Federal Highway Administration (FHA). 2009. National Household Travel Survey (NHTS). Tabulation created on the NHTS website: http://nhts.ornl.gov.

Feenberg, Daniel Richard, and Elizabeth Coutts. 1993. “An Introduction to the TAXSIM Model.” Journal of Policy Analysis and Management, vol. 12 no. 1, pp. 189–194. http://users.nber.org/~taxsim/feenberg-coutts.pdf

Internal Revenue Service (IRS). 2012. IRS Announcement 2011-116,“2012 Standard Mileage Rates.” http://www.irs.gov/pub/irs-drop/n-12-01.pdf

National Bureau of Economic Research (NBER). 2013. TAXSIM Model Version 9.2 with ATRA. http://nber.org/~taxsim/taxsim-calc9/

Office of Management and Budget (OMB). 2009. “Update of Statistical Areas Definitions and Guidance on Their Uses.” OMB Bulletin No. 10-02. http://www.whitehouse.gov/sites/default/files/omb/assets/bulletins/b10-02.pdf

U.S. Census Bureau. 2013. Census Regions and Divisions of the United States. https://www.census.gov/geo/www/us_regdiv.pdf

U.S. Department of Agriculture Center for Nutrition Policy and Promotion (USDA). 2013. “Cost of Food at Home: U.S. Average at Four Cost Levels” [data tables, June 2012 annual average]. http://www.cnpp.usda.gov/USDAFoodCost-Home.htm

U.S. Department of Health and Human Services (HHS). 2013a. Medical Expenditure Panel Survey. MEPS HC-138: 2010 Full Year Consolidated Data File [microdata]. http://meps.ahrq.gov/data_stats/download_data_files_detail.jsp?cboPufNumber=HC-138

U.S. Department of Health and Human Services (HHS). 2013b. Medical Expenditure Panel Survey. “Table IX.A.2 (2011), Average Total Premiums and Employee Contributions (in Dollars) for Private-Sector Establishments for Areas Within States: United States, 2011.” http://meps.ahrq.gov/mepsweb/data_stats/summ_tables/insr/state/series_9/2011/tixa2.pdf

U.S. Department of Housing and Urban Development (HUD). 2013. Fair Market Rents dataset, county-level data file. http://www.huduser.org/portal/datasets/fmr.html