Abstract
This article examines how firms respond to federal affirmative action regulation when it is temporary. As intended, affirmative action increases a firm’s black share of employees. Strikingly, the black share continues to grow at a similar pace even after a firm is deregulated. This persistence is driven in part by affirmative action inducing employers to improve their methods for screening potential hires.
In 2019, the topic of affirmative action policy is as electrified and fraught with controversy as it has ever been. Originally designed to increase diversity among employees, students, politicians, or businesses by advantaging candidates from under-represented social groups, these rules have become a frequent target of criticism. They are a part of a national conversation about disadvantaged groups as well as the fight over defining exactly which groups are, in fact, disadvantaged. Detractors have been emboldened by a surge of energy during the Trump administration, as TV pundits and talk radio hosts complain that any regulation that incentivizes diversity results in the victimization of white males. In 2018, Fox News host Tucker Carlson aired a segment with a bold graphic suggesting that affirmative action policies represent “America’s War on Standards.” 1 However, there are many reasons to look at affirmative action positively. The problem is that our understanding about the effectiveness of these regulations has never painted an accurate picture about how they do—or do not—work.
From detractors to apologists, there is a broad spectrum of opinion on this subject, and the differences between the beliefs of those on both sides (and in the middle) go beyond passion on the topic of political affiliation. Very often the key lies in how they perceive that the mechanism of affirmative action works: an incentive for workers, unfair quota-filling (i.e., “America’s War on Standards”), or changing the way that people see members of under-represented groups. In spite of the many frames through which this issue can be viewed, one of its greatest hurdles, even among advocates, is the seemingly well-proven argument that affirmative action policies just do not work in the ways that we wish they would. Often introduced or supported only as temporary remedies for existing social inequalities, 2 these regulations are rarely advocated as the long-term solution to inequity, even by ardent believers in their merits. The hope is that a temporary affirmative action program that enhances diversity and reduces inequality can persistently alter outcomes that unfairly disadvantage under-represented groups. That, in itself, is an issue because there is little evidence on the topic to demonstrate that it is effective. Therein lies the problem. If these policies should only be enacted temporarily and yet temporary affirmative action may not work in the long-term, then what is the point?
The history of affirmative action in the United States has been one of courtroom battles, executive orders, and bureaucratic skirmishes that still affect millions of people. On the education front, Harvard currently holds the nation’s attention. The university will likely see a ruling from the Supreme Court regarding its affirmative action admissions policies. The plaintiffs complain that the Ivy League school is intentionally curbing its admission of Asian American students. The case will likely bring about ramifications for affirmative action policies at all American universities. Regardless of the outcome, the case embodies the precariousness and controversy inherent to affirmative action policies.
Although often confused with each other, the affirmative action regulations that are mandated in labor markets are distinct in policy and enforcement from those in the realm of education. On the labor side, these rules are steeped in as much heated debate as their education counterparts. For instance, in the past 2 years alone, we have seen a litany of civil rights rollbacks, many of which affect the implementation of affirmative action in employment contexts. In 2018, the Justice Department argued that discrimination based on sexual orientation is not covered by the Civil Rights Act of 1964, showing the relevance that this act, which is the inspiration for federal affirmative action policy, still exerts on today’s marketplace. For the fiscal year of 2018, the Trump administration proposed cutting the Office of Federal Contract Compliance Programs (OFCCP), which was seen by some as an “effort to minimize civil rights efforts.” 3 This move echoed one by the Reagan administration decades ago, in which the agency charged with overseeing affirmative action rules in the workplace was so significantly defunded that it effectively took the teeth out of the enforcement of these policies on federal contractors.
All of the fight to keep affirmative action in place might lead one to believe that there is overwhelming evidence for positive outcomes from these programs. But at least in the area of temporary affirmative action enforcement, there has been no significant study that could prove that they have even made a difference. This has been the case as scholars examine the effects of Executive Order 11246, signed by Lyndon B. Johnson in 1965. The order prohibits federal contractors who do more than $10,000 worth of business with the government each year from certain forms of discrimination, and it mandates contractors to take affirmative action to ensure the employment of applicants and the fair treatment of workers. It is also the same order that was declawed by the Reagan administration in the 1980s.
As is true for the diverging theories on why and how affirmative action works as a mechanism for change, the answer to whether or not temporary regulation actually works seems to be in how we look at it. For this article, I study the dynamic effects of Executive Order 11246 from a new vantage point. I demonstrate that a transitory intervention can have persistent effects through employer-level mechanisms that affect the racial composition of employee flows. For example, temporary affirmative action can kick-start change in an employer’s recruitment and screening practices or in the composition of its referral applicants. I label these investments to improve an employer’s ability to screen potential workers as “screening capital.”
By shifting the focus to the employer-level and implementing an event study research design, I found results that should be heartening to advocates of affirmative action policies and that should provide a path to composing a more diverse workforce for non-regulated firms who, for a variety of reasons, set a goal of more equitable hiring practices. In fact, affirmative action sharply increases an establishment’s black share of employees, with the share continuing to increase over time. Five years after an establishment is first subject to the regulation, its black share of employees increased by an average of 0.8 percentage points. To put this magnitude in perspective, note that a 0.8 to 1.3 percentage point increase of black share of the U.S. workforce would eliminate the black-white jobless gap over this period.
Strikingly, I find that the black share of employees continues to grow even after an employer is deregulated. In the 5 years after an establishment is last observed as a contractor, its black share of employees increases by an additional 0.8 percentage points. This persistence is evident more than a decade following deregulation. The observed persistence is difficult to reconcile with existing economic models of affirmative action, which focus on human capital channels. 4 In particular, because the policy variation exploited here is not consistent across individual employers, it should have minimal effects on the human capital investment incentives workers face in the broader labor market. Rather, any response is likely driven by changes at the employer level. Promisingly, these types of changes are available to all firms that wish to diversify their workforces.
Some History and Context
Affirmative action, as a concept, is always inextricably linked with the political and social mind-set of the time in which it is implemented. The evolution of these policies, from their entry into legislation to the judicial and financial hurdles that they have faced in the public arena, follows the arc of America’s attitude on race and equity. Executive Order 11246 presents a fascinating opportunity for the study of affirmative action in the workplace. The order is part of the story of the changing (and sometimes unchanging) American attitudes about race—and the role of the federal government in striving for standards of equality. It is arguably one of the most controversial labor market interventions in U.S. history, and yet we know surprisingly little about its effect on the labor market. 5 The order is rooted in Title VII of the Civil Rights Act of 1964, a measure that still shapes hiring practices to this day. To understand how an executive order from the Johnson era that only affects a limited scope of employers in the United States can still have far-reaching implications, it is important to understand the history and context of its inception.
Affirmative action policies have roots in the culture wars and landmark change of the 1960s. Employees, voters, and students today are still directly affected by the legacy of its marches and calls for action through the enforcement of the Civil Rights Act of 1964. John F. Kennedy first proposed sweeping legislation to battle Jim Crow laws and other social injustices. Lyndon B. Johnson continued this work, as Southern states in the U.S. Senate fought hard against the new law, and almost succeeded in their attempts to thwart its passage. The battle that ensued resulted in the civil rights guidelines that have come to shape the United States and in cultural battles that still headline our news.
On September 24, 1965, President Lyndon B. Johnson signed Executive Order 11246 to formally set the rules regarding discrimination by large federal contractors and put affirmative action policies in place to guide their hiring practices. While Title VII of the Civil Rights Act of 1964 outlaws discrimination on the basis of race, color, religion, sex, or national origin in all but the smallest private firms, Executive Order 11246 requires that firms with federal contracts make active efforts to prevent discrimination. This regulation applies to firms that have sizable contracts or subcontracts with the federal government. According to United States Department of Labor estimates, these firms employ about a quarter of the U.S. workforce, 6 so even though the employers affected by this act are narrow in scope, the number of employees who are touched by it remains significant.
Not in the Affirmative
The seismic shift that characterized civil rights during the Reagan years cannot be underestimated. This era introduced the language of some popular arguments against affirmative action that are still in use today. Decrying the use of quotas and timetables in the hiring process, some members of the Reagan administration set about dismantling Executive Order 11246. These attempts were unsuccessful in terms of repealing the act itself, so detractors instead began gutting the processes through which it was enforced. For instance, the Reagan administration slashed the budget for the OFCCP so that the order became more difficult to enforce, along with other measures intended to mute the effects of regulation.
It is worth noting that these changes continue in the contemporary labor market. The Trump administration’s attempts to eliminate the OFCCP might have ushered in an era of increased erosions in the enforcement of affirmative action. Speculation abounded that the office would be more lenient on employers under the administration. This situation took a surprising turn when record-breaking enforcements and fines shook federal contractors starting in 2017. This demonstrates the precariousness and the fluidity that have persisted in the history of affirmative action policies in the United States.
How Affirmative Action Affects Hiring Processes
In national conversations, we often see an oversimplification of the issue of affirmative action, both as a process to be implemented in workplaces and as a philosophical approach to bring equity to the labor market. The Reagan-era talk of quotas still affects the way that affirmative action is viewed by many in the United States. There is a misconception that the sole function of affirmative action is to make sure that a given number of under-represented candidates are employed by a set of given firms. According to this logic, affirmative action would make up for discriminatory hiring practices by ensuring that an adequate number of under-represented employees attain viable work. Although there is some truth to this view, its over-simplification of the logic of affirmative action tends to benefit detractors who argue that these policies hinge on the act of taking jobs away from advantaged candidates. If affirmative action were merely a numbers and quotas game, then temporary interventions would be very unlikely to have a long-term impact. It would only benefit members of under-represented groups who gain employment through these programs before they cease operation.
In the literature of affirmative action, the view is more nuanced. These policies can be viewed as an impetus for individuals and groups who might benefit from their implementation or—more accurately—suffer from their absence. Existing economic models of affirmative action argue that such policies can have persistent effects by incentivizing members of disadvantaged groups to better invest in their own skills. The idea is that, in a world where firms make hiring decisions based on group stereotypes, members of disadvantaged groups may have no incentive to invest in their own skills, because their skills won’t be rewarded by the market. This can generate a “self-fulfilling” prophecy in which firms believe members of disadvantaged groups are not skilled, and consequently do not hire them. From this perspective, fears of prejudiced hiring practices hold the potential to disincentivize members of disadvantaged groups in a way that makes negative stereotypes accurate. An affirmative action policy can correct this problem by forcing firms to hire from disadvantaged groups. This oversight can correct the incentive problem faced by minority workers, because they will (hopefully) invest in building necessary skills based on the perception that they have a chance of being hired. This makes sense in the context of a labor market-wide affirmative action policy, because if one’s chances of being hired anywhere change dramatically, this can plausibly affect an individual’s behavior when they are investing in their own skills, for instance, in high school or college or training programs.
Rather than focusing on the labor market or changes in employee behavior, I examine employers who observe the black share of their employees, both before and after regulation. It is through this approach that I was able to demonstrate the persistent effects of temporary affirmative action interventions. To understand the data that are unique to this study, it is important to outline the specific regulatory constraints that firms must meet in order to comply with existing federal hiring guidelines.
Rules, Regulations, and Outcomes
Although many firms voluntarily participate in affirmative action hiring practices and all firms with more than 25 employees are prohibited from taking part in anti-discriminatory practices, only federal contractors meeting certain criteria are required to hire in accordance with United States Department of Labor affirmative action policy. Executive Order 11246 still guides many of these practices. While firms are required to comply with the order by recruiting and hiring women and all minorities, research on affirmative action has demonstrated that there has been little improvement for the employment prospects of many groups. For instance, previous studies have found that affirmative action regulation has had a negligible impact on female employment. 7 Leonard posits that this finding may result from the historical prioritization of minority employment in enforcement or the secular growth of female employment.
I limit my study to the effect of Executive Order 11246 on the employment of black workers, one of its original targets, the largest minority group over my period of study, and a group that is typically the focus of affirmative action research. 8 I focus on firms that are temporarily subject to affirmative action regulation. Using administrative data from 1978 to 2004 which notably include the Reagan-era cuts and their aftermath, I examined temporary federal contractors that were regulated by affirmative action policies.
Since employers are the focus of my study, a thorough explanation of how Executive Order 11246 affects them should inform any analysis of the rule’s effectiveness. This order mandates that firms with 50 or more employees and holding a federal contract worth $50,000 or more meet specific requirements. These contractors must identify underutilization of minorities and women in any occupation group by considering “the availability of minorities having requisite skills in an area in which the contractor can reasonably recruit.” 9 Contractors are also required to make “good faith” efforts to rectify underutilization, including the use of numerical goals with timetables. Broadly speaking, affirmative action mandates that federal contractors make a real effort to employ minorities at rates (at least) proportional to the shares of the local and qualified workforce, although local and qualified are not guided specifically. This regulation applies to all establishments under the firm, regardless of whether the particular facility is executing any portion of the contract.
The mechanisms through which these regulations are enforced have evolved and shifted over time. Initially, 13 federal contracting agencies—for example, the Department of Defense and the General Services Administration—were responsible for enforcing affirmative action regulation. Enforcement responsibilities were generally assigned on the basis of a contractor’s industry, irrespective of the agency contracting with the firm. 10 Although the Office of Federal Contract Compliance was established in the United States Department of Labor to advise and coordinate enforcement activities among firms, there was wide variation across agencies in both the scope and quality of their enforcement. The lack of consistency in reporting and regulation during this period affects the clarity and reliability of available data. In 1978, the Carter Administration consolidated affirmative action regulation enforcement activities under the renamed Office of Federal Contract Compliance Programs (OFCCP), the same office that faced closure under the Trump administration last year. In my analysis, I use only data from 1978 onward, so employer reporting to the OFCCP is crucial to this study.
Two sources of data completed by employers are key to implementation and study of affirmative action and its function in the labor market. As part of the Civil Rights Act of 1964, private sector firms meeting certain size requirements are required to complete EEO-1 forms annually and submit them to the Equal Employment Opportunity Commission (EEOC). 11 Firms are required to report their overall racial, ethnic, and gender composition and the racial, ethnic, and gender composition of each of their establishments meeting size requirements, disaggregated by nine major occupation groups. 12 Employers are instructed to base demographic classifications on worker self-identification or visual inspection, where the former is the preferred method. There is no distinction between race and ethnicity in the data; in particular, Hispanic workers are classified as a distinct, non-overlapping group. While a firm is a contractor, it is also required to write an Affirmative Action Plan (AAP) for each establishment. An AAP must describe the organizational structure of the firm and establishment, identify underutilization of minorities by job group, and detail strategies, goals, and timetables for eliminating underutilization in the next year and beyond. Each AAP must be updated annually while the firm is a contractor. Contractors must also “maintain and have available records for each job on all applicants, hires, promotions, terminations, and any other selection decisions” disaggregated by minority group. 13
To enforce the regulation, the OFCCP conducts compliance evaluations, reviews of a small fraction of covered establishments each year (about 1% of covered establishments in 2004) 14 to determine whether their AAPs are sufficient and whether they made good faith efforts to implement their plans. Compliance evaluations consist of a desk audit and a possible site visit. As part of the desk audit, compliance officers determine whether an establishment’s AAP is adequate and whether the establishment made sufficient efforts to improve minority utilization, relying on the submitted personnel data and EEO-1 forms. If potential violations are identified during the desk audit, the OFCCP may conduct an on-site review at the establishment. During a site visit, compliance officers further investigate potential violations, verify the firm’s efforts to implement its AAP, and obtain information needed to work with the contractor to resolve any violations. Officers accomplish this in part by inspecting the contractor’s facilities and reviewing its personnel files. 15
If the OFCCP finds that a contractor is not in compliance, it will seek a letter of commitment for minor violations or a conciliation agreement for major violations. Some of these agreements include financial settlements that involve back pay to alleged individual victims of discrimination. In 2004, the OFCCP collected $34.5 million from settlements on behalf of more than 9,000 workers. If the OFCCP and a contractor fail to resolve affirmative action violations, the OFCCP may take legal actions to penalize the contractor. The ultimate punishment for a contractor is to be debarred from doing business with the federal government, sometimes permanently. However, this outcome is quite rare. Only 43 companies were debarred up to 2001. 16 About half refused to develop an AAP or submit personnel data, while the other half did not make sufficient efforts to implement plans or violated an existing conciliation agreement. About 60% of debarred firms were later reinstated, and for those contractors the median period of debarment was 9.5 months. 17
Since my study focuses on regulated contractors and I found an upward trend in their black employee share, it might be tempting to assume that these contractors were more likely to employ a greater number of workers from disadvantaged groups before they gained contractor status. But, critically, the earlier racial composition of a firm does not affect whether it acquires a contract or not. This is because awards of federal contracts are administered separately from affirmative action enforcement. Large contracts are one potential exception because these firms may be subject to “pre-award” compliance evaluations, which take place before a firm can formally initiate the contract. 18 But in practice, very few contracts are large enough to require pre-award compliance evaluations, and they are even less common for the firms I focus on in the analysis, which are not perennial contractors. Moreover, there is no requirement for an establishment to be in compliance when it does not have a federal contract.
These rules affect outcomes of the hiring process and help shape the racial makeup of employees who work for regulated federal contractors. From a research standpoint, it is also important to note that they affect the decision-making process of contractors. From the ability to navigate these complex regulations after the allocation of contractor status to maintaining good standing and avoiding penalties, firms must take these mandates into account, even before they apply for a federal contract.
A Deeper Dive into the Research
We have been studying affirmative action in the United States almost since its inception. After all, such a large-scale experiment with the American workforce did not come with guaranteed results. Existing research tends to focus on the potential for affirmative action to reduce inequality by incentivizing members of disadvantaged groups to invest in their own skills. 19 This argument hinges on the accumulation of human capital for disadvantaged groups as a key to labor market-wide change. From a social perspective, one might assume that affirmative action policies should have some effect through altering attitudes and misguided notions about disadvantaged workers. For example, we might expect affirmative action to work by eradicating prejudiced views or exposing hiring managers to employees from disadvantaged groups.
The absence of persistent effects from temporary measures would reflect negatively on affirmative action as a vehicle to drive equity in the workplace, and, perhaps more problematically, demonstrate that hiring managers do not update stereotypes about members of disadvantaged groups after working with them. In fact, there are some studies that suggest affirmative action can backfire and exacerbate negative stereotypes. 20 Alternately, if disadvantaged members of the workforce continue to pursue the skills and education that make them more attractive to hiring managers because of affirmative action regulation, these policies should have a legacy in a more representative workforce. Altogether, the theory on how temporary affirmative action policies will work is ambiguous, and there is little empirical evidence to distinguish between narratives.
Previous research has substantially understated the causal impact of affirmative action regulation on federal contractors. To explain these outcomes, it is necessary to note a central innovation in my approach: the comparison of “ever” contractor to “never” contractor. It seems that we should be able to compare firms that are currently contractors to firms that are currently non-contractors and observe the differences in order to see if these policies make an impact. On the surface, this approach makes perfect sense. If a new set of policies go into effect and they are only applied to federal contractors that meet a narrow set of criteria, we should be able to test the outcomes of these rules by comparing those contractors to unregulated firms. According to this logic, firms that are regulated should demonstrate the effectiveness of these policies by maintaining a higher share of black employees than firms that are not regulated.
In this view, firms that are regulated and firms that are not are seen as closed ecosystems that do not overlap in any way. In actual fact, regulated contractors represent a group that is almost always in flux, and an acknowledgment of the fluidity of contractor status is the only way to truly understand how the mechanism of affirmative action functions in the real world.
A New Model
The set of companies from which the government purchases goods and services is always changing, which means that comparisons of contractors to non-contractors is problematic. Some non-contractors who previously operated under contractor status may still experience the effects of regulation. A better tool for assessing the regulation’s impact is a comparison of employers that have ever been contractors and those that have never been contractors. If affirmative action policies continue to affect the minority share of employees after they are no longer regulated by Executive Order 11246, this unanticipated result reveals an additional shortcoming in our previous approach to the study of this topic because it means that the full impact of the regulation has been drastically understated. Although these two issues are intertwined, each one exerts a distinct impact on the way that previous models have underestimated the persistence of affirmative action regulation on federal contracts. First, by grouping previously regulated firms with firms that are never regulated, persistence in the black share of employees will skew results when we compare them with currently regulated firms. That’s because the regulation may still affect the black share of employees at previously regulated firms. Comparing ever contractors and never contractors, though, requires a new approach.
An event study research design can approximate an experiment by looking at how employment outcomes change at a firm following some event—in this case, becoming a regulated contractor—relative to otherwise similar firms that do not experience the event (firms that do not become regulated contractors). If we see that firms that will become contractors (“eventual contractors”) look similar to other non-contractors prior to the event, but then sharply more diverse precisely following the event, that provides strong evidence that the regulation itself is causing this divergence.
The work of McCrary 21 and Miller and Segal 22 both offer precedents for my work in terms of research and design. In the 1970s, federal courts imposed racial hiring quotas on municipal police departments and other law enforcement agencies. When compared with national trends, these measures made a significant impact on the black share of workers employed at these agencies. 23 This outcome might be expected, given that we know rigorously enforced policies tend to boost the minority share of employees at regulated firms. Although these court orders were terminated because of the shifting attitudes about affirmative action and successful reverse discrimination challenges, the authors found that the minority share of employees did not, in fact, erode, although it did stagnate over time. 23
Screening Capital
As previously noted, I found that federal contractors did not see a decline in the black share of their employees after deregulation. Instead, I observed an increase in the black share of employees after temporary affirmative action intervention. I posit that investments in screening capital account for many of these observed trends. 25 Affirmative action can drive employers to make (partially) irreversible screening investments and that this may be a driver of persistence. My model also has two main predictions in reference to screening capital investments and diverse hiring practices, which are supported by the data. First, the model suggests that screening investments see a higher return because of regulation, and, second, that between-group differences in hiring rates are reduced with screening investments. These predictions suggest somewhat of a symbiotic relationship between affirmative action policies and screening capital investments. Screening capital provides an effective way to comply with affirmative action regulations, and affirmative action pushes firms to make those screening capital investments. These findings are not only relevant to the way that we view affirmative action policies; they can act as guideposts for non-regulated firms that have a goal of increased diversity or more effective screening practices.
If firms continue to increase the black share of employees post-deregulation, they must view it as profitable to their business. I argue that investments in screening capital, including the training and employment of personnel specialists, the development of job tests, harnessing referral networks, utilization and relationship building with intermediaries (employment agencies, schools), and learning through practice and experimentation, represent a key to the persistent effects of temporary affirmative action intervention. Previous research has posited that employers are often less able to judge the quality of minority candidates than they are for majority candidates. My screening model helps demonstrate that improved screening precision benefits candidates from under-represented groups. Employers who prefer to hire the best employee must choose from a set of workers in a pool of candidates. If they are unclear about each one’s productivity, it becomes difficult to judge the quality of minority candidates, especially if employers lack experience in this area. Investment in improved screening capital helps employers make better decisions during the screening process and to decrease the hiring gaps between groups. I interpret “screening” broadly as choosing the “best” candidates from a set of potential workers, including both recruitment and selection components of the hiring process. Regulated employers use more screening methods than otherwise comparable unregulated employers, particularly “formal” methods such as personnel specialists, job tests, credential checks, and intermediaries. 26 Employer size is also an important predictor of more equalized group representation among employees. This supports the thesis that investments in screening practices play a crucial role in equitable hiring since previous research has documented that larger employers invest more time and money in screening capital. 27
How It Works (and How It Doesn’t)
It is important to note that before the regulation event, establishments in the regulation and overlapping samples exhibit employee black shares that are very similar to non-contractors. But the black share of employment continues to grow after the deregulation event at a rate that is similar to the one that emerges following the initial regulation (see Figure 1). The results in this study suggest that affirmative action has a sizable impact on the personnel composition of regulated establishments.
Regulation and black employment growth.
Anticipatory Behavior and Multiple Channels
The issue of anticipatory behavior is a concern for anyone interpreting such results. For instance, employers in the deregulated sample may believe that they will pursue a federal contract at a later date and assume that an increased black share will help their case or they could try to safeguard against adjustment costs by hiring practices that were first influenced by regulation. Concerns about anticipatory behavior as a factor in this process seem to be negated by the actual circumstances that surround the allocation of federal contracts. As previously noted, these contracts are awarded and regulated separately by the OFCCP, so there is no immediate advantage for firms who adhere to regulation prior to winning a federal contract. This observation is borne out in the data. For instance, ever contractors and never contractors face a similar likelihood of acquiring contracts. Similarly, data do not support the hypothesis that option value drives persistence: for previously regulated employers, the degree of persistence is independent of whether they win a contract at a later date. Size declines among firms and occupation changes among minority employees similarly do not explain black share increases after deregulation.
There are other potential channels through which temporary affirmative action may persistently increase black employment. Exposure to workers from under-represented groups may spur employers to correct negative stereotypes they held about those groups. If employers rely on referrals made by their current employees when hiring, then temporary affirmative action can have persistent effects by diversifying the referral pool. 28 It is nearly impossible to distinguish between all potential mechanisms that increase the black share of employees over time because of limitations on available data. My approach has been to focus on one potential channel and demonstrate through available evidence that this channel is, indeed, empirically relevant. However, given the variation in regulation explored in this research, it is unlikely that the increase in black employment I observe is driven by workers accumulating human capital. The regulation of a particular firm should have little influence on incentives for workers who presumably search for work in a broader market. Regardless of the channel, the fact that a temporary intervention spurs a long-term impact on a given employer’s trajectory suggests that there exist multiple equilibria for the racial composition of its workforce.
Conclusion: New View, Old Problem
If we change the way we compare contractors and non-contractors, the results tell an entirely different narrative than the one we are used to hearing. Rather than simply quota-filling or incentivizing skill investments, affirmative action policies seem to work through an alternate mechanism. If it is not driven by changes in the skills of disadvantaged workers or in checking off boxes by hiring managers, what force acts as a catalyst for this change? Although multiple factors may be at work in this long, complex process, screening capital investments play a vital role.
These findings should help reshape our ideas about whether or not affirmative action “works” as a way to introduce more equity to the hiring process. Now, the answer is not only firmly in the affirmative—it can be seen as part of an elaborate and perhaps mysterious process that has acted as a near-constant force on portions of our labor market for decades. Out of the long debate about the fairness of policies that aim for fairness, there has emerged a cultural battle that is often ugly and is, even at its most aspirational, sometimes without much hope. If temporary affirmative action has, indeed, failed to affect the minority share of employment in the United States, even ardent supporters might want to throw in the towel. By changing the way we look at the data, we can explore important drivers of change and begin to understand the mechanisms through which these policies are most effective.
For firms who wish to bring more equity to their workforces, this new narrative about affirmative action regulation offers a pathway to more diverse hiring practices. Improved screening precision seems to act as a roadmap for those who have a goal of increasing the minority share of their employees. While these prospects should inspire some degree of optimism, it is important to remember that challenges have defined affirmative action throughout its brief historical span. These policies have always been perched on the edge of conflicting American beliefs about race, fairness, and opportunity. My research has demonstrated that affirmative action policies not only work—they work in a way that is different from the paradigms through which we have previously seen this issue. But history has repeatedly shown us that the road to increased fairness in our hiring practices is a long, uphill one.
Footnotes
Acknowledgements
Based upon Conrad Miller, “The Persistent Effect of Temporary Affirmative Action,” American Economic Journal: Applied Economics, 9/3 (July 2017): 152-190. Copyright American Economic Association; reproduced with permission of the American Economic Journal: Applied Economics.
Notes
Author Biography
Conrad Miller is an assistant professor at the University of California, Berkeley, Haas School of Business, and a faculty research fellow at the National Bureau of Economic Research (email: