Boosting Wages Archives

Addressing gender and racial disparities in the U.S. labor market to boost wages and power innovation

Posted on January 14, 2021May 13, 2021 by Maria Monroe

This essay is part of Boosting Wages for U.S. Workers in the New Economy, a compilation of 10 essays from leading economic thinkers who explore alternative policies for boosting wages and living standards, rooted in different structures that contribute to stagnant and unequal wages. The authors in the new book demonstrate that efforts to improve workers’ access to good jobs do not need to be limited to traditional labor policy. Policies relating to macroeconomics, to social services, and to market concentration also have direct relevance to wage levels and inequality, and can be useful tools for addressing them.

To read more about Boosting Wages for U.S. Workers in the New Economy 20 and download the full collection of essays, click here.

Overview

Allowing people to pursue their talents and interests is essential to individual well-being, but it also is a crucial part of any market economy. U.S. laws and society too often limit people from developing their potential, harming those individuals and the overall economy in the process. Policies that encourage more equal participation for women and African Americans could boost economic growth, reduce inequality, and power innovation.

The costs of misallocating talent in the U.S. economy are increasingly being identified in the economics literature. Four scholars at the University of Chicago’s Booth School of Business and Stanford University recently analyzed the gender and racial distribution of workers in highly skilled occupations over the past 50 years.¹ They show that the change in the occupational distribution since 1960 suggests that a substantial pool of innately talented women and African Americans in 1960 were not pursuing their comparative advantage, and that this misallocation of talent affects aggregate productivity in the economy. They find that one-quarter of growth in aggregate output from 1960 to 2010 can be explained by an improved allocation of talent.

Fact sheet: Addressing gender and racial disparities in the U.S. labor market to boost wages and power innovation Download

Other recent research finds that women’s underrepresentation in engineering and in jobs involving development and design explains much of the patent gap between men and women.² Closing this divide could increase per capita U.S. Gross Domostic Product by 2.7 percent.³ And using data from the National Science Foundation’s Survey of Earned Doctorates, I and my co-author Yanyan Yang, now an economist at the University of Massachusetts Boston, estimate that GDP per capita could rise by 0.6 percent to 4.4 percent if more women and African Americans were included in the initial stages of the innovation process.⁴

Whatever their source, gender and racial disparities exist at each stage of the innovation process. From education and training to the practice of invention, and then onto the commercialization of invention, these disparities are costly to the U.S. economy. These disparities also can lead to increased income and wealth inequalities at each stage of the innovation process for those who are unable to participate fully. Reducing barriers to participation in the innovation process could affect productivity, as well as both the level and the distribution of income.

In this essay, I demonstrate why innovation and the commercialization of invention are both desirable and necessary in modern economies and why the benefits of doing so are not evenly distributed among those who are powering innovation today.⁵ Despite numerous initiatives to train and cultivate innovators, women and African Americans continue to participate at each stage of the innovation process at lower rates than their counterparts. I then propose four policies to close these race and gender innovation divides, particularly at the education and practice-of-invention stages:

Improve mentoring programs with additional government policymaking and fiscal support
Facilitate early education exposure to invention opportunities
Engage in blind patent reviews by patent examiners
Address the climate in high-tech workplaces to attract and retain women and African Americans in the places where invention and innovation happen

Women and African Americans have not enjoyed their proportionate share of innovation’s ample economic benefits.⁶ Fundamentally, innovation is critical for economic growth, wealth generation, and higher living standards. Innovation can substantially affect each component of economic growth—labor, capital, and total factor productivity.

Why the innovation process is important to U.S. economic growth and well-being

From a number of perspectives, innovation is a good thing. Economists have long recognized that the generation and implementation of ideas drives economic growth.⁷ Historians also have demonstrated the positive relationship between innovation, industrialization, and economic activity in studies of early U.S. inventors and entrepreneurs and in the creation of the patent system.⁸ Statisticians provide additional evidence of the importance of innovation to the U.S. economy: From 1960 to 2013, the number of workers in innovation jobs grew 3 percent annually, compared to 2 percent growth for the broader workforce.⁹

Economists measure innovation’s contribution to the economy with increasing precision, and it is clear that innovation’s importance is growing.¹⁰ In 2017, the National Science Foundation calculated that there were roughly 7 million to 25 million workers engaged in jobs related to the innovation process.¹¹ These workers earn substantially more than the median income for all workers. In 2017, the median worker in innovation-related jobs earned $85,390, compared to $37,690 for all workers.

Innovation jobs also are growing faster than jobs in other sectors, and unemployment rates are lower in the sector, too. In 2017, the unemployment rate for scientists and engineers was 2.7 percent, compared to 3.1 percent for all college graduates and 4.9 percent for the United States overall.¹² During the Great Recession of 2007–2009, moreover, when the U.S. workforce contracted, the innovation workforce was less affected by the overall economic contraction.¹³ Amid that recession, the income gap between innovation workers and the general labor force also widened. In 2012, median innovation economy earnings were double those of other workers. By 2014 the median innovation worker was earning 2.3 times more than the general labor force.¹⁴

Thus, across a number of measures, the science-based innovation workforce provides a tremendous boost to the overall economy, with better pay and job security going to those who work in the innovation sector.

Measuring the race and gender participation divide in U.S. innovation

Since the 1960s, both women and African Americans have obtained an increasing (though still not equal) share of bachelor’s degrees and advanced degrees in fields most associated with invention—the so-called STEM fields of science, technology, engineering, and mathematics. Despite this progress, I and my co-author Chaleampong Kongchareon at Michigan State University in a recent paper do not observe a similar increase in patenting activity among these groups.¹⁵

In general, women and African Americans remain underrepresented in the workplaces and boardrooms of high-tech firms in the United States. Today, both the lack of diversity in the venture capital industry and the paucity of women and African Americans who serve as executives and board members at high-tech companies receive regular attention.¹⁶ This innovation divide represents a lost opportunity and is a discriminatory drag on the U.S. economy. These distributional issues provide further evidence of the wide income and wealth gaps in the United States.

Economists draw on a wide range of metrics to define and measure participation in the innovation process.The National Science Foundation defines the science and engineering, or S&E, workforce in one of three ways:

By the parts of the U.S. economy measured by workers in S&E occupations
By the number of holders of S&E degrees
By the use of technical expertise on the job¹⁷

The National Science Foundation collects data on science and engineering students, graduates, and workers using a variety of surveys and sources, including its Survey of Earned Doctorates and the National Center for Education Statistics’ Integrated Postsecondary Education Data System Completions Survey. Demographic data, such as gender, race, and ethnicity, are among the data collected. In addition to collecting data on fields of study, I have assembled NSF data on S&E doctoral degrees earned by women from 1966 to 2018 and African Americans from 1968 to 2019.

Another way to measure what I call the “pink and black” innovation divide between women and men and between Black Americans and White Americans is via patent data. Data on patents, recorded and disseminated by the U.S. Patent and Trademark Office, are available from 1790 to the present and thus provide a relatively consistent historical metric.¹⁸ Demographic data, such as gender, race, and ethnicity, are not explicitly recorded in patent data. To address this issue, my co-authors and I have developed sophisticated methods for inferring which historical and contemporary patents were granted to women and African Americans.¹⁹ I turn to this in the next section.

Understanding the reasons for these race and gender divides in innovation

Within the innovation process in the U.S. economy, both participation and salaries vary greatly by gender, race, and ethnicity. Importantly, these racial and gender gaps are manifested throughout different stages of the innovation process. In this section of my essay, I provide longitudinal, quantitative evidence to outline the nature and scope of these gaps over time.

I then complement this aggregate, statistical picture with historical and contemporary examples from individual women and African American innovators who were impacted by racial and gender discrimination during the innovation process. This analysis across different scales illuminates both the aggregate, macroeconomic impact and the intimate lived experience of the innovation gap in pink and black. And, consistent with the time frame of available NSF education data, I focus on patent data from 1966 to 2014.

An individual participates in the innovation economy by passing through three stages of the innovation process. First, innovation typically begins with formal education or training, such as an apprenticeship, in a chosen technical field, often but not exclusively in a STEM field. Second, workers in the innovation economy participate in actual invention in university or federal laboratories, corporate research facilities, government agencies, or less formal workspaces.

Finally, innovation, or the commercialization of invention, occurs when inventors sell or license their patents or launch a new start-up or business unit to profit directly from the development of the invention. Let’s now examine more closely these first two stages—the preparation and education divide and the invention divide—which lay the foundation for bridging the race and gender income divides in science and engineering jobs and the race and gender patent divides.

The preparation and education divides

Women and African Americans have enjoyed significantly improved access to technical training over the past few decades, but lingering education gaps remain. Women and African Americans have increasingly been involved at the beginning of the innovation process—for example, by getting doctorates in the sciences and doing basic research that undergirds changes in the stock, flow, and direction of knowledge.

In 1970, only 9 percent of all doctorates in the science and engineering fields were awarded to women. By 2014, the share going to women was nearly 42 percent. In 1970, only 1 percent of all S&E doctorates went to African Americans. By 2014, the share going to African Americans was roughly 4 percent. For context, African Americans represent just more than 13 percent of the population.²⁰ The trends are similar for master’s and bachelor’s degrees, and are comparable through 2014.²¹ (See Figures 1 and 2.)

Figure 1

The share of science and engineering doctorates earned by women, by field of study, 1966–2014

Figure 2

The share of science and engineering doctorates earned by African Americans, by field of study, 1966–2014

Increases among women and African Americans, however, have not been uniform across collegiate fields of study. Psychology starts off with the largest share of female science and engineering doctorate recipients in 1966 at 22 percent, and finishes with the largest share in 2016 at 75 percent, according to the most recent NSF data.²² Apart from the field of psychology, women have traditionally received the highest share of doctoral degrees in the life sciences (more than half of all degrees in biological sciences in 2016) and one of the lowest shares in engineering (24 percent in 2016).

This is important because engineering is the field most closely associated with patenting. There is a large literature that examines why few women enter the field of engineering, and how and why those few leave.²³ Similarly, among STEM fields, the highest share of African American doctorates was in psychology (8 percent in 2014) and the lowest was in engineering (2 percent in 2014). African Americans also traditionally earned the highest share of doctorates in the life sciences and the lowest share in the physical sciences. In the 2000s, apart from psychology and the social sciences, the share of doctoral degrees going to African Americans has hovered between 2 percent and 3 percent.

With respect to education and training, women and African Americans are participating in increasing numbers over time. For both groups, a divide remains, however, and there is considerable heterogeneity of representation across fields. Examples of persistent barriers to women and African Americans pursuing degrees in STEM fields abound. Jennifer Selvidge, a former honors student in materials engineering at the Massachusetts Institute of Technology, captured the experiences of many women and African Americans. In her 2014 article, she reports that she was told “hundreds of times” that, as a woman, she did not deserve to be there, and that metallurgy was a “man’s field.”²⁴

She also witnessed male professors attempting to publicly humiliate the small number of female professors, and sexual harassment by teaching assistants. In addition to observing people of color being actively advised to change majors and leave the department, she also was subject to a teaching assistant arguing that “Black Americans are genetically inferior due to slavery-era breeding practices.”²⁵

The invention divides

The second stage in participating in the innovation process requires being involved in actual invention. Women and African Americans also have faced pervasive barriers to invention. For centuries, individual women and African Americans have had to battle the perception that they were mentally inferior and technically incompetent. Consequently, women and African Americans were not welcome in the White, male culture of corporate research and development labs. They were also barred from joining professional scientific and engineering societies until the mid-20th century, thus depriving them of the social capital and connections required to advance their careers and develop their inventions.²⁶

Contemporary measures of invention activity among women and African Americans simultaneously reveal evidence of stifled dynamism in the U.S. labor market for inventors, despite their increased participation, due to lingering barriers to access and inclusion. Women’s participation in this invention stage is definitely on the rise. Between 1993 and 2010, the share of women working in a science and engineering field rose from 31 percent to 37 percent. Over the same period, women in science and engineering occupations rose from 23 percent to 28 percent.²⁷ By 2017, women made up 29 percent of these workers, and the percent of African Americans working in these fields had increased to 13 percent.²⁸

Yet both female and African American scientists and engineers are more likely to work in nonscience and engineering occupations than directly in these occupations. More than two-thirds of psychologists were women in 2015, and women are less concentrated in the computer and mathematical sciences and engineering, compared to men. In 2010, 25 percent of the workforce in computer and mathematical sciences were women, and in engineering, 13 percent were women; in 2017, these shares were 27 percent and 16 percent, respectively.²⁹ Similar patterns are evident among African American scientists and engineers, though with several marked differences. (See Figure 3 and Figure 4.)

Figure 3

Percent of employed women scientists and engineers in a range of innovation occupations, 2015

Figure 4

Percent of employed African American scientists and engineers in a range of innovation occupations, 2015

Concretely examining specific occupations within a field reveals even more telling data on these race and gender divides. More than half the people in S&E-related occupations are women. Among them, women constitute 71 percent of workers in health-related occupations, more than half of S&E precollege teachers, more than half of technologists and technicians in the life sciences, but just one-fifth of S&E technologists and technicians.

Female scientists and engineers constitute half of scientists and engineers in non-S&E occupations. Women often start their careers working in occupations that are part of the innovation process but then leave for various reasons, including the need to provide child care due to the lack of family-leave policies and because of intolerable workplace environments.³⁰ Such departures have implications for the earnings of these scientists and engineers. For one thing, women’s wages will, on average, be lower in noninnovation occupations relative to wages within them. Furthermore, those lower wages will exacerbate inequality that exists between the innovation and noninnovation economies.

African American scientists and engineers make up just 4.8 percent of workers employed in S&E occupations. Among S&E occupations, African American scientists and engineers are more concentrated among social science-related occupations and among computer and math scientists and analysts than they are in other S&E occupations. Among S&E-related occupations, African American scientists and engineers, similar to the female scientists and engineers discussed above, are more concentrated in health-related occupations and in precollege teaching than in other S&E occupations. Almost twice as many African American scientists and engineers are in non-S&E occupations as are in S&E occupations.

Unemployment rates among scientists and engineers reveal discrimination in the innovation jobs in the U.S. labor market, too. Unemployment for African American and Hispanic men, at just more than 4 percent, is higher than for White and Asian men, and higher than the average for all scientists and engineers.³¹ (See Figure 5.)

Figure 5

Unemployment rates among scientists and engineers, 2015

Although not illustrated above, the unemployment rate for African American women in science and engineering occupations is higher than the unemployment rate overall—nearly double that of all scientists and engineers, and more than double that of White female scientists and engineers. Similar to the data on occupations, the data on enemployment indicate that the consequences of discrimination within science and engineering occupations are higher income inequality within these sectors of the economy. Unemployed scientists and engineers will likely be poorer and less able to accumulate wealth, compared to their employed counterparts.

Digging deeper into the data on sectors of employment among scientists and engineers by race and ethnicity reveals further divides in participation in the innovation process. Most scientists and engineers are employed in business or industry. For African Americans and Hispanics, the second and third sectors of employment are education and government. On average, government and education salaries are lower than those in business or industry, further deepening the income inequality among S&E workers. (See Figure 6.)

Figure 6

Share of employment by sector among scientists and engineers, by race and ethnicity, 2015

Just as incomes vary between occupations in the innovation process and jobs in the rest of the economy, incomes also vary among those within innovation occupations themselves. Among other things, they differ by gender and race. The median salary for men in the innovation economy in 2010 was $80,000, but it was only $53,000 for women, or 66 percent of the median male salary.³² In 2017, the median salary for scientists and engineers was $90,000 for men, yet it was only $66,000 for women, or 73 percent of the median male salary.³³

Some of this wage divide is attributable to the different occupations people perform across race and gender lines, with more White men in S&E occupations, which tend to be higher-paid. If considering only S&E occupations, the share of female-to-male median salary narrows to 81 percent and ranges from 77 percent for ages 29 and younger to 85 percent for ages 50 to 75. The share of female-to-male median salary is slightly higher in S&E-related occupations, 73 percent, and slightly lower for non-S&E occupations, 69 percent. “Mathematical scientist” is the only occupation in which the median female salary exceeds the median male salary, and the ratio of female-to-male median salary is 1.13.³⁴

Indeed, the earnings or income divide between workers engaged in the innovation process and those employed across the overall economy is substantial. Innovation workers earned 63 percent more than the average U.S. worker in 2014, the most recent year for which complete data are available.³⁵ Yet the gender and racial earnings divides among innovation workers and the overall economy is telling. The salaries of African American female scientists and engineers, for example, are 87 percent of what White women earn in all occupations, yet there is salary parity in science and engineering occupations for African American and White women.

Similarly, the median salary for African American women in S&E-related occupations also is at parity with White women, yet the median salary for African American women in non-S&E occupations is 83 percent of the median salary for White women. The largest gaps within S&E occupations are among psychologists (83 percent) and computer scientists (87 percent). Among mathematical scientists, the ratio of the median salary of African American women-to-White women is 1.21.³⁶

The gap between the median salary for African American and White workers is not as large as it is between men and women. In 2010, the median salary for White full-time workers with the highest degree in an S&E field was $72,000, and for their African Americans counterparts, it was $56,000, or 78 percent of the median salary for White workers.³⁷ In 2015, this share had moved only slightly, to 79 percent. For S&E occupations, this share narrows to 92 percent. Among S&E occupations, the gap is widest among psychologists (65 percent) and physical scientists (67 percent). There is parity in engineering, and, like women, mathematical scientists’ median salary for African Americans is higher than the median salary for White workers, with a ratio of 1.13.³⁸

In 2015, the share of median African American salary-to-White salary for S&E-related occupations is also 92 percent. As is the case for women, this share is lowest in non-S&E occupations, at 70 percent.³⁹ With respect to employment and salary data, the gaps in participation that existed even 7 years ago are closing. Yet gaps remain with respect to gender and race.

The patent divides

Legal access to the U.S. patent system offered greater, but still limited, opportunities for women and African Americans. There was no language in the original Patent Act of 1790 limiting patentees based on gender, race, age, or religion. Consequently, in the decades before emancipation and universal suffrage, women and (free) African Americans could—and did—invent and earn U.S. patents.⁴⁰

Still, women and African Americans did not have equal protection under the patent laws. While free African Americans were allowed to obtain patents, the U.S. Patent and Trademark Office refused to grant patents to enslaved African Americans. Moreover, laws in many states assigned all marital property rights to husbands, which effectively prohibited married women from owning or controlling patents in their own names. These draconian social norms and policies deterred many women and African Americans from even becoming inventors.⁴¹

Patent data provide another, albeit imperfect, means of measuring invention activity.⁴² In earlier research, my colleagues and I demonstrated that women and African Americans lag far behind other U.S. inventors with respect to patent activity. Using USPTO data from 1970 to 2006, we calculated that patent output for all U.S. inventors is 235 patents per million. But for women, the number is 40 patents per million. And for African Americans, it is 6 patents per million.⁴³

Moreover, researchers find that a propensity to patent is correlated with prior exposure to invention activity, and multigenerational income and wealth disparities. Children from high-income families who grow up around other inventors are more likely to patent, while children from low-income families with limited exposure to emerging technology are less likely to patent.⁴⁴

Taken together, these findings and others suggest a misallocation of resources that could lead to suboptimal levels and rates of economic growth that could persist across generations. For example, patent teams made up of both men and women are more productive than single-sex teams with respect to the most valuable patents. Patent teams, firms, and the economy will continue to perform at suboptimal levels without diverse teams and inclusion more generally.⁴⁵

The potential for discrimination to contribute to the invention divides was on public display in the summer and fall of 2017. A Google engineer, James Damore, wrote a memo that leaked and went viral. This memo, directed at diversity initiatives at the company, argued that women were underrepresented in technology careers because of “inherent psychological differences” between the genders.⁴⁶ Google dismissed Damore for “perpetuating gender stereotypes.”⁴⁷

A few weeks later, a former Google software engineer, Kelly Ellis, and two other women sued Google, alleging discrimination in both the pay and promotion of women. Coincidentally, the U.S. Department of Labor, in an ongoing investigation of Google’s gender gap in salaries, reported its finding of systemic discrimination of women at Google in the spring of 2017.⁴⁸

There is a large literature that suggests that unequal salaries and promotions could depress women’s interest in pursuing S&E degrees and careers, resulting in underrepresentation of women. The number of complaints and lawsuits against tech firms related to discrimination related to pay and promotions based on gender and race is increasing, which suggests both of these issues are being taken more seriously—why incur the costs associated with filing a complaint or lawsuit otherwise?—and that there is a role for policymaking in addressing these race and gender hurdles to stronger economic growth.

Policies to increase participation and wages in the innovation economy

The U.S. economy and society will never fully realize its scientific potential and ever-higher economic growth and living standards without including more women, African Americans, and others who are still actively or passively discouraged from earning degrees in STEM fields and training for STEM careers. There are four areas in which policymakers can make changes to resolve this problem in our society and our economy:

Mentoring
Early education about inventing
Blind patent reviews
High-tech workplace climates

Lets examine each policy idea in turn.

Mentoring

Mentoring is one broadly suggested tool to address the gender and race divides in STEM careers.

As detailed in the analysis above, the income, race, and gender gaps in invention are primarily due to barriers in acquiring human capital—the lack of mentoring and exposure to careers in science and innovation in childhood—and not due to differences in ability.⁴⁹ What’s more, the effectiveness of mentoring is recognized beyond academic papers and university programs, with programs designed to make a difference.

One such program is the Makers + Mentors Network (formerly US2020), an organization focused on programming that supports underserved and underrepresented students. Its mission is to change the trajectory of STEM education in the United States by dramatically scaling the number of STEM professionals engaged in high-quality STEM mentoring with youth. The Makers + Mentors Network is building a community of companies, organizations, schools, government agencies, and cities to participate in mentoring. It seeks to encourage 1 million science, technology, engineering, and math professionals to mentor students in Kindergarten through graduate school.⁵⁰

The Makers + Mentors Network currently operates in 21 communities across the country, serving more than 150,000 students and 20,000 mentors annually.⁵¹ The program also places AmeriCorps VISTA members within the organization and with its local community efforts, and is launching a Maker Fellows program in partnership with local community colleges and historically black colleges and universities. The Makers + Mentors Network connects students with STEM professionals within their communities through official mentorship programs, hands-on learning experiences, and afterschool and summer programs. More public investment and partnerships with these and other mentorship efforts can further strengthen the foundation of innovation and design for students at key moments in their education and career choices.

Early education about invention

Exposing children to invention and innovation is becoming a more recognized method of increasing participation. Doing so could foster more representation for women and African Americans early in the education pipeline and eventually at high levels in STEM fields.

An example of this kind of public-private engagement in early eduation and invention is the Spark Lab at the Lemelson Center for Invention and Innovation at the Smithsonian Institution. This is an activity space that allows children to create an invention and to help them think about making the invention useful. We recommend targeting low-income children, children of color from underrepresented groups, and female children for such activities, with more fiscal resources deployed to take this kind of program national to science museums around the nation.

Blind patent reviews

As detailed above, inequality in patent applications is a legacy of historic racial and gender discrimination that persists to this day. A recent paper in Nature finds that, all else being equal, patent applications with women as lead inventors are rejected more often than those with men as lead inventors.⁵²

An easy fix would be for the U.S. Patent and Trademark Office to engage in the blind review of patent applications by patent examiners, so that names of patent applicants are not visible to reviewers, as this allows for the possibility of discrimination by race and gender.

High-tech workplace climates

Workplace issues for women and African Americans go beyond the opportunity to participate in invention and innovation. Other issues now stand in stark relief due to recent events related to workplace climate, such as recent activities, protests, and discussions at Google and Microsoft. Among the issues identified are ones that have been reported on about the climate in similar workplaces, such as lack of transparency (including forced arbitration for sexual harassment claims), workplace culture, and pay and opportunity inequality.

Most patented inventions occur at firms. Therefore, at public companies, shareholders need to hold CEOs more accountable for workplace climate, and, for private companies, boards and CEOs should do the same. Congress also could play a role in bolstering the U.S. Equal Employment Opportunity Commission to investigate such complaints and help to minimize the frequency and intensity of hostile workplaces for women and African Americans.

As is the case for the rest of the U.S. economy, greater representation for workers in the innovation process seems warranted to effectively and sustainably address workplace-related issues. The recent unionization of Google workers is a promising step in helping address discrimination, harassment, and other workplace climate issues, as well as creating more inclusive organizations more broadly.

Conclusion

In a recent paper, I outline proposals to address the commercialization phase of the innovation process, including collecting demographic data on inventors, enhancing the Small Business Administration’s programs related to innovation to promote diversity and inclusion, and improving the climate in the fields and workspaces where these innovations take place.⁵³ Creating a more inclusive innovation ecosystem will increase the participation of women, African Americans, and other underrepresented demographic groups at every stage of the innovation and commercialization process, from early education and training to the practice of invention and the later economic gains from those breakthroughs.

Taken together, these proposals could augment the participation of women and African Americans in the innovation process and boost wages for them across the innovation sector. Doing so would boost innovation and thus the productivity of the U.S. economy, and ensure the fruits of economic growth are more broadly shared—thus reinforcing more sustainable economic growth.

—Lisa D. Cook is a member of the Washington Center for Equitable Growth’s Steering Committee and professor of economics and international relations at Michigan State University.

Acknowledgments

Place-conscious federal policies to reduce regional economic disparities in the United States

Posted on January 14, 2021June 2, 2021 by Maria Monroe

To read more about Boosting Wages for U.S. Workers in the New Economy 20 and download the full collection of essays, click here.

Overview

Over the past four decades, geographic inequality between regions of the United States has grown dramatically. A handful of metropolitan areas, largely along the coasts, have become some of the richest economic regions—cohesive groups of counties linked by strong economic ties, such as those between a city and its suburbs—in world history. At the same time, large swaths of the country have been trapped in economic decline, struggling with deindustrialization, stagnant incomes, and rising unemployment.

Fact sheet: Place-conscious federal policies to reduce regional economic disparities in the United States Download

These economic challenges have contributed to a host of social and political problems, among them public health crises,⁵⁴ declining social mobility,⁵⁵ racial inequality,⁵⁶ and political polarization.⁵⁷ They also reduce national economic growth by reducing investment in local public goods, limiting aggregate demand, and lessening the effectiveness of federal economic policy. Absent an innovative and muscular policy response, the coronavirus pandemic and its accompanying economic collapse are likely to only worsen the outlook for struggling economic regions.⁵⁸

After decades in which regional economic challenges were largely considered a problem for state and local civic leaders alone, today, there is an increasing appetite for federal policy action to reduce this form of geographic inequality—distinct from the different challenge of economic inequality within regions. Most recent proposals for reducing interregional inequality have tended to take the form of “place-based policies” that target government investment or subsidies to economically struggling cities or neighborhoods.⁵⁹

This essay argues that to successfully boost wages and employment levels in struggling areas, policymakers must instead adopt a “place-conscious” approach that treats reducing interregional inequality as a priority throughout federal policymaking rather than something to be remedied after the fact. Recent research shows that the growth in geographic inequality since 1980 stems in large part from changes made to federal policy that seem, at first glance, to be geographically neutral, such as the relaxation of antitrust enforcement,⁶⁰ the lowering of trade barriers,⁶¹ and the deregulation of the transportation and communications industries.⁶² Although these policies were the same everywhere, they interacted with the existing spatial patterns in the economy in ways that systematically advantaged some places while harming others.⁶³

Beyond place-based: Reducing regional inequality with place-conscious policies

June 30, 2021 2:00PM – 3:30PM

Learn More

Meaningful and sustained reductions to interregional inequality will require structural changes to the U.S. economy made through federal action. Almost every domain of federal policy, from food policy to military spending to public finance, has a spatial footprint. Policies that take the geographic consequences of federal action seriously—that seek to alter the economic structures that create disparities to begin with—are much more likely to succeed than policies that seek to ameliorate geographic income disparities after the fact while leaving the underlying structures that generate them intact.

To illustrate how place-conscious federal policies can reduce interregional inequality, I consider four examples of such policies:

Universal anti-poverty programs
Restored regulation of key sectors, notably transportation and communications
State and local finance reform
Direct investment to meet national priorities, such as climate resilience

Each of these policies would have the effect of reducing interregional inequality, without explicitly targeting struggling regions for subsidies or investment. Taken together, they represent a revival of the federal government as a proactive force for uniting disparate regions into one national economy—a role it has repeatedly played throughout U.S. history, from the development of the U.S. Postal Service to the building of the interstate highway system.

The problem: Rising interregional inequality and its consequences

Over the past four decades, regions of the United States have diverged economically from one another. In 1980, most U.S. commuting zones had mean family incomes roughly in line with the national average.⁶⁴ While there was certainly inequality within these areas, average incomes were roughly the same across most of the country—with important exceptions in the rural South, the Rio Grande Valley, and other long-identified areas of persistent rural poverty.⁶⁵ By 2013, there were far more places with mean incomes that were either much higher or much lower than the nation as a whole.

Overall, the income gap between the richest 10 percent of U.S. commuting zones and the poorest 10 percent grew by almost 50 percent during this period. The fraction of the U.S. population living in areas more than 20 percent richer or poorer than the national mean almost tripled, from 12 percent to 31 percent.⁶⁶ (See Figure 1.)

Figure 1

Mean family incomes relative to the nation, by U.S. commuting zones, 1980 and 2013

Growing interregional economic inequality, and the local economic dislocations that contribute to it, are major contributors to many of the most pressing social, economic, and political challenges facing the United States today. Public health crises—from the opioid epidemic ⁶⁷ to exposure to environmental toxins ⁶⁸—are disproportionately concentrated in economically struggling areas. Deindustrialization has contributed to a host of other social problems, from family instability to declining upward mobility.⁶⁹ Regional divergence also contributes to political disfunction, both by altering the material conditions in different places such that their material interests diverge, making compromise more difficult, and by contributing to growing anger, political extremism, and polarization.⁷⁰

Regional disparities are both a driver and a consequence of racial inequality. Black Americans in particular disproportionately live in the large, multistate regions—especially the Midwest and the South—that have higher-than-average poverty rates and that were hit hardest by the economic shocks of the past several decades.⁷¹ The effects of these economic dislocations were felt earliest and most strongly in communities of color, although they have now spread to many White communities as well.⁷²

That said, some regional economic challenges can be traced directly to racist policies at all levels of government. Local and state governments in parts of the country with large Black populations—historically in the South, but also in Northern cities that received large numbers of Black migrants during the Great Migration of the 20th century—historically spent less on public goods such as education and infrastructure that contribute to regional prosperity.⁷³ After the passage of the Voting Rights Act of 1965, state and local governments in the South began making such investments at greater rates in response to electoral pressure from African American voters, which had the effect of boosting the economy of the entire region.⁷⁴

At the federal level, early examples of place-conscious policies, such as the Social Security Act of 1935 and the Fair Labor Standards Act of 1938, were originally designed to exclude African Americans, reducing their impact on regional disparities while entrenching racial inequality.⁷⁵

Regional economic struggles undercut national economic performance

The economic consequences of interregional inequality are not limited to struggling regions, but are felt nationally through slower overall growth and greater macroeconomic volatility. Within regions, local economic multipliers mean that when workers lose their jobs, those job losses build on themselves: Each job lost in an initial shock from, say, a tariff reduction or a factory closing reduces demand for local services, creating further job losses that can persist for decades.⁷⁶

These shocks are exacerbated by our current system of fiscal federalism, which precludes deficit spending by state and local governments and thus requires them to implement pro-cyclical fiscal policy—strengthening booms and worsening recessions.⁷⁷ During the first few months of the coronavirus recession, for instance, state and local governments collectively laid off or furloughed 1.5 million workers, and layoffs are expected to continue in the absence of new federal fiscal aid.⁷⁸ Cutting government spending in the middle of a recession is the definition of macroeconomic malpractice, yet it is consistently done in the United States because of how our cities and states are funded.

At the federal level, geographic inequality makes macroeconomic policymaking more challenging because the same set of federal policies must meet the needs of regions with very different economic circumstances. Interest rates have strong regional economic impacts, for example,⁷⁹ and the interest rate that best serves high-cost areas such as San Francisco and New York is likely to be very different from that which meets the needs of areas struggling with unemployment. But because our country is a monetary union, we must set just one interest rate for our shared currency.⁸⁰

Current place-based policies are welcome, but are limited and fragile

For decades, regional economic challenges were largely considered a problem for local civic leaders alone. But today, there is an increasing appetite for federal policy action to reduce geographic inequality. Initial proposals have largely taken the form of “place-based policies” that target struggling neighborhoods, cities, or regions for investment or subsidies with the goal of increasing incomes and employment.⁸¹ Recent proposals for place-based policies include geographically targeted tax breaks tied to job creation⁸² and direct federal investment in research and development located in struggling regions.⁸³

Place-based policies have historically been the subject of skepticism on several grounds. First of all, deciding which places deserve to be targeted with extra subsidies is both technically challenging and politically fraught. Secondly, there is no guarantee that residents of a targeted place will be the final beneficiaries of these investments, since newly created jobs are often filled by in-migrants rather than current residents. Finally, such policies often appear to have the effect of simply moving economic activity from one place to another rather than increasing the total amount.⁸⁴

Recent interest in place-based policies is welcome because it indicates support for federal action to reduce interregional inequality. But long-identified weaknesses remain major challenges. On the technical front, effective place-based policies require constantly identifying and judging which places are truly in need and merit federal investment. This is genuinely difficult and can lead to perceptions of corruption or that the government is “picking winners.” Any missteps cast doubt on the entire project.

This concern is prominent in critiques of the Opportunity Zones tax credit program, created in the 2017 tax bill, which has come under fire from both the left and the right for directing federal subsidies to people and areas argued not to need the help.⁸⁵ Further, given the sheer magnitude of geographic inequality, the amount of subsidies required to meaningfully reduce interregional income disparities—as opposed to merely offering some relief to struggling regions—is likely to be enormous.

The solution: Universal, place-conscious policies

The key fact about regional economic divergence is that while it appears to be the result of many different local trends, it is fundamentally a national process driven in large part by national economic trends and federal policies. As recent research shows, much of the increase in interregional inequality since the 1970s is attributable to federal policy changes that seem geographically neutral at first glance.⁸⁶ These include:

Lowering of trade barriers ⁸⁷
Relaxation of antitrust enforcement ⁸⁸
Deregulation of the transportation and communications industries⁸⁹

Though implemented nationally, these policies interacted with existing spatial patterns in the economy in ways that systematically helped some places while disadvantaging others.⁹⁰

Addressing interregional inequality, then, will be most effectively accomplished at the federal level through what might be termed “place-conscious policies.” Unlike place-based policies, place-conscious policies are implemented everywhere but are designed in such a way that their benefits disproportionately fall on those places that need the most help. In this, they are the spatial analogue of the “targeting within universalism” that has long been one of the most successful ways to sustain support for programs that benefit disadvantaged individuals.⁹¹

Rather than attempting to identify and compensate regions that are “losing” the economic competition after the fact, as place-based policies do, place-conscious policies seek to even the economic playing field and lower the stakes of interregional competition to begin with. This will make them easier to administer, less susceptible to critiques for being unfair or corrupt, and ultimately more effective.

A place-conscious approach can be applied in almost every area of federal policy, from trade to taxes to military spending. Here, I briefly consider four types of place-conscious policy:

Universal anti-poverty programs
Re-regulation of key industries
State and municipal finance reform
Direct investment

Let’s look at each of them in turn.

Universal anti-poverty programs

The growing economic disparities between regions of the United States are fundamentally intertwined with the growth of economic inequality more generally. In fact, more than 50 percent of the cross-regional divergence in mean family incomes since 1980 is strictly attributable to rising overall economic inequality rather than any spatial reallocation in who lives where.⁹²

Any federal program that benefits lower-income Americans, then, will have the effect—without explicit spatial targeting—of reducing interregional income disparities. This applies to existing transfer programs such as Social Security, disability insurance,⁹³ and the Earned Income Tax Credit; but it would also apply to new or expanded programs such as child allowances or an expanded Child Tax Credit—proposals that have received bipartisan support ⁹⁴—or even a full-fledged Universal Basic Income.⁹⁵

The geographic impacts of such universal antipoverty programs can be substantial. One recent study found that in Brazil, the Bolsa Família and Benefícios de Prestação Continuada programs of conditional cash transfers to families in poverty were responsible for a full 24 percent of the reduction in cross-state inequality from 1995 to 2006, despite comprising just 1.7 percent of disposable household income.⁹⁶ An earlier study in the Netherlands found that social insurance programs, taken together, reduced regional inequality by 40 percent.⁹⁷

Antitrust and regulated industries

Governments set the rules and determine the playing field in which markets operate. Through much of U.S. history, the federal government used its power to create a playing field that would disperse economic activity and integrate geographically isolated regions into the national economy. These efforts began with the creation of the U.S. Postal System in 1775,⁹⁸ and continued through the regulation of railroads with the Interstate Commerce Commission in the late 19th century,⁹⁹ rural electrification in the 1930s,¹⁰⁰ and the first 70 years of air travel.¹⁰¹

Proximity to natural resources, trade routes, and other economic advantages has always mattered for regional economic performance, but these federal efforts had the aim and consequence of making geography less determinative of economic success, reducing the advantages that large urban centers or centrally located towns would otherwise receive due to their location alone.¹⁰² The strong American tradition of antitrust enforcement and skepticism of “bigness” similarly sought to reduce economic concentration in general, often with the effect of spatially distributing economic activity.¹⁰³

Since the 1970s, deregulation—particularly of the transportation and communications industries—and weakening antitrust enforcement have shifted the economic playing field in favor of large corporations and large, well-connected urban centers. That is, market concentration has contributed to the geographic concentration of prosperity and, conversely, disadvantage. The airline industry offers a useful illustration. Under the old regulatory regime, price and firm entry regulations created a system in which profits from well-traveled routes cross-subsidized service on less-profitable routes, consistent with the policy goal of providing similar levels of service to similarly sized cities.¹⁰⁴ Following deregulation in the 1970s, airlines shifted their focus to the cities and routes with the highest traffic, cutting service to small and mid-size cities.¹⁰⁵

With lax antitrust enforcement, airlines also consolidated, reducing competition and the number of hub airports that offer large numbers of direct flights.¹⁰⁶ This change to transportation geography, driven primarily by changes to federal regulation, triggered a reshuffling of corporate headquarters. Companies have moved from cities that lost service, such as Cincinnati or St. Louis, to cities currently blessed with extensive air connections, such as Charlotte or Atlanta.¹⁰⁷

A place-conscious approach to regulation would return the federal government to its historic goal of binding the disparate parts of the country into one national economy and society.¹⁰⁸ This would primarily involve once again requiring that the key industries that connect people and places—telecommunications and transportation in particular—do so on an equal footing. Reinvigorated antitrust enforcement, meanwhile, would also reduce the stakes of geographic competition: When more firms exist in an industry, small cities have more opportunities to land a headquarters or foster a successful entrant.

State and municipal finance reform

The current U.S. system of fiscal federalism is inefficient, unequitable, and almost unique among high-income countries.¹⁰⁹ In the United States, states and localities are the primary providers of many standard government services, yet they are denied access to many of the most important tools of governance—running deficits, controlling the entry and exit of capital and labor, and printing money.¹¹⁰ This harms individuals, economically struggling regions, and the country as a whole.

For individuals, it makes access to basic government services contingent on the wealth of one’s community, entrenching inequality.¹¹¹ It also creates economic distortions by artificially bundling choices about fiscal policy with the decision of where to live.¹¹² For regions, it advantages already-wealthy areas by allowing them to provide more public services at lower tax rates.¹¹³ This, in turn, encourages race-to-the-bottom tax and incentive competitions.¹¹⁴

For the country as a whole, the current system of fiscal federalism makes recessions worse, because most state and local governments are legally prohibited from deficit spending .¹¹⁵ This frequently requires that cities and states, which employed about 20 million Americans prior to the coronavirus pandemic,¹¹⁶ cut spending and lay off workers in the middle of recessions—a highly costly emergency measure that creates further job losses through multiplier effects, worsening economic volatility.

There are ways to fix the problems baked into fiscal federalism. As recently as the 1980s, municipal governments in the United States received more than 15 percent of their revenue from federal grants. Today, that share has fallen to less than 5 percent.¹¹⁷ New proposals to increase federal grants to state and local governments—especially ones that disproportionately help lower-income parts of the country—enjoy support across the ideological spectrum and would provide a large and immediate boost to national economic activity—one that is greatly needed to combat the coronavirus recession.¹¹⁸

Direct investments and employment

A final example of federal policy to reduce interregional economic inequality is straightforward public investment. When the federal government makes investments around the country to meet national policy priorities, it contributes employment and spending to local economies. Unless these investments deliberately target only prosperous areas, the net effect will be to reduce interregional inequality.

Historically, the largest such investment has been military spending, which dramatically reshaped the economic geography of the postwar United States and continues to provide the economic base for many communities around the country.¹¹⁹ Federal infrastructure investments in the 1940s and 1950s were likewise a key driver of regional convergence in that era, particularly the growth of the South.¹²⁰ In the future, investments in climate resilience and clean energy offer a similar opportunity to reboot lagging regional economies while meeting an overriding national goal. More broadly, federal agencies can consider spatial implications whenever they make investment or staffing decisions.¹²¹

Conclusion: National action for a national problem

Place-conscious structural fixes along the lines of what I have described—from reinvigorated antitrust, transportation, and communications regulation to fiscal federalism reform to universal anti-poverty programs—are likely to be both more effective at reducing interregional inequality and more durable than explicitly targeted place-based policies. Because they are implemented everywhere simultaneously, place-conscious policies do not require politically fraught decisions about which places qualify for help.

This universality also makes them more agile, able to respond instantly to future changes in economic geography. And because place-conscious policies would produce clear benefits to residents of all parts of the country, they offer the potential for broad-based political coalitions.

A common concern about universal programs is that they are too costly. This concern is misguided when it comes to place-conscious policies for regional development. First of all, several of the programs I outline involve changes to regulation, which would entail minimal costs to the federal government. More importantly, though, the ongoing economic and budgetary costs of local economic struggles around the country are enormous. Each person who is unemployed against their will represents tens of thousands of dollars of lost economic output each year. Investments in reviving the economies of struggling areas, then, offer an enormous economic upside for the national economy and the federal budget.

The four policy areas I explore in this essay are only the beginning of what is possible. A place-conscious approach to policymaking can and should be adopted throughout the federal government. Every policy has a spatial footprint, and while spatial considerations will rarely be the primary or sole consideration in policymaking, they deserve to be articulated and considered.

There have been multiple recent proposals to formalize this consideration of spatial equity in federal policymaking. One such proposal would implement an executive order requiring that agencies consider the geographic impact of the regulatory choices they make, much as they are currently required to consider the impact on federalism.¹²² Another recent proposal would subject major federal discretionary spending decisions to a place-conscious equity review,¹²³ building on past efforts to consider geographic equity throughout the federal budget.¹²⁴

At the broadest level, addressing interregional inequality requires embracing two fundamental truths. First, the United States is one country and will ultimately prosper or perish as a unit. Second, very little in economics is inevitable, and Americans have the collective power, through the federal government, to make the economy serve the needs of all parts of the country.

—Robert Manduca is an assistant professor of sociology at the University of Michigan.

Acknowledgments

Sovereignty and improved economic outcomes for American Indians: Building on the gains made since 1990

Posted on January 14, 2021May 28, 2021 by Maria Monroe

To read more about Boosting Wages for U.S. Workers in the New Economy 20 and download the full collection of essays, click here.

Overview

The exercise of American Indian tribal sovereignty over the past 30 years resulted in more economic growth and improved well-being for American Indians than during any other point in the more than 500-year history post-contact with European colonists and settlers. Increased self-governance over tribal lands and resources created new economic and employment opportunities for American Indians and for non-American-Indians on or near tribal lands and resources as well.¹²⁵

Fact sheet: Sovereignty and improved economic outcomes for American Indians: Building on the gains made since 1990 Download

These advances are the foundation upon which policymakers can build to develop policies to address persistent earnings and income inequality faced by American Indian workers and their families. Even though American Indians are a relatively small proportion of the United States as a whole, in certain jurisdictions, they comprise a relatively large proportion of the population. In states such as Oklahoma, New Mexico, South Dakota, Montana, North Dakota, and Arizona, the American Indian population ranges from 5 percent to 13 percent of these states’ populations. Some tribal governments control (relatively) large amounts of land areas in these states, which should result in meaningful influence over the demand and supply of labor and job creation in these states.

But, as this essay details, large hurdles remain before American Indian tribal governments can fully realize their economic and political opportunities to provide full employment (through public or private enterprises) for their citizens. Significant reductions in family poverty rates and unemployment rates almost doubled real per capita income between 1990 and 2015, yet still-significant income gaps remain, largely because American Indians residing on reservations are less likely to be employed full time than their White counterparts. After detailing these findings—presented against the backdrop of the recent gains in tribal self-governance—I conclude with a look at three areas that hold the most promise for improving earnings and employment for the American Indian population on reservation lands:

Support tribal sovereignty and industry innovation
Reduce barriers to economic development
Improve data collection for American Indians

While not all of these recommendations detailed below focus directly on individual American Indian workers, their families, or their employers, they are important in stimulating economic development and eliminating existing obstacles in and around tribal lands. This economic development and the well-being of American Indians are not only important in their own right but are also key to greater economic growth in these mostly rural and underdeveloped regions of the United States.

The problem: Economic and labor conditions of American Indian workers and their families

There is surprisingly very little research on the determinants of earnings and wage growth for Indigenous peoples in the United States. The vast majority of existing research focuses on the American Indian reservation-based population, which is what I focus on in this essay, excluding the urban American Indian population in my analysis.¹²⁶ Nonetheless, systematic analysis and evaluation of policies intending to improve the earnings and employment opportunities of American Indians are few and far between.¹²⁷ Existing longitudinal datasets often do not have a sufficient number of American Indians in order to conduct the standard analyses employed for other races or ethnic groups around the nation.

As a result, researchers do not have the rigorous set of studies over time for these populations. Therefore, there are many opportunities for future researchers to expand upon this report and investigate determinants of earnings growth and employment growth for the American Indian population. Identifying policy solutions to improve the earnings and employment of American Indians is not a trivial task.

First of all, there is little rigorous evaluation of labor force training programs or policies available, compared to other race and ethnic groups in the United States. Additionally, data for this group are notoriously difficult to come by due to the small proportion of American Indians in the overall U.S. population. Still, there are some broad datasets that enable researchers to present a broad picture of the economic conditions of American Indian individuals and their families.

Economic conditions for American Indians residing on tribal reservations are often depicted as dire. In fact, conditions are quite variable, depending upon the tribe and the time period examined. There are 324 reservations or joint-use areas—areas where multiple American Indian reservation governments share legal and political jurisdictions—in the lower 48 states, with wide geographic, regional, and economic variations. Yet despite improvements, family poverty rates, unemployment rates, and labor force participation rates for American Indians are substantially higher, compared to the United States as a whole.

The average poverty rate for American Indians (including Alaska Natives and American Indians living outside of tribal reservations) in 2015, the most recent year for which complete data are available, was 28.8 percent; the unemployment rate in 2015 was 11.6 percent; and the labor force participation rate was 55 percent. All of these rates are generally worse than the United States as a whole, even though economic conditions have improved for American Indians since 1990. (See Table 1.)

Table 1

Real per capital income, poverty rate, unemployment rate, and labor force participation rate of American Indians and Alaska Natives, compared to overall U.S. rates, 1990–2015

Recent research shows that income inequality also increased in recent decades for American Indians, as is the case for the United States as a whole.¹²⁸ Measures such as income mobility, or the ability to move up and down the income distribution, stagnated in recent years for American Indians.¹²⁹

American Indian families also possess little wealth, though relatively little is known about asset ownership across the American Indian population as a whole. One perhaps indicative study of two tribes in Tulsa, Oklahoma, however, finds that the total net worth of American Indian families enrolled in the Cherokee and Muscogee Creek tribes stood at 85 percent and 25 percent, respectively, of the net worth of White families in areas nearby their reservations.¹³⁰ Examining only home ownership, Richard Todd and Federico Burlon at the Federal Reserve Bank of Minneapolis find that home ownership rates for American Indian families overall stood at 56 percent, compared to 71 percent for White families, in 2006, according to the U.S. American Community Survey.¹³¹

How work, wages, and incomes of American Indians intersect with state and federal governments

American Indians and other Indigenous peoples in the United States occupy a fairly unique space. They are both a racial group and a political group. The U.S. Constitution and several U.S. Supreme Court cases have determined that Indigenous peoples and governments in the United States are domestic-dependent nations. These populations possess rights that extend to self-governance, exercise of treaty rights, and other laws specific to their political status. These are unique tools and opportunities available for American Indians for creating jobs and boosting wages and incomes.

Historically, the federal government maintained significant control over American Indians residing on reservations. Until they were granted U.S. citizenship in 1924 under the Snyder Act, American Indians were required to get the approval of the federal agent from the U.S. Department of the Interior’s Bureau of Indian Affairs before they were allowed to leave their reservations. In subsequent decades, the federal government enacted additional policies for this population. One was the urban relocation program, which aimed to create incentives for mostly rural American Indians to seek employment in urban centers such as Los Angeles, Minneapolis, Seattle, and Chicago. The program resulted in the relocation of tens of thousands of American Indians from rural to urban populations between 1950 and 1970.

A second program implemented during this time aimed to completely assimilate American Indians by terminating their tribal statuses as political entities. The so-called Indian Termination era—1940 to the 1960s—aimed to pay American Indian tribes to accept a settlement of money in exchange for the termination of their political status for themselves and their descendants. Researchers have yet to evaluate the impact of these two programs on the well-being, earnings, and employment of these relocated American Indian families and their descendants. Anecdotally, however, there do not appear be any marked improvements, as many American Indians in these urban settings are living in poverty.¹³²

More recently, the federal government took a different approach. It allowed and encouraged tribal governments to expand economic and political control over their jurisdictions and resources. The modern-day federal-tribal relationship is known as the self-determination era and is marked by the expansion of tribal governments into business operations and in the direct provision of government services.

Previously, for example, the federal government would administer programs on the reservations, from schools to health clinics to housing programs. In the self-determination era, under the Indian Self-Determination and Education Assistance Act of 1975, tribal governments increased their administration of these programs through direct funding from the federal government, which opened up employment and service opportunities to tribal citizens.

In addition, the passage of the Indian Gaming Regulatory Act in 1988 provided a unique opportunity for American Indian tribal governments to expand into the gaming industries. As a result, tribal governments expanded their economic development activities into this new area on their own terms over the decades of the 1990s and 2000s. Assessing how these new efforts by tribal governments to boost employment, wages, and human capital investments is the first step toward formulating new policy proposals to address the still-deep economic inequality faced by American Indian workers and families living on reservations across the United States.

Jobs and wages in and around tribal reservations

The basic employment and earnings data for American Indians in the United States as a whole and among those residing on American Indian reservations indicates overall that labor force participation and real per capita income for American Indians increased over the past 30 years. What’s more, the earnings growth for American Indians broadly kept pace with White Americans over this same time period. There are level differences in earnings, however, that persist—even after accounting for some basic human capital and demographic characteristics between these groups over time. And there are stark differences in the earnings of full-time versus part-time workers for American Indians, too.

Consider first the median earnings of American Indians and non-American-Indians for the country as a whole and for nonmetropolitan areas as a proxy for reservation locations from 1988–2019. The median earnings of American Indians are indeed lower than those of White Americans for the country as a whole, as well as for the nonmetropolitan/reservation locations.¹³³ The median earnings of American Indians did post a pronounced increase in 1998, but then flatlined before declining amid the Great Recession of 2007–2009 and in the subsequent slow recovery. (See Figure 1.)

Figure 1

Median earnings of American Indian and Alaska Native workers in rural and metropolitan areas, compared to White workers, real 2019 dollars, 1988–2019

But there is another measure over the same time period that is useful for understanding the median wages of American Indians—those who are full-time employees working 40 hours per week or more. Earnings for full-time American Indian workers and White workers increased for both groups by about $20,000—though full-time employment, on average, accounts for approximately 57 percent of White employment but only 49 percent for American Indian workers. And there was an upward trend in real earnings for all race groups and for all locations. (See Figure 2.)

Figure 2

Median full-time earnings of American Indian and Alaska Native workers in rural and metropolitan areas, compared to White workers, real 2019 dollars, 1988–2019

Some of this upward trend in earnings, however could be due to differences in educational attainment, age, and gender in addition to full-time status. Preliminary analysis suggests that there is a persistent difference between White and American Indian workers for all locations, even after controlling for these important human capital and demographic characteristics.

The upshot: Earnings among American Indian workers residing on reservations are persistently lower than that of White workers in the United States, particularly those American Indian workers who are employed full time, despite some specific economic gains detailed above.

Employment opportunities in and around tribal reservations

Increasing full-time employment opportunities, then, could well be an important way to increase earnings for American Indian workers. And indeed, one measure of job opportunities on and off reservations demonstrates there are ways to lift employment, and thus the wages and incomes, of American Indians.

Specifically, there is a large share of jobs on reservations (relative to off-reservation locations) in the following industrial categories: arts and recreation, accommodations and food services, and public administration, according to the Longitudinal Business dataset. These align with the relatively large gaming, tourism, and public administration activities on tribal lands. In contrast, there are relatively fewer jobs on reservations (compared to off-reservation locations) in agriculture, construction, manufacturing, retail, and health services.¹³⁴ (See Figure 3.)

Figure 3

The kind of jobs and the share of jobs on a nearby tribal reservation, 2010

But how many businesses survive in and around tribal reservations? It’s important to assess and understand where policies can drive job creation. The data show, perhaps surprisingly, that reservation-based business establishments during the Great Recession of 2007–2009 were more resilient than in adjacent counties. In particular, establishments in the arts and entertainment, education, accommodations and food services, public administration, and wholesale/retail sectors fared quite well.¹³⁵ (See Figure 4.)

Figure 4

Differences in the probability of reservation-based businesses surviving, by industry, 2007–2012

One potential reason for more successful businesses on tribal reservations may well be that many of these establishments are also tribally owned and operated. Thus, they may have been maximizing the employment of tribal citizens and local residents over profits during the Great Recession. As a result, they may have made different business decisions than private business owners might embrace. This is one of several findings that inform the policy solutions presented next in this essay.

Solutions to improve earnings and employment in American Indian communities

The data analysis above helps identify potential pathways to encourage and support earnings and employment for American Indians residing on tribal reservations. In this section, I will detail three broad policy proposals that could strengthen the overall conditions for improving employment and earnings on American Indian reservations. Specifically, policies that:

Support industry innovation and tribal sovereignty to boost jobs growth and wage gains
Reduce barriers to economic development to spur more job-creation and nonwage income
Improve data quality and collection to improve economic program and policy evaluation

Let’s consider each in turn.

Support industry innovation and tribal sovereignty

Supporting the full exercise of tribal sovereignty over land, resources, and citizens is an important step in improving the economic conditions of American Indians residing on reservation lands. Under the Indian Self-Determination and Education Assistance Act of 1975, tribal governments could apply to take over housing, education, healthcare, and the management of tribal lands themselves. The federal government provided funding for this, and the tribal government hired and managed the employees.

Tribal governments consequently became more responsive to the needs of their tribal communities and were able to hold employees responsible for their actions. Employees had a greater incentive to do their jobs effectively, too. The quantity and the price of tribal timber resources, for example, increased once tribes took control of their timber resources from the federal agencies.¹³⁶ Similar changes occurred in other agencies; however, rigorous evaluations of those effects in housing, healthcare, and education require better data collection.

The Indian Gaming Regulatory Act of 1988 established the requirements for tribal governments to pursue gaming operations on their tribal lands. As a result, gaming operations expanded tremendously throughout American Indian reservations in the 1990s and 2000s. Not all tribal reservations benefited equally from tribal gaming operations, but a large proportion have done quite well. Tribal gaming revenues increased from zero to almost $30 billion annually by 2015, the most recent year for which complete data are available, compared to commercial, non-American-Indian gaming industry earnings of approximately $38 billion.¹³⁷

Tribal gaming revenues are used to augment existing funding for programs, employment, and service delivery to tribal members. There is some evidence that tribal casino operations also have positive spillover effects to neighboring communities by providing employment options and demand for small business services.¹³⁸ The gaming industries and provision of federal government services by tribal nations are two examples of tribal nations exercising their own sovereignty and authority in new economic endeavors. They serve as guideposts for future innovation by tribal governments.

Create new innovative industries

New policies, for example, could encourage the funding of new and innovative economic activities that further strengthen and diversify tribal economies in sustainable ways. Expansion into novel industries such as climate mitigation and renewable energy generation would reinforce existing aims and goals of tribal governments while diversifying the economic base in tribal reservations. And expanding tribal jurisdiction and ownership of reservation lands would improve the ability of tribal governments to enact economic development projects on larger scales that may be necessary for nonmarket-based economic activities.

Given centuries of the taking of tribal lands, a concerted effort to restore and increase existing land bases would provide tribal governments with the resources to improve employment and earnings opportunities. This would include efforts to reduce the fractionated ownership of tribal and individual American Indian lands. Additionally, it would provide tribal citizens, potentially, with the option to earn a living in the nonmarket economy through traditional subsistence activities, which is an often-cited desire of tribal citizens.

These policy ideas are already being tested. In Washington state, the Lummi tribe recently created the first tribally owned commercial wetland-mitigation bank in the United States. The landbank provides credits for non-American-Indian developers in other parts of the region that create unavoidable adverse impacts to the aquatic environment from permitted projects. It is an offset program that puts specific areas on the Lummi reservation in perpetual conservation mode. The non-American-Indian developers are required to provide or pay for these credits under several environmental laws. The tribal government earns a significant amount of revenue from this activity annually, and it uses those funds to hire and train wildlife biologists, conservation technicians, and flood specialists. Tribal members are often trained on the job, as well as provided scholarships for college and graduate school to specialize in these occupations. As a result, the labor force has expanded and diversified dramatically.

Similarly, the Yakama, Umatilla, Nez Perce, and Warm Springs Tribes formed the Columbia River Inter- Tribal Fish Commission to improve salmon spawning to the upper regions of the Columbia River. After decades of blockages due to dams, the commission advocated for fish passes at existing dams, as well as negotiated the return of water to some of the smaller tributaries of the Columbia River so that salmon are able to spawn well into Idaho on Nez Perce lands. These results significantly increased the availability of salmon resources inland and provided a return to a traditional way of life and diet for many people. Additionally, fishing revenues for tribal members have increased twofold as a steadier supply of salmon made it more lucrative to specialize in salmon fishing and selling.

Emerging opportunities—such as the 2020 Supreme Court decision in McGirt v. Oklahoma, which reaffirms the jurisdiction of the Muscogee Creek lands in the state of Oklahoma—should be viewed as an opportunity to improve economic conditions for the Muscogee Creek and their neighbors. Indeed, this decision may have lasting implications for other Oklahoma tribes and reservation lands. The ruling re-emphasizes the importance of existing treaties that delineate tribal authority and resources; many of these treaties have been ignored or unenforced for generations.

These types of policy solutions could include a new funding mechanism through the Bureau of Indian Affairs that specifically encourages diversifying economic activities on reservation lands. Awards could be selected in a “race to the top” program, where the most innovative and implementable projects are awarded and serve as an example for other communities and tribal governments.

Tribal tax-exempt bonds could supplement these novel activities. Tribal governments have the authority to issue these bonds, but recent data indicate that only 17 percent of tribal governments have used this method to finance investment opportunities.¹³⁹ These types of bond issuances are the same ones used by municipal governments for local infrastructure investments and economic development activities all across the United States—they should be encouraged by other tribal nations as an extension of their sovereignty and an opportunity to develop novel industries on reservation lands.

Enact a tribal jobs guarantee program

Given the high levels of poverty and unemployment, on average, on American Indian reservations, a federal jobs guarantee for tribal citizens and residents would provide an important mechanism to improve earnings on reservations. There is significant justification for this program, given the high levels of existing unemployment and the additional impact of the coronavirus pandemic on employment.

The jobs needed should be determined at the tribal level, with the flexibility to adjust employment decisions to strengthen tribal sovereignty and long-run economic development planning for new and innovative industry development on tribal lands. A tribal jobs guarantee program alsowould align employment activities with large-scale climate mitigation efforts and/or ecosystem restoration efforts in tribal nations. Further, the jobs guarantee could also focus on other areas where there are well-identified deficits in service or program provision such as healthcare services.

An ancillary policy reform to encourage jobs growth would be to encourage the adoption of uniform commercial codes on American Indian reservations. These codes allow for a more seamless flow of goods and services across these borders. While all 50 U.S. states have adopted these codes, which facilitate trade in goods and services, not all tribal governments have adopted these codes.¹⁴⁰ Taking this step across all tribal reservations would boost jobs.

Restore tribal lands to tribal nation governance

The stealing of Indigenous peoples’ lands is too complex a story to discuss in detail here. Briefly, however, over the centuries, European settler-colonial populations stole, purchased, and seized land from American Indians. Then, during the tribal land allotment era (1887–1934) large parcels of remaining lands were deemed “surplus” by the Bureau of Indian Affairs and often sold at below market value. As a result, almost two-thirds of American Indian-controlled lands, as of 1887, were no longer under tribal or individual American Indian control by 1934.

Efforts to improve the land area controlled by American Indians is an important step toward improving the earnings of American Indians. Today, some American Indian reservations are “checkerboarded” in terms of land ownership, where one square block may be owned by a tribal government or an individual American Indian landowner while the adjacent parcels are held by non-American-Indians. This makes governing difficult, and it makes economic development and investment especially difficult because the parcels may not be large enough to support economies of scale.

As a result, there may be large areas that are not particularly productive because the land use and planning are not well-coordinated. This may also be the case for the preservation of forests, rivers, and lakes. Depending upon the size of parcels, there may be little incentive to protect and preserve large swaths of natural resources.

This, in turn, affects tribal members in a second manner. Tribal lands do not just serve as a place to start a business or to build a home. Reservation lands and the surrounding areas are often protected lands and regions such as forests, mountains, bogs, marshes, bays, lakes, or rivers. Increased access to lands could increase American Indians’ nonwage earnings through hunting, fishing, or gathering. These food resources may be an important component of Indigenous peoples’ diets, depending upon region and time of year.¹⁴¹

Tribal reservations need to be contiguous land to increase the productivity and use of those lands for hunting, fishing, trapping, or other nonmarket-based uses. The U.S. Congress should fund the purchase of reservation lands that have been previously sold to non-American-Indians and expand existing programs such as the tribal land buy-back program, which buys land from tribal members for tribal government use, including the buy-back of nontribal, member-owned reservation lands.¹⁴² This would increase the available tribal lands for economic development, land conservation, and nonmarket-based economic activities.

Reduce barriers to economic development

There are significant obstacles to economic development on reservation lands, and tribal nations have taken significant steps to improve those conditions in the past few decades. I briefly describe three specific areas that deserve more attention and are crucial to future economic activities:

Increase access to capital
Invest and expand infrastructure on tribal lands
Boost educational attainment and access

Investments in these three areas would create potential opportunities on tribal lands to increase retail and health services at least to reach parity with that of the adjacent, off-reservation locations and enable the development of agriculture and manufacturing on reservation lands due to the endowment of natural resources.

Increase access to capital

Several of the prior policy proposals refer to programs for tribal governments. Yet access to capital is essential to increase the opportunities for entrepreneurship for individual tribal citizens as well. There are several policy options that would either increase asset ownership or access to credit for tribal members. The first one is to fully fund and extend the Section 184 Indian Home Loan Guarantee Program administered by the U.S. Department of Housing and Urban Development. This program is an important source for home mortgages on American Indian reservations. The expansion of this program has put home ownership in reach for many more American Indians and may serve as an important method to assist in asset ownership for this population.¹⁴³

A second policy is to boost U.S. Small Business Administration loan guarantee programs, specifically the 7a program, which has been used extensively by American Indian-owned businesses in recent years. It is an important source of capital for business start-up and expansion funds.

A third proposal is to increase the capitalization of American Indian-owned and American Indian-serving Community Development Financial Institutions. In general, CDFIs are regulated by the U.S. Department of the Treasury and were set up in the Community Development and Regulatory Improvement Act of 1994. They are designed to improve the lending and financial services for underserved people and communities in the United States. Importantly, there is a Native American CDFI Assistance Program that facilitates the Native American-serving CDFIs.¹⁴⁴

The demand for credit often exceeds the supply from these CDFIs in and around American Indian reservations. These CDFIs are often composed of American Indian board members and community members who are better equipped to assess and conduct business with this population. These CDFIs increased their lending over the past decade where little to none had previously existed, and the number of Native American-serving CDFIs grew from just 14 in 2001 to 74 by 2016.¹⁴⁵ More funding from the U.S. Congress to expand the capital and lending power of these institutions would be money well-invested.

Invest and expand infrastructure on tribal lands

Increasing investment in infrastructure on American Indian reservations is an important means to improving access to employment and educational opportunities. There are two policies that can significantly upgrade or establish different types of infrastructure on tribal lands—improve the physical infrastructure on reservation lands and increase broadband access in rural and reservation communities.

Roads on reservations are often poorly maintained due to a lack of resources. They are often unpaved or severely potholed. Poorly maintained roads impede travel to and from schools, health clinics, and employment. These have long-term effects on the ability to live and work on or off of a tribal reservation.¹⁴⁶ Even worse, housing on tribal lands have some of the highest levels of incomplete plumbing and kitchen facilities.¹⁴⁷ Tribal homes have a higher likelihood of using wood or coal to heat homes in the wintertime, which often contributes to considerable respiratory illnesses affecting school attendance and employment. Appropriations for the creation and maintenance of tribal infrastructure come from the U.S. Congress. These appropriations are often based on existing treaties and agreements with tribal governments. These responsibilities should be honored by the U.S. government.

Second, there is an increasing need for increasing broadband access in rural and reservation communities. Prior to the coronavirus pandemic, internet access on reservations was low, compared to the United States as a whole, with about 61 percent of households on the median reservation with internet access, compared to about 69 percent in adjacent counties.¹⁴⁸ Some largely population reservations, however, fall below 55 percent access, such as the Navajo Nation. Broadband and reliable internet connectivity has taken on even more prominence amid the coronavirus pandemic, as many school-age children and college students rely on the web for their current educational pursuits. For some workers, broadband connection has meant that they are able to work from home while shelter-in-place mandates have been implemented.

In the absence of these connections, this would deal a severe blow to the working population and primary school, secondary school, and college students. The U.S. Congress needs to provide additional funding for the U.S. Department of Agriculture’s Tribal Broadband program because current funding allocations and plans are not comprehensive and cover only a small proportion of the uncovered tribal rural areas.

Educational attainment and access

An important determinant of earnings is educational attainment and experience. Educational attainment, on average, is lower for American Indians than the average U.S. citizen.¹⁴⁹ Additionally, there is evidence that school quality lags behind that of other race and ethnic groups in the United States.¹⁵⁰

That’s why Congress should increase federal funding for reservation-based schools and funding for the Bureau of Indian Education as required by dozens of U.S. treaties with American Indian nations. The Bureau of Indian Education is a direct mechanism to improve educational quality and access on reservations, while fully funding existing tribal colleges and universities would improve access to higher education in culturally relevant settings.

Improve data collection for American Indians

A pervasive obstacle to diagnosing and tracking economic development and earnings growth of the American Indian population is the lack of timely and disaggregated data. Due to their relatively small population size in the United States, American Indians comprise anywhere from 1 percent to 2 percent of the U.S. population, depending upon the definition used for “American Indian.” National longitudinal surveys tend to have very few American Indian observations in their samples. As a result, the only reliable data for this population tends to be data from the U.S. Census Bureau’s decennial censuses and the American Community Surveys.

While these two Census Bureau datasets provide useful information about the American Indian population at single points in time, the same individuals are not linked across time or place, and thus make it difficult to evaluate the impact of employment, training, and educational programs aimed at improving economic outcomes or earnings for American Indians. Therefore, there are far fewer credible research findings for this population, which diminishes the opportunities for advocacy and improved policy implementation.

The U.S. Congress should increase funding for the creation of longitudinal datasets focused on the American Indian population. Alternatively, existing surveys such as the U.S. Bureau of Labor Statistics’ National Longitudinal Study of Youth could oversample for these populations so that there would be a usable sample population. Additional longitudinal datasets such as the University of Michigan’s Health and Retirement Study could do the same. These efforts would all lead to increased tools for assessment of the earnings and well-being of the American Indian population over time and under different policies and programs.

Conclusion

Improvements since 1990 in the general well-being of American Indians residing on reservations detailed in this essay need to be expanded in three key ways. The first is the support and exercise of tribal sovereignty and the expansion of innovative industries on reservations. The second is to reduce the barriers to economic development on tribal lands and provide funding for educational institutions on tribal reservations to reduce a persistent barrier to the economic development and well-being of American Indian workers and their families. The third is to increase the data collection on the American Indian population in nationally representative datasets. Without these longitudinal datasets, we are unable to conduct the standard evaluation and analysis of existing training or employment programs on the success of American Indians. Investing in these longitudinal datasets will go a long way to improving our ability to assess which programs work and which do not.

—Randall Akee is an associate professor in the Department of Public Policy and American Indian Studies at the University of California, Los Angeles and a research associate at the National Bureau of Economic Research.