Report | Inequality and Poverty

What labor market changes have generated inequality and wage suppression?: Employer power is significant but largely constant, whereas workers’ power has been eroded by policy actions

Download PDF

A version of this report was presented at Unrigging the Market: Convening to Restore Competitive Labor Markets, Labor and Worklife Program, Harvard Law School, June 13, 2018.

What this report finds: Labor markets in capitalist economies are fundamentally tilted against individual workers’ ability to bargain effectively with employers. Policy does not have to be rigged for employers to give them particular clout in labor markets; instead, the very nature of these labor markets gives them clout. In the past, when economic growth was broadly shared across the population, it was because policymakers understood this basic asymmetry and used policy levers to bolster the leverage and bargaining power of workers. Conversely, recent decades’ rise of inequality and anemic wage growth has resulted from a stripping away of these policy bulwarks to workers’ labor market power.

Why it matters: Recent research on “monopsony power”—the leverage enjoyed by employers to set their workers’ pay—is a valuable contribution to our understanding of the asymmetry inherent in labor markets. However, “monopsony power” is often a confusing term to even the most savvy economic writers and researchers, and too often it is used only to describe markets that are concentrated (i.e., where there are relatively few employers). Market concentration can indeed suppress workers’ wages, but employer power exists even in markets with lots of employers. If only the narrow conception of “monopsony power” is recognized and policymakers focus only on interventions that target the effect of market concentration (antitrust, for example), then other measures that could more effectively restore the balance of power in labor markets might not get the consideration they should.

What can be done about it: There is no one panacea for restoring workers’ leverage and bargaining power in labor markets. Policymakers must be committed to working on every available margin, including restoring genuine full employment as a macroeconomic policy priority; reforming labor law so that workers who want to form a union to collectively bargain to improve their wages and working conditions are able to do so; raising the minimum wage; and strengthening enforcement of labor standards and workplace civil rights laws.


Since 1979, the bottom 90 percent of the American workforce has seen their pay shrink radically as a share of total income. Figure A shows total labor compensation for the bottom 90 percent as a share of all market-based income in the American economy. In 1979, this share was 58 percent, but as of 2015 it had shrunk to just under 47 percent. The amount of money this loss represents is staggering; had the 1979 share held constant, the bottom 90 percent of the American workforce would have had roughly $1.35 trillion in additional labor income in 2015, or about $10,800 per household.

Figure A

Pay for the bottom 90 percent loses ground: Labor income of the bottom 90 percent as share of total personal income, 1979–2015

Year Bottom 90 percent labor share
1979 58.0%
1980 57.6%
1981 57.5%
1982 56.8%
1983 55.5%
1984 54.7%
1985 54.3%
1986 51.8%
1987 54.1%
1988 51.7%
1989 52.3%
1990 52.9%
1991 53.6%
1992 52.6%
1993 53.4%
1994 53.4%
1995 52.5%
1996 51.1%
1997 50.2%
1998 49.6%
1999 48.1%
2000 47.1%
2001 50.3%
2002 51.6%
2003 51.1%
2004 49.5%
2005 47.4%
2006 46.5%
2007 45.8%
2008 49.0%
2009 51.1%
2010 49.0%
2011 48.9%
2012 46.1%
2013 47.5%
2014 46.6%
2015 46.6%
ChartData Download data

The data below can be saved or copied directly into Excel.

Note: Labor income for the bottom 90 percent includes cash wages, employer-provided benefits, and employer-side payroll taxes, as well as labor’s imputed share of unemployment and corporate taxes.

Source: Authors’ analysis of data from the Congressional Budget Office (2018).

Copy the code below to embed this chart on your website.

What happened in the American economy that drove this collapse in pay for the bottom 90 percent? We suggest that a good metaphor is a tug-of-war, where the bottom 90 percent of workers is on one side and corporate managers and capital owners (shorthand these two groups simply as employers) are on the other.1 What matters for the final distributional outcome of this tug-of-war, of course, is simply the relative strength of each side, and on these grounds the graph makes it obvious that the bottom 90 percent has lost enormous ground to their employers. But this raises three key questions:

  1. Did the force brought to bear by employers in this distributional tug-of-war get stronger, or did the force exerted by typical workers get weaker? While the final outcome of bargaining just hinges on relative strength, making policy decisions aimed at giving typical workers a fair shot at achieving wage gains requires a correct and precise diagnosis of what has changed in recent decades.
  2. What role has policy played in granting additional advantages to one side or the other? Did the relative strength of employers in recent decades grow because of apolitical technological changes? Did policymakers hand employers spiked cleats to give them extra purchase in the distributional tug-of-war (e.g., by allowing industry concentration to march forward unchecked by antitrust action)? Or did policymakers grease the floor under workers (e.g., by allowing labor law to wither as an effective guarantor of workers’ rights to organize collectively)?
  3. Can we fix the problem simply by breaking up employer market concentration—to make the playing field more “competitive”—and then allowing labor markets to self-correct? A new and exciting economics literature on market concentration has sometimes been characterized by economic observers as providing the dominant explanation for adverse wage trends in recent decades.2 If this were true, allowing for greater competition to break up market concentration could be a silver bullet for getting typical workers’ wages growing again. So is it possible to achieve a competitively “fair” contest between workers and employers that will lead to more equitable distributional outcomes simply by trying to tame employer power? Or are capitalist labor markets intrinsically tilted against workers, and can the distributional contest never be truly “fair” without policymakers being willing to weigh in on the side of workers?

Our short answers to these questions are:

  1. The biggest change in relative power between typical workers and their employers in recent decades has been a collapse of workers’ power. There is some evidence of increasing absolute employer power (e.g., through increased market concentration), but our view is that the bigger change remains the collapse of workers’ power.
  2. This collapse of worker power has been overwhelmingly driven by conscious policy decisions that have intentionally undercut institutions and standards that previously bolstered the economic leverage and bargaining power of typical workers; it was not driven simply by apolitical market forces.3
  3. No. The lodestar for economic policy should be balanced—not necessarily competitive—labor markets. Many of the policy changes that have undercut workers’ power cannot be characterized as simply being “uncompetitive” per se. In competitive markets in economics textbooks, both employers and employees lack power. But in real-world labor markets, employers rarely lack for power, and our strong view is that policymakers should care more about balancing labor market power between employers and workers than about trying to create labor markets that are competitive in the textbook sense of the word.

In the rest of this brief, we expand on these answers and also explore how the new economics literature on the effect of market concentration fits into our understanding of the sources of rising inequality and labor market power imbalances. Our conclusion regarding this new literature is that it is rigorous and eye-opening and largely reinforces the answers to our questions above rather than overturning them.

Claim 1: The biggest change in relative power between typical workers and their employers has been a collapse in workers’ power

The collapse over the last four decades in the share of national income going to the labor earnings of the bottom 90 percent, described above, has been accompanied by rising inequality and near-stagnant pay for most workers. This dynamic is arguably best represented by the divergence between the growth of compensation for the typical U.S. worker and the growth in economywide productivity. Figure B shows this divergence.4 In the 1950s and 1960s, hourly compensation grew at roughly the same rate as productivity. But from 1973 to 2016, productivity grew six times as fast as compensation for typical workers, with the vast majority of this gap driven by rising inequality. The gap since the 1970s between economywide productivity growth and growth in the typical worker’s pay is the footprint of an economy in which the benefits of growth are largely being captured by those at the top of the income distribution, leaving most workers behind.

Figure B

The gap between productivity and a typical worker's compensation has increased dramatically since 1973: Productivity growth and hourly compensation growth, 1948–2016

Year Hourly compensation Productivity
1948  0.00% 0.00%
1949 6.25% 1.55%
1950 10.48% 9.33%
1951 11.76% 12.35%
1952 15.04% 15.63%
1953 20.85% 19.55%
1954 23.52% 21.56%
1955 28.74% 26.46%
1956 33.95% 26.66%
1957 37.14% 30.09%
1958 38.16% 32.78%
1959 42.55% 37.64%
1960 45.50% 40.05%
1961 47.99% 44.36%
1962 52.48% 49.79%
1963 55.03% 55.01%
1964 58.51% 59.99%
1965 62.47% 64.94%
1966 64.90% 70.00%
1967 66.90% 72.05%
1968 70.74% 77.16%
1969 74.67% 77.88%
1970 76.61% 80.37%
1971 82.02% 87.10%
1972 91.25% 92.05%
1973  91.31% 96.75%
1974 86.98% 93.66%
1975 86.86% 97.92%
1976 89.68% 103.44%
1977 93.15% 105.79%
1978 95.98% 107.79%
1979 93.45% 108.14%
1980 88.59% 106.57%
1981 87.61% 111.02%
1982 87.79% 107.88%
1983 88.37% 114.13%
1984 86.96% 119.73%
1985 86.33% 123.43%
1986 87.34% 127.99%
1987 84.62% 129.12%
1988 83.87%   131.78%
1989 83.72% 133.65%
1990 82.24% 136.98%
1991 81.90% 138.89%
1992 83.07% 147.56%
1993 83.41% 148.37%
1994 83.84% 150.75%
1995 82.73% 150.86%
1996 82.81% 156.92%
1997 84.82% 160.50%
1998 89.19% 165.71%
1999 91.94% 172.08%
2000 92.92% 178.50% 
2001 95.59% 182.84%
2002 99.40% 190.72%
2003 101.66% 200.17%
2004 100.87% 208.21%
2005 100.07% 213.58%
2006 100.24% 215.48%
2007 101.73% 217.70%
2008 101.74% 218.24%
2009 109.72% 224.75%
2010 111.56% 234.28%
2011 109.09% 234.67%
2012 107.30% 236.51%
2013 108.34% 237.57%
2014 109.22% 239.30%
2015 112.81% 241.08%
2016 115.14%  241.76% 
ChartData Download data

The data below can be saved or copied directly into Excel.

Notes: Data are for compensation (wages and benefits) of production/nonsupervisory workers in the private sector and net productivity of the total economy. “Net productivity” is the growth of output of goods and services less depreciation per hour worked.

Source: EPI analysis of unpublished Total Economy Productivity data from Bureau of Labor Statistics (BLS) Labor Productivity and Costs program and of wage data from the BLS Current Employment Statistics, BLS Employment Cost Trends, BLS Consumer Price Index, and Bureau of Economic Analysis National Income and Product Accounts

Updated from Figure A in Raising America’s Pay: Why It’s Our Central Economic Policy Challenge (Bivens et al. 2014)

Copy the code below to embed this chart on your website.

Why growing employer concentration is unlikely to explain long-run trends

A recent empirical literature has attracted much attention for its hypothesis that growing market concentration has boosted the power of employers and suppressed wage growth. This literature has examined concentration in both product markets (monopoly power) and labor markets (one form of monopsony power).5 While this examination of market concentration is exciting and welcome, a close look at the empirical findings would argue that market concentration by itself is unlikely to be able to explain larger trends in American labor markets—like the growing gap between economywide productivity and pay for typical workers.6

Evaluating the role of market concentration in this growing pay–productivity gap since the 1970s can largely be informed by assessing how growing concentration affects the two channels through which increased productivity can bypass the pay of typical workers: (1) the erosion of labor’s share of income, i.e., a decline in the percent of the total income in the economy that is received by workers in wages and benefits and an increase in the percent that is received by owners of capital; and (2) increasing inequality in compensation, i.e., a decline in the percent of labor’s share of income that is received by low- and moderate-wage workers and an increase in the percent that is received by workers at the top of the pay distribution.

An increase in monopoly power means that firms can raise the prices that consumers pay, increasing corporate profits. This results in a shift in national income toward owners of capital and away from workers, i.e., an erosion of labor’s share of income, not increasing inequality of compensation within labor’s share of income. An increase in monopsony power means that firms can set wages lower than they would be able to in a more competitive labor market, which also results in a shift of national income away from workers. In other words, if increasing concentration were a key driver of the increase in the productivity–pay gap since the 1970s, we would expect to see a substantial erosion of labor’s share of income.

A caveat to this analysis is that it assumes that labor market concentration affects all workers equally and hence it does not increase compensation inequality. But if, for example, labor market concentration is more pronounced in economic sectors that disproportionately employ less-credentialed workers, then concentration could in theory contribute to rising compensation inequality. An empirical examination of the effect of labor market concentration on compensation inequality is hence a prime candidate for further research.

This caveat notwithstanding, it certainly seems fair to use the shift from labor to capital incomes as a proxy for the scope of concentration’s impact on the labor market. A decomposition of these two components of the productivity–pay gap (the erosion of labor’s share of income and increasing inequality of compensation) shows that typical workers lost substantial ground through both channels since the 1970s—but that rising inequality of compensation made up the vast majority of the gap. In particular, between 1973 and 2014, rising inequality of compensation made up 83.5 percent of the growth of the productivity–pay gap, whereas the erosion of labor’s share of income explains about one-sixth of the gap (16.5 percent).7 Importantly, if rising product and labor market concentration were primary drivers of the rise in inequality over these four decades, it seems that the erosion of labor share should have played a much larger role in the growth of the productivity–pay gap. The contribution of labor share’s erosion to this gap post-2000, however, is larger than over the entire period, so it is theoretically possible that rising concentration over that time period is pulling more weight in explaining large wage trends.

Bivens, Mishel, and Schmitt (2018) examine some of the recent research on labor market concentration; they find that the results of this research imply that increased labor market concentration between 1979 and 2014 reduced wage growth only by enough to explain about 3.5 percent of the total increase in the productivity–pay gap over this period. They also look at recent research on product market concentration and find that concentration in product markets likely reduced overall wages only by enough to explain 10 percent or less of the productivity–pay gap between 1979 and 2015. This suggests that, while it is clear from compensation trends since the 1970s that employers do wield more relative power vis-à-vis their workers, it is not clear that growing employer concentration explains a lot of this growing relative imbalance.

Employers’ market power (or ‘monopsony power’) does not just stem from market concentration

Growing concentration is, of course, just one possible manifestation of growing employer power. Other forms are highlighted in the “dynamic monopsony” framework best elucidated by Manning (2003). In this framework, labor market frictions (e.g., workers lacking complete information about employment alternatives, spatial mismatches, and the need for flexible schedules) give employers some measure of wage-setting power over workers. For example, there may be plenty of fast-food restaurants employing workers within a few miles of a given low-wage worker’s house, but only one of those restaurants is on a direct bus route that makes the commute manageable. Or only one allows for an efficient pairing of commuting and dropping kids off at school. Such frictions accumulate and grant employer-side power in the labor market.

While this dynamic monopsony approach is enormously valuable for understanding why many labor markets (particularly those for low-wage workers) don’t behave the way textbook models of competitive labor markets claim they ought to (as we discuss in more detail later), there has been little evidence so far mobilized to suggest that these frictions granting dynamic monopsony power have increased dramatically in recent decades. (An increase is what would be required for these factors to explain the rising productivity–pay gap.) It’s worth noting that growing attention is being focused on employer practices that create job frictions—such as requiring employees to sign noncompete agreements, setting up nonpoaching agreements with other employers, and using mobility-restricting temporary immigrant work visas.8 This attention is more than warranted and will almost certainly indicate that these practices have indeed increased in recent decades. Yet it seems to us that these developments (however worrisome) are likely dwarfed by clearly visible degradations in employee-side power in the labor market. Further, the degradation in employee-side power may in many cases be the thing that paves the way for employers to be able to adopt such practices.

The evidence clearly shows that erosion of worker-side power is the dominant factor in the power imbalance in the labor market

Likely the most important factor behind the collapse in workers’ bargaining power has been the erosion in the share of workers in a union, which fell from 24 percent in 1973 to 10.7 percent in 2017. Research demonstrates that this erosion has had a substantial impact on middle-wage workers, including both union and nonunion workers (Rosenfeld, Denice, and Laird 2016).9

Another post-1979 development that has undercut the bargaining position of low and moderate-wage workers is the rise in the average level of unemployment. Between 1949 and 1979 the unemployment rate averaged 5.2 percent; between 1979 and 2017 it averaged 6.3 percent. This is not just a result of the Great Recession—between 1979 and 2007 unemployment averaged 6.1 percent. Research has demonstrated clearly that low- and middle-wage workers’ pay is much more responsive to unemployment than the pay of highly paid workers. The post-1979 increase in average unemployment hence has predictably contributed to rising inequality and slow pay growth for the bottom 80 percent.

Finally, economic theory and evidence clearly indicates that growing trade with low-wage countries should boost wage inequality in the United States and lower wages for workers without a four-year college degree. This type of trade grew significantly since the 1970s. Imports from low-wage countries were equal to 0.7 percent of U.S. GDP in 1973, but 6.3 percent in 2016.

All in all, the direct evidence indicating a substantial decline in workers’ power in the labor market is much easier to see in the data than evidence indicating an increase in employers’ power. This, of course, does not mean that employer power in labor markets is trivial or should be ignored. It may not have changed dramatically in recent decades, but it has been an ongoing fact of labor markets for decades. It has simply become more visible in recent years: as the countervailing power of workers has been stripped away, the relative strength of employer power has increased, contributing to substantial slowdowns in wage growth.

The view that labor market concentration and other specific sources of employer power have always been present, but were tamed in previous decades by countervailing worker power, is consistent with the empirical findings of Benmelech, Bergman, and Kim (2018); they find that the wage-suppressing effect of labor market concentration is lessened when union coverage is strong. So if labor market concentration has been relatively constant, but the countervailing force imposed by unionization has eroded, this combination could well have led to significant compensation losses.

Claim 2: The collapse of workers’ power has been driven by conscious policy choices

The prior section suggests that it has always been the case that American labor markets are riven with forces—concentration and other frictions—that impede competition and, all else equal, give employers the power to set wages lower than the competitive wage. This section will show that the key difference between the post-1970s period (when the pay of most workers and economywide productivity diverged) and previous decades (when pay and productivity grew in tandem) is that in the earlier period, these sources of employer power were more likely to be compensated for by institutions and policies that provided countervailing power to workers. In the recent period, many of these institutions and policies have been eroded or rolled back, with nothing to replace them as sources of countervailing worker power.

Take, for example, the higher average unemployment rate characterizing the post-1979 period (versus the three prior decades, as discussed in the previous section). This is not just a sad accident. Instead, macroeconomic policy (particularly monetary policy) has prioritized steady and very low inflation over low unemployment in recent decades. Even by too-conservative standards set by official estimates of the natural rate of unemployment, macroeconomic policy has failed to secure full employment for the large majority of these years. It is no coincidence, in our view, that the only period of strong, across-the-board wage growth since 1979 was during the late 1990s and early 2000s, when unemployment was allowed to fall far below levels that had previously been thought to lead to accelerating inflation.

Similarly, the steady erosion of union coverage is not the natural evolution of a modern economy, as is often claimed. Instead, it is the result of a sustained policy assault on workers’ right to effectively organize. Almost half (48 percent) of workers polled said they’d vote to create a union in their workplace tomorrow if they got the chance, indicating that union decline was not driven by workers’ preferences (Bivens et al. 2017). Instead, a failure of policy to keep the playing field level—between workers hoping to organize and employers willing to engage in ever more aggressive practices to stymie these campaigns—has driven this decline (Bronfenbrenner 2009).

For the bottom end of the labor market, the policy assault on their bargaining position is obvious: the federal minimum wage is now roughly 25 percent lower in inflation-adjusted terms than it was at its height in 1968, even though productivity has nearly doubled and low-wage workers have become far more educated in the intervening years (Cooper 2017). Notably, policymakers have failed to enact sufficient increases in the federal minimum wage despite growing economic evidence that most minimum wage increases since 1990 (at the federal or state level) have not caused measurable employment loss, contrary to predictions of competitive labor market models (Cooper, Mishel, and Zipperer 2018). This finding of no measurable job loss is consistent with low-wage labor markets that are characterized by monopsony power held by employers. In models of monopsony, legislated wage increases can lead to higher wages and greater—not reduced—employment.

Further, employers have pursued an aggressive host of practices meant to limit workers’ bargaining position, and policymakers have not, for the most part, taken the actions necessary to curb these practices through strengthened standards or enforcement. These aggressive employer practices include (among other things) requiring workers to sign mandatory arbitration agreements with class and collective action waivers as a condition of employment, misclassifying workers as independent contractors, and not providing workers with predictable schedules.10

Finally, the rise of globalization that has pressured American workers’ wages is often presented as inevitable—simply driven by technology and other countries’ decisions to join the global trading system. There is some truth to the latter, and these developments were likely always going to be hard on American workers’ wages. But this trade-induced redistribution has been amplified by policy decisions. Trade agreements in recent decades have sought to maximize labor market competition between workers in the U.S. and abroad while simultaneously boosting protections for corporate profits.11 Further policy actions that amplified the wage-depressing effects of globalization include tolerance of an overvalued dollar, which has led to trade deficits and manufacturing job losses. Globalization’s pressure on American wages, in short, clearly has deep policy roots that look like intentional assaults on the economic leverage of typical workers.

Claim 3: The lodestar for economic policy should be balanced, not necessarily competitive, labor markets

The previous sections highlight that it is primarily a collapse of workers’ power, rather than an increase in employer power, that has imbalanced the distributional tug-of-war in recent decades and led to a rise in inequality. We have also shown that the fingerprints of intentional policy decisions are all over this collapse in workers’ power.

However, many of the key policy changes that undercut workers’ power cannot necessarily be characterized as making the labor market less “competitive.” The textbook conceptions of competitive labor markets are those where both employers and workers lack power. In fact, an implication of a perfectly competitive model of the labor market is that policy interventions that intentionally reduce union power and/or the purchasing power of legislated minimum wages push labor markets closer to the textbook definition of competitive.

The recognition that “deviations” from competitive models do not occur only on the worker side of labor markets is an obvious motivation for the new attention many authors are giving to market concentration. However, as we note above, the empirical bite of concentration on wages does not seem large enough to explain a large portion of the gap between typical workers’ pay and economywide productivity. Relatedly, there is little evidence indicating that concentration has increased significantly enough in recent decades to provide a compelling explanation of these wider trends in wages. We are therefore pessimistic that labor market outcomes for workers can be improved by relying on tools that tame employer power.

Our pessimism also stems from our assessment that bargaining between employers and workers always takes place on an unlevel playing field—even in nonconcentrated markets that have not been riven by noncompete agreements or other explicit aids to employer power. Labor markets are generally tilted against individual workers simply because workers have only one job to lose while employers typically have access to plenty of workers, so workers are naturally hit harder by employment relationships that dissolve. The fact that power is exercised—even in labor markets characterized by free entry of employers—has often been overlooked or even denied by economists. Alchian and Demsetz (1972) argue, for example, that

The firm…has no power of fiat, no authority, no disciplinary action any different in the slightest degree from ordinary market contracting between any two people…Wherein then is the relationship between a grocer and his employee different from that between a grocer and his customer? (777)

Hart (1989) provides one of the clearest statements of how labor markets are different from product markets, and how employers are indeed likely to be able to exercise power over their employees:

…the reason an employee is likely to be more responsive to what his employer wants than a grocer is to what his customer wants is that the employer…can deprive the employee of the assets he works with and hire another employee to work with these assets, while the customer can only deprive the grocer of his custom and as long as the customer is small, it is presumably not very difficult for the grocer to find another customer. (1771)

In the end, our reading of the evidence on labor market trends is that employer power is ubiquitous and that it persists even when there is free entry into markets, i.e., even in labor markets that are not concentrated. This employer power is simply a fact of modern capitalist labor markets. Perhaps the best recent mobilization of evidence in support of the view that labor markets’ “natural” state is characterized by significant employer power is provided by Naidu and Yuchtman (2016), who look at American labor markets in the Gilded Age (late 19th and early 20th century). They note:

The American Gilded Age labour market came extraordinarily close to the archetypical labour market taught in Economics 101. One might think that in a world without regulatory red tape, the labour market would simply equate supply and demand, establishing a “market wage” for a unit of labour, eliminating non-competitive rents, and diminishing the stakes from conflict. One might further assume that in such a world, the institutions resolving such conflict would be irrelevant. However, we provide evidence that despite the lack of regulation, economic frictions in the labour market generated rents, and costly and violent conflicts over these rents were pervasive.

These findings reinforce our judgment that it is not possible to construct a labor market that conforms to competitive ideals found in textbooks. Instead, the “prelapsarian” state of capitalist labor markets is precisely one in which power rules and employers have disproportionate power. The only period that has seen sustained growth up and down the wage distribution, the decades following World War II, was one during which political mobilization led to strong policy buttresses for labor’s bargaining position.12 We worry that this point might be obscured by a recent flurry of interest in one facet of employer power—market concentration—that may be ameliorable in some cases with policy action (antitrust, for example). Restricting policymakers’ attention only to pushing back against violations of competitive markets that empower employers will not lead to a restoration of healthy and equitable wage growth.


Some economists and policymakers might express unease at the view that the downsides of one deviation from “competitive” markets (either labor market frictions or market concentration or some other source of employer power) should be countered by introducing another market imperfection (e.g., unions or a binding minimum wage). But this unease is unwarranted. The “theory of the second best” clearly argues that once markets depart at all from perfect competition, efficiency may well be increased by further departures. For example, in the case of monopsony power in low-wage labor markets, legislated minimum wage increases can potentially move wages closer to efficient levels and increase employment.

At the macroeconomic level, this claim that stripping away bulwarks to workers’ power has failed to lead to efficiency gains seems extremely well supported by the evidence. While many of the policy changes that have limited workers’ market power since the 1970s were done explicitly in the name of efficiency-seeking, the rate of productivity growth (a measure of how much income is generated in an hour of work—the most common macroeconomic measure of economic efficiency) slowed radically in the years after the mid-1970s, as seen in Figure B.13 This slowdown was briefly reversed in the late 1990s by the large investment in information and communications technologies associated with the widespread adoption of the internet and by a period of tighter labor markets (Bivens 2017b). But this brief surge soon failed and productivity continued growing much more slowly than in previous periods—when policy had consciously supported the leverage of typical workers. In short, the policy movement to disempower workers not only led to less equal growth, but was also associated with significantly slower growth.

When we assert that most of the policy change that led to inequality and slower growth was focused on disempowering workers—and that policy going forward needs to work to reempower workers—we certainly do not mean to imply one should ignore potential policy opportunities that could erode employer power (e.g., through more robust antitrust enforcement). But the larger opportunities are likely those that lead to more labor market balance in the power between employers and workers by increasing worker power—not trying to move the labor market toward a competitive ideal that is not attainable.


1. It is true that in the real world distributional conflicts are more complicated than a contest between two parties, but tug-of-war can be multipolar as well, as highlighted in Wells 2013.

2. Covert (2018) and Weissman (2018) are examples of extremely savvy economic journalists who have made large claims about the potential of market concentration to explain long-run wage trends.

3. An obvious corollary is that no new institutions or standards that provided such leverage and bargaining power were put in place to replace those that were lost.

4. In Figure B, we use compensation for production and nonsupervisory labor to represent a typical worker’s pay. Production and nonsupervisory labor account for roughly 80 percent of the private-sector workforce.

5. Readers should not be misled by the “mono” in both monopoly and monopsony. Markets—whether product or labor markets—can be concentrated even when there is more than one buyer or seller.

6. See Bivens, Mishel, and Schmitt 2018. We should note that the finding that concentration alone cannot explain large wage trends in recent decades is not a criticism of this literature. The papers themselves generally make no such claim and the authors tend to emphasize a broad portfolio of shifts in market power that have affected wages. However, commentary and analysis of these papers often do make these claims, both implicitly and explicitly. It is also worth noting that not all of the rigorous papers in the recent literature find increasing concentration in labor markets. For example, Rinz (2018) finds that industrial concentration at the national level was roughly the same in 1976 as in 2015, while concentration at the local level was generally declining over that period.

7. See Bivens and Mishel 2015 for these decompositions. It is, however, worth noting that the erosion of labor’s share of income became more important after 2000, though it still explains less than half of the productivity–pay gap over this period. Declining labor share explains 46.3 percent of the gap from 2000 to 2014 compared with 53.7 percent explained by rising compensation inequality.

8. See the “Immigration” and “Good Jobs” sections of EPI’s policy agenda (EPI 2018).

9. The erosion of unions harms the wage growth of nonunion workers by depriving this latter group of wage standards set by unionized firms in industries and geographies where unions are strong. The extension of union wage standards to nonunionized firms in a given industry is sometimes called a “union threat effect” or “spillover.”

10. See McNicholas, Sanders, and Shierholz 2018 for an explanation of how employment practices make jobs unfair to workers from the first day of work and how policy changes can remedy this.

11. These protections included expansions of American intellectual property protections to other countries and new legal forums for multinational corporations to contest regulatory actions that could reduce their profits. See Bivens 2017a and Rodrik 2018 for further discussion of how trade agreements have become a vehicle for corporate rent-seeking.

12. It goes without saying that this earlier postwar period between 1947 and 1979 also saw many terrible inequalities—in particular along dimensions of race and gender. Pointing out the more equitable growth along income or wage dimensions of that period is certainly not a call to “make America great again” or romanticize the past. But we should not ignore the clear evidence that more progressive class-focused policies in the earlier period worked as intended, often by explicitly seeking to remedy imbalances in labor market power between employers and employees. It should also be noted that these class-based policies provided disproportionate aid to workers of color. For example, median income growth for black families between 1947 and 1979 was significantly faster than for white families (2.8 vs. 2.4 percent). Some of that relative gain was likely driven by political mobilization to remedy race-based discrimination in society at large and the labor market in particular, but a significant portion of this gain can be explained by the fact that policies aiming to boost wages and incomes for the bottom 90 percent of the population will, all else equal, always provide disproportionate benefit to nonwhite workers, who make up a disproportionate share of this bottom 90 percent.

13. Between 1973 and 2017, net productivity grew half as fast as it had from 1948 to 1973.


Alchian, Armen A., and Harold Demsetz. 1972. “Production, Information Costs, and Economic Organization.” American Economic Review 62, no. 5 (December): 777–795.

Benmelech, Efraim, Nittai Bergman, and Hyunseob Kim. 2018. “Strong Employers and Weak Employees: How Does Employer Concentration Affect Wages?” Working paper, January 2018.

Bivens, Josh. 2017a. Adding Insult to Injury: How Bad Policy Decisions Amplified Globalization’s Costs for American Workers. Economic Policy Institute, July 2017.

Bivens, Josh. 2017b. The Potential Macroeconomic Benefits from Increasing Infrastructure Investment. Economic Policy Institute, July 2017.

Bivens, Josh, Lora Engdahl, Elise Gould, Teresa Kroeger, Celine McNicholas, Lawrence Mishel, Zane Mokhiber, Heidi Shierholz, Marni von Wilpert, Valerie Wilson, and Ben Zipperer. 2017. How Today’s Unions Help Working People: Giving Workers the Power to Improve Their Jobs and Unrig the Economy. Economic Policy Institute, August 2017.

Bivens, Josh, Elise Gould, Lawrence Mishel, and Heidi Shierholz. 2014. Raising America’s Pay: Why It’s Our Central Economic Policy Challenge. Economic Policy Institute, June 2014.

Bivens, Josh, and Lawrence Mishel. 2015. Understanding the Historic Divergence Between Productivity and Typical Workers’ Pay: Why It Matters and Why It’s Real. Economic Policy Institute, September 2015.

Bivens, Josh, Lawrence Mishel, and John Schmitt. 2018. It’s Not Just Monopoly and Monopsony: How Market Power Has Affected American Wages. Economic Policy Institute, April 2018.

Bronfenbrenner, Kate. 2009. No Holds Barred–The Intensification of Employer Opposition to Organizing. Economic Policy Institute, May 2009.

Congressional Budget Office. 2018. The Distribution of Household Income, 2015. Supplemental data available in a downloadable Excel file.

Cooper, David. 2017. Raising the Federal Minimum Wage to $15 by 2024 Would Lift Wages for 41 Million Workers. Economic Policy Institute, April 2017.

Cooper, David, Lawrence Mishel, and Ben Zipperer. 2018. Bold Increases in the Minimum Wage Should Be Evaluated for the Benefits of Raising Low-Wage Worker’s Total Earnings: Critics Who Cite Claims of Job Loss Are Using a Distorted Frame. Economic Policy Institute, April 2018.

Covert, Bryce. 2018. “When Companies Supersize, Paychecks Shrink.” New York Times, May 13, 2018.

Economic Policy Institute (EPI). 2018. Policy Agenda. December 2018.

Hart, Oliver. 1989. “An Economist’s Perspective on the Theory of the Firm.” Columbia Law Review 89, no. 7 (November): 1757–1774.

Manning, Alan. 2003. Monopsony in Motion: Imperfect Competition in Labor Markets. Princeton, N.J.: Princeton Univ. Press.

McNicholas, Celine, Samantha Sanders, and Heidi Shierholz. 2018. First Day Fairness: An Agenda to Build Worker Power and Ensure Job Quality. Economic Policy Institute, August 2018.

Naidu, Suresh, and Noam Yuchtman. 2016. “Back to the Future? Lessons on Inequality, Labour Markets, and Conflict from the Gilded Age, for the Present.” VOXeu website, August 2016.

Rinz, Kevin. 2018. “Labor Market Concentration, Earnings Inequality, and Earnings Mobility,” CARRA Working Papers 2018-10, Center for Economic Studies, U.S. Census Bureau.

Rodrik, Dani. 2018. “What Do Trade Agreements Really Do?Journal of Economic Perspectives 32, no. 2 (Spring): 73–90.

Rosenfeld, Jake, Patrick Denice, and Jennifer Laird. 2016. Union Decline Lowers Wages of Nonunion Workers: The Overlooked Reason Why Wages Are Stuck and Inequality Is Growing. Economic Policy Institute, August 2016.

Weissman, Jordan. 2018. “Why Is It So Hard for Americans to Get a Decent Raise?Slate, January 16, 2018.

Wells, Jeremy. 2013. “International Tug of War: Simulating Polarity and Alignment” (blog post), PAXsims, September 24, 2013.