Download PDF

Systematic erosion of workers’ power relative to their employers has suppressed US wages

Politicians in both US political parties now acknowledge wage stagnation and have adopted narratives claiming that “the system is rigged.” Some focus on the number of immigrants and on what they see as unfair trade with China. Others focus on monopolies charging higher prices and reaping huge profits. There is, however, no agreement on what, and who, rigged the system.

In fact, as my new paper with colleagues Josh Bivens and Heidi Shierholz, “Explaining Wage Suppression” shows, wages have been kept low in the United States because workers have been systematically disempowered as a result of corporate practices and economic policies that were adopted—or reforms that were blocked—at the behest of business and the wealthy. This lack of worker power has caused wage suppression, increased wage inequality, and exacerbated racial disparities. The specific mechanisms behind this shift in power are excessive unemployment, globalization, eroded labor standards and their lack of enforcement, weakened collective bargaining, and corporate structure changes that disadvantage workers. To reestablish patterns of growth that benefit the vast majority requires new policies that center on rebuilding worker power.

Over the past four decades, the earnings of the top 1 percent and 0.1 percent increased 158 percent and 341 percent, respectively. While economy-wide productivity rose almost 70 percent, hourly compensation for typical workers increased less than 12 percent. And since 2000, labor’s share of income has steadily eroded.

These wage patterns and economic outcomes can be explained by the cumulative effect of the following policy decisions and corporate practices, which have systematically undercut the bargaining power of the majority of workers.

Contractionary policy

Monetary, fiscal, and trade policies have led to excessive unemployment, defined as unemployment above full employment. Since 1979, Federal Reserve Board policymakers have worried too much about inflation risks and have not embraced the benefits of full employment. This approach has created excessive unemployment, which has hurt low- and moderate-wage workers the most, an impact concentrated among Black workers.

If we had maintained full employment of 5 percent between 1979 and 2007, median wages would have been between 18 and 28 percent higher by 2007 (Bivens and Zipperer 2018). The corrosive effects of the extended period of high unemployment following the Great Recession, caused by austere federal and state fiscal policy, compounded wage problems further. This excessive unemployment alone explains the 9 percentage point increase in the wage gap between Black and White workers since 1979.

Globalization on capital’s terms

Globalization has proceeded, profoundly shaped by intentional policy decisions that maximized its wage-suppressing effects, with Black and Latinx workers disproportionately negatively affected. Bivens (2013) finds that trade flows with low-wage nations were likely reducing wages for workers without a four-year college degree by roughly 5.6 percent—almost a $2,000 annualized loss. Other studies confirm this finding.

Globalization driven solely by technological change and political changes in our trading partners was always likely to depress wage growth. But US policy failures significantly amplified these damaging effects: failing to secure reasonable compensation or a countervailing domestic boost to bargaining power for those on the losing end; failing to address currency misalignments causing large trade deficits and manufacturing job losses; and passing trade agreements that undercut workers’ economic leverage while protecting corporate profits. This encouraged US-based employers to substitute imports for the production formerly carried out by US workers.

Weakened unions

The key wage-setting institution for middle-wage workers has been collective bargaining, so the erosion of union representation has been the major factor depressing wage growth in the middle for men (whose unionization fell far more than among women). That collective bargaining leads to more equal wage outcomes, and declines lead to inequality, was firmly established in academic literature published in the mid-1980s. Fortin, Lemieux, and Lloyd (2019) show that union erosion explains from 29 percent to 37 percent of male wage inequality growth and 37 percent of the growing gap between high-wage and middle-wage men. This implies that eroded unions lowered the male median wage by 0.33 percentage point each year. There was a smaller impact on women’s wage inequality, explaining roughly 10 to 13 percent of the overall growth and 13 percent of the growth of the top-half wage gap.

The decline in private sector union coverage (from about 23 percent in 1979 to just 7 percent in 2019) can be explained primarily by changes in corporate practices and legal judgments that undercut workers’ ability to unionize and bargain. Polling evidence points to a huge unmet demand for collective bargaining: in 2017, almost half of nonunion workers wanted collective bargaining, an underrepresentation equivalent to approximately 58 million workers.

Weakened labor standards

A multidimensional reduction in labor standards and their enforcement has helped suppress wages.

The dramatic failure to maintain an appropriate value of the minimum wage has undercut the earnings of the bottom third of workers, heavily affecting Black, Hispanic, and women workers. The federal minimum wage in 2019 was 25 percent below its peak value in 1968, even though productivity doubled. The erosion of the minimum wage explains most of the shift in the wage gap between low- and middle-wage workers.

Overtime protections have also been weakened. While nearly half of salaried workers in 1975 were covered by the overtime threshold—salaried workers below this level automatically qualify for overtime—that share had fallen to just 10 percent of salaried workers by 2014.

Reduced enforcement of labor standards compounds wage problems by allowing extensive theft of wages: wage theft occurs when employers fail to pay workers the wages they are entitled to, including paying below minimum wage or failing to pay overtime. Low-wage workers lost more than $50 billion to wage theft in 2016, far exceeding the loss of stolen property in robberies. Wage theft falls hardest on low-wage and immigrant workers and represents about 3 percent of the total wages of the bottom 60 percent of earners. It is not possible to gauge how much wage theft has grown over four decades, but experts believe that it is now epidemic—as such, it is reasonable to say that an additional 1.5 to 2 percent of wages are now lost as a result of wage theft compared with 1979.

Immigration policy generates labor-standard-free zones in the labor market. Roughly 6 percent of the workforce, including undocumented and guest workers, lacks full labor protections from employers’ exercise of market power. This growth of an exploitable immigrant workforce undercuts wage and employment standards and puts downward pressure on wages. Note that the focus is not on immigrants taking jobs from others; rather, the focus is on weak labor standards and protections that leave immigrants open to exploitation.

Employers have also come up with innovative agreements that workers are forced to sign, limiting their job prospects and their ability to challenge employers in courts and with government agencies: these agreements suppress wages, as they are intended to do. Noncompete agreements, for instance, bar workers at one company from going to work for a competing business and now dim the prospects of between 28 percent and 46 percent of private sector workers. Forced arbitration provisions compel workers to take discrimination charges, wage and hour law violations, and other matters to corporate-dominated arbitration—frequently as individuals and not as part of a collective action—rather than to courts. These agreements covered 56 percent of nonunion private sector workers in 2018.

Millions of workers are being deliberately misclassified as independent contractors when they are actually employees, denying them social insurance protections, workplace protections (anti-discrimination, collective bargaining), and health and pension benefits. This is the case not only with leading ride-sharing firms such as Lyft and Uber but also in trucking, construction, and janitorial services.

Corporate structures

Changes in corporate structure have shifted power, and income, between firms and between employers and employees.

The most pronounced way employers have shaped labor market outcomes to their advantage is through “fissuring.” Fissuring is domestic outsourcing with dominant firms maintaining tight controls over prices and outcomes via standards and other mechanisms. This redistributes profits to the dominant firm and worsens wages and working conditions at the subcontracted firms. Between a fifth and a third of the economy is characterized by fissuring, a much larger share than four decades ago, when it was likely half that. A speculative gauge of the impact of fissuring is that a shift of 15 percentage points of employment into fissured workplaces where workers earn 15 percent less would imply a 2.25 percent overall decline in wages.

Major firms increasingly coerce their suppliers into giving them low prices. Wilmers (2018) quantified dominant buyers’ ability to squeeze supplier profits and lower wages. The share of nonfinance suppliers’ revenue obtained from dominant buyers increased from 5 percent in 1979 to 19 percent in 2014 overall, including from 6 percent to 26 percent in manufacturing and logistics. Wilmers estimates that this lowered annual earnings by 3.4 percentage points among publicly owned nonfinancial firms. The impact among low- and moderate-wage workers was almost surely larger.

Congress deregulated airlines, trucking, interstate busing, telecommunications, utilities, and railroads in the late 1970s, which lowered the compensation of blue-collar workers. Nine percent of the workforce in the 1980s was affected by industry deregulation, which eroded middle-wage jobs and lowered wages so much that it explains the almost 7 percent higher male wage inequality between 1979 and 1988.

Both monopoly (dominance of a few sellers in a product market) and employer concentration (a few employers dominating the job market) lead to lower wages, but these factors have not significantly suppressed wages.

Employer concentration affects primarily rural areas and does not seem to have grown over the past four decades. Labor’s share of income has declined, but it is not evident that this is the result of monopoly profits rather than eroded worker power. The rise in labor’s share of income as unemployment fell during the recent recovery suggests a strong role for eroded workers’ power, since monopoly power does not necessarily decline with lower unemployment. The enormous profits of the five big tech firms do represent a monopoly problem, although the role in wage suppression is difficult to assess and has not been explored.

Road ahead

The majority of people will benefit substantially from future growth only if policies are enacted that center on restoring bargaining power to the typical worker. Keep unemployment low. Provide adequate labor standards, including family leave, sick leave, and other policies. Effectively enforce laws. Enable worker options by eliminating forced arbitration and noncompete agreements. Rebuild collective bargaining. These policies will not only reestablish economic fairness, they will also invigorate civic engagement and democracy and promote freedom in and outside the workplace.

The conventional wisdom of much economic punditry has comforted the elite for many decades by proclaiming that the wage stagnation of the past four decades was simply the unfortunate by-product of economically progressive forces, such as globalization and automation. The prima facie case for an automation explanation has been lacking for two decades: college graduate wages have faltered, and the rate of automation has been historically low. Globalization could have been handled differently. Moreover, this comforting narrative conveniently overlooks the superlative growth of income for executives and others in the top 1 percent, which did not result from their special skills or from automation. The policy debate in both parties has moved beyond this “automation-driven skills deficit” narrative, and there is now wide agreement that politics and policy will determine whether workers get a fairer share of economic growth.

Opinions expressed in articles and other materials are those of the authors; they do not necessarily reflect IMF policy.