How are wages set in the United States? The standard neoclassical economic model asserts that the wage reflects the marginal revenue for the firm produced by the worker. In a defense of welfare payments that compensate for low wages for workers in The Week, James Pethokoukis invokes this theory:

… [E]conomics won’t be ignored. If workers at a big profitable company only generate $10 an hour of revenue, then the company won’t pay them $15 an hour.  A company should not be expected to lose money on a worker, especially to hit a wage number pulled out of thin air by politicians.

Among the many problems with this theory is the fact that it implicitly rules out supply and demand in a free market in labor.  For the sake of argument, let’s say that a welder at an oil and gas company engaged in fracking is paid $21 an hour (which happens to be the median at present). Suppose there is a fracking boom and the scarcity of welders allows the same welder to charge $30 an hour, merely to do the same work while the firm earns the same profits as before. Alternately, suppose the market collapses. While the firm’s revenues are steady, mass unemployment among welders means that the same welder is willing to do the same job as before for $15 an hour. Are we really to believe that the “objective” contribution of the welder to the firm is worth $15 an hour, no matter what labor market conditions are? If the firm agrees to pay the worker $30 an hour because welders are scarce, even though the worker’s contribution is “only” worth $15 an hour, is the firm committing economic suicide? If the firm pays the desperate worker during the bust $15 an hour, is the firm, to use Marxist language, expropriating the “surplus value” added by the welder, in the amount of $6 an hour, the difference between the worker’s new, reduced hourly wage and the worker’s “objective” hourly contribution to the firm?

Remember, this theory of compensation holds that the exact amount that a particular employer will pay just happens to be exactly what an individual employee is “worth” (in objectively-measured contribution to the enterprise). But between upper limits at which they would bankrupt the enterprise and lower limits at which no workers would accept them, wages are indeterminate. Even among employees, how does Amazon determine the share of gross profit from each sale of The Road to Serfdom that was “generated” by the software engineer at headquarters and how much by the stockpicker in the warehouse? And how did they get paid at all during the many years when Amazon turned no profit?

Where the wage is ultimately set within that band of possible wages is determined not by objective, measurable, impersonal market forces but rather by the relative bargaining power of workers and employers.   This was recognized not only by the classical economists Adam Smith and J.S. Mill but also by one of the founders of modern neoclassical economics, Alfred Marshall: “It is this unfairness of bad masters which makes unions necessary and gives them their chief force.”

Something like a free market in labor does exist in occupations like welder, gardener, maid, nanny, and construction day laborers, in which there are many small employers and a large pool of low-wage workers with little bargaining power, a pool made larger by illegal immigrants in the workforce, who are disproportionately represented in these jobs relative to U.S. citizens and law-abiding immigrants. Notwithstanding economic theory, employer lobbies are convinced that tight labor markets lead to higher wages by increasing the bargaining power of workers, even when workers are not represented by unions. That is why these lobbies spend huge sums on propaganda claiming that the U.S. is afflicted by “labor shortages” in this and that sector, shortages which can only be cured by increasing the labor force—by allowing more immigration, by delaying the retirement age, by encouraging both parents in a household to join the workforce. Conversely, organized private sector unions in the U.S., for generations before American organized labor became an obedient wing of the Democratic party, traditionally favored restrictive immigration, along with breadwinner wages and maternalist and protective laws that limited the participation of mothers in the work force, to make labor markets tighter.

Rational employers want loose labor markets and a buyer’s market in labor, to drive down wages. Rational workers want tight labor markets and a seller’s market in labor, to drive wages up. This is simple to understand, even if many academic economists and libertarians take pains not to understand it.

But the role of supply and demand (as distinct from marginal worker contribution to revenue) in determining the price of labor is relevant only in highly competitive sectors in which something like a free market in labor actually exists. Many Americans, including almost all of those who are better-paid, work in oligopolistic sectors characterized by large firms and imperfect competition in which the link between labor market supply and demand is attenuated, if it exists at all. For the more than half of Americans who work for corporations with 500 or more employees, compensation is set largely by price-fixing by their employers on the basis of collusion with other firms in the same business.

It’s not called price-fixing, of course. Under U.S. antitrust law, it is illegal for employers to band together and agree on the same employee compensation policies, in order to allow the firms to present a united front to job-seekers who might hope for a bidding war among employers competing to hire them.

In 2016, the Justice Department issued an “Antitrust Guide for Human Resource Professionals,” reminding the HR staffs of firms that they cannot get together to set common prices for labor, to undercut the bargaining power of workers:

From an antitrust perspective, firms that compete to hire or retain employees are competitors in the employment marketplace, regardless of whether the firms make the same products or compete to provide the same services. It is unlawful for competitors to expressly or implicitly agree not to compete with one another, even if they are motivated by a desire to reduce costs. Therefore, HR professionals should take steps to ensure that interactions with other employers competing with them for employees do not result in an unlawful agreement not to compete on terms of employment. Any company, acting on its own, may typically make decisions regarding hiring, soliciting, or recruiting employees. But the company and its employees should take care not to communicate the company’s policies to other companies competing to hire the same types of employees, nor ask another company to go along.

The Department of Justice guidance continues:

An individual likely is breaking the antitrust laws if he or she:

  • agrees with individual(s) at another company about employee salary or other terms of compensation, either at a specific level or within a range (so-called wage-fixing agreements), or
  • agrees with individual(s) at another company to refuse to solicit or hire that other company’s employees (so-called “no poaching” agreements).

Let’s set aside non-compete agreements for another time and look at the first violation of antitrust law in compensation policy. In practice, many U.S. firms get around this restriction by hiring third parties like consultants to tell them what other companies are doing, rather than asking other companies directly. In other cases, HR officers set salary bands on the basis of Bureau of Labor Statistics (BLS) data about average salaries in particular occupations. (I doubt that the founders of BLS intended for its statistics to be used to create hiring cartels.)

Astron Solutions, a consulting firm that specializes in HR issues, provides a list of ways that firms can avoid running afoul of antitrust law:

It can be difficult for Human Resources to explain and adhere to these anti-trust implications when a manager fears losing talented staff to a competitor. Here are some viable, proactive steps you can take now to obtain data and stay in compliance with the law:

  1. Work with local SHRM, WorldatWork, or other Human Resource organization chapters to offer confidential annual and quarterly “hot job” salary surveys, with the data compiled by a disassociated third party.
  2. Work with the local Chamber of Commerce or Industrial Association to develop a similar survey process.
  3. Invite competitors to meet with a disassociated third party to create a confidential survey group.
  4. Incorporate “new salary” questions into your exit analysis process to track where employees are going and for what salary level.
  5. Survey potential employees who reject offers to determine if salary played a part in their decision.

As this shows, the practice of most of corporate America (as well as the nonprofit and government sectors) is utterly at odds with the neoclassical theory of wages promoted by the well-funded libertarian intellectual movement. In the real world of American business, firms do not analyze the specific contribution of each worker to the firm’s profits or productivity, according to some objective metric—if that were even possible. And in the real world of American business, wages are not even set by supply and demand in competitive markets. Instead, large firms typically hire workers into predetermined salary bands, based on the salary bands of other firms in the same business.

Within these salary bands, compensation is so divorced from individual merit that a frequent theme of HR literature (I have read it, so you don’t have to) is the crisis that occurs when particularly meritorious workers run up against the upper limit of the salary band for their particular job descriptions. At that point, the HR literature tells us, if a high-performing worker asks for a raise, the company should politely but firmly say no. Giving the employee a raise might shatter the glass ceiling of that particular salary band! The very foundations of the corporate bureaucratic hierarchy might tremble and crack! Essential secretaries might demand higher pay than useless vice presidents! Unthinkable.

Instead of giving the talented employee a raise, here is what the Society for HR Management (SRHM) recommends:

“Red circle rates” are salaries/wages that are above the maximum rate the organization has established for the position’s salary range. Strategies to rectify red circle rates include the following:

  • In lieu of base salary increases, offer star employees a bonus that is roughly the amount of what the pay increase would have been. This allows for recognition of an employee’s outstanding performance without raising his or her base pay even more.
  • Explore developmental opportunities to facilitate promotion into the next pay grade.
  • Restrict further salary increases by freezing pay.

The visible hand of HR has replaced the invisible hand of the free market in labor. Nothing could be further from the fantasy of a free labor market in which all individuals are paid on the basis of their objectively-measured personal contributions to firm productivity and sales than this military-style system of bureaucratic ranks within major companies, in which all workers performing similar work are paid a salary within a particular range between a minimum and a maximum.

To return to Pethokoukis:

If workers at a big profitable company only generate $10 an hour of revenue, then the company won’t pay them $15 an hour.  A company should not be expected to lose money on a worker, especially to hit a wage number pulled out of thin air by politicians.

The answer to this is that many if not most wage numbers at big profitable companies are already pulled out of thin air by HR. Individual firms do not determine the salary bands on the basis of each particular employee’s isolated contribution to sales. Rather, on the basis of information provided by consultants, BLS or other third parties,  firms collude with each other to fix the costs of labor by using informally-shared salary bands to create de facto hiring cartels.

Salary bands with maximum salary caps reduce the ability of low-level workers to bargain for higher wages at hiring or promotion, to the benefit of all firms that collude in this informal wage-fixing system. At the same time, this method of compensation can help greedy CEOs lobby for raises or more stock options for themselves. All the CEO needs to do is pay a consulting firm to provide a report to the company’s board of directors showing what CEOs are paid at “comparable” commercial firms or nonprofit institutions. It would be a foolish consulting firm indeed that told the Board that the CEO who hired it is making too much money, not too little.

Far be it from me to criticize salary bands. Their existence confirms what intelligent economists like Smith, Mill and Marshall knew all along—namely, that within broad ranges wages are indeterminate and set by bargaining. Informal bargaining among firms, mediated by third party consultants or the BLS, to set salary bands that are the same for all workers in the same category in the same industry results in one form of wage determination. Collective bargaining among managers and employers is another.

Indeed, the existing intra-industry salary band system lends itself to collective bargaining at the level of an entire sector, rather than individual firms. The salary band system can be thought of as a kind of informal sectoral bargaining among firms, to fix wages for particular occupational categories to the detriment of labor. The parties are the managers and HR departments of individual firms, plus third parties like consulting firms which are used to evade antitrust law. Given this existing system of sectoral inter-firm wage setting, it would be easy to slot in representatives of organized labor.

The problem with the salary band system is not that it ties wages to particular generic occupational categories in medium to large firms. There is no realistic alternative. The idea that it is possible to precisely calculate the exact value added by the unique “human capital” of each individual adds on a case-by-case basis in a firm with 50 or 500 employees, down to dollars per hour, is a fantasy that exists only in the dream world of neoclassical academic economics and the propaganda of libertarian think tanks (which, I should note, like other think tanks typically compensate their fellows and staff on the basis of…salary bands).

To paraphrase the architect Robert Venturi on the architecture of Las Vegas, the salary band system is almost all right. The main problem with the salary band system is that representatives of workers are not included in the wage  determination process. The existing system should be modified by the inclusion of representatives of workers in some form, not smashed by the hammer of antitrust law. Organized labor has traditionally been skeptical about antitrust policies, both because large firms tend to pay better than small ones and because for generations antitrust was used by governments to prosecute labor unions as conspiracies against the supposed free market in labor. In this realm as in others, antitrust theory is based on simple-minded libertarian axioms about free markets in goods, services, and labor that are at odds with reality.

In much of the U.S. economy, a free, highly competitive market in labor does not exist and will never exist. Instead, in many industries, wages are set by informal collusive coordination among all of the firms in a sector. Only a small step is required to convert the existing informal system of  collusive bargaining among managers to set wages in a sector into a formal system of collective bargaining among managers and labor in the same sector.

Michael Lind
Michael Lind is a columnist at Tablet, a fellow at New America, and the author of more than a dozen books, including Hell to Pay: How the Suppression of Wages is Destroying America (2023).
Recommended Reading
A New Centrism Is Rising in Washington

Call it neopopulism: a bipartisan attitude that mistrusts the free-market ethos instead of embracing it.

Oren Cass on Curbing the Free Market

On the DishCast with Andrew Sullivan, Oren Cass goes deep on conservative economics and much more.

Republicans’ appetite for government

Most want expansion, not cuts, of entitlement programs, according to American Compass survey.