Goodhart’s law states that “When a measure becomes a target, it ceases to be a good measure.” This is what I think might be happening with the distribution of income or wealth. Economists, in particular, have a singular focus on the density plots and time-series graphs measuring income inequality, but sometimes we miss the underlying problems of which inequality is a symptom. When focused on inequality as the target of our interventions, we might emphasize solutions like guest worker programs, means-tested transfers, or private-sector payroll subsidies—all of which may compress the income distribution, but none of which may do much to redistribute power or alleviate oppression. Rather than focusing on lowering the 1 percent’s share, we might want to explore how a broad economic democracy will have a compressed distribution of income as a byproduct.

What do I mean by economic democracy? Analogous to its political namesake, it is an equal distribution of economic—rather than just civic—liberties, and foregrounds the autonomy, respect, and independence that people want in their economic lives. Instead of interpreting freedom or liberty as simply the size of the choice set, my working definition of economic liberty borrows a lot from “liberty as non-domination” as articulated by recent work in “neo-republican” political philosophy. Like freedom from arbitrary decisions of police and government officials, it means having some autonomy from your bank, landlord, or boss, as well as being able to quit your job or change your grocery store when you want. It also means interacting in markets where purchasing power is distributed with rough parity, so that the money-backed preferences of a few, however rightfully earned, do not have an outsized impact on the overall allocation of scarce resources. Instead of reducing income and wealth inequality to target some desired pattern, perhaps we should consider economic inequality as a diagnostic for an economy that is undermining the “social bases of respect” good societies should deliver to their citizens.


Let’s begin with economic liberty in contracts. Economic liberty is not simply liberty to contract, but also should restrain the discretion that principals (e.g., creditors, employers, landlords) have inside the contract. Ronald Coase, Herbert Simon, and legions of economic theorists have stressed the importance of authority in the employment contract, extending Marx’s insight that formal equality in the market becomes de facto inequality inside the firm. In efficiency wage models, contingent employment rents increase the income that workers get relative to being unemployed, but also give employers a credible threat to fire the worker (or not promote them). By threatening loss of employment rents (hence the contingency), employers extract effort, and other non-contractible services that employees can provide. As a stark example, prior to the secret ballot employers in Chile used these rents to induce inquilino workers to vote for right-wing parties. In 1958, the secret ballot was introduced in Chile, which unsurprisingly lowered the right-wing votes in high-inquilino municipalities. More surprisingly, the price of land fell faster in those municipalities, as the value of land had previously capitalized the controlled votes of the inquilinos on it. Asset prices incorporated political control as well as economic yields. Recent anecdotes and research, such as that of Alexander Hertel-Fernandez, suggest that employer mobilization of votes may be becoming more prevalent in the United States. Policies that raise employment rents (e.g., the Earned Income Tax Credit [EITC]) may increase earnings, but also could increase the power employers have over workers.

“No one should have their fundamental interests depend on the fickle goodwill of a superordinate.”

“If your boss was a total jerk, how miserable could he or she make your life?” is a question that could be posed in a survey. This captures what Philip Pettit has described as “republican freedom,” or the autonomy one has from the arbitrary whims of others. This is fundamentally different from the distinction between “positive” or “negative” freedom Isaiah Berlin burned into our normative lexicon years ago in “Two Concepts of Liberty.” A person can have abundant negative and positive liberties (e.g., free to move and speak, plenty of food and health care), but if those freedoms can be revoked at the whim of another, they do not have republican freedom. Pettit further gives a test of republican freedom: the “eyeball test,” where members of society can all look each other in the eye. I like to think of this as the “Don’t have to laugh at unfunny jokes of superiors” test, but they capture the same idea: no one should have their fundamental interests depend on the fickle goodwill of a superordinate, be they well-meaning philanthropists or foremen. Enslaved people lack freedom even if they serve the most benevolent master.

I wonder if the kinds of contingent employment rents that impinge on republican economic liberty thrive in highly unequal economies, where many have to work and few have the prerogatives of owners and managers. How many subordinate groups endure extra abuse from managers when outside options are scarce? Sexual harassment, workplace injuries, quasi-voluntary overtime, and innumerable petty indignities thrive with contingent employment rents. Recent research documents that workers exert more effort when the labor market is slack, and many models tie wages to bargaining power and outside options. This is the promise behind tight labor markets, public employment programs, and unconditional income grants: the threat of employee dismissal loses its bite. The petty tyrannies experienced on the late shift, outside the view of customers, are blunted by the credible threat of exit.

“Market society and income inequality have a menacing complementarity.”

Outside of the firm, but inside the market, economic democracy also means that purchasing power is widely distributed and monopolies are not tolerated. A simple result from general equilibrium theory is that when markets are complete and competitive, the economy implements something like a wealth-weighted Pareto-optimal allocation. Markets translate ownership of something scarce, like land, a good reputation, or useful skills, into money that enables command of all kinds of goods and services. This includes what Richard Musgrave called “merit goods,” things that should be allocated according to need rather than ability to pay. In an unequal society, the more things we allocate on the market, like health care, personal security, legal services, art, and education, the more it will reflect the interests of a narrow group. Some of the innovation at the top will trickle down, but unequal incomes quite simply mean that equal distributions of high-quality vital commodities will not occur. If public goods are provided by local government, housing prices will make sure they are best provided by enclaves of the wealthy. Market society and income inequality have a menacing complementarity.

Consider next the interaction of inequality and market power. Without limitless entry, businesses aim their goods and prices to the part of the distribution where the money is, and ration out a larger share of the population. This could also imply that the direction of technical change and innovation becomes slanted towards the preferences of the rich. Indeed, some research finds that recent inflation has been lower for the rich: the high-quality goods that rich Americans buy have fallen faster in price than other comparable goods. We get iPhones, TaskRabbits, and tailored organic food, while real estate prices and health care stay expensive.

The interaction also goes the other way, where market power, pervasive in many sectors, increases inequality. Monopolies reduce supply of production so as to increase prices, resulting in some customers who are willing to pay more than the cost of production but unwilling to pay the price charged by the company. Data-intensive targeting, which allows firms to target customers precisely, might mitigate inefficiencies due to market power, but this only increases profits. Since the distribution of shareholding is even more unequal than the distribution of wealth (because housing is relatively equally held), the additional profits realized by monopolized industries increase inequality even further. The rise in concentration may be a concern not just for inequality, but for growth: a potential explanation for the lack of productivity growth despite rapid technological progress in the past fifteen years is that increased concentration (which has been documented in virtually every industry) has allowed businesses to lower costs but keep prices high by not expanding production. One wonders if the generous definition of “competition” and narrow concern with aggregate consumer welfare espoused by recent antitrust jurisprudence has had a role to play.

Finally, there is the frequently noted pathology of political representation, where money influences representation and policy implementation. This problem, easily summarized as an oligarchic feedback loop, where income inequality generates pro-rich policy that generates more inequality, is the anti-democratic mechanism that is possibly most dangerous. This is widely discussed, and the only thing to add is that it is difficult to encourage markets in everything, particularly speech, and not expect it to lead to generalized capture of political expression.


“A classic set of institutions did a good job taking labor, in particular, out of the market: unions.”

A classic set of institutions did a good job taking labor, in particular, out of the market: unions. It is considered stubbornly naïve to propose rebuilding the labor movement as an object of policy. Wonkish technocrats tend, not without basis, to think of the labor movement as nice and historically necessary in its idealized form, corrupt and inefficient in its actual form, and a doomed anachronism in any case. But I think unions pushed on all three of the dimensions of economic power I sketched above. Shop stewards and grievance procedures addressed domination within the firm, collective bargaining and the higher earnings for regular workers addressed issues of market power, and voter mobilization and lobbying counteracted the political power of business. Union power is also a core challenge to liberal institutions of private property: every unfair labor practice filing, collective bargaining agreement, and strike is a claim to control a firm that is legally owned by shareholders and run by managers. There are a number of concrete policy ideas that would shift the balance of power in workplace politics and thus contribute toward economic democracy. None of these substitute for on-the-ground innovations in organizing at economic chokepoints, which is the only truly transformative strategy available, but they can tilt the odds. And the evidence, while mixed, suggests that unions shrank the top income share as a positive byproduct. It is hard for a company facing the prospect of a strike at the next collective bargaining session to hand out huge bonuses to management.

Such reforms include a comprehensive right to strike, where replacement workers are prohibited and secondary solidarity strikes are allowed—this is the most important tool in a wage bargain, and its efficacy has been muted over time, thanks to judicial precedent and the conventions of collective bargaining (e.g., binding arbitration). Indeed, fines might be imposed on corporations who use replacement workers.

Next is a federal repeal of right-to-work laws along with the elimination of exclusive representation, which allows for minority unions. Other changes would include promoting electronic card-check union recognition, streamlining the internecine jurisdictional conflicts among unions, and a massive increase in fines for intimidation of union organizing efforts. These reforms are not necessarily friendly to existing unions. The existing labor movement, and many of its pathologies, is a calcified creature of eighty years of manufacturing decline, seventy years of Taft-Hartley, and thirty-five years of employer and government attacks, and it may have to give way to new organizations able to energize and channel the demands of a broad, service-sector-oriented working class.

“An ideal system protects employees rather than jobs.”

This is no panacea. An ideal system protects employees rather than jobs. Firm-based unions are sometimes grotesquely parochial, despite strong incentives for an inclusive, productive membership to increase density. In the more fragmented collective bargaining environments, like the United States, narrow sectoral unions fail to internalize the costs their bargaining strategies might impose on other workers (nowhere is this more true than public sector unions). Another real problem for the labor movement is that the production process has become more dispersed across the value chain, so there are few industries where low-wage workers are employed by high-profit firms (Walmart and Verizon are exceptions, I suspect). And so it is hard to improve low wages by bargaining over profits when these profits are largely paid out to landowners, financiers, and other upstream suppliers. This suggests an organizing model and complementary National Labor Relations Board (NLRB) structures that account for the full value chain, bringing workers at the bottom into negotiations with the owners at the top, despite the many layers of contracts between them.

Finally, in a period of intense nationalism, it is still important to defend an expansive idea of who gets to share in any given prosperity. National equality should not automatically be prioritized over international equality, but nor should the arenas where agents can exercise political voice be ignored. Some trade-based defenses of inequality suggest that inequality within the US population is the price required to close the gap between countries. For example, Branko Milanovic and others pose the trade-off between 300 million Chinese workers catapulted out of subsistence since 1975 versus 300 million first-world workers experiencing relative affluence on the world scale, but stagnating relative incomes at the national scale. Others look instead at immigration, and suggest that staggering domestic inequality, for example the kind seen in the Gulf countries between citizens and migrants, is the best way to deal with global inequality. This cosmopolitanism, for better or worse, rips up any social contract that may have existed between states and incumbent citizens, and demands that we explicitly confront trading off the well-being of (desperately poor) noncitizens with (somewhat poor) citizens.

This is a difficult problem. I don’t have a solution, but I’d like to confront it with the aim of maximizing economic freedom rather than income alone. The problem with cross-country inequality is that it allows the rich agents (countries or corporations or philanthropists) to dominate the poor agents (via imperialism or sweatshops or charity). Reducing the vast international inequalities that exist is thus an urgent goal for neo-republicans. But casting the objective as non-domination means some trade-offs look worse than others. China’s growth in GDP is impressive, and its successful fight against poverty is to be nothing but applauded, but equally spectacular is the consolidation of its autocracy and extra-legal enrichment of its officials, both of which will likely slow its future growth. Guest worker programs can raise the incomes of the poor, but the brutal subordination endured by migrants shouldn’t be worn lightly because of that, and may impede investments in human capital or the development of less coercive labor markets that would pay off in the long run. Priorities of democracy and economic prosperity should be traded off at extremely steep rates, both at home and abroad.