One of the enduring faiths of modern progressive thought is that omniscient policy makers can cancel out the errors of one form of economic intervention by implementing a second. That lesson was brought home to me when I was a third year student at Yale Law School, whenever discussion turned to the perennial debate over the minimum wage. The charge against the minimum wage was that it had to introduce some measure of unemployment into labor markets by raising wages above the market-clearing price. “Not to worry,” came the confident reply. The way to handle that imperfection is to raise the level of welfare benefits in order to remove the dislocations created by the minimum wage. If one government program had its rough edges, a second government program could ride to the rescue. Implicit in this argument was the tantalizing, but fatal, assumption of economic abundance: The government has the power to tax, and with that power, has access to a cornucopia of public funds that never runs empty—at least until it does.
This abundance-based argument is not confined solely to the minimum wage, but has been extended to countless programs of state intervention in labor, or indeed, any market. Thus in 1935, American labor law created a system of collective bargaining whereby employees bargain with a single voice. That system allows unions to seek, and often obtain, monopoly profits for their members. That system in turn reduces the number of workers hired by the unionized firms. So what is to be done with the excess workers? They should be shepherded into job-training programs, funded by the public, which would allow them to reenter the labor force with other jobs.
Illustration by Barbara Kelley
For example, job training is the solution for those workers in the Northwest whose skilled jobs in the timber industry have been decimated by a variety of environmental diktats, of which the Endangered Species Act of 1973 is only the most notable. That same two-pronged strategy is evident in the American Jobs Act. Key provisions require recipients of government expenditures (of at least $50 billion) to adopt Buy American programs or to pay prevailing wages, both of which hobble the recipient firms. One predictable offset is tax credits to employers who hire long-term unemployed workers, coupled with yet another “Bridge to Work” job-training program.
The two-sided programs so popular in the United States also play a large role in the European Union, which has strongly collectivist labor policies. There, employers find it next to impossible to fire workers, to whom they owe a rich set of statutory benefits covering everything from maximum hours to minimum vacations, funded, of course, by government tax revenues. Meanwhile, displaced or unemployed workers receive generous welfare benefits, which reduce their incentive to find new work. The upshot is chronic levels of unemployment in countries like Greece, Italy, Spain, and Portugal, whose fragile financial conditions have put into doubt the survival of the Euro and indeed the European Union.
The massive level of economic dislocation both at home and abroad offers conclusive evidence that this venerable two-part strategy does not, and cannot work. Pinpointing its systematic errors is critical to avoid expanding on past mistakes. The proper approach is simple to state but hard to execute: Always seek “first-best” solutions. The correct response to any restriction on capital or labor is its prompt removal. A “second-best” effort to introduce some offsetting program only makes matters worse. The two errors do not cancel out. They cumulate.
A job-training program is a weak substitute for real work experience.
The point is made by looking at the interaction between tough minimum wage laws and high unemployment benefits. The minimum wage law introduces two immediate distortions into labor markets, which grow as the gap between the market-clearing wage and the minimum wage increases. First, it imposes huge administrative burdens on the Department of Labor and other state and federal agencies that have to enforce the law, and the private firms who have to be sure that they act in compliance with its commands. It is no easy thing to define an “hour” for the full range of jobs. There are no uniform answers for dealing with statutory exemptions, commuting time, work breaks, sick leave, or jobs away from home. Overtime pay is a world unto itself. The complex regulations needed to implement this one provision of the labor code require large investments from firms who need to avoid the heavy exposure to government sanctions and private lawsuits from noncompliance. None of these costs are eliminated by the adoption of any program of unemployment benefits or job creation, each of which imposes its own distinct, and costly, administrative overlay.
These costs have to be borne by someone else, and most of them are in fact covered by a combination of general revenues and specific unemployment taxes on current workers. Both of these programs produce additional distortions. Any tax on general revenues reduces the income available for both investment and consumption in all sectors of the economy. Unlike taxes that are imposed to create sensible infrastructure and other public goods, these unemployment taxes do not generate benefits for the parties taxed that equal even a small fraction of the costs that they impose. Firms do not benefit by paying income taxes to government agencies whose job it is to oversee their operations. The same can be said of specific taxes geared to fund unemployment programs, which hit most heavily those firms that have expanded their workforce in ways that reduce the ranks of the unemployed. Make no mistake about it: Any effort to cushion the blow of unemployment also functions, in both the short and long run, as an impediment to job creation. The effort to cushion the blow of unemployment necessarily adds to the ranks of the unemployed.
The two-part strategy also fails as a long-term measure. The consequence of higher rates of unemployment is the detachment of workers from the workforce. One of the serious mistakes of much labor market regulation, including the minimum wage law, is to assume that the only compensation given to employees is found in wages and benefits. But a sounder understanding of labor markets indicates that workers at all levels of the workforce also gain additional marketable skills from working on a steady job. For workers at the bottom of the ladder, those key skills could be as simple as knowing how to keep to a schedule, how to dress for work, how to take instruction, how to work in teams, and how to balance a ledger. For workers up and down the income distribution, idleness means a deterioration of work skills that reduces the potential for job advancement down the road.
Labor and capital markets are manifestly not free today.
Government job-training programs are a feeble substitute for real work experience. Labor markets are always dynamic while job-assistance programs are designed by agency bureaucrats who have all the flexibility of a Soviet bureaucracy. These agencies lack a profit motive, they are heavily budget-constrained, and they specialize in the use of outdated equipment for jobs that will have disappeared before the training program is completed. It has long been known that most graduates of these programs don’t get jobs. That trend continues today, especially in fields like energy.
If program participants do get jobs, they rarely lead to long-term employment. Another baleful consequence of the minimum wage is that it inhibits the job-training programs that employers give to their own prospective employees. It makes good economic sense for workers to accept a reduced wage, or even no wage, during a period in which they are in training for jobs that the employer will eventually offer. In these situations, the mismatch between training and placement is negligible relative to that of a government-training program, which necessarily lacks the tight connection between today’s training and tomorrow’s labor.
The adverse effects of these labor market restrictions are not confined to labor markets. They also negatively affect taxation and fiscal policy. One of the most dominant features of modern Keynesianism is the belief that, left to their own devices, markets in both goods and labor will not clear, so that there can be long-term forms of structural unemployment, which a government stimulus program could help correct. In dealing with this issue, Keynes rejected “Say’s Law,” first promulgated in 1803. In its most famous formulation, that law holds that supply creates its own demand, which applies to labor as well as goods. Keynes’s solution to this problem is the use of government stimulus programs to supply the additional demand for goods and services that markets, left to their own devices, cannot generate for the want of demand.
Of course, boom and bust cycles have been common both before and after the Great Depression. But it hardly follows that the source of all economic difficulty lies in the stunted operation of unregulated labor and capital markets. Most markets do not involve the simple barter of goods and services. Since ancient times, everyone has recognized that barter is a highly inefficient mode of exchange, for it is rare that the purchaser of one set of goods or services has for sale a market basket of goods and services that fits the needs of this or that particular purchaser. The introduction of a standard unit of exchange—money—in effect allows me to buy from A without having to sell to A a market basket of goods of comparable value. Vast new levels of efficiency are introduced into the market by the ability to sell goods to A for cash (or credit), which A can then use to buy goods and services from someone else.
The establishment of money and credit as mediums of exchange introduces vast efficiencies into all markets. But, at the same time, it introduces a pervasive form of government intervention into the private markets as long as the dollar (or any other currency) is subject to government control. Monopolies are always dangerous because they are unchecked. That general proposition applies to the government control of cash and credit. There are, at best, imperfect institutional constraints against flooding the market with cheap dollars, or by starving the market by taking too many dollars out of circulation. The former happened with the housing bubble of 2003-2008, while the latter happened with the deflation of the depression years.
It costs very little to eliminate elaborate forms of regulation.
In general, the effort to remove these government distortions by disciplining monetary policy and reducing overall levels of taxation will serve us better than government stimulus programs. A stable monetary policy will not lead to the long-term waste of capital. Real production will be in; busy-work will be out. Yet stimulus programs raise busy-work to an art form and thus lack the benefit of ordinary economic exchanges. The wage gains to workers are not paired with an improvement in the overall capital stock of the nation. This in turn depletes the remaining wealth available for investment or consumption in all future periods. Stimulus programs are like eating seed corn.
Learned economists write reams about whether the imperfections of private labor and capital markets justify imperfect stimulus programs. But they tend to assume that little can be done to fix the current imperfect operation of capital and labor markets. This would be the case if they were generally unregulated, but labor and capital markets are manifestly not free today. The obstacles to gains from trade in labor markets are larger than they have ever been before in the United States. In addition to regulations like minimum wage, overtime pay, and collective bargaining, the markets face countless other restrictions, like antidiscrimination and quota laws, insurance, other health-care mandates, and family leave policies.
The Keynesians have very weak explanations for why labor markets break down: They focus on “sticky” markets. “Stickiness” occurs when transaction costs prevent the quick readjustment of prices and wages to changes in conditions of supply and demand. But it is highly unlikely that this supposed phenomenon could carry the huge weight that is attributed to it, especially in unregulated markets. The point seems wildly overstated, especially today when there is near instantaneous re-pricing of everything from airplane tickets, to rental housing, to hotel rooms, to gasoline, to Google ads, and so on.
These profound microeconomic shifts are driven by information technology. In recent times, this added measure of price flexibility has largely removed any supposed macroeconomic obstacle to full employment. There is nothing about neoclassical economic theory that predicts that labor markets will always clear. In fact, it predicts the precise opposite. Burden these labor markets with the barrage of taxes and restrictions that they now face and it is virtually certain they will fail to clear. That simple insight gives a prescription for current policy. Government treasuries do not offer a cornucopia of riches that can be used to fix the short-term dislocations in labor markets that are caused by government intervention. Those resources will always run out, and when they do, there is no other government that can play the role of deus ex machina.
Rather than go down the road of foolish grand public initiatives that bankrupt nations, government policy, both in the U.S. and the E.U., should start at the other end. It costs very little to eliminate elaborate forms of regulation. A thorough-going program of deregulation will reduce the wealth that is directed into costly compliance and unwise transfer systems, and increase the total level of goods and services produced in the overall economy, which in turn should reduce the fiscal drain of government transfer programs.
Any comprehensive program that addresses labor markets at large cannot focus exclusively, or even mainly, on expanding job stimulus or transfer programs, both of which hamper the growth that is necessary to escape the current malaise. The systematic deregulation of labor markets offers the best, last hope of tackling unemployment. If only our political leaders understood that simple and powerful message. Unfortunately, they don’t. So expect more stagnation going forward.
Great essay, as always. The mountain of fixes upon fixes upon patches is indeed the biggest problem we face.
What I find particularly inspiring about his approach in this instance is the way it illuminates the incremental and synergistic impact of government interventions. Not only are they subject to ratchet effects—no bureaucracy ever abolished itself—but they also compete with, and confound, each other. If the economy were an airframe, most of what government does is add dead weight at random on positively undesirable locations, reducing the "design margin" and forcing the craft to operate ever closer to its limits and beyond. And when the "stall" buzzer sounds, the political leadership panics, points fingers, and hauls back even harder on the stick. The trajectory goes non-linear, and there is less and less altitude or airspeed in which to recover.
Mr. Epstein splendidly describes the current system and its costs. I read his article during a break in training a new employee on how to prepare certified payrolls (a US Department of Labor form used to report wages and hours on construction projects to which Davis Bacon regulations apply). When done, I will review some other employment related forms and then look up some old tax returns for some information needed to complete the current year’s return. Then I will distribute new applications to all employees regarding new regulations which change (not for the better, in my humble opinion) the withholding of local wage tax from employees. A check of my to-do list reveals several more days worth of similar projects. When I have completed all of these tasks, government employees will review and summarize the data for some unknown purpose. All in a day’s work in America, land of the free.