In 1906, President Theodore Roosevelt, unwisely reading Upton Sinclair’s muckraking novel The Jungle (1905) at the breakfast table, was so revolted by the depictions of what went on in the meatpacking industry that he flung the book out the White House dining-room window.

You will be tempted to do the same with Philip K. Howard’s new book, Not Accountable: Rethinking the Constitutionality of Public Employee Unions. The impulse will be due not to horror but to anger at what public service in the United States today has become. For, as Howard makes all too clear, it is now a veritable conspiracy between elected officials and public-service workers to defraud the taxpayers.

The system siphons off billions of dollars a year to its beneficiaries—the members and officers of public-sector unions—while giving very short shrift to the public at large. The nation’s children, all too often trapped in schools that are organized strictly for the benefit of the teachers, suffer the most.

It is an old maxim in political science that “today’s reform is tomorrow’s problem.” When new rules are put in place to eliminate corruption, people find ways to exploit the new situation and corruption begins to creep back into the system in new ways.

The Founding Fathers, aware of how entrenched bureaucrats made government inefficient and corrupt in Europe, hoped that democratic elections would prevent such entrenchment in this country. Elections, however, did not prevent a descent into malfeasance, and by the 1820s, as Howard notes, “Federal jobs had little turnover and . . . had become sinecures rife with inefficiency and corruption.”

When Andrew Jackson became president in 1829, he tried to reform the situation by introducing what he called “rotation in office.” This measure, he hoped, would align the bureaucracy with the country’s majority party and prevent bureaucratic deadwood. But it soon degenerated into the spoils system, where each change of parties in the White House resulted in a wholesale turnover of government jobs, which were filled by people chosen for their party loyalty and not their competence.

Demands for reform began to build again, especially after President James A. Garfield was assassinated by a disappointed office seeker in 1881. The new president, Chester A. Arthur, shepherded the Pendleton Act through Congress in 1883. It created a civil-service system that was supposed to remove politics from public service, allowing public employees to keep their jobs regardless of party as long as they performed adequately. In other words, it was emphatically not a tenure system. The president retained the power to fire employees who did not do their jobs.

At first, the civil-service system applied to only about 10 percent of the federal workforce. But later presidents, in order to lock in their own appointees to federal jobs, kept expanding the system to cover more and more employees. The various states, plagued with the same problems, also began adopting the civil-service model.

Without the discipline imposed by the need to make a profit, few public employees will ever be as hardworking as those in profit-seeking corporations. But the system worked reasonably well. Then, beginning in the 1960s, the federal government as well as many state governments made the worst public-policy mistake of the twentieth century. They began to allow public employees to unionize and collectively bargain over wages (although not in the federal government) and working conditions.

Franklin Roosevelt was decidedly pro-labor, and the passage of the National Labor Relations Act of 1935, often called the Wagner Act for its chief congressional sponsor, Senator Robert F. Wagner, gave the labor movement a major boost. By the early 1950s, unions in the private sector represented about 35 percent of all workers (today that figure is down to about 6 percent). But fdr adamantly opposed the idea of allowing public employees to unionize, as did George Meany, the head of the afl and the afl–cio after the merger of the two labor organizations in 1955.

In the private sector, management and labor collectively bargain, in effect, over how to divide the profits both sides worked to create. But governments don’t create profits. So at best, when government and public employees bargain over wages, they are deciding how much of other peoples’ money (i.e., the taxpayers’ money) to spend. And as Milton Friedman famously pointed out, no one spends other peoples’ money as carefully as he spends his own.

Worse, in the private sector neither management nor labor has any influence over who sits on the other side of the negotiating table. But public-service unions can, and frequently do, make massive campaign contributions to politicians. They expect, and receive, a handsome return on their investment. As governor of New Jersey, Jon Corzine told a rally of government workers that “We will fight for a fair contract!” having apparently forgotten that his job was to minimize the state government’s labor costs.

But public officials are elected by taxpayers, and so salaries offered in collective bargaining are limited by public opinion. The public, however, pays much less attention to fringe benefits, such as pensions and health care, because the economic consequences are down the road, not up-front.

As a result, pensions in the public sector are now much more generous than in the private sector, which has largely converted from defined-benefit to defined-contribution plans anyway.

Retirement in the private sector is usually at age sixty-five. In the public sector it can be much earlier. Some firefighters and policemen can retire with full benefits after twenty years of service. Often these early retirees go right back to work for government and end up getting double pensions, known as double-dipping. After one village official in Illinois finally retired for good, he was the recipient of no fewer than three pensions.

In the private sector, unused sick days expire at the end of the year. In the public sector they often accumulate and can be used to pad pension benefits. It is common for public workers in their last year on the job to run up extensive overtime in order to maximize their pensions still further, their pensions being based on their final-year compensation.

In order to keep up current services, many governments have put off funding future pension benefits. The city of Detroit was forced into bankruptcy in 2013, when unfunded pension liabilities made up 40 percent of its $18 billion debt. Howard reports that “Illinois’ state pension liability (not even including municipal pensions) was so high that every household in the state would have to pay $65,000 to cover the difference.”

Besides negotiating outsized pay and benefits, public-employee unions have been able over time to impose work rules that run up costs. It costs two to three times as much to collect garbage in Chicago as it does in other cities with fewer union-imposed rules. And these rules have essentially ended the most powerful tool any management has to require employees to do their jobs: the power to fire those who don’t.

Howard provides many examples of the results of these work rules. For instance, in 2017 Reuters compiled a list of police officers who had repeatedly abused innocent people. One officer had severely beaten a college student who was guilty of only the minor infraction of drinking beer in public. Reuters showed how the officer had been the subject of no fewer than forty similar complaints. But he could not be dismissed, for the union contract required that prior complaints be expunged from the record after only a few months. So, Howard says, “it’s almost impossible for supervisors to terminate repeat offenders.” A Washington Post report on police departments in thirty-seven large cities calculated a dismissal rate of only two-tenths of 1 percent.

Among teachers, the dismissal rate is even lower in many places. In New York City in 2006–07, only eight teachers out of 55,000 were terminated for poor performance. Why? Because as one supervisor said, “Dismissing a tenured teacher is not a process. It’s a career.” New York City has a number of teachers who, unfit for the classroom but not subject to termination, sit all day in what are called “rubber rooms,” doing nothing while receiving full pay and benefits.

As Howard explains,

Public unions have created a modern spoils system. Just as the spoils system ran government for the benefit of campaign supporters of the winning party, public unions control government for the benefit of public employees. Like the old Tammany machine of New York, public unions have consolidated their political might to advance policies aimed at keeping public employment as a sinecure, unmanageable and unreformable.

But it is even worse than that. Under the spoils system the people could, from time to time, “throw the rascals out.” The public-union spoils system, however, is “encased in legal entitlements and powers.”

What can be done? In 2011, the Republican governor of Wisconsin, Scott Walker, with large majorities in both houses of the legislature, was able to ram through—over ferocious opposition from Democrats—reforms that limited public-sector collective bargaining to base pay only.

Howard points out that public-sector collective bargaining violates both the U.S. Constitution and all state constitutions in at least two ways.

First, it violates the non-delegation doctrine, which forbids either the executive or legislative branches from delegating the powers given them by the people under the Constitution. When a labor contract goes to arbitration, it is the arbiters, not the government, that make public-policy decisions, such as pay scales.

Further, Howard thinks that public-sector collective bargaining violates the “Guarantee Clause” of Article IV of the Constitution, which guarantees to every state a republican form of government. James Madison defined such a government as one “which derives all its powers directly or indirectly from the great body of the people . . . not from . . . a favored class of it.” The Supreme Court has usually shied away from ruling on whether something was a “republican form of government,” saying that the issue is a political one and thus nonjusticiable. But Howard notes that in Baker v. Carr (1962), which enshrined the doctrine of one man, one vote, the court ruled that the “nonjusticiability of such claims has nothing to do with their touching upon matters of state governmental organization.”

Philip K. Howard has written a short, important book on an urgently needed reform that gets only more urgent with every new labor contract negotiated by governments and public-sector unions. For, as the late economist Herbert Stein noted in his famous “Stein’s Law,” “if something cannot go on forever, it will stop.”