Fixing Broken Government: Put Humans in Charge
The Veterans Affairs scandal of falsified waiting lists is the latest of a never-ending stream of government ineptitude. Every season brings a new headline of failures: the botched roll-out of Obamacare involved 55 uncoordinated IT vendors; a White House report in February found that barely 3 percent of the $800 billion stimulus plan went to rebuild transportation infrastructure; and a March Washington Post report describes how federal pensions are processed by hand in a deep cave in Pennsylvania.
The reflexive reaction is to demand detailed laws and rules to make sure things don’t go wrong again. But shackling public choices with ironclad rules, ironically, is a main cause of the problems. Dictating correctness in advance supplants the one factor that is indispensable to all successful endeavors — human responsibility. “Nothing that’s good works by itself,” as Thomas Edison put it. “You’ve got to make the damn thing work.”
Responsibility is nowhere in modern government. Who’s responsible for the budget deficits? Nobody: Program budgets are set in legal concrete. Who’s responsible for failing to fix America’s decrepit infrastructure? Nobody. Who’s responsible for not managing civil servants sensibly? You get the idea.
Modern government is organized on “clear law,” the false premise that by making laws detailed enough to take in all possible circumstances, we can avoid human error. And so over the last few decades, law has gotten ever more granular. But all that regulatory detail, like sediment in a harbor, makes it hard to get anywhere. The 1956 Interstate Highway Act was 29 pages and succeeded in getting 41,000 miles of roads built by 1970. The 2012 transportation bill was 584 pages, and years will pass before workers can start fixing many of those same roads. Health-care regulators have devised 140,000 reimbursement categories for Medicare — including 12 categories for bee stings and 21 categories for “spacecraft accidents.” This is the tip of a bureaucratic iceberg — administration consumes 30 percent of health-care costs.
Legal detail skews behavior in ways that are usually counterproductive. Why did VA officials regularly falsify waiting times? Bureaucratic metrics required them to meet waiting time deadlines — or else they would forfeit a portion of their pay. Why didn’t they just do a better job? Compliance was basically impossible: Congress had mandated more VA services but only modestly expanded resources. Undoubtedly, better efficiency could have been squeezed out of available resources, but that would require liberating VA officials from civil-service straitjackets so they could manage other civil servants. Rigid bureaucracy, not the inexcusable dishonesty of VA officials, was the underlying cause of the VA scandal.
“Clear law” turns out to be a myth. Modern law is too dense to be knowable. “It will be of little avail to the people,” James Madison observed, “if the laws be so voluminous that they cannot be read, or so incoherent that they cannot be understood.” The quest for “clear law” is futile also because most regulatory language is inherently ambiguous. Dense rulebooks do not avoid disputes — they just divert the dispute to the parsing of legal words instead of arguing over what’s right. Indeed, legal detail often undermines the regulatory goal. “The more exact and detailed a rule, the more likely it is to open up loopholes, to permit by implication conduct that the rule was intended to avoid,” Judge Richard Posner observed.
What’s the alternative? Put humans back in charge. Law should generally be an open framework, mainly principles and goals, leaving room for responsible people to make decisions and be held accountable for results. Law based on principles leaves room for the decision-maker always to act on this question: What’s the right thing to do here?
“The more exact and detailed a rule, the more likely it is to open up loopholes, to permit by implication conduct that the rule was intended to avoid.”
Until recent decades, law based on principles was the structure of most public law. The Constitution is 10 pages long and provides basic precepts — say, the Fourth Amendment prohibition on “unreasonable searches and seizures” — without trying to define every situation. The recent Volcker Rule regulating proprietary trading, by contrast, is 950 pages, and, in the words of one banker, is “incoherent any way you look at it.”
Legal principles have the supreme virtue of activating individual responsibility. Law is still supreme. The goals of law are centralized, but implementation is decentralized. Every successful regulatory program works this way. New airplanes, for example, must be certified as “airworthy” by the FAA. There are no detailed regulations that set forth how many rivets per square foot are required. It’s up to the judgment of FAA officials. This system works pretty well. Which would you trust more, a plane approved by experts at the FAA or a plane that was allowed to fly merely because it satisfied a bunch of rules, many outdated?
Simplifying regulation — replacing thick volumes of rules with guiding principles — has two more virtues as well. First, democracy is effective only when there’s someone to hold accountable. Second, principles are coherent. People generally know what’s expected of them. Doctrines such as “unreasonable risk” or a “nutritious meal” or “industry standards” have practical meaning and can be enforced by reference to social norms. “Standards that capture lay intuitions about right behavior,” Posner notes, “may produce greater legal certainty than a network of precise … non-intuitive rules.”
Potentially, simplifying regulation can appeal to both sides: to liberals because it offers regulators more leeway, and to conservatives because it simplifies government and avoids mindless compliance costs.
Here are three examples of how regulation could be simplified:
Oversight of social services: Today, nursing homes, day-care centers, and similar social-service providers are regulated with a maze of input-oriented regulations. “Food shall be stored not less than 15 cm above the floor”; “there shall be .09 recreational workers per resident” — about a thousand rules in most states for nursing homes.
Australia had a similar regulatory structure. But in the wake of scandalous revelations of poor nursing homes in the late 1980s, it abandoned the thick rule book and replaced it with 31 general principles, for example to provide “a homelike environment” and to honor residents’ “privacy and dignity.” The result was an almost immediate transformation for the better. Nursing-home employees started acting on their instincts of right and wrong, instead of trudging through dreary bureaucratic checklists. Regulators and family members engaged in regular dialogues with nursing homes on how to improve things. Nursing homes became nice.
They abandoned the thick rule book and replaced it with 31 general principles. Nursing home became nice.
Environmental review: Environmental review and other infrastructure approvals can last a decade or longer in America. Even projects with virtually no environmental impact can last years, as project sponsors jump through scores of bureaucratic hoops.
The benefits of streamlining approvals would be enormous: several million new jobs, a greener environmental footprint, and enhanced global competitiveness. Replacing America’s antiquated power grid, for example, would save at least 7 percent of electricity — equivalent to the output of 200 coal-burning power plants.
Today the process is interminable, because any naysayer can complain that some pebble was left unturned — and who knows what will happen in court? Far better to give an environmental official responsibility to decide when important facts have been set forth instead of letting the process spin its wheels for a decade and then end up in court. For other permits — for instance, for land-use regulations, navigable-waters approval, landmarks review, and the like — there should also be a “one-stop shop” — a lead agency with the job of coordinating all regulatory concerns. That’s how other greener countries such as Germany are able to approve new infrastructure projects in a fraction of the time it takes in the United States.
Civil Service: More than 20 million people work for federal, state, and local government. Most of them perform needed services. But the accretion of antiquated and unjustifiable work rules has rendered them practically unmanageable.
Hiring and promotion is largely based on written tests, not demonstrated competence. Promoting an exemplary employee is often impossible. Work rules can prevent supervisors from asking workers to pitch in. In New York City, how to use a new copying machine and who can use it is subject to collective bargaining. Firing an incompetent employee under civil-service bureaucracy is almost impossible.
Any critique of this regulatory jungle is met with sanctimonious remonstrations about workers’ rights and the return of the spoils system. But the only relevant criterion for any regulatory structure should be whether it is in the public interest. By that standard, the current civil-service system is indefensible.
The solution is straightforward. Scrap the system and replace it with principles designed to achieve the original goal of a merit system. Avoiding spoils is not hard: Funnel hiring through an independent agency. Work rules should be replaced by general principles, overseen by a neutral review board. Eliminate the presumption of lifetime service, as recommended by the Partnership for Public Service. Terminating a public employee should trigger a safety net, not years of litigation.
Principles, ironically, are less susceptible to abuse of state power and gamesmanship than precise rules. One of the many paradoxes of “clear law” is that no one can comply with thousands of rules. With principles, a citizen can stand his ground to an unreasonable demand and have a good chance of being supported up the chain of authority.
In the civil service, promoting an exemplary employee is often impossible.
There is still a place for precise rules. Rules are effective in situations where the protocol is more important than context and balance — say, with age limits or effluent discharges. Management expert Brenda Zimmerman makes the distinction between the legal framework for “complicated” activities — such as engineering or rocket launches, where a small error might have disastrous results — and “complex” activities, such as running a health-care system or regulating nursing homes. For “complicated” activities, rules and checklists can impose the discipline to avoid disastrous error. For “complex” activities, general principles are far superior, because they allow people to adapt to many moving parts. The more complex the area of oversight the simpler and more flexible the regulatory framework must be.
But what about human error and venality? Does law based on principles mean we must trust people? Of course not. That’s why accountability is still important. Moreover, for important decisions, a structure can require approval of several people. Nothing can get done sensibly or fairly, however, until we reconstruct government with a legal framework which liberates people to roll up their sleeves and make things happen.
Originally published at www.theatlantic.com on September 22, 2014.