5

9.4.7 The rewards of failure

On 11 September 2001, America suffered the most devastating terrorist attack in history. Four airplanes were hijacked and destroyed, the Pentagon was attacked, and the World Trade Center was destroyed, resulting in close to three thousand fatalities. This was many times worse than any other terrorist attack ever suffered by the U.S. or any other country.

What were the consequences for the United States government? First, the president’s approval rating underwent an immediate and enormous jump, from about 55 percent to almost 90 percent. Over the following seven years, it dropped steadily, finally ending below 30 percent in 2008.2© Though this remains a matter for speculation, it is plausible that George W. Bush would not have been reelected in 2004 if not for the terrorist attack.

On the face of it, this is paradoxical. If there was anyone whose job it was to protect Americans from this sort of attack, it would be the executive branch of the U.S. government. As the head of that branch, George W. Bush might have been expected to come under criticism. As an analogy, imagine that you have hired a company to provide security for your building. You have just learned that last night, vandals broke into the building and destroyed thousands of dollars’ worth of equipment. Someone now asks you what you think of the job that your security company is doing. What do you say? ‘Best security company ever!’?

Why did Americans so enthusiastically approve of Bush after the attacks? Partly, it was because they felt that it was patriotic to support the president, and that is partly because people tend to confuse the country with the government and the government with the government’s officeholders.

My aim in raising this example is not to blame the government for 9/11. My aim is to examine the incentives applying to the government. What happens when the government fails to achieve its purposes? Often, the result is that the government is more rewarded than punished. In the present case, Bush reaped the highest approval ratings of his career, along with the opportunity to expand executive power in ways that otherwise would have met with much greater resistance.24

Similar things are true of other parts of government. Suppose that a city suffers from a crime wave. What will be the effects on the police department? It is much more likely that the police department will receive greater funding to combat the problem than that their funding will be cut. Or suppose that a society suffers from a severe increase in poverty. If the government has no agencies designed to combat poverty, one or more will probably be established. If some already exist, they will most likely receive greater funding rather than less. Few public figures would have the temerity to argue that the poverty programs should be cut because there is too much poverty. The general lesson is that if some part of the government fails in its function, it will most likely be given greater funding and power. Of course, the purpose of this is not to reward failure; the thinking would be that more money and power will enable the agency to solve the problem. But the effect is that government grows when social problems grow, and thus it is not in the government’s interest to solve society’s problems.

My claim is not that government agents actually try to fail. | do not think, for example, that the Bush administration actually wanted 9/11 to occur. My claim is twofold: first, that government agencies simply do not try as hard as they could to succeed at their assigned tasks, because they do not suffer from the negative consequences of failure. Second, that unsuccessful government programs tend to persist and grow, with the result that over a period of decades the government will come to be dominated by such programs.

Of course, other anecdotes could be told about government officials’ losing their jobs because of some prominent failure. My claim is not that failure is always rewarded; my claim is that failure tends to be rewarded in a large range of cases, leading to serious problems in democratic states. If a government official is guilty of some simple and demonstrable malfeasance, with large and well- publicized negative consequences, then that official will probably lose his job. If a government agency that is felt to be dispensable - for example, NASA or the NEH - fails in simple and well-publicized ways, then that agency’s funding will likely be cut, and the agency might even be eliminated. But suppose that some essential part of the government - say, the court system or law enforcement - fails chronically in ways that are complicated, difficult to understand, and not readily traceable to specific actions by a small number of individuals. Then that part of the government will more likely be rewarded than punished. People will feel that eliminating the dysfunctional agency or branch is not a feasible option, and since there are no specific individuals to blame, none will lose their jobs over it. The members of that part of the government will blame inadequate funding, and ill-informed voters are likely to find that explanation more understandable than the complicated truth. Thus, over time, the kinds of failures we should expect to see accumulating are low-profile and systemic failures in essential government services.