When he opened the Guardian Bank and Trust Company in the Cayman Islands in 1986, John Mathewson had no experience, not many clients, and only a cursory knowledge of how banks really worked. But, in his own peculiar way, he was a visionary. What Mathewson understood was that there were many American citizens with lots of money that they did not want the Internal Revenue Service to know anything about, and that these Americans would pay hefty sums if Mathewson could keep their money safe from the prying eyes of the IRS.
So Mathewson obliged them. He showed his clients how to set up shell corporations. He never reported any of the deposits he received to the IRS. And he gave his clients debit cards that allowed them to access their Guardian accounts from anywhere in the United States. Mathewson charged hefty fees for his services— $8,000 to set up an account, $100 for each transaction—but no one seemed to mind. At its peak, Guardian had $150 million in deposits and two thousand clients.
In 1995, Mathewson left the Caymans after a dispute with a government official, and moved to San Antonio to enjoy his retirement. It didn’t last long. Within a few months, he was arrested for money laundering. Mathewson was an old man. He did not want to go to prison. And he had something valuable to trade for his freedom: the encrypted records of all the depositors who had put money into Guardian Trust. So he cut a deal. He pled guilty (and was sentenced to five years’ probation and five hundred hours of community service). And he told the government everything he knew about tax cheats.
The most interesting information Mathewson had to offer was that offshore banks were no longer catering only to drug dealers and money launderers. Instead, these banks served many Americans who had earned their money honestly but simply didn’t want to pay taxes on it, As Mathewson told a Senate panel in 2000, “Most of [Guardian’s] clients were legitimate business people and professionals.” A typical Mathewson client was someone like Mark Vicini, a New Jersey entrepreneur who ran a computer company called Micro Rental and Sales. Vicini was, by all accounts, a respected member of his community. He put his relatives through college. He gave generously to charities. And, between 1991 and 1994, Vicini sent $9 million to the Caymans, $6 million of which he never mentioned to the IRS. This saved him $2. 1 million in unpaid taxes. (It also eventually earned him a five-month stint in federal prison, where he was sent after pleading guilty to tax evasion.)
Mathewson’s clients were not alone, either. In fact, the nineties saw a boom in tax evasion. By the end of the decade, two million Americans had credit cards from offshore banks. Fifteen years earlier, almost none did. Promoters, who often used the Internet to push their scams, advertised “layered trusts,” “offshore asset protection trusts,” and “constitutional pure trusts.” A small but obstinate (and obtuse) group of tax evaders advised people that they didn’t have to pay their taxes because the income tax had never actually been passed by Congress. And old standbys—keeping two sets of books, incorporating yourself as a charity or a church and then writing off all your expenses as charitable contributions—stayed alive. All these schemes did have an important downside: they were illegal. But rough estimates suggested that they were costing the United States as much as $200 billion a year by the end of the decade.
The vast majority of Americans never experimented with any of these schemes. They continued to pay their taxes honestly, and they continued to tell pollsters that cheating on your taxes was wrong. But there’s little doubt that the proliferation of these schemes—and the perception that many of them were successful—made average Americans more skeptical of the fairness of the tax system. Adding to those doubts was the ever-increasing complexity of the tax system, which made it more difficult to know what your fair share of taxes really was, and the 1990s boom in corporate tax shelters, which was responsible for what the Treasury Department called, in 1999, “an unacceptable and growing level of tax avoidance.” The title of a 2001 Forbes article on the tax system captured what more than a few Americans were wondering about themselves: ARE YOU A CHUMP?
Why did this matter? Because tax paying is a classic example of a cooperation problem. Everyone reaps benefits from the services that taxes fund. You get a military that protects you, schools that educate not only your children but the children of others (whom you need to become productive citizens so that they will grow up to support you in your old age), free roads, police and fire protection, and fundamental research in science and technology You also get a lot of other stuff you perhaps don’t want, too, but for most people the benefits must outweigh the costs, or else taxes would be lower than they are. The problem is that you can reap the benefits of all these things whether or not you actually pay taxes. Most of the goods that the government provides are what economists call nonexcludable goods—meaning, as the name suggests, that it’s not possible to allow some people to enjoy the goods while excluding others. If a national missile defense system is ever built, it will protect your house whether or not you’ve ever paid taxes. Once 1-95 was built, anyone could travel on it. So even if you think government spending is a good thing from a purely self-interested perspective, you have an incentive to avoid chipping in your fair share. Since you get the goods whether or not you personally pay for them, it’s rational for you to free ride. But if most people free ride, then the public goods disappear. It’s Mancur Olson’s theory all over again.
We may not normally think of taxpaying as a matter of cooperation, but at its core that’s what it comes down to. Taxpaying is obviously different from, say, being a member of an interest group in one important sense: not paying your taxes is against the law. But the truth is that if you cheat on your taxes, the chances that you’ll get caught have historically been pretty slim. In 2001, for instance, the IRS audited only 0.5 percent of all returns. In purely economic terms, it may actually be rational to cheat. So a healthy tax system requires something more than law. Ultimately, a healthy tax system requires people to pay their taxes voluntarily (if grudgingly). Paying taxes is individually costly but collectively beneficial. But the collective benefits only materialize if everyone takes part.
Why do people take part? In other words, why, in countries like the United States where the rate of tax compliance is relatively high, do people pay taxes? The answer has something to do with the same principle that we saw at work in the story of Richard Grasso: reciprocity. Most people will participate as long as they believe that everyone else is participating, too. When it comes to taxes, Americans are what historian Margaret Levi calls “contingent consenters.” They’re willing to pay their fair share of taxes, but only as long as they think that others are doing so, too, and only as long as they believe that people who don’t pay their taxes have a good chance of being caught and punished. “When people start to feel that the policeman is asleep, and when they think others are breaking the law and getting away with it, they start to feel like they’re being taken advantage of,” says Michael Graetz, a law professor at Yale. People want to do the right thing, but no one wants to be a sucker.
Consider the results of public-goods experiments that the economists Ernst Fehr and Simon Gächter have run. The experiments work like this. There are four people in a group. Each has twenty tokens, and the game will last four rounds. On each round, a player can either contribute tokens to the public pot, or keep them for himself. If a player invests a token, it costs him money. He invests one token, and he personally earns only 0.4 tokens. But every other member in the group gets 0.4 tokens, too. So the group as a whole gets 1.6 tokens for every one that’s invested. The point is this: if everyone keeps their money and invests nothing, they each walk away with twenty tokens. If everyone invests all their money, they each walk away with thirty-two tokens. The catch, of course, is that the smartest strategy ordinarily will be to invest nothing yourself and simply free ride off everyone else’s contributions. But if everyone does that, there will be no contributions.
As with the ultimatum game, the public-goods games are played in a similar fashion throughout the developed world. Most people do not act selfishly at first, instead, most contribute about half their tokens to the public pot. But as each round passes, and people see that others are free riding, the rate of contribution drops. By the end, 70 to 80 percent of the players arc free riding, and the group as a whole is much poorer than it would otherwise be.
Fehr and Gächter suggest that people in general fall into one of three categories. Twenty-five percent or so are selfish—which is to say they are rational, in the economic sense—and always free ride. (That’s close to the same percentage of people who make lowball offers in the ultimatum game.) A small minority are altruists, who contribute heavily to the public pot from the get-go and continue to do so even as others free ride. The biggest group, though, are the conditional consenters. They start out contributing at least some of their wealth, but watching others free ride makes them far less likely to keep putting money in. By the end of most public- goods games, almost all the conditional consenters are no longer cooperating.
The key to the system, then, is making sure the conditional consenters keep cooperating, and the way to do that is to make sure they don’t feel like suckers. Fehr and Gächter tweaked the public- goods game to demonstrate: this time, at the end of every round, they revealed what each person had or had not contributed to the public pot, which made the free riders visible to everyone else. Then they offered people the opportunity to punish the free riders. For the price of a third of a token, you could take one token away from the free rider. Two things happened as a result. First, people spent money to punish the evildoers—even though, again, it was economically irrational for them to do so. Second, the free riders shaped up and started contributing their fair share. in fact, even during the last rounds of these games, when there was no reason to keep contributing (since no punishment could be inflicted), people continued to chip in.
When it comes to solving the collective problem of how to get people to pay their taxes, then, there are three things that matter. The first is that people have to trust, to some extent, their neighbors, and to believe that they will generally do the right thing and live up to any reasonable obligations. The political science professor John T. Scholz has found that people who are more trusting are more likely to pay their taxes and more likely to say that it’s wrong to cheat on them. Coupled with this, but different from it, is trust in the government, which is to say trust that the government will spend your tax dollars wisely and in the national interest. Not surprisingly, Scholz has found that people who trust the government are happier (or at least less unhappy) about paying taxes.
The third kind of trust is the trust that the state will find and punish the guilty, and avoid punishing the innocent. Law alone cannot induce cooperation, but it can make cooperation more likely to succeed. If people think that free riders—people not paying taxes but still enjoying all the benefits of living in the United States—will be caught, they’ll be happier (or at least less unhappy) about paying taxes. And they’ll also, not coincidentally, be less likely to cheat. So the public image of the IRS can have a profound impact on the way conditional consenters behave. Mark Matthews, head of the agency’s Criminal Investigative Division, was keenly aware that the success of criminal investigations was measured not just by the number of criminals caught but also by the public impact of its work.. “There is a group of people that could be tempted by these scams, a group that could let aggressive tax planning become too aggressive. We need to convince them before that happens that it doesn’t make sense,” Matthews said. “A huge part of the agency’s mission is making sure that people believe the system works.”
Getting people to pay taxes is a collective problem. We know what’ the goal is: everyone should pay their fair share (this says nothing, of course, about what a fair share is). The question, then, is how? The U.S. model—which is, by global standards, successful, since despite Americans’ vehement anti-tax rhetoric they actually evade taxes far less than Europeans do—suggests that while law and regulation have a key role to play in encouraging taxpaying, they work only when there is an underlying willingness to contribute to the public good. Widespread taxpaying amounts to a verdict that the system, in at least a vague sense, works. That kind of verdict can only be reached over time, as people—who perhaps first started paying taxes out of fear of prosecution—recognize the mutual benefits of taxpaying and institute it as a norm.
Another way of putting this is to say that successful taxpaying breeds successful taxpaying. And that positive-feedback loop is at work, I’d argue, in most successful cooperative endeavors. The mystery of cooperation, after all, is that Olson was right: it is rational to free ride. And yet cooperation, on both a small and a large scale, permeates any healthy society. It’s not simply the obvious examples, like contributing to charities or voting or marching on picket lines, all of which are examples of collective action that people participate in. It’s also the subtler examples, like those workers who, by all rights, could shirk their responsibilities without being punished (because the costs of monitoring them are too high) and yet do not, or those customers who leave tips for waitresses in restaurants in distant cities. We can anatomize these acts and explain what gives rise to them. But there is something irreducible at their heart, and it marks the difference between society on the one hand and just a bunch of people living together on the other.