8

Government’s End

IN THE SUMMER 1989 issue of The National Interest magazine, an obscure State Department functionary named Francis Fukuyama proclaimed the end of history. One way to achieve renown is to be audaciously misunderstood, and Fukuyama certainly achieved that. Obviously, history had not come to an end; not even the sort of great-power geopolitical history of which Fukuyama was writing had come to an end. Fukuyama, however, was using the word “history” in a particular and rather curious sense. He was a thinker in the tradition of G.WF. Hegel, the early nineteenth-century German philosopher who viewed history as a steady progress through successive stages of human organization, with freedom gradually increasing. In the Hegelian tradition, history itself could be seen as unfolding through a dialectical process, as conflicts between and within governmental and economic regimes worked themselves out. Thesis, antithesis, synthesis.

But what if there no longer was any coherent, sustainable alternative to the Western system of liberal democracy? What if the last big conflict had been resolved? Then history, in Hegel’s peculiar sense, was over. True, there continued to be feudal states, monarchies, tyrannies, military governments, dictatorships, theocracies, and even, here and there, the odd spot of lingering communism. The great ideological battle over how best to manage a human society, however, was finished. “The exhaustion of viable systematic alternatives to Western liberalism,” Fukuyama wrote, had led to “the end point of mankind’s ideological evolution and the universalization of Western liberal democracy as the final form of human government.”

In effect, he was saying that if you want to know how the most successful, sustainable sort of human society will be organized in the future—well, look around. This is it. True, there will be plenty of variations on the theme, many local and regional differences, and many struggles and much change. The mixture of elements within the structure of liberal societies will be constantly in flux, and there will be more than enough worldly events to fill the history books of the future. But the big question, said Fukuyama, the (literally) history-making question, had been settled.

Fukuyama’s idea made a splash partly because it was audacious and partly because, with the fall of communism and the Berlin Wall, its timing was good. But partly it also quite aptly captured the common feeling that something had changed fundamentally in the world: that the arrow of history had, with the evolution of liberal capitalism, reached a point from which there would be no turning back. Having emerged from many millennia of struggle for control of human society, Western liberalism had now established itself as not just a phase in that struggle but as its one really successful outcome.

I am no Hegelian, and I take sweeping claims, including Fukuyama’s, with a grain of salt. Still, by the time Fukuyama’s thesis had aged a decade, it seemed to have worn pretty well. It also served as a useful reminder. History is turbulent and change is constant, but deep social forces can conspire to produce long periods of equilibrium, from which radical change in any direction is difficult, perhaps even impossible without the stimulus of some outside shock or catastrophe (shades, here, of Mancur Olson). And it is possible, even probable, that, in the structural sense, the American government has reached such a stage: a kind of steady state from which it may be dislodged by some powerful outside force but from which it cannot dislodge itself.

In other words, it is possible that the government has reached its end, in a sense somewhat like Fukuyama’s concept of “end.” No, government isn’t dead, as is clear to anyone who pays taxes or draws Social Security. But its overall scope and shape are no longer negotiable. They have evolved to a state from which they cannot, if you will, unevolve.

True, there will be change in the mix of programs and in the claims on the federal purse of various needs and functions. As baby boomers age, pension spending will grow as a share of the total, and education spending may diminish. Defense will rise and fall and rise and fall. Washington will change, sometimes wrenchingly, and will face difficult choices between quite different sorts of policies. Politicians and activists and interest groups will tussle and shout as though the very future of democracy were at stake. There will be sound and fury aplenty. But signifying how much? If you want to know basically how the American government will look for the indefinite future—well, look around. Taxpayers will not allow the government to do a great deal more than it does. Sorry, liberals. The client groups, however, will not allow the government to do much less than it does. Sorry, conservatives. Goldwaterian reformers who dream of making Washington a small town again, or even a significantly smaller town, might as well leave the pumpkin patch and go home. Above all, politicians and the electorate can tinker with parts of the government, but they cannot make it coherent or rational, and only now and then can they reform it more than marginally. Sorry, voters.

In effect, having passed through a minimalist stage in its first 150 or so years and then an expansionist stage in its next fifty years, Washington has trapped itself in a new equilibrium. It has become a sort of sprawling organism, its many parts not particularly good at solving social problems but extremely good at surviving from year to year and at serving assorted clients. And this evolution cannot be undone.

This is, I’m the first to agree, a fairly strong claim. It is certainly not the only reading of the situation, and if it is true, it seems bleak. Actually, it’s not so bleak, but I’ll get to all that in Chapter 10. Here, let me focus on the elements of the case for government’s end, in that particular evolutionary sense of “end.”

Dinner at Mancur’s

Start with a thought experiment, one that goes back to Mancur Olson’s problem of collective action, in which, for reasons as fundamental as arithmetic, large and diverse populations have trouble working together even when doing so is strongly in their interest. You are one of, say, four people eating together at a restaurant. The four will share the bill equally. The bill is presented, and everyone is horrified. It seems that you noticed that if you ordered a $10 appetizer of snail-darter soup, you could eat the whole thing while paying only $2.50 for the privilege. But the other three made the same calculation, so everybody indulged. And so everybody ordered dessert, and wine, and aperitifs, and entrees of beer-fed, hand-massaged beef, and more.

This problem is not hard to solve. The four diners could talk to one another as they order and exchange information about what they’ll have. If someone suggests something very expensive, someone else might cough or frown. They might reach an understanding of about how much their dinner should cost and tailor their choices accordingly.

Now, however, change the scenario in only one respect: Put the dinner under, say, the New Orleans Superdome, and put, say, 10,000 diners around the table. Each diner knows what he orders, can see what the five or ten people to either side of him order, and can talk to those people and maybe a few others. Now the problem is greatly magnified—so greatly, in fact, that it becomes a different sort of problem, in two ways.

First, the incentive to overdo your order is, arithmetically, much larger. If you order an extra dessert that costs $10, you pay only an additional tenth of a cent ($10 divided by 10,000 diners). No one who isn’t paying special attention to your order will notice the difference. In fact, for a tenth of a cent, no one would even bother to find out what you’re ordering. Second, not only is the problem much more serious, but the solution is much more elusive as well. If you wanted to deter dessert ordering, what would you do? Giving up your own dessert, if that’s all that happens, is unsatisfactory: You lose your dessert, but your share of the bill falls by only a tenth of a cent. Maybe you can persuade the five people on either side of you to give up dessert. If all eleven of you do so, each of you saves 1.1 cents. Congratulations. If you stand on the table and shout, maybe you can persuade a hundred people to skip dessert. Now you’re getting somewhere: Each of these heroic sacrificers will lose his dessert while reducing his dinner bill—and everybody else’s—by a dime.

Note a further problem. The more people you try to persuade, the more patchy the results will be. If I make a dessert-skipping deal with the man to my left and the woman to my right, chances are that I’ve had some friendly chat with them over appetizers and have established at least some rudimentary level of trust. Plus, of course, each of the three of us sees what the others are ordering, so we can verify that all three of us have kept the bargain. But for a cluster of a hundred to know or even watch each other is much harder. You might have to set up some sort of monitoring process, and even then the incentive to cheat would be large, because if Jones breaks the deal and orders dessert while others keep their promises, Jones gets dessert and saves a dime on his dinner tab. So saving even this one dime each by organizing a dessert boycott among a hundred diners, though possible, is difficult. And, of course, a hundred diners are still only 1 percent of the total, and a dime is still only a dime.

Now try to imagine overcoming the dessert problem around the whole table of 10,000 diners, half of whom (say) cannot even clearly see the other half. To make the situation challenging, add a further condition. Rather than being limited to 10,000, this dinner is open to anyone who pays a small—and, proportionately, shrinking—gate fee. (Remember, setting up a pressure group or hiring a lobbyist becomes relatively easier every year, as technology improves and communications costs fall and so on.) In fact, more and more diners are attracted to the meal, because if you don’t join in the festivities you may not eat. (Remember, failing to play the lobbying game against your competitors can cost you dearly.) So the dinner grows and grows, and the newcomers are strangers to everybody present. Why trust them? Under the circumstances, there’s no very compelling reason for anyone to do anything except order the best meal he can.

Remember, what you’re up against here is arithmetic: If you sacrifice, you know what you lose, but you may gain nothing or next to nothing, because you can’t be assured that more than a handful of others will follow suit. No, correction: You can be assured that most people will not follow suit. Most people will look around the table and, being patriotic or communitarian but not blind or feebleminded, make the same calculation: Ms. Patriot’s sacrifice is unlikely to be emulated by more than a few other people. After all, everyone understands the arithmetic. Even if you sacrifice, the gesture is likely to be wasted, because many other people won’t. What would have been your dessert will end up on someone else’s plate. Not being in the business of making futile gestures, you decline to sacrifice. Because everybody else makes the same calculation, the prophecy is self-fulfilling.

By now you no doubt see where this analogy is going. Government is the dinner table, and its client groups are the diners. I’ve often emphasized that lobbies aren’t evil, and now that point comes more sharply into focus. In fact, viewing lobbies as rapacious or unpatriotic completely misconstrues the situation, by turning a very difficult arithmetical problem into a facile moral one. To reach the bad result in which collective action fails, you don’t need to assume the diners are greedy or selfish or out to hurt their fellow citizens. You need only assume they’re rational.

In the example of the Superdome dinner, the alert reader may have recognized a case of the problem that Mancur Olson worked through in 1965 in The Logic of Collective Action—the same problem I discussed briefly in Chapter 2. As Olson himself put the case, “The fairly small (or intermediate) group has a fair chance that voluntary action will solve its collective problems, but the large, latent group cannot act in accordance with its common interests so long as the members of the group are free to further their individual interests.”

Another useful way to think about this lamentable situation is in terms of a classic economic conundrum that economists have called “the tragedy of the commons.” The problem appears where you have a limited resource that’s open to all comers. Suppose, for instance, you have a common forest and many independent loggers. Each logger will rush to cut down and sell as many trees as possible, before everyone else takes all the lumber. The forest is badly overlogged. In fact, chances are that it will soon be destroyed. This problem arises again and again: The overhunting of the buffalo was a classic and devastating case; a more recent one was the overhunting of elephants in Africa. In a common-resource situation, if everyone tries to win, everybody loses. Note the apt use of the word “tragedy” for this situation instead of “conspiracy” or “immorality.” Most people in the commons are acting not out of greed or depravity, but out of the impulse to survive in the world as they find it. Good intentions, or at least honest intentions, breed collective ruin.

Something similar happens with a run on government, which is what the Olsonian problem creates. The universe of public policies is a kind of commons. If you see others rushing to lobby for favorable laws and regulations, you rush to do the same so as not to be left at a disadvantage. But the government can do only so much. Its resource base and management ability are limited, and its adaptability erodes with each additional benefit that interest groups lock in. In fact, the more different things it tries to do at once, the less effective it tends to become. Thus if everybody descends on Washington hunting some favorable public policy, government becomes rigid, overburdened, and incoherent. Soon its problem-solving capacity is despoiled. Everybody loses.

So what is the answer? There is no answer. In the Superdomedinner situation as I’ve set it up, or in the forest situation, the combination of rational actors and Olsonian arithmetic is not beatable by any quantity of good intentions. Asking all the world’s fishermen to please, pretty please, voluntarily limit their catch may help a little around the margins, but, as we know, it cannot save the world’s fishing stocks. Instead, you need to change the situation.

Leadership and Followership

Mancur Olson, needless to say, appreciated this problem. He understood the difficulty of reforming a government consisting of countless thousands of programs and benefits and transfers “owned” by countless thousands of clients. A couple of years before he died, I asked him what to do. Well, he said, each of the tiny minorities represented by a client group can be easily beaten by the united opposition of a large majority. The trick was to assemble the majority. To do that, you would need two things. First, members of the broad public—the people supporting the lobbies around the dinner table, so to speak—need to understand the trap they’re in, so that they appreciate that there really is a general problem and a collective payoff if everybody gives up dessert. Second, you need a leader with the vision to assemble a reform package, and with the political skills to sell the deal and keep it together.

In effect, you need to add a new element to the Superdome dinner: the possibility that somebody with a megaphone will have either the authority or the prestige to lead the diners in making a deal to forgo dessert. Ideally, this leader would also be in a position to enforce the deal or at least to offer some general assurance that it would stick. “I won’t rescind dessert until I have a written commitment from at least 90 percent of our diners,” this leader might say, “and then I’ll appoint monitors who will report to me on what everybody is ordering.”

Consider the case again from a slightly different angle: Escaping the tragedy of the commons requires that some person or institution or group would assume ownership or stewardship of the common resource. For a forest, that could mean ownership by a private party or stewardship by the Forest Service; for the fisheries, it could mean some sort of treaty commitment (as with whales). For the government in Washington, it means giving reformers enough political power to “own” the government, as it were, long enough to assemble a reform package and make it stick. With committed leadership and committed followership, getting a handle on the government should be possible.

Now some good news. We know, for a fact, that a reform along those lines is possible, because it has been done.

In 1986, Congress (then split between a Republican Senate and a Democratic House) and President Reagan did what many old-timers in Washington said could never be done: They pushed through a wholesale cleanup of the tax code. Jimmy Carter, in the 1970s, called the tax code a disgrace to the human race. If that was true, then by the mid-1980s the code was also a disgrace to chimpanzees and hyenas. It had become a microcosm of the government as a whole: a sprawling mess of handouts and favors and gimmicks and shelters, a deluxe condominium for parasites. As the code descended further into incoherence and overcomplexity, enlightened Democrats began calling for reform. Republicans, not wanting to lose the issue, jumped on board. In 1985, the Reagan administration proposed sharply curtailing loopholes and using the money to reduce tax rates. In 1986, with a hard push from the president, the tax-reform bill passed. At one blow, it swept away years’ worth of accumulated tax breaks that had distorted the economy and acted as a full-employment program for tax lobbyists.

Before the reform, there were so many tax shelters that people with $100,000 incomes were paying only half the official 50 percent marginal rate on their last dollars earned. After their loopholes had been plugged and their rates reduced, they paid as much as ever, and the poor paid less. On balance, tax reform was a takeaway from the lobbying industry, not a giveaway to the rich. With tax rates lower, people had less reason to lobby for loopholes. After all, once your top tax rate is cut from 50 percent to 25 percent, a tax shelter is only half as valuable to you—so a tax lobbyist is only half as much worth hiring.

In all, the 1986 tax reform was the most brilliant antiparasite stratagem of its time (incomparably more successful than Stockman’s budget efforts). True, lobbies never give up and go home, and politicians can never resist buying votes by offering tax breaks to favorite lobbies. In 1988, George Bush promised fistfuls of tax breaks in his election campaign—for capital gains, for oil and gas, for rural development, you name it. Bill Clinton was even worse, proposing bunches of new tax breaks. In fact, nobody could resist using the tax code as a pork barrel. In 1997, the Republican Congress passed a “Taxpayer Relief Act” that added 800 amendments to the tax code, 290 new sections, and 36 retroactive provisions. Over time, the 1986 tax reform eroded. Nonetheless, it proved a point. Politicians can do a big housecleaning.

There is a “but,” however: Some things need to go right. With the 1986 tax reform, Republicans in the White House, led by Reagan, and Democrats on Capitol Hill, led by the chairman of the House Ways and Means Committee (Dan Rostenkowski), were in broad ideological agreement on what to do. The Republicans liked reducing the rates, and the Democrats liked closing the loopholes for the rich. Well, if they closed the loopholes, they could reduce the rates. Consensus helped neutralize the partisanship that could have blown apart a deal. With both sides in general accord, the party leaders could act as monitors and disciplinarians.

Moreover, both sides realized they needed to develop and enact a big package that would bestow large, visible, and thus politically attractive rewards upon ordinary voters. Going after only this loophole or that one was a sucker’s game, because no one would care or even notice except the lobby that was offended. But the tax reformers rounded up whole herds of loopholes and pushed them over the cliff, and then used the money to buy highly visible rate cuts. The promise of a pot of gold at the end of the rainbow—but only if the deal held together—sufficed to neutralize the voters, who might otherwise have been rallied to opposition. The voters were never particularly enthusiastic about the tax reform; they tended to be suspicious of anything Washington did. But they assented to it or decided they didn’t care.

There were still the lobbies to worry about. But the reformers realized that by cutting rates they could turn the lobbies against each other, because for every group whose taxes rose, there would be another whose taxes fell. In fact, many lobbies found themselves neutralized by a dilemma: Keeping a loophole by killing the tax bill might also mean losing the benefits that would come with lower rates. With the politicians united, the voters apathetic, and the lobbies divided, the job got done.

It worked once. It ought to work again. So what keeps going wrong?

The flaw isn’t lack of leadership. I made clear in the previous chapter that, whatever else may have been missing in the reform era of the 1980s and 1990s, creative and energetic leadership was on hand. For all their various shortcomings, Reagan and Stockman and Clinton and Gingrich were not lacking in talent or vision. Popular cynicism notwithstanding, the American system is good at cultivating and bringing forward leaders of quality and substance. Something else keeps going wrong. The problem, it turns out, is not poor leadership but poor followership. Consider closely, in this connection, the case of Newt Gingrich, and how the transfer-seeking game, which he sought to subvert, subverted him instead.

A Revolutionary’s Blueprint

In hindsight (always the most discerning kind of vision), Gingrich seems to have been an overweening idealist who pushed his luck too far. But defeat appears inevitable only after the war. Gingrich did not enter office as House speaker without a plan. He explained it in January 1995, and it was not a stupid plan.

He was no newcomer. Like David Stockman, who was himself a former House member familiar with the “iron triangle” (bureaucracy, client lobby, congressional patrons) that protected every program, Gingrich went in with his eyes wide open. The power structure on Capitol Hill, he told the Washington Times as he assumed the speakership, had “ossified into a straitjacket. That is not partisan or ideological—these guys and their staffs had networks of power and networks of relationships and habits and things that they weren’t going to break for a mere president. They’d ignored Nixon, Ford, Carter. They had blocked Reagan and beaten Bush.” Moreover, “every time you mention something which ought to be shrunk or zeroed, twenty-five people who are making money off of it jump up to explain why it is a wonderful institution and they should continue to make money off it.”

Gingrich’s response, his battle plan, is instructive, because on paper it was plausible. First, he would mobilize his supporters, the fiery voters who had demolished Democrats and tossed out a reigning House speaker to put Gingrich and his reformers in charge. “The point we’re going to make to people is, you’d better call your representative and tell them you want them to help pass the constitutional amendment to require a balanced budget—with a tax-increase limit. We’re going to use every bully pulpit we have. . . . And we’re going to tell every conservative group in the country and every group that wants smaller government, you’d better talk to your representatives.” The intensity of the government reformers was high, Gingrich knew, so they could mobilize some of the same merciless spot pressure as the interest groups.

As for the Democrats, the 1994 election had thrown them into disarray. The president sounded chagrined, humbled, the wind knocked out of him. “I agree with much of what the electorate said yesterday,” he said the day after the election. “They still believe that government is more often the problem than the solution. They don’t want any party to be the party of government. They don’t want the presumption to be that people in Washington know what’s best. . . . I accept responsibility for not delivering to whatever extent it’s my fault we haven’t delivered.” This humbled president would still wield a veto, but he would be presented with a stream of bills passed on the Republicans’ terms in the wake of an election that had given them a mandate. If he refused to deal, he would risk seeming obstructive and deaf to the voters’ demands. Anyway (said the Republicans to themselves), this was not a president who had shown a lot of backbone.

The lobbies, of course, could be counted upon to try to block or emasculate everything. Gingrich’s response: swamp them. Attack so many programs at once that the Democrats and liberals and establishmentarians would have to choose the programs they wanted to save. The rest, the Republicans would knock off. The Democrats would have to “figure out which fights to stay and fight,” Gingrich said. Gingrich was hoping to invert the usual Washington pattern, in which reformers were required to focus their energies on a few programs and let the rest of their program slide away. By attacking on a broad front, he would force the defenders to concentrate their fire. The Republicans would not get everything, but they would get a great deal.

Finally, Gingrich knew that at each stage of the process— House deliberations, Senate deliberations, House-Senate conferences, negotiations with the White House, presidential vetoes—he would lose bits and pieces of his agenda. A month or a year wouldn’t be enough, a point he went out of his way to emphasize. Instead, he would start in 1995, running a flying wedge through the Washington power structure, and then come back again and again after that, widening the breach. There could be no “Mao-style revolution,” he said. “I want to get to a dramatically smaller federal government. I think you do that one step at a time, but you insist on steps every year. . . . The reason I keep telling people to study FDR is if you take fourteen steps successfully you’re a lot farther down the road than this guy next to you if he’s trying to get all fourteen steps in one jump.”

The trouble was, of course, that he never got to the second step. Why?

The Paradox of Particulars

Voters in the polling booth vote for “change” in the abstract. But presidents and members of Congress can’t. “In Congress, we don’t get to vote on the abstraction,” Republican representative Vin Weber told Time in 1992, shortly before retiring from office. “We have to vote for or against actual programs.” That means confronting actual constituencies. Gingrich’s hope to invert this equation foundered on the fact that in the case of any particular program or subsidy or perquisite of whatever sort, there is almost always far more energy on the defensive side than on the offensive side.

Say someone in Gingrich’s position as a House leader hoped to reform or abolish a thousand programs. No one of those programs is essential to his effort. If he must, he can always drop twenty or thirty or even a hundred or two. There is no overwhelming incentive to go after any particular constituency. To the defender of the subsidy for left-handed screwdrivers, however, only one program matters: his own. He will spare no effort. For that defender, and for each of the others, it’s life or death.

Gingrich understood this but thought he could count on his zealous Republicans to hold the line across a broad front. The discipline he was expecting, however, was superhuman. The temptation to help out this one group, or that one, was not Democratic or liberal; it was universal. After all, the clients understood that if one congressman would not help them, another might. Every congressman understood this, too. Why let someone else do the rescuing and take the credit? And every congressman also understood that every other congressman understood. And so, at every stage in the process, Democrats and Republicans demanded that this or that program be let off the hook. “I’d love to support you, Mr. Speaker, but I tell you, I am just taking a beating from those left-handed screwdriver people in my district—you’ve got to cut me a break.” Facing this inevitable onslaught, Gingrich found that it was he, not the Democrats or the liberals, who was swamped.

Stockman had run into the same problem, and had reacted with contempt for the gutless Republicans who were all for cutting government except the bits they wanted to save. Stockman, however, missed the point: Given the calculus of the game, the gutless Republicans were doing the only rational thing. The same sort of calculus had wrecked Clinton’s health-care package. In fact, what was remarkable in 1995, arguably, was not how much the reform package was watered down in Gingrich’s House (with significant program terminations shrinking by a factor of ten) but how large a tattered remnant actually survived.

Gingrich understood the importance of public mobilization. He counted on it to push his program past the Democrats and Clinton. Clinton’s bus campaign for health care was a similar attempt to go “over the head” of Congress. In Gingrich’s case, and also in Stockman’s and Clinton’s, the reformers depended on the public to rally around when political hackery began to prevail over the spirit of reform. And, sure enough, the public always did rally—but to the wrong side.

It turns out to be surprisingly easy for the protectors of programs to spook the public by screaming bloody murder. The public wants the government to be leaner, but not at the expense of students, farmers, bankers, workers, veterans, retirees, homeowners, artists, teachers, train riders, or cats and dogs. The people cannot abide the ghoulish shrieks and moans that are heard the moment the reformers’ scalpel comes out. The same narrow focus and intense commitment that make lobbies so adept at defending themselves on Capitol Hill also make them good at alarming the public with “red alert” mailings and scary television ads (as with “Harry and Louise”). When all else fails, there is the old “Don’t hurt our children” ploy. In 1993, when Congress managed to abolish the wool and mohair subsidy, the reformers were all the more courageous for having faced down pleas like the one from Nelda Corbell, whose parents raised mohair in Texas: “I am eight years old and I want to know why the government wants to take away our living.” What kind of monster would hurt little Nelda?

Now and then, politicians manage to turn public opinion against a particular lobby, or at least they manage to exploit a change in public opinion, as the tobacco lobby found out. But usually they can’t even do that. In his 1996 presidential campaign, when Bob Dole tried to mobilize public sentiment against the teachers’ unions, he was judged quixotic. The public is nervous, often rightly, when politicians try to demonize some faction or other. Public nervousness makes the climate of opinion flammable; all that remains is to light a spark.

Rational Paranoia

In May 1981, President Reagan, on Stockman’s advice, proposed a package of modest reductions in Social Security: reduced benefits for early retirees, a three-month delay in the cost-of-living adjustment, and so forth. The result was what Congressional Quarterly described as a “tempest in Congress.” The Democrats until then had been helpless against Stockman, but they knew that this time he had stumbled onto vulnerable ground. The House Democratic caucus promptly and unanimously passed a resolution denouncing Reagan’s “unconscionable breach of faith” and swearing not to “destroy the program or a generation of retirees.” Democrats in the Senate promised to use “every rule in the book” to stop the proposal. “Democrats waged their assault with obvious glee,” said Congressional Quarterly, and they kept waging it through the 1982 elections, when they gained twentysix seats in the House and regained effective control there. Painting Reagan and the Republicans as scourges of Social Security received a good deal of the credit (the economic recession received most of the rest).

In 1995, Newt Gingrich’s Republicans, responsibly and courageously, undertook to propose some modest but significant reforms of the Medicare program for the elderly. That the program’s finances were in trouble, and that reductions would have to be made one way or another, were facts known to everybody in Washington, including President Clinton. He proposed reducing the growth of Medicare’s costs from more than 4 percent a year for six years to 2.7 percent. The Republican plan, in not exactly sharp contrast, proposed reducing the growth path to 1.5 percent, with some larger structural reforms than Clinton preferred. In dollars, the difference between the plans was about 7 percent in the last year, 2002. But that was enough for the Democrats. Through the 1996 campaigns, they hammered the Republicans for “cutting” Medicare. “The Republicans are wrong to want to cut Medicare benefits,” a voice-over intoned in one Democratic ad, as the faces of Bob Dole and Newt Gingrich danced on the screen. “And President Clinton is right to protect Medicare, right to defend our decision as a nation to do what’s moral, good, and right by our elderly.” The campaign became known as Mediscare, and it was accounted a great success. The public was quite willing to believe that Gingrich and his crew were out to gut Medicare. Despite their pleas of innocence, the Republicans never recovered.

In 1993, Bill Clinton proposed a health-care reform package. In 1994 came Harry and Louise and plenty of others like them. But you know that story. The point is that the trick works for both parties.

It works, you may say, because the public is ignorant and easily frightened. That explanation is true, to some extent. But it fails to give the public quite enough credit, because the public’s suspicions were rational in each case. When the Gingrichites tried to make changes in Medicare, they plausibly argued, just as someone might argue to overeaters at the Superdome dinner, that the (small) pain they were imposing on one group would be more than offset by the benefits to everybody from lower deficits, lower taxes, and a solvent Medicare program. But at that stage, the Democrats and the lobbies, acting as a swing vote, did exactly as the playbook suggests. They recast the debate as group versus group rather than as group versus nation. They stood on a box with a megaphone and warned: “Don’t believe those Republicans! They’re not going to give anything back to you once they’ve cut Medicare. They’re financing tax cuts for the rich! They’re just taking from you to give to their friends!”

Most Americans will sacrifice for a larger public good, but few will sacrifice for a competing group. The larger public loses interest in reducing Medicare, or in reducing anything else, if it believes that the only result will be to shift resources from one group to another. By kindling suspicions that the Republicans were acting in the interests of their favorite clients rather than of the nation as a whole, the Democrats and their allied lobbies had no trouble sinking the Republicans’ Medicare deal. On health reform, the Republicans and the plan’s other opponents used the same tactic against Clinton and the Democrats. “This plan doesn’t mean more care at lower prices,” they said. “It means poorer care for you and better care for other people, with huge new bureaucracies in the bargain.”

Alas, this trick of kindling mistrust can almost always be used by somebody, because the charges, though overdrawn and often misleading, are usually plausible and partly true. The Republicans were trying to cut Medicare while also reducing taxes for better-off citizens. The Democrats were relying on bureaucratic controls to constrain choices for the middle class and expand health access for the poor. In 1981, the Reagan administration was trying to use Social Security reductions to help pay for upper-class tax cuts. In a democracy, parties do not get things done (or win elections) unless they favor their supporters, which means that the other side of any argument can always cry foul. And the voters’ cynicism, which admittedly is often justified, makes them quick to believe charges that the system will double-cross them. The cynicism, of course, is self-fulfilling.

So here is the final conundrum of collective political action. If you assume that everyone else will act in his rational self-interest, you have every reason to support politicians who put dollars or benefits or protections in your pocket, and little or no reason to support politicians who remove them. Although it is certainly possible to neutralize the opposition party and divide the lobbies and win the public’s support, no sensible politician or voter ought to expect it to happen. Far more likely is the fate of the reformers of the 1980s and 1990s, who found themselves, after starting out well, suddenly staring at a coalition of opponents that comprised the opposition party, the lobbies, and the broad public. Against that array of forces, there is simply no hope. Reformers are crushed.

I have described in this book many dynamic asymmetries, forces acting in lopsided ways on complex systems—groups, governments, societies—that are themselves in constant motion. There is Olson’s asymmetry of collective action, giving narrow groups such large advantages in furthering their common good. There is Olson’s asymmetry of group-formation, as groups prove hard to form but much harder still to eliminate. There is the asymmetry of demosclerosis itself, in which programs likewise are hard to form but nearly impossible to excise. To that already rather daunting pile can now be added one more: For all the reasons discussed in this chapter, it is quite easy to mobilize political support for “reform” in the abstract but much easier still to mobilize political opposition to reforms in the particular. Reform, sure. But not this reform! Not that reform! Some other reform! Something less hurtful to me, or to my friends, or to any deserving party, or to anybody at all.

In the movie The African Queen, a famous scene has the protagonists’ boat hopelessly stuck in a marsh—only a few yards, it turns out, from open water. Today’s government is in a similar plight. Dissatisfaction ought, by rights, to open the path to comprehensive change. But it does not. The African Queen was lifted from the quagmire by the tide. But in the case of the American government, the boat cannot be lifted. The government is, of its nature, inseparable and inalienable from the million commitments it has made and the million interest groups it has spawned. They now form its environment. It cannot lift itself above them. With the replacement of Carter with Reagan, Bush with Clinton, and Clinton (for a while) with Gingrich, the restive electorate outside Washington showed that it could still radicalize politics, at least temporarily, and shake the very ground of the capital. Notwithstanding all the little gray groups and politicians and lobbyists and claques that occupy and ossify the government, the broad electorate proved more than able to coil itself and strike back. What was lacking in the system was not energy or leadership but the ability to focus reformist energy on any particular program of reform. Converting the electorate’s shuddering waves of discontent into the hundreds or thousands of alterations to programs affecting specific groups is like converting earthquake energy into steam power: possible in theory but elusive in practice.

Hey! That’s Mine!

So far, I’ve kept the discussion at the systemic level: the level of the game, so to speak, in which the politics of reform finds itself. The game problem, of course, is the Olsonian arithmetic of collective action, inasmuch as the broad public never cares about eliminating any given program remotely as much as its beneficiaries care about saving it. But that mechanical imbalance is rooted in, and compounded by, an attitudinal imbalance. Consider what I have come to think of as the entitlement mentality.

At the core of the entitlement mentality is this presumption: Whatever benefits me (or my group) belongs to me. Programs are property. Maybe, in a pinch, I’ll let you reduce my benefits. But if you cut my program without my permission, that’s theft. I have every reason to be outraged.

I’m not saying that entitlements have no place in government. One reason we have government is to offer security in an uncertain world. I also am not saying that entitlement thinking is always wrong or inappropriate. Social Security should not be overhauled twice a year in hopes of making it perfect. On the other hand, though, the politics of entitlement should not become so prevalent as to obliterate the politics of experiment. But that is exactly what has happened.

In Washington, a standard and generally useful distinction is between entitlement programs and discretionary ones. In the argot of budgeting, entitlements are permanent grants written into law; unless the law is changed, you automatically receive your benefits every year. Discretionary programs are the ones that come up for annual review; unless the law is renewed, the programs stop spending. Anyway, that’s the theory. With it goes the conventional wisdom that the big entitlements—the federal social-insurance programs, like Social Security and Medicare—operate on a different level of politics from the many smaller subsidies that serve smaller groups. And this conventional wisdom is largely true. In my effort to schematize, I’ve tended to treat the politics of Social Security (say) as interchangeable with the politics of maritime subsidies (say). In reality, a broad social-insurance system engages the public’s attention and pushes political hot buttons in ways that quiet little programs can never do. Cutting Medicare is difficult because it inevitably becomes a national political issue, whereas cutting the peanut protectorate is difficult because it inevitably does not become a national political issue.

I don’t want to dismiss this distinction, which is important. Yet to settle for it is to breeze past a curious and underappreciated truth. The politics of broad programs and the politics of narrow programs, for all their differences, finally converge to a pronounced similarity at the operational level, because, in the real world today, every federal program is an entitlement. Not every program technically is an entitlement, of course, but every program behaves as though it were. It lasts forever, and it’s viewed as a right by the people who benefit from it. Some constituencies are very large, and others are very small, but they all claim and exercise ownership rights over whatever piece of the government “belongs” to them.

If the collective-action problem is the mechanism of sclerosis, then the entitlement mind-set is the morality. The collective-action arithmetic induces everybody around the table to hold onto whatever he possesses, since, after all, nothing any one actor can do by himself will have any discernible effect on the overall situation. Meanwhile, the entitlement mentality provides moral support for this same tendency. After all, no one should be required to give up what is rightly his, at least not until everyone else has agreed to do the same. If others get to keep theirs, I get to keep mine! Of course, since everybody knows that everybody else feels equally entitled, no one has any sane reason to be the sucker who sacrifices. Checkmate. And so the entitlement mentality and the Olsonian obstacles to collective action merge, finally, into a single seamless problem.

The result is a peculiar, though entirely understandable, sort of disconnect. The American public is made up of good and patriotic people who want the best for their country and are more than willing to sacrifice—okay, not too much sacrifice, but within reason—in order to make things better. But they do not see why any particular person or group, least of all oneself, should be singled out to give.

In a particularly revealing vein, in 1994 pollsters for ABC News asked people what politicians should and should not do. Predictably, a heavy majority (77 percent) blamed “special-interest groups” for Congress’s failure to get more done; an almost equally heavy majority (63 percent) piously said that their representative should be more interested “in doing what’s best for the country” than “in doing what’s best for your own congressional district where you live.” No surprise. People forever demand that politicians push the special interests aside and do what’s best for the country as a whole—which is what politicians forever promise to do. But generally people decline to believe that there may be a conflict between their program or district and the national interest. Pressing on, the ABC pollsters asked again what it is that members of Congress should be doing, but this time they did not imply a conflict between parochial interests and the larger good. In effect, they asked the question the way it is asked in the voting booth, without an “either-or.” (“For each item I name, please tell me if that’s something your own representative in Congress should or should not be doing.”) Should your representative be “trying to direct more government spending to your congressional district”? Yes! said 58 percent. Should he be “trying to bring federal projects to your district”? Yes! yes! said 73 percent. And should he be “trying to help create jobs in your district”? Yes again! said 90 percent.

“I don’t feel the people of this country have any control over what’s going on, even if we voted in the person we wanted,” a thirty-five-year-old Indiana woman told The New York Times in 1994. By the 1990s, Americans increasingly had come to believe that Washington was in business for itself—which, increasingly, it was. But they spoke as though Washington was in business only for itself rather than brokering endlessly escalating conflicts between ever proliferating groups—as though the people themselves had nothing to do with this. Here, then, was the central paradox of American democracy as the century turned: The public was, at one and the same time, deeply implicated and utterly alienated.

Survival Skills

What to conclude, then? First, what not to conclude.

From the beginning, I have emphasized that this book tries to be a dynamic analysis of government. The problems in modern government don’t have to do with the system’s standing inert or producing a paucity of new laws and regulations, because government emphatically does not stand inert or produce a paucity of new laws and regulations. Rather, it is in a constant state of frenzied motion. To suppose that the combination of unbalanced forces that I’ve described allows no change or reform would be altogether wrong. The significance, rather, is in the nature and direction of the changes that get made, and—what this argument is headed toward—the implications of that sort of change for political control of the government.

To see this, begin by recalling something I said in Chapter 6: “In a Darwinian sense, the collectivity of federal policies is ceasing to evolve.” Because almost no program can be ended, the collectivity of government activities becomes like a company producing every product it ever made, with the government as a whole sclerotic and incoherent and the parts often working at cross-purposes. The message of this chapter and the previous one is that comprehensive, top-down change—change that attempts to reshape the government according to some sort of coherent plan or set of principles—is inherently much harder to accomplish than a casual glance would suggest. Even coherently reshaping some piece of the government (grain subsidies, welfare, banking laws), though possible and from time to time successful, is difficult and requires a burst of energy and some good luck.

Here, however I want to deepen the evolutionary metaphor. We know that every year Congress and the White House review and reauthorize and readminister thousands of programs that are lovingly tended by thousands of client groups. All these people and groups are not keeping themselves busy all the time by doing nothing. We also know that every so often, for ideological or political reasons, some program is attacked by a president or a congressional committee chairman who sees a need to do things differently, or who wants to stir the pot and generate some lobbying activity that will generate some campaign contributions, or both. Of course, almost invariably the program survives, but it may be cut (or expanded) or changed. Its lobbyists may need to scramble to accommodate the new skeptical Californian who replaced the friendly Ohioan on the House Left-Handed Screwdriver Committee.

In this process, adaptation unquestionably occurs. But this is adaptation with a difference. The peanut program, as you’ll recall, was a temporary program that turned out not to be temporary. It was designed to cope with a transitory problem in the 1930s agricultural marketplace, using the heavy-handed market controls in fashion at the time. Its original purpose—to keep family peanut farmers in business through Depression-era turmoil—became archaic, as agribusiness replaced family farms and as economic and marketplace conditions changed. The original problem, to the extent it was a problem, went away, or at least became less pressing than any number of other problems. The program, however, turned out to be very well adapted—not well adapted to meeting any pressing national need in any particularly efficient way, mind you, but well adapted simply at surviving.

In other words, the subsidies and programs and their client groups adapt, all right, but they do so in a way that makes them more tenacious and helps them to lock out other, possibly better ventures that might work better or be more important. The government as a whole stagnates, failing to relinquish the old and thus failing to evolve toward solving problems in any effective or coherent way. From the top-down point of view of someone who wants to use national programs to solve national problems, the trial-and-error process, and therefore evolution, fails. But from the bottom-up point of view of each client group, government evolves constantly. To be specific, it evolves to keep its collection of client services intact from one year to the next.

To see what I mean, it helps to consider an example from life: a federal program known, these days, as the Market Access Program. I say “these days” because it is now on its third name. But I’m getting ahead of myself.

How to Become an Eel

The creature was born in 1985, popping into existence pretty much out of nowhere. No plan was laid out, nor was there any clearly identified national problem to be solved. There was, however, a political problem. That year Congress was passing a budget-busting farm bill for the Midwest, which was in a deep agricultural recession. “Hey, what about us!” said the fruit and vegetable and nut and processed-products industries, many of them based in California. They argued that foreign subsidies were giving them headaches in export markets overseas, and they persuaded a California senator to throw them a bone. And so, to buy the California interests’ support for the farm bill, what was then called the Targeted Export Assistance program was born.

The program was, in effect, a subsidy for advertising. The money was given to agricultural groups and agribusinesses, which used it to market their products overseas. Typically, the federal government required companies or groups to apply for funding and put up some money of their own (anywhere from 10 percent to half). Of course, advertising their products is something that businesses typically do in the ordinary course of things. But they argued that farm products were different, since so many of them were sold generically. No one cotton farmer could begin to finance a campaign hawking, say, American cotton in Turkey. So a government subsidy could be justified.

Or not. To conduct such campaigns on behalf of their members was exactly why trade associations and farm co-ops existed in the first place. Why the Wine Institute (granted about $3.8 million from Washington in 1998), the USA Dry Pea and Lentil Council ($554,000), the Popcorn Institute ($342,000), Ocean Spray International ($247,000), or the American Jojoba Association ($89,000) needed taxpayers’ help to do their job was never quite clear. Moreover, a significant share of the program’s budget went to advertise branded products, for which there was no problem of free ridership, even in principle. Through the Poultry and Egg Export Council, McDonald’s got $465,000 from the taxpayers in fiscal 1991 to promote Chicken McNuggets. No one offered a very good reason why, if promoting Chicken McNuggets made economic sense, McDonald’s needed a subsidy to do it. Other recipients ranged from small players (Southern Pride Catfish) to some of the biggest names in American corporate agriculture: Sunkist, Ernest and Julio Gallo, Tyson Foods, Sun Maid, M&M/Mars. The sorts of promotions being funded sometimes seemed, let us say, picturesque. In 1997, Welch’s Foods, a big grape-growers’ cooperative, received $800,000 (reported National Public Radio) to pay for newspaper supplements and free juice samples at supermarkets in Latin America and Asia. In 1996, the government (reported the Fresno Bee) “paid about $21,000 to fund a California winery tour for nine English wine writers and buyers.” Forty British journalists tasted American wine at a posh London hotel, courtesy of American taxpayers. And so on.

Most programs, the vast majority, labor in welcome obscurity. They are under little evolutionary pressure and remain much the same over time, as the scorpion and shark have done. But helping advertise Chicken McNuggets or sponsoring wine-tasting tours was bound to attract attention. As you might gather from the news stories just cited, almost from its inception this subsidy was a whipping boy, not only for free-market groups but for environmentalists and urban Democrats and, not least, populist liberal opponents of “corporate welfare.” People who needed “wasteful” government programs to denounce made this one their first stop. Year after year, the program was spotlit by groups with names like Citizens Against Government Waste and Taxpayers for Common Sense.

Perhaps a good way to think of this program is as a big, juicy crawdad sitting in full view in shallow waters. It needed survival skills. In the real world, of course, a crawdad is a crawdad; but in Washington, a crawdad can, within only a very few years, become an eel, lean and slippery and elusive.

How? First, the program’s supporters and its administrators in the Agriculture Department poured forth a cascade of studies purporting to show how subsidies for marketing had increased American farm exports by untold billions of dollars. When outside experts looked at the studies, they noted, as any good economist would, that simply counting up the exports flowing from subsidized products revealed nothing at all. Congress’s General Accounting Office, on at least two occasions, sifted through this mass of paper and unearthed nothing worthwhile. In 1997, the GAO said it “found no conclusive evidence that these programs have provided net benefits to the aggregate economy. Government export programs largely reallocate production, employment, and income among sectors.” In other words, the program gave to farmers and took from taxpayers and consumers, making a deposit with the politicians and lobbyists along the way. But this was less important than the fact that the program’s friends could wave studies touting jobs created and the like.

Second, the program enjoyed plenty of support in both parties. The California delegation—Congress’s largest—loved it and teamed up with farm-state Republicans, who had their own subsidies to protect. Enough recipients received enough money in enough states so that Congress was liberally salted with friends. Free-market Republicans and urban Democrats who tried to get rid of the program inevitably ran into opposition from both parties.

Third, and most important, the program adapted. It underwent a series of reforms that made it smaller, to be sure, but that also greatly narrowed its profile. Its annual budget grew from the initial $110 million in 1986 to $200 million by the early 1990s, then fell to about $90 million, where it stayed. This reduction of more than half—from a temporary peak, of course—was touted endlessly as evidence that the program had already given its “fair share” (the entitlement mentality, again). The program also changed its protective plumage. In 1990, the name was changed from Targeted Export Assistance (“assistance” sounded like handouts) to Market Promotion Program. In 1996, the name was changed yet again from Market Promotion Program (“promotion” sounded like corporate welfare) to Market Access Program. Who could be against access to foreign markets?

Other, more significant changes took aim at those headlinemaking horror stories about “corporate welfare” flowing to giant companies. In 1993, Congress gave some preferences to small companies and nonprofit groups. By 1998, the brand-name promotions by big companies were cut out of the program altogether.

Politically, this move made sense. It would reduce the outrageous “corporate welfare” headlines. But, to the extent the program had any rationale at all, the idea was to use tax dollars as efficiently as possible to push American products into foreign markets—and, of course, bigger companies tend to be the most efficient at marketing, overseas or anywhere. “The ability of small companies to spend their branded money is nowhere near that of the larger companies,” grumbled an official of the Wine Institute to the Bee. True enough: If the goal was to open foreign markets, then a bulldozer would work better than a garden trowel. Big companies like McDonald’s could sell a lot more chicken or beef overseas than could Pop Granger’s chicken farm. But it was the big brands that were winnowed. Meanwhile, the promotional money continued flowing to huge trade associations and co-ops, like Sunkist, the giant citrus group, which posted annual sales of more than $1 billion and received about $10 million from the Market Access Program just from 1994 to 1998. No one should be surprised to learn that the said recipients gave reliably to their supporters in Washington.

In short, to become more politically rational, the program became less economically rational. It became more about keeping the money flowing and keeping the press away, and less about doing the job in the most efficient way possible.

In 1992, the Democratic chairman of the House Budget Committee, Leon Panetta, looked at the (then) Market Promotion Program and said, “You almost have to consider eliminating programs like this one.” The next year, Panetta was President Clinton’s new budget director, and he didn’t dare touch the program. In 1992, Richard Armey, a free-market House Republican, denounced the (then) Market Promotion Program as “a completely outrageous case of corporate welfare.” By 1995, Armey was the House Majority Leader—Newt Gingrich’s second-in-command. Still, the Republican Congress pondered all those votes in California and decided that it, too, would take a pass.

After that, matters degenerated into kabuki warfare. Every year the opponents would introduce an amendment to kill the Market Access Program. Every year the defenders would rise to say, as the chairman of the Senate Agriculture Committee said in 1997, that the program had already been “reformed, reformed, and reformed.” Out would come the studies showing jobs purportedly created, and again would be heard the recitations of unfair trade practices abroad. At least one congressman simply made more or less the same speech every year. Others added the odd dramatic touch. A New York Republican said that the program was vital to apple growers in his own Hudson Valley. “We are up there, and the temperatures drop down to 30 or 40 below zero. It is tough enough to make a living as it is.” Deeply moved, every year a compassionate majority of both parties, in both chambers, reaffirmed support for the Market Access Program.

What, you may wonder, was the end being served? Remember, this particular program was not created to solve any particular social problem. It was created to solve a political problem (getting a budget-busting farm bill through Congress in 1985). There were never any firm criteria by which to measure its performance, even if the partisans could have agreed on how to do the measuring. But on one criterion it performed well: It learned how to stay alive. It evolved to become less conspicuous and to lose protuberances that predators could grab. And that is how a crawdad becomes an eel.

As I mentioned, the Market Access Program was under more pressure than most, so it had to be particularly agile. Still, evolutionary feats of this sort are not uncommon. You may recall that in Chapter 6 I mentioned the venerable Rural Electrification Administration, still alive and well sixty years after it was created and forty or more years after its original mission had faded to a memory. By the 1990s, it had become what amounted to a government-aided system of utility cooperatives, competing with private and municipal utility companies (which were themselves subsidized in various ways). In 1993, President Clinton proposed eliminating the rural electric subsidy. To lobby for it, three thousand rural citizens descended on Washington, where they were briefed by the National Rural Electric Cooperative Association’s staff, armed with information packets and buttons, then sent up to Capitol Hill to buttonhole members. After a horde of them circled the desk of Representative Melvin Watt, a North Carolina Democrat, he said, “It’s certainly powerful to see this many people visit you on an issue. It’s got to have an impact on you.” Something else doubtless had an impact, too. In December 1993, only a few weeks after Congress rescued the rural electric program, the National Rural Electric Cooperative Association announced the name of its new chief executive officer: none other than Representative Glenn English of Oklahoma, who, as chairman of the subcommittee on rural development, had been the program’s leading protector in the House. (“Throughout his congressional career,” sang the association, “he has shared our goals.”) Apparently in a hurry to slip through the revolving door, English left halfway through his term to work for the group whose program he had just saved—a cheeky move even by Washington’s slippery standards.

Still, the program was reduced by 40 percent, and, for the first time in its long history, larger subsidies were aimed at needier borrowers—on the whole, meaningful reforms. Then, in 1994, the anachronistic-sounding Rural Electrification Administration was renamed (does this sound familiar?) when it was conjoined with some other Agriculture Department programs. It was called the Rural Utilities Service, with a mission that included financing water, wastewater, and other such local facilities. All of this was done with the support of the subsidies’ recipients, who knew they needed new pastures to graze.

At one level, the change was to the good. Here was an archaic program being gradually modified to go where the need was. If the program was to exist, better it should undertake activities that the relevant lobby and members of Congress agreed were suitable. Yet the change could just as rightly be viewed as what old Washington hands call “mission creep,” which is when a lobby and a bureaucracy go in search of new ways to secure funding. Both views, in fact, are perfectly correct.

But the larger point is that two simplistic—but opposed—views of the situation are wrong. On the one hand, it is not the case that things stay the same in government. To the contrary, things survive by changing all the time. On the other hand, however, it is equally wrong to think that adaptation of this sort is directed in any coherent way, or solves any particular social problem, or comports to any vision of what government should or shouldn’t do, or even notices what’s going on in any other part of the government. Rather, what’s happening is that particular programs are changing to cling to particular niches. The Rural Utilities Service, as the General Accounting Office noted in a 1997 report, still directed much of its subsidized credit to places that were neither notably rural nor notably needy. Once a utility cooperative qualified as serving a rural area, it didn’t need to requalify; for instance, said the GAO, one electric company that qualified for loans in 1945 was still borrowing in 1996, by which point, with 140,000 customers in areas that were now part of a big city, it was hardly “rural.” Spreading around subsidies to plenty of congressional districts, including urban ones, offered good political protection. The program evolved to survive by flinging its seeds everywhere.

Law of the Jungle

“Well,” you say, “of course! So what else is new?” Fair enough. As long ago as 1969, in his influential book The End of Liberalism (whose title, like Francis Fukuyama’s, I tip my hat to), the political scientist Theodore J. Lowi noticed how groups and their programs entrenched themselves as independent fiefdoms—microcosmic regimes or, so to speak, duchies. “Each administrative organization becomes a potent political instrumentality,” he wrote. In agriculture, for example:

Each of the self-governing local units becomes one important point in a definable political system which both administers a program and maintains the autonomy of that program in [the] face of all other political forces emanating from other agricultural systems, from antagonistic farm and nonfarm interests, from Congress, from the secretary, and from the president.

The politics of each of these self-governing programs is comprised of a triangular trading pattern, with each point complementing and supporting the other two. The three points are: the central agency, a congressional committee or subcommittee, and the local or district farmer committees. The latter are also usually the grass-roots element of a national interest group.

Lowi’s commentary on the results was scathing. These self-propagating little units of government, he said, are worrisome for at least three reasons. First, since only an expert could ever understand or even know about any of the minigovernments, they shut out the public, resulting in “the atrophy of institutions of popular control.” Second, the little duchies “tend to create and maintain privilege,” favoring insiders and tight-knit groups over outsiders with greater need or better ideas. Third, and by the same token, this insiders’ world is deeply conservative, inasmuch as the established clients can be counted on to resist any attempts to upset their cozy arrangements. In consequence of all of this, “Liberal governments cannot plan.” (Here he used “liberal” to refer to what he called interest-group liberalism, the giant collectivity of duchies.) “Liberalism replaces planning with bargaining,” Lowi said. “In a pluralistic government there is, therefore, no substance. Neither is there procedure. There is only process.”

Around the time Lowi was writing those words, in the late 1960s, Mancur Olson’s thinking on the arithmetic of collective action was beginning to win attention, and the American government was at the high noon of its dramatic postwar expansion. The years from 1966 to 1976 brought, to name only a few, the National Highway Traffic Safety Administration, the age-discrimination laws, the Consumer Credit Protection Act, the Clean Air Act, the noise- and water-pollution laws, the Environmental Protection Agency, the Occupational Safety and Health Administration, the Consumer Product Safety Commission, the Equal Employment Opportunity Act, and many more. If anyone, back then, had managed to integrate the ideas of Olson and Lowi and then to draw out the implications, he might have seen trouble ahead. In retrospect, what happened is all too clear. The sudden and enormous expansion of the government’s reach—often for the best of reasons—led, as a side effect, to the wild proliferation of groups trying to get and stay on Washington’s good side. Those groups, as Olson’s theory predicted, proved mostly impossible to root out or circumvent, and they established countless new principalities in government—all, of course, on top of the many old ones. By 1981, a reformist president was finding himself almost powerless to reshape the system—if “system” is actually the right word—that resulted. After that, each successive reform wave (Stockman, Clinton, Gingrich) fared worse than the last.

In fact, “system” is not quite the right word. “Ecosystem” is a better word. “System” implies control. But, just as Lowi predicted, today’s government is shaped mainly by its own internal dynamics and by the occasional outside shock (war, mass demographic change) rather than by anyone or anything in particular. In Chapter 6, I spoke of the government as a city in which every house has to be built on top of some existing house. From the point of view of an outside problem-solver, that’s an apt metaphor. If, today, you set out to build a sensible health program for the elderly, or you decided to provide some kind of public insurance for farmers, no one would dream of creating anything like Medicare or the Agriculture Department’s termite nest of subsidies. However, for the people who actually need to work the government, whether as politicians or administrators or lobbyists, there is no “outside the system”: no pristine vantage point from which to assess and redesign. From this operational point of view, a better metaphor for government in the age of demosclerosis is, I think, a jungle.

I was brought up believing that American government was something like an airplane. The voters were the passengers, and the politicians were at the controls. Maybe the airplane wasn’t easy to turn on a whim; in fact, it was designed to be hard to turn. But there was a chain of command. If the politicians didn’t do what the electorate wanted, they would be fired and other, more responsive politicians would take their places. And, one way or another, the whole of government worked that way. It was, in that sense, democratically accountable through and through.

That civics-book view is, of course, not all wrong. But the implication of Olson and Lowi and the rise of the parasite economy, with all that they imply, is that at any given moment only the most public, the most controversial, or the most dangerously dysfunctional bits of the government are accessible to coherent reform or direct public pressure, never mind who the reformer happens to be. Once in a while, perhaps once or occasionally twice in a decade, a piece of the government—welfare, say, or maybe Social Security—rises to the forefront of politics, and then a big-picture reform guided by the election returns is, at least, possible (though hardly assured). Nearly always, though, almost all of the government goes its own way. Politicians talk, and think, as though they are pilots in a cockpit. In fact, they are more like elephants in the jungle: highly visible and capable of throwing their weight around or toppling the odd tree, but mostly at the mercy of the intricate ecology that surrounds them. It is the aphids and earthworms and dung beetles and termites and algae, the busy little creatures all in their niches—the farm lobbies and veterans’ lobbies and real-estate lobbies and education lobbies and a million other species—that shape the jungle’s topography. They are neither evil nor benign, and their unimaginably variegated activities have all sorts of effects on all kinds of things, good or bad depending on your point of view. They are merely doing what they do.

As for the voters, they too are simply a part of the landscape. They are, of course, important, but not in any consistent or reliable way. They are an unpredictable force that once in a while generates an inchoate upheaval, like a flood or a fire. Facing such an upheaval, the bugs and worms and mice and (of course) elephants scurry to adjust; whereupon, after an interruption, all settle in again to the familiar patterns. No, the voters are not helpless. Politicians are not helpless. They just matter, most of the time and in most decisions of ordinary governance, much less than they suppose. Voters and politicians influence the governmental ecology, but they do nothing like control it. No one controls it. In the universe of government, there is no God.

In The End of Liberalism, Theodore Lowi said that interest-group liberalism in America had so transformed governance as to have established what he called a new regime, a Second Republic of the United States. The First Republic lasted through the early years of the twentieth century and was marked by a relatively weak and limited federal center. With the rise of interest-group pluralism beginning around the time of the New Deal, the government became increasingly a collection of subsidies and guarantees parceled out to groups, giving rise at last to the Second Republic. Just what he meant by that term is not altogether clear to me, but never mind: His language is interesting and suggests a way of looking at things.

In ideological terms, conservatives see government as properly a guarantor of individual rights, and possibly also as a watchman for the interests of enterprise. For 150 years or so, American government conformed largely to their vision. By today’s standards, it was very small and very weak, and the country’s many associations were of the voluntary, nonlobbying kind that were familiar to Tocqueville. Call that (this is me, now, not Lowi) a First Republic.

Liberals see government as properly a solver of national problems, and possibly also as a builder of a more nearly ideal society. For thirty or forty or fifty years, beginning around the time of the New Deal, the liberals had their day: The government was ambitious, undertook all sorts of commitments to pensioners and veterans and students and consumers, and seemed often successful in meeting them. But with the growth of the programs came the jungle, the burrowing and flying and stinging creatures; and with the growing perception of failure—with farmers being paid not to grow food, the welfare culture expanding, the tax code becoming spaghetti, lawyers and lobbyists overrunning Washington, inflation, deficits, bureaucracies—came the backlash and the era of reform. Call the expansionist period and the accompanying backlash the Second Republic.

And now, at last, comes this, what you see around you: the perpetual stalemate of evolutionary equilibrium, in which the clients and the calculus of collective action will not allow the government to become much smaller or to reorganize its basic functions, while the taxpayers will not suffer it to grow much bigger. The borders of the jungle are more or less as they will be. From a distance, in macrocosm, the jungle seems an immovable mass, unchanging from year to year and impenetrably dense, whereas up very close, in microcosm, it is a constant turmoil of digging and scurrying and eating and mating. But it exists primarily to survive from year to year and to feed its clients. Its clients—we—draw sustenance from it but yield control. Call that the Third Republic.

In the end, it is not the conservative vision of government or the liberal vision that prevailed. It is no vision at all that prevailed. The client groups prevailed. And that is the end of government. To see the future, look around.