The Accidental Theorist

by Krugman, Paul

  Finally, “Looking Backward” may need some explaining. It was written for a special centennial issue of the New York Times Magazine; the assignment given to each of the authors was to write about his or her specialty as if looking back from the perspective of the year 2096. Incredibly, out of the fifteen or so authors only two were willing to play that game—the rest defied instructions and wrote boringly straight essays about “What I predict will happen over the next century.” It’s a mystery to me, because I thought the idea was great fun—although the essay is more serious than it looks.

  Earth in the Balance Sheet: Economists Go for the Green

  Like most people who think at all about how much burden their way of life places on Spaceship Earth, I feel a bit guilty. But on Earth Day in 1997 my conscience was clearer than usual—and so were those of 2,500 other economists.

  A few months earlier, an organization called Redefining Progress enlisted five economists—the Nobel laureates Robert Solow and Kenneth Arrow, together with Harvard’s Dale Jorgenson, Yale’s William Nordhaus, and myself—to circulate an “Economists’ Statement on Climate Change,” calling for serious measures to limit the emission of greenhouse gases. To be honest, I agreed to be one of the original signatories mainly as a gesture of goodwill, and never expected to hear any more about it; but the statement ended up being signed by, yes, more than 2,500 economists. Whatever else may come of the enterprise, it was an impressive demonstration of a little-known fact: Many economists are also enthusiastic environmentalists.

  Partly this is just because of who economists are: Being by definition well-educated and, for the most part, pretty well-off, they have the usual prejudices of their class—and most upper-middle-class Americans are sentimental about the environment, as long as protecting it does not impinge on their lifestyle. (I bring bags and bottles to the recycling center in my gas-guzzling sports utility vehicle.) But my unscientific impression is that economists are on average more pro-environment than other people of similar incomes and backgrounds. Why? Because standard economic theory automatically predisposes those who believe in it to favor strong environmental protection.

  This is not, of course, the popular image. Everyone knows that economists are people who know the price of everything and the value of nothing, who think that anything that increases gross domestic product is good and anything else is worthless, and who believe that whatever free markets do must be right. (I’m sorry to say that some of the people at Redefining Progress published an impressively ill-informed diatribe along these lines in the Atlantic back in 1995.)

  But the reality is that even the most conventional economic doctrine is a lot more subtle than that. True, economists generally believe that a system of free markets is a pretty efficient way to run an economy, as long as the prices are right—as long, in particular, as people pay the true social cost of their actions. Environmental issues, however, more or less by definition involve situations in which the price is wrong—in which the private costs of an activity fail to reflect its true social costs. Let me quote from the textbook (by William Baumol and Alan Blinder) that I assigned when I taught Economics 1 last year: “When a firm pollutes a river, it uses some of society’s resources just as surely as when it burns coal. However, if the firm pays for coal but not for the use of clean water, it is to be expected that management will be economical in its use of coal and wasteful in its use of water.” In other words, when it comes to the environment, we do not expect the free market to get it right.

  So what should be done? Going all the way back to Paul Samuelson’s first edition in 1948, every economics textbook I know of has argued that the government should intervene in the market to discourage activities that damage the environment. The usual recommendation is to do so either by charging fees for the right to engage in such nasty activities—a.k.a. “pollution taxes”—or by auctioning off rights to pollute. Indeed, as the extraordinary response to the climate-change statement reminds us, the idea of pollution taxes is one of those iconic positions, like free trade, that commands the assent of virtually every card-carrying economist. Yet while pollution and related “negative externalities” such as traffic congestion are obvious problems, in practice, efforts to make markets take environmental costs into account are few and far between. So economists who actually believe the things they teach generally support a much more aggressive program of environmental protection than the one we actually have. True, they tend to oppose detailed regulations that tell people exactly how they must reduce pollution, preferring schemes that provide a financial incentive to pollute less but leave the details up to the private sector. But I would be hard pressed to think of a single economist not actually employed by an anti-environmental lobbying operation who believes that the United States should protect the environment less, not more, than it currently does. (The signers of the climate-change statement, incidentally, included thirteen economists from that temple of free-market theory, the University of Chicago.)

  But won’t protecting the environment reduce the gross domestic product? Not necessarily—and anyway, so what?

  At first sight, it might seem obvious that pollution taxes will reduce GDP. After all, any tax reduces the incentives to work, save, and invest. Thus a tax on exhaust emissions from cars will induce people to drive cleaner cars or avoid driving altogether. But since it will also in effect lower the payoff to earning extra money (since you wouldn’t end up driving the second car you could buy with that money anyway), people will not work as hard as they would have without the tax. The result is that taxes on pollution (or anything else) will, other things being equal, tend to reduce overall monetary output in the economy—which is to say, GDP.

  But things need not be equal, because there is already a whole lot of taxing and spending going on. Even in the United States, where the government is smaller than in any other advanced country, about a third of GDP passes through its hands. So existing taxes already discourage people from engaging in taxable activities like working or investing. What this means is that the revenue from any new taxes on pollution could be used to reduce other taxes, such as Social Security contributions or the income tax (but not, of course, the capital-gains tax). While the pollution taxes would discourage some activities that are counted in the GDP, the reduction in other taxes would encourage other such activities. So measured GDP might well fall very little, or even rise.

  Does this constitute an independent argument for taxing pollution, quite aside from its environmental payoff? Would we want to have, say, a carbon tax even if we weren’t worried about global warming? Well, there has been an excruciatingly technical argument about this, mysteriously known as the “double dividend” debate; the general consensus seems to be no, and that on balance pollution taxes would be more likely to reduce GDP slightly than to increase it.

  But so what? “Gross domestic product is not a measure of the nation’s economic well-being”—so declares the textbook as soon as it introduces the concept. If getting the price of the environment right means a rise in consumption of nonmarket goods like clean air and leisure time at the expense of marketed consumption, so be it.

  Isn’t this amazing? Not only do thousands of economists agree on something, but what they agree on is the warm and cuddly idea that we should do more to protect the environment. Can 2,500 economists be wrong? Well, yes—but this time they aren’t. The Great Green Tax Shift—a shift away from taxes on employment and income toward taxes on pollution and other negative externalities—has everything going for it. It is supported by good science and good economics, as well as by good intentions.

  Inevitably, then, it appears at the moment to be a complete political nonstarter. The problem, as with many good policy ideas, is that the Great Green Tax Shift runs up against the three I’s.

  First, there is Ignorance. In 1996 Congress rushed to cut gasoline taxes to offset a temporary price rise. Not many voters stopped to ask where the money was coming from. So what politician will be foolish enough
to take the first step in trying to institute new taxes on all-American pollution, even with the assurance that other taxes will be lowered at the same time? (My friends in the Clinton administration tell me that the word “taxes” has been banned even from internal discussions about environmental policy.)

  Then there are Interests. It is hard to think of a way to limit global warming that will not gradually reduce the number of coalmining jobs. As labor-market adjustment problems go, this is a pretty small one. But the coal miners and the energy companies are actively opposed to green taxes, while the broader public that would benefit from them is not actively in support.

  Finally, there is Ideology. It used to be that the big problem in formulating a sensible environmental policy came from the Left—from people who insisted that since pollution is evil, it is immoral to put a price on it. These days, however, the main problem comes from the Right—from conservatives who, unlike most economists, really do think that the free market is always right—to such an extent that they refuse to believe even the most overwhelming scientific evidence if it seems to suggest a justification for government action.

  So I do not, realistically, expect the Economists’ Statement to change the world. But then I didn’t expect it to go as far as it has. Certainly those of us who signed it did the right thing; and maybe, just maybe, we did our bit toward saving the planet.

  Taxes and Traffic Jams

  Why do we spend so much time talking about tax reform? Why don’t we try to eliminate traffic jams instead?

  This is not a silly question. The case for doing something about traffic is based on impeccable free-market economics, every bit as solid as the argument for reforming our tax system. The difference is that whereas in practice tax reform is an iffy business—the plans currently being peddled would probably do more harm than good—traffic reform is a $40 billion sure thing. The fact that traffic congestion, along with a number of similar issues like pollution management and water rights, goes unmentioned in current political discussion—and that when such issues do come up conservative politicians are often on the wrong side—tells you something important about the blinkered vision of many people who imagine that they are champions of free markets.

  Let’s talk for a moment about tax reform. The air is thick with schemes for flat taxes, value-added taxes, national sales taxes, and so on. Serious advocates of such proposals point out that the current system has two main flaws from the point of view of economic efficiency. First, some people pay a “marginal” tax rate as high as 40 percent—that is, of every extra dollar they earn forty cents goes to the I.R.S. This surely discourages people from working as hard as they might. Second, because the I.R.S. taxes interest and profits, the system discourages people from saving for the future. So theorists have devised alternative tax systems that might lead to greater work effort and higher savings, and might therefore expand the American economy.

  But how big would these gains be? A lot of savings are already tax exempt, because of special tax breaks for retirement accounts. And we can’t do away with taxes altogether: Like it or not, we still need to pay for the government services we want. This limits the scope for reducing marginal rates. For example: A realistic flat tax, one that would raise as much money as the current income tax, would still have to involve a marginal rate of well over 20 percent. That’s lower than the 40 percent rate some people now pay, but most people don’t pay that rate anyway. And even that twentysomething rate is possible only if we eliminate the tax deduction for interest on home mortgages—which means that while the economy might gain from the new tax, millions of middle-class families would lose.

  The fact is that serious tax analysts believe that the net benefits from even a complete overhaul of the tax system would be modest, and that while most people might gain, many would lose. In practice, the prospect of middle-class outrage means that schemes like the flat tax are usually sold with the promise of unrealistically low tax rates. And while a realistic tax reform would be good for economic growth, the typical political scheme—which invariably promises to provide large tax cuts for the rich without any increases for the middle class, would create a massive budget deficit—thereby doing the economy far more harm than good.

  Now let’s talk about traffic. Traffic congestion is not a minor annoyance. Last year Americans lost more than eight billion hours to traffic delays, at a total economic cost of more than $80 billion—mainly in the form of wasted time, but also through extra consumption of gasoline, wear and tear on autos, and so on.

  But aren’t traffic jams just a fact of life? No: In large part they are the result of a system that, like the tax system, encourages people to make economically inefficient decisions.

  Consider, for example, my own antisocial actions one day in 1996, when I was living and working at Stanford. A colleague who lived nearby and I were both going to a meeting in San Francisco, thirty miles away. At the cost of some minor inconvenience we could have traveled together. But we didn’t, and by taking my own car I added marginally to the already world-class traffic congestion on Highway 101. I probably didn’t delay any individual’s morning commute by more than a fraction of a second—but that tiny delay was imposed on each of thousands of cars crawling along the freeway behind me. It’s a good bet that the total delay that other people suffered for the sake of my slight convenience was more than an hour.

  There ought to have been a way to make a deal: I wouldn’t clog the freeway, and other drivers would compensate me for the inconvenience of carpooling. Since the cost I imposed on other people by driving during rush hour was much greater than the benefit I derived from so doing, such a deal could have made everyone happier. Of course it is impractical to make such a deal directly; but we can try to reproduce its results.

  The classic economist’s prescription for dealing with traffic is for the government to impose “congestion fees,” tolls for using roads during periods of overcrowding. With modern technology such tolls could even be collected without tollbooths: An electronic sensor could pick up the signal from a tiny gizmo on your dashboard, or a low-powered laser could read a bar code on your windshield. The fees could then be rebated to the public. Realistic estimates suggest that such fees could cut the cost of traffic congestion dramatically, and leave the great majority of people better off. Most of those deterred from rush-hour driving would be more than compensated for their inconvenience through the rebate scheme. Those who continued to drive would be compensated for the extra tolls with a much faster trip to work.

  I know what some conservative readers are thinking—that this is just another government intrusion into daily life. But think of it another way: What we have here is a problem of inadequately defined property rights. Space on Highway 101 during rush hour is a scarce resource, just like waterfront real estate. Unfortunately, nobody holds clear title to that resource, and so it gets overused.

  So why don’t we try to establish a properly functioning free market in road space? Suppose we issue every registered driver in a metropolitan area with a specified number of “rush hour points,” which he is free to sell to other drivers at his discretion, and require anyone who drives during rush hour to present (electronically) the appropriate number of points. Let us also create a market in these points. Some people will continue to drive every day; they will need to purchase extra points. Others will find other ways of getting to work, and will sell their points for whatever price the, ahem, traffic will bear. How market-oriented can you get? And such a scheme, like congestion fees, will make most people better off—either because they get extra money by selling their points, or because the cost of buying points is more than offset by a quicker commute.

  The reason creating a market in road space and imposing congestion fees would produce similar results is, of course, that—as any bright Econ 1 student can tell you—they are essentially equivalent. Either way, what you are doing is creating a market incentive for people like me to take into account the costs we impose on others b
y driving during rush hour.

  The benefits of such a scheme would not be small change. Traffic experts tell us that a sensible system of congestion fees could easily cut the annual cost of traffic delays by $30 or 40 billion. Against these savings one must set the inconvenience to those who are deterred from driving; but the net benefits could be $15 billion a year or more. Add fees on heavy trucks that damage highways, and on air travel to crowded airports, and a middle-of-the-road estimate (and what other kind of estimate would you want?) is that traffic reform could enrich the American public by $40 to 50 billion every year.

  There are other places where properly defined property rights could yield a big payoff. One example is water—a very scarce resource in the American West. Under the current system thirsty cities are rationed even while desert farmers feed irrigated alfalfa to their cattle. Why not create a free market in water rights? Another example is the airwaves, which used to carry only radio and television but are increasingly in demand to allow wireless communication among people and computers. Broadcast rights are now a valuable commodity, which could be traded in a free market, but—despite the recent auction of a limited slice of the spectrum—mostly aren’t, with the result that bandwidth that could be carrying vital business data is carrying infomercials instead.

  So why isn’t there a major conservative politician out there campaigning on a platform of extending free-market principles, of creating property rights where they should exist but don’t? Such a politician could promise to raise national income by $60, 80, even 100 billion a year—and he or she could do so honestly, without the slippery arithmetic that underlies the promises we are actually hearing.