Naked Economics (7 page)

Or would it? Using a car seat requires that a family buy an extra seat on the plane, which dramatically increases the cost of flying. Airlines no longer offer significant children’s discounts; a seat is a seat, and it is likely to cost at least several hundred dollars. As a result, some families will choose to drive rather than fly. Yet driving—even with a car seat—is dramatically more dangerous than flying. As a result, requiring car seats on planes might result in more injuries and deaths to children (and adults), not fewer.

Consider another example in which good intentions led to a bad outcome because the incentives were not fully anticipated. Mexico City is one of the most polluted cities in the world; the foul air trapped over the city by the surrounding mountains and volcanoes has been described by the
New York Times
as “a grayish-yellow pudding of pollutants.”
⁶
Beginning in 1989, the government launched a program to fight this pollution, much of which is caused by auto and truck emissions. A new law required that all cars stay off the streets one day a week on a rotating basis (e.g., cars with certain license plate numbers could not be driven on Tuesday). The logic of the plan was straightforward: Fewer cars on the road would lead to less air pollution.

So what really happened? As would be expected, many people did not like the inconvenience of having their driving days limited. They reacted in a way that analysts might have predicted but did not. Families who could afford a second car bought one, or simply kept their old car when buying a new one, so that they would always have one car that could be driven on any given day. This proved to be worse for emissions than no policy at all, since the proportion of old cars on the road went up, and old cars are dirtier than new cars. The net effect of the policy change was to put more polluting cars on the road, not fewer. Subsequent studies found that overall gas consumption had increased and air quality did not improve at all. The policy was later dropped in favor of a mandatory emissions test.
⁷

Good policy uses incentives to some positive end. London has dealt with its traffic congestion problems by applying the logic of the market: It raised the cost of driving during the hours of peak demand. Beginning in 2003, the city of London began charging a £5 ($8) congestion fee for all drivers entering an eight-square-mile section of the central city between 7:00 a.m. and 6:30 p.m.
⁸
In 2005, the congestion charge was raised to £8 ($13), and in 2007, the size of the zone for which the fee must be paid was expanded. Drivers are responsible for paying the charge by phone, Internet, or in selected retail shops. Video cameras were installed in some 700 locations to scan license plates and match the data against records of motorists who have paid the charge. Motorists caught driving in central London without paying the fee are fined £80 ($130).

The plan was designed to take advantage of one of the most basic features of markets: Raising prices reduces demand. Raising the cost of driving discourages some drivers and improves the flow of traffic. Experts also predicted an increase in the use of public transit, both because it is a cheap alternative to driving, but also because buses would be able to move more quickly through central London. (Faster trips lower the opportunity cost of taking public transit.) Within a month, the results were striking. Traffic fell 20 percent (settling after several years at 15 percent lower). Average speed in the congestion zone doubled; bus delays were cut in half; and the number of bus passengers climbed 14 percent. The only unpleasant surprise was that the program had such a significant deterrent effect on car traffic that revenues from the fee were lower than expected.
⁹
Retailers have also complained that the fee discourages shoppers from visiting central London.

Good policy uses incentives to channel behavior toward some desired outcome. Bad policy either ignores incentives, or fails to anticipate how rational individuals might change their behavior to avoid being penalized.

The wonder of the private sector, of course, is that incentives magically align themselves in a way that makes everyone better off. Right? Well, not exactly. From top to bottom, corporate America is a cesspool of competing and misaligned incentives. Have you ever seen some variation of the sign near the cash register at a fast-food restaurant that says, “Your meal is free if you don’t get a receipt. Please see a manager”? Does Burger King have a passionate interest in providing a receipt so that your family bookkeeping will be complete? Of course not. Burger King does not want its employees stealing. And the only way employees can steal without getting caught is by performing transactions without recording them on the cash register—selling you a burger and fries without issuing a receipt and then pocketing the cash. This is what economists call a principal-agent problem. The principal (Burger King) employs an agent (the cashier) who has an incentive to do a lot of things that are not necessarily in the best interest of the firm. Burger King can either spend a lot of time and money monitoring its employees for theft, or it can provide an incentive for you to do it for them. That little sign by the cash register is an ingenious management tool.

Principal-agent problems are as much a problem at the top of corporate America as they are at the bottom, in large part because the agents who run America’s large corporations (CEOs and other top executives) are not necessarily the principals who own those companies (the shareholders). I own shares in Starbucks, but I don’t even know the CEO’s name. How can I be sure that he (she?) is acting in my best interest? Indeed, there is ample evidence to suggest that corporate managers are no different from Burger King cashiers—they have some incentives that are not always in the best interest of the firm. They may steal from the cash register figuratively by showering themselves with private jets and country club memberships. Or they may make strategic decisions from which they benefit but shareholders do not. For example, a shocking two-thirds of all corporate mergers do not add value to the merged firms and a third of them leave shareholders worse off. Why would very smart CEOs engage so often in behavior that seems to make little financial sense?

One partial answer, economists have argued, is that CEOs benefit from mergers even when shareholders are left with losses. A CEO draws a lot of attention to himself by engineering a complex corporate transaction. He is left running a bigger company, which is almost always more prestigious, even if the new entity is less profitable than the merged companies were when they were on their own. Big companies have big offices, big salaries, and big airplanes. On the other hand, some mergers and takeovers make perfect strategic sense. As an uninformed shareholder with a large financial stake in the company, how do I tell the difference? If I don’t even know the name of the CEO of Starbucks, how can I be sure that she (he?) is not spending the bulk of her day chasing attractive secretaries around her office? Hell, this is harder than being a manager at Burger King.

For a time, clever economists believed that stock options were the answer. They were supposed to be the CEO equivalent of the sign near the cash register asking if you received your receipt. Most American CEOs and other important executives receive a large share of their compensation in the form of stock options. These options enable the recipient to purchase the company’s stock in the future at some predetermined price, say $10. If the company is highly profitable and the stock does well, climbing to say $57, then those stock options are very valuable. (It is good to be able to buy something for $10 when it is selling on the open market for $57.) On the other hand, if the company’s stock falls to $7, the options are worthless. There is no point in buying something for $10 when you can buy it on the open market for $3 less. The point of this compensation scheme is to align the incentives of the CEO with the interests of the shareholders. If the share price goes up, the CEO gets rich—but the shareholders do well, too.

It turns out that wily CEOs can find ways to abuse the options game (just as cashiers can find new ways to steal from the register). Before the first edition of this book came out, I asked Paul Volcker, former chairman of the Federal Reserve, to give it a read since he had been a professor of mine. Volcker read the book. He liked the book. But he said that I should not have written admiringly about stock options as a tool for aligning the interests of shareholders and management because they are “an instrument of the devil.”

Paul Volcker was right. I was wrong. The potential problem with options is that executives can do things to goose the firm’s stock in the short run that are bad or disastrous for the company in the long run—after the CEO has sold tens of thousands of options for an astronomical profit. Michael Jensen, a Harvard Business School professor who has spent his career on issues related to management incentives, is even harsher than Paul Volcker. He describes options as “managerial heroin,” because they create an incentive for managers to seek short-term highs while doing enormous long-term damage.
¹⁰
Studies have found that companies with large options grants are more likely to engage in accounting fraud and more likely to default on their debt.
¹¹

Meanwhile, CEOs (with or without options) have their own monitoring headaches. Investment banks like Lehman Brothers and Bear Stearns were literally destroyed by employees who took huge risks at the firm’s expense. This is a crucial link in the chain of causality for the financial crisis; Wall Street is where a bad problem became disastrous. Banks across the country could afford to feed the real estate bubble with reckless loans because they could quickly bundle these loans together, or “securitize” them, and sell them off to investors. (A bank takes your mortgage, bundles it together with my mortgage and lots of others, and then sells the package off to some party willing to pay cash now in exchange for a future stream of income—our monthly mortgage payments.) This is not inherently a bad thing when done responsibly; the bank gets its capital back right away, which can then be used to make new loans. However, if you take the word “responsibly” out of that sentence, it does become a bad thing.

Simon Johnson, former chief economist for the International Monetary Fund, wrote an excellent postmortem of the financial crisis for
The Atlantic
in 2009. He notes, “Major commercial and investment banks—and the hedge funds that ran alongside them—were the big beneficiaries of the twin housing and equity-market bubbles of this decade, their profits fed by an ever-increasing volume of transactions founded on a relatively small base of actual physical assets. Each time a loan was sold, packaged, securitized, and resold, banks took their transaction fees, and the hedge funds buying those securities reaped ever-larger fees as their holdings grew.”
¹²

Each transaction carries some embedded risk. The problem is that the bankers making huge commissions on the buying and selling of what would later become known as “toxic assets” do not bear the full risk of those products; their firms do. Heads they win, tails the firm loses. In the case of Lehman Brothers, that’s a pretty accurate description of what happened. Yes, the Lehman employees lost their jobs, but those most responsible for the collapse of the firm don’t have to give back the huge bonuses they made in the good years.

One other culpable party deserves mention, and again misaligned incentives was a key problem. The credit rating agencies—Standard & Poor’s, Moody’s, and others—are supposed to be the independent authorities that evaluate the risk of these newfangled products. Many of the “toxic assets” now at the heart of the financial meltdown were given stellar credit ratings. Part of this was pure incompetence. It didn’t help, however, that the credit rating agencies are paid by the firms selling the bonds or securities being rated. That’s a little like a used car salesman paying an appraiser to stand around the lot and provide helpful advice to customers. “Hey Bob, why don’t you come over here and tell the customer whether he is getting a good deal or not.” How useful do you think that would be?

These corporate incentive problems remain unresolved as far as I can tell, both for senior executives in public companies and for other employees taking risks with their firm’s capital. There is a fundamental tension that is tough to resolve. On the one hand, firms need to reward innovation, risk, insight, hard work, and so on. These are good things for the firm, and employees who do them well should be paid handsomely—even astronomically in some cases. On the other hand, the employees doing fancy things (like designing new financial products) will always have more information about what they are really up to than their superiors will; and their superiors will have more information than the shareholders. The challenge is to reward good outcomes without creating incentives for employees to game the system in ways that damage the company in the long run.

One need not be a corporate titan to deal with principal-agent problems. There are plenty of situations in which we must hire someone whose incentives are similar but not identical to our own—and the distinction between “similar” and “identical” can make all the difference. Take real estate agents, a particular breed of scoundrel who purport to have your best interest at stake but may not, regardless of whether you are buying or selling a property. Let’s look at the buy side first. The agent graciously shows you lots of houses and eventually you find one that is just right. So far, so good. Now it is time to bargain with the seller over the purchase price, often with your agent as your chief adviser. Yet your real estate agent will be paid a percentage of the eventual purchase price. The more you are willing to pay, the more your agent makes and the less time the whole process will take.