Every May, Chicago public school students take a standardized test. Students are used to being tested, graded, and rewarded accordingly, but beginning in May 1996, teachers and principals had a lot more than usual on the line: Schools with low scores would be closed, teachers reassigned, and principals fired. The idea, of course, was to give educators stronger incentives to work harder and better. If grading was good for the students, why not for the teachers?

Stronger incentives do give teachers and principals an incentive to put in extra hours and search for better teaching methods. But how else can teachers raise the grades of their students? Here’s a hint: Some students also use this method. That’s right—they cheat. Indeed, teachers can cheat a lot better than students because they know which answers are correct! Two economists who understand incentives, Brian Jacob and Steven Levitt (the latter of Freakonomics fame), started to look carefully at test data and asked: Would teachers really cheat to raise student grades?1 Sure enough, Jacob and Levitt found odd patterns in the data—students who got easy answers wrong and difficult answers right, groups of students who had exactly the same right and wrong answers, and students who received high grades during a test year but low grades the year after. Most telling for an economist was that the indicators of cheating were much stronger after the penalty for low-performing schools went into effect than before!

410

Perhaps you think that teachers’ cheating to raise student grades is a good idea! But it wasn’t what the proponents of strong incentives for teachers had in mind. Not all teachers cheated, but cheating was surprisingly common. Jacob and Levitt estimated that cheating occurred in at least 4% to 5% of classrooms. Other researchers have found that after the introduction of strong incentives, a lot more students are declared learning disabled.2 Why? Test scores of students called “learning disabled” are usually not counted when it comes to rewarding teachers and principals.

Does all this mean that strong incentives for teachers are a bad idea? Not necessarily. Students who learn more, earn more. If strong incentives for teachers do increase true scores, even by a small amount, maybe it’s a good idea even if some of the better scores are due to cheating.3

A similar example of incentives for cheating comes from corporate finance. In the 1980s, chief executive officers (CEOs) were given much stronger incentives to increase their firm’s stock price. Instead of being paid a straight salary, they were awarded stock options. These are complicated financial instruments, but what you need to know is that they pay off only if the stock rises above a certain price. As with strong incentives for teaching, strong incentives encouraged CEOs to work harder and smarter. It also encouraged them to cheat by manipulating earnings reports to make their firms appear more profitable than they really were. Enron and the other scandals of the 1990s and first decade of the 2000s were, in part, the result. Were strong incentives worth it? If the shareholders believed that on average the costs of cheating exceeded the benefits of encouraging harder work, they would offer their CEOs fewer options and other strong incentives. But so far most of these incentives have stayed in place, albeit with more monitoring of potentially bad behavior.

Shareholders, however, are not the only ones who can be harmed when a company like Enron or Lehman Brothers collapses, so their choice of CEO incentive scheme may not reflect everyone’s interests and may not be best for society as a whole. (Recall our discussion of externalities in Chapter 10.) Incentive schemes, for example, that give executives big bonuses for very good performance but don’t penalize them very much for very bad performance can encourage executives to take on too much risk. In part for this reason, investment banks such as Bear Stearns and Lehman Brothers took on lots of risk in mortgage securities, and when they collapsed, that helped lead to the financial crisis of 2008. Executive compensation, therefore, has become a subject of political controversy. We will return to executive compensation later on in this chapter.

When designing an incentive scheme, remember this: You get what you pay for. That sounds good but there is a problem. What if what you pay for is not exactly what you want? If you pay for higher test scores, you will get higher test scores. But test scores are an imperfect measure of what you really want—more productive teachers and more knowledgeable students. What you pay for is higher stock prices, but what you really want is a more profitable firm. Usually, stock prices reflect a firm’s fundamental value, but even the market can be fooled sometimes!

411

The closer “what you pay for” is to “what you want,” then the more you can rely on strong incentives. Careful design of an incentive scheme can narrow the gap between what you want and what you can pay for. After Jacob and Levitt published their results, the administrators of Chicago Public Schools, to their credit, fired some teachers, and instituted new procedures to make cheating more difficult. After the Enron scandal, investors demanded more independent financial audits. The stronger the incentives, the more it pays to invest in careful measurement and auditing, and vice versa.

If you can’t bridge the gap between “what you pay for” and “what you want,” then weak incentive schemes can be better than strong incentive schemes.

Should the management of prisons be contracted out to the private sector? The owners of a private firm have a strong incentive to cut costs and improve productivity because they get to keep the resulting profits. If a public prison cuts costs, there is more money in the public treasury but no one gets to buy a yacht, so the incentive to cut costs is much weaker.

In 1985, Kentucky became the first state to contract out a prison to a for-profit firm. Private prisons today hold about 128,000 prisoners in the United States, about 8% of all prisoners. Should efficient private prisons replace inefficient public prisons? Three economists—Oliver Hart, Andrei Shleifer, and Robert Vishny (HSV)—say no. HSV don’t question that the profit motive gives private prisons stronger incentives than public prisons to cut costs—HSV say that’s the problem! Suppose that we care about costs but we also care about prisoner rehabilitation, civil rights, and low levels of inmate and guard violence. What we pay for is cheap prisons, but what we want is cheap but high-quality prisons. If we can’t measure and pay for quality, then strong incentives could encourage cost cutting at the expense of quality.

The principle is a general one, a strong incentive scheme that incentivizes the wrong thing can be worse than a weak incentive scheme. One car dealer in California advertises that its sales staff is not paid on commission.4 Why would a store advertise that its sales staff do not have strong incentives to help you? The answer is clear to anyone who has tried to buy a car. High-pressure dealers who pounce on you the moment you enter the showroom and bombard you with high-pressure sales tactics (“I can get you 15% off the sticker, but you have to act NOW!”) may sell cars to first-time buyers, but the strategy is too unpleasant to win many repeat customers. Car dealers who rely on repeat business usually prefer a low-pressure, informative sales staff.

In theory, a car dealer could have strong incentives and repeat business by paying its sales staff based on their “nice” sales tactics, but in practice, it’s too expensive to monitor how salespeople interact with clients. Cheating by the sales staff would be difficult to detect and thus would be common. Paying the sales staff a salary instead of a commission calms them down a bit. Of course, there is a price to be paid for weak incentives. Imagine that Joe’s Honda pays its sales staff on commission, while Pete’s Subaru pays its staff a straight salary. Which dealership do you expect to be open late at night and on Sundays?

412

What about prisons? Are HSV correct that weak-incentive public prisons are better than strong-incentive private prisons? Not necessarily. HSV assume that cutting quality is the way to cut cost. But sometimes higher quality is also a path to lower costs. Low levels of inmate and guard violence, for example, are likely to reduce costs. And respect for prisoner’s civil rights? That can save on legal bills. When quality and cost cutting go together, a private firm has a strong incentive to increase quality.

HSV may also underestimate how well quality can be measured. Measuring intensively pays off more when incentives are strong. Unsurprisingly, therefore, private prison companies and government purchasers have made extensive efforts to measure the quality of private prisons.

Finally, don’t forget that weak incentives reduce the incentive to cut costs but they don’t increase the incentive to produce high quality! Public prisons might use their slack budget constraints to offer high-quality rehabilitation programs, or they might instead offer prison guards above-market wages. Which do you think is more likely?

Nevertheless, whether HSV are right or wrong about private prisons, their argument is clever. The usual argument against government bureaucracy is that without the profit incentive, public bureaucracies won’t have an incentive to cut costs. HSV suggest this is exactly why public bureaucracies may sometimes be better than private firms.5

A majority of workers are paid by the hour but a significant number are paid by the piece. An hourly wage pays workers for their inputs (of time); a piece rate pays workers for their output. Agricultural workers, for example, are often paid by the number of pieces of fruit or vegetable that they pick. Garment workers are often paid per item completed. Salespeople are often paid, in part, by the number of sales that they make. When should workers be paid by the hour and when should they be paid by the piece?

Piece rates increase the incentive to work hard and can work well when output is easy to measure so “what you pay for” is close to “what you want.” Piece rates are common in agricultural work because it’s easy to measure the number of apples picked and this is close to what the employer wants. Even in agricultural work, however, the employer wants not just apples but ripe and unbruised apples so piece rates usually require some form of quality control. Piece rates do not work well when quality is important but quality control is expensive.

In the early days of computing, IBM paid its programmers per line of code. Can you see the problem? When IBM paid by the line, IBM programmers produced lots of code, but in their rush to earn more money, the programmers often wrote low-quality code. IBM’s incentive scheme rewarded what was measurable—lines of code—at the expense of what IBM really wanted but what was difficult to measure, high-quality code. IBM quickly stopped paying its workers by the line and switched to hourly wages. Hourly wages reduced the incentive to work hard but also reduced the incentive to rush the work before it was ready.

The advantage of piece rates is that, if used properly, they can greatly increase productivity. The auto-glass installer Safelite Glass Corporation switched from an hourly wage system to a piece rate. Safelite was able to handle the quality control issue by linking every job with a worker so that if a quality problem arose, the worker who was responsible for that windshield installation had to fix it on his or her own time. Productivity quickly improved by an astonishing 44%.6 About half of the increase in productivity was due to the same workers working harder, including lower absenteeism and fewer sick days, but the other half of the productivity increase was due to another important effect of piece rates. A piece rate system attracts more productive workers.

413

Consider two firms, one of which pays workers by the piece and the other pays workers by the hour. Now consider two workers, one of whom can install five windshields a day, the other just three. Which worker will be attracted to which firm? The piece rate firm will attract the more productive worker because piece rates give productive workers a chance to earn more money. The hourly wage plan will attract workers who are relatively less productive or even “lazy.”

The differences between workers in productivity can be surprisingly large. One California wine grower switched from paying grape pickers by the hour to paying by the pound. Previously, the firm had paid its workers $6.20 per hour. Under piece rates the average pay was effectively $6.84 per hour, about the same as before, but some workers were making as much as $24.85 an hour.

When some workers are more productive than other workers, piece rates will tend to increase inequality in earnings. Under the hourly wage, every grape picker earned $6.20 an hour. Under the piece rate, some earned $6.84, while others earned $24.85. Information technology is making it easier to measure the output of all kinds of workers, not just grape pickers. As a result, performance pay (piece rates, commissions, bonuses, and other rewards tied directly to output) is becoming more common in the U.S. economy and this is one reason why the inequality of earnings has increased.7

The increase in effective pay under piece rates explains why both firms and employees can benefit from piece rates. Under hourly wages, workers don’t have an incentive to work harder even when they can do so at low cost. Piece rates benefit productive workers by giving them an opportunity to use their skills to make more money. Piece rates also benefit firms by increasing productivity more than wages.

Even though firms and workers can both benefit from piece rates, piece rates are sometimes not implemented because of distrust. Workers fear that if they respond to a new piece rate plan by increasing productivity (and thus wages), the firm will respond by reducing the piece rate in the next period (e.g., paying less per pound of grapes picked). In the old Soviet Union, factory managers who increased productivity in response to new incentives were often denounced because their increased performance proved that they had previously been lazy! Of course, this greatly reduced the incentive to increase productivity. Similarly, workers won’t work harder if they expect that higher productivity will be punished with lower piece rates. Firms that want to introduce piece rates must build trust with their workers.