Chapter 20: The Problem of Moral Hazzard Ordering Information: Betty Jung Marketing Specialist, Finance/Economics/Decision Sciences South-Western | Cengage Learning 5191 Natorp Boulevard, Mason, OH 45040 The ISBN for your 2e book alone is:  1439077983   The Bundle ISBN for your 2e book + the printed access card for MBA Primer is: 0538771240 Managerial Economics: A Problem Solving Appraoch (2nd Edition) Luke M. Froeb, luke.froeb@owen.vanderbilt.edu Brian T. McCann, brian.mccann@owen.vanderbilt.edu Website, managerialecon.com COPYRIGHT © 2008; Slides prepared by Lily Alberts for Professor Froeb Thomson South-Western, a part of The Thomson Corporation. Thomson, the Star logo, and South-Western are trademarks used herein under license. Slides prepared by Lily Alberts for Professor Froeb

Summary of main points Moral hazard refers to the reduced incentive to exercise care once you purchase insurance. Moral hazard occurs in a variety of circumstances: Anticipate it, and (if you can) figure out how to consummate the implied wealth-creating transaction (i.e., ensuring that consumers continue to take care when the benefits of doing so exceed the costs). Moral hazard can look very similar to adverse selection—both arise from information asymmetry. Adverse selection arises from hidden information about the type of individual you’re dealing with; moral hazard arises from hidden actions.

Summary of main points (cont.) Solutions to the problem of moral hazard center on efforts to eliminate the information asymmetry (e.g., by monitoring or by changing the incentives of individuals). Shirking is a form of moral hazard. Moral hazard in loans: Borrowers prefer riskier investments because they get more of the upside while the lender bears more of the downside. The problem is worse for borrowers who have nothing to lose.

Introductory anecdote: TripSense In 2004, the Progressive Direct Group of Insurance Companies introduced TripSense – a service that provided a free device to record mileage, speeds and times driven in a vehicle. Progressive then used this information to offer discounted renewal policies to customers who drove fewer miles at slower speeds during non-peak hours. This helps the insurance company solve two problems. Adverse selection, from the last chapter, and moral hazard, the focus of this chapter. The decision of how frequently, how far, or how fast to drive is equivalent to choosing your probability of having an accident. The cost of having an accident goes down when you buy insurance. Drivers respond to this reduced cost by “choosing” to have more accidents. This is called “moral hazard.”

Introduction: Moral hazard Once you have insurance, the cost of an accident is reduced, which also reduces the cost of the risky behavior. This is the problem of “moral hazard” and exists in many contexts, not just in the market for insurance. Moral hazard and adverse selection are closely related problems. Both, are caused by information asymmetry: moral hazard results from hidden actions; while adverse selection results from hidden information The cost of managing both problems can be reduced by reducing uncertainty (gathering more information).

Moral Hazard in Insurance To illustrate moral hazard, lets return to the bicycle insurance example of chapter 19. Suppose that bike owners stand a 40% chance of theft when parking their bike on the street overnight. However, if the bike owner exercises care (locks the bike), the chance of theft is reduced to 30%. Suppose the cost of taking care (buying a lock) is $5. For uninsured bike owners, the benefit of exercising care is (0.40 - 0.30)($100) = $10 and is greater than the costs of exercising care, $5. Moral hazard suggests that once customers purchase insurance, they exercise less care because there is less incentive to do so. Is this really the case? Benefit of care: change in probability multiplied times the cost of the bike

Moral Hazard in Insurance (cont.) In our example, The cost of bike theft is reduced when an insurance policy is purchased. Sp, the consumer stops taking the extra time to lock up the bicycle every night once she buys insurance. The probability of theft then increases from thirty back to forty percent. The insurance company anticipates this moral hazard, and now charges $45 for every policy it sells. If you do NOT anticipate that the probability of theft will increase from 30% to 40%, you will lose money on the insurance you sell. In other words, anticipate moral hazard and protect yourself against it. Discussion: benefits of exercising care are still higher than the costs; but consumers don’t exercise care because the benefits do not accrue to them

Creating wealth with moral hazard Moral hazard can also represent an unconsummated wealth-creating transaction. This opportunity exists because the benefits of taking care are bigger than the costs of taking care. But how can the insurance company induce the bike owner to take care? If the insurance company could observe whether the customer was exercising care, then it could lower the price of insurance to those taking care. This is exactly what Progressive’s MyRate/TripSense system tries to do. It could also purchase the lock for the bike owner. Note that these kinds of prevention and wellness programs do NOT reduce health care costs.

Moral Hazard or Adverse Selection? To distinguish between moral hazard and adverse selection, ask whether: Information is hidden (adverse selection) or the action is hidden (moral hazard) The problem arises before a transaction (adverse selection) or after (moral hazard) Discussion: Give a moral hazard and an adverse selection explanation for each the following: Drivers with air bags are more likely to get into traffic accidents. Volvo drivers are more likely to run stop signs. At all-you-can-eat restaurants, customers eat more food. Air Bags: The adverse selection explanation is that bad drivers are more likely to purchase cars with airbags. If you know you are likely to get into an accident, it makes sense to purchase a car with airbags. The moral hazard explanation is that once drivers have the protection of airbags, they take more risks and get into more accidents. If you don’t believe that people change behavior in this way run a simple experiment – next time you drive somewhere, do not wear a seat belt. See if you drive more carefully. Volvo Drivers: Adverse selection – people who are more likely to run stop signs will want to buy safer cars, so they will be more likely to buy Volvos. Moral hazard – once you own a Volvo, you feel safer so you will be more likely to run a stop sign Restaurant: Adverse selection – attracts people who are more likely to be big eaters. Moral hazard – once you have paid for all you can eat, you are more likely to eat more.

Shirking as moral hazard Because it’s difficult to monitor an employee’s actions after they are hired, employers anticipate shirking. Problem: What commission rate is required to induce hard work? Suppose the benefit of working hard is the higher probability of making a sale, e.g., probability of a sale rises from 50% to 75%. The cost of working hard is $100 To induce hard work, (0.25) x (Commission) > $100, i.e., the commission must be bigger than $400 Unless the contribution margin on the item is at least $400, you can’t afford to pay a $400 commission. You make more money by letting the salesman shirk, i.e., it doesn’t pay to address the moral hazard problem. If there is no solution, there is no problem! Note also that you can use the law of large numbers to design a compensation scheme over a year, e.g., if you are expected to sell one unit/day, then it becomes easier to distinguish shirking from working hard because a hard worker would expect to sell 75 units/100 days and a shirker 50 units/100 days. Adopt incentive compensation only for those who sell more than 70 units. .75C - $100 > .5C; equivalently .75C-.5C > $100; equivalently .25C > $100; equivalently C > $400

Shirking as Moral Hazard (decision tree)

Shirking (cont.) Another potential solution is to try to get a better indicator of effort than sales. Suppose that by incurring costs of $50, you could observe whether the sale person was working hard. Would it be profitable to hire someone to monitor the salesperson’s behavior? Expected benefit of inducing hard work is the increased probability of making a sale (twenty-five percent) times the margin. If the item’s margin is at least $200, then it pays to monitor the worker. The company could also pay $50 more for a worker that has a reputation for working hard, whether or not she is being monitored. Remember: A reputation for working hard without monitoring is valuable to both companies and workers. Think of rewarding the salesperson for effort directly, with either a stick (work hard or get fired) or a carrot (work hard and earn a reward). If you have a performance metric like this, then almost any incentive compensation scheme will work. The new performance evaluation metric allows you to put the salesperson’s entire compensation or job at risk. If the benefits of keeping a job and earning a salary are bigger than the costs of exerting effort, the salesperson will exert effort. Our discussion of “framing” in Chapter 12 suggests that a carrot is likely to be a more effective motivator for the employees.

Shirking (cont.) Moral hazard injures both parties to a transaction. If firms anticipate moral hazard, they will be less willing to transact; or put a lower value on the transaction. Example: A consulting firm is paid on an hourly rate. Given the rate structure, and the inability of the client to monitor what the consultant is doing, the client expects the consultant to shirk by billing more hours than the client would prefer, or by working on projects that are valuable to the consultant but not the client. Are there solutions to this problem? Discussion: Clients anticipate this kind of shirking and are understandably reluctant to transact, unless the consulting firm can figure out how to address the problem. Both parties end up transacting less than they otherwise would if moral hazard did not exist. Solutions: The consultant can solve this moral hazard problem by establishing a good reputation, by accepting a portion of the contract on a fixed-fee basis, and by providing information to the client documenting the usefulness of the work that is done.

Moral Hazard in Lending Banks face a moral hazard in loans: borrowers who are least likely to repay loans are the most likely to apply for them. Example: a $30 investment opportunity arises. The investment has a 50% chance of a $100 payoff and a 50% chance of a $0 payoff The bank offers a $30 loan at 100% interest based on the expected value of the investment. If the investment pays off, then the bank gets $60, if the investment fails the bank gets $0. Bank’s expected payoff is $30 = (.5)($60) + (.5)($0) Borrower’s expected payoff is $20 = (.5)($40) + (.5)($0)

Moral hazard in lending (cont.) However, after the loan is made, the borrower “discovers” a different investment. This second investment pays off $1000 but only has a 5% probability of succeeding. Here the borrower receives more if the investment pays off, so the bank receives a smaller payoff, $ 3 = (.05)($60) + (.95)($0) The lender prefers the less risky investment because she receives a higher expected payoff. But, the borrower prefers the riskier investment.

Moral hazard in lending (cont.) Moral hazard is a problem for both the lender and the borrower in this situation. If the bank anticipates moral hazard they will be less willing to lend, or demand a higher interest rate. This incentive conflict is only made worse when the borrower can put other people’s money at risk. Borrowers take bigger risks with other people’s money than they would with their own. To control this, lenders must find ways to better align the incentives of borrowers with the goals of lenders. Banks sometimes do this by requiring borrowers to put some of their own money at risk. This is why banks are much more willing to lend to borrowers who put a great deal of their own money at risk, but it also leads to the complaint that banks lend money only to those who don’t need it. Real-world example In the early 1980s in Texas, a bust in the real estate market reduced the value of many Savings and Loan institutions (S&L’s). S&L’s are basically banks, who borrow from depositors and lend to homeowners. When the real estate market collapsed, many homeowners stopped repaying their loans and forfeited their real estate to the S&L. The value of the S&L’s assets (the real estate) fell below that of their liabilities (the money they owed depositors). With nothing to lose, these S&L’s borrowed even more money from depositors at very high interest rates and “bet” heavily on junk bonds – the most risky investment available to them. And just as in our loan example, this decreased the expected payoff to the lender. Since deposits were insured by the US government, US taxpayers were stuck with the $200 billion cost of repaying depositors.

Moral hazard in financial crisis Regulators try to reduce the costs of moral hazard by requiring banks to keep about 10% of their equity in case depositors want their money back. But when the value of assets fall by more than 10%, (as they did in 2008) banks become insolvent and the risk of moral hazard increases. In late 2008, the US treasury guaranteed short-term loans to help banks make riskier loans – if loans payoff, the bank profits; but if they fail, taxpayers cover the loss. A better solution may have been to simply give the banks more equity. The govt. would own equity and thus share in the upside gain, i.e., should the loans payoff. Companies that are “too big to fail,” such as AIG, take bigger risks because they know the government will bail them out. AIG example: One of AIG’s main business lines was selling credit default swaps. These swaps allowed banks that purchased them to reduce their risk profiles by transferring the risk to AIG. AIG, however, did not hedge any of their own swaps, which had a notional value of around half a trillion dollars at their peak (notional value refers to what AIG would have owed if all its bets went bad). A lot of these loans did go bad, and now the U.S. taxpayer is paying off AIG’s debts.

Moral hazard in 2008 (cont.) Bailing out homeowners also causes moral hazard. Foreclosure bailouts helps irresponsible homeowners who made risky investments that they couldn’t afford; responsible borrowers wouldn’t need the bailout assistance. The bailouts end up hurting responsible borrowers. Those who were less cautious are now getting to keep their risky investment while taxpayers (including those cautious borrowers) pay for the bailouts. AND, the new rules favoring borrowers increase the cost of making loans. So responsible borrowers who had no part in the real estate collapse pay the higher loan rates caused by new regulations; AND pay for the bailout.

Extra anecdote: Driver Tracking Regional phone company using GPS to track driver location Designed to deploy repairmen more efficiently. Used to investigate slow response time Led to surprising conclusions on source of problem (drivers having extra marital affairs) Example of moral hazard Similar to adverse selection (but post-contractual or “hidden action”) Caused by same information asymmetry

Managerial Economics - Table of contents 1. Introduction: What this book is about 2. The one lesson of business Benefits, costs and decisions 4. Extent (how much) decisions 5. Investment decisions: Look ahead and reason back 6. Simple pricing Economies of scale and scope 8. Understanding markets and industry changes 9. Relationships between industries: The forces moving us towards long-run equilibrium 10. Strategy, the quest to slow profit erosion 11. Using supply and demand: Trade, bubbles, market making 12. More realistic and complex pricing 13. Direct price discrimination 14. Indirect price discrimination 15. Strategic games 16. Bargaining 17. Making decisions with uncertainty 18. Auctions The problem of adverse selection The problem of moral hazard 21. Getting employees to work in the best interests of the firm 22. Getting divisions to work in the best interests of the firm 23. Managing vertical relationships 24. You be the consultant EPILOG: Can those who teach, do? Managerial Economics - Table of contents