Incentives and Behavior Prof. Dr. Heiner Schumacher KU Leuven 12. - - PowerPoint PPT Presentation

Incentives and Behavior

• Prof. Dr. Heiner Schumacher

KU Leuven

• 12. Libertarian Paternalism
• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 12. Libertarian Paternalism

1 / 32

Introduction

Additional Literature Thaler, Richard, and Cass Sunstein (2008): Nudge - Improving Decisions About Health, Wealth and Happiness, Penguin Books, London, England.

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 12. Libertarian Paternalism

2 / 32

Introduction

Paternalism is de…ned as measures taken by the state, which limit some person’s liberty or autonomy for their own good. In the last chapter we saw that a certain degree of paternalism may well be warranted when it comes to …nancial decision making (e.g. force people to save for retirement). However, economists (in particular, the famous Chicago school) usually oppose state paternalism. Individuals are assumed to know their own good better than the state does. In this chapter, we will see that there is a third way, “libertarian paternalism”, that may causes people to make better decisions (like paternalism), but without limiting their freedom (hence, “libertarian”).

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 12. Libertarian Paternalism

3 / 32

Introduction

As a motivating example, consider the problem facing the director of a company cafeteria who discovers that the order in which food is arranged in‡uences the choices people make (recall “framing”). She has three alternative strategies: (1) she could make choices that she thinks would make the customers best o¤; (2) she could make choices at random; (3) she could make choices that she thinks would make the customers as obese as possible. She has to make one choice. So what is the optimal one? Option 1 seems paternalistic, but the other options clearly make no sense. Option 1 does not a¤ect the customers’ freedom, it just makes them to behave in a more healthy way. This is the idea behind libertarian paternalism.

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 12. Libertarian Paternalism

4 / 32

Introduction

Overview Defaults Framing and Information Provision Social Norms Product Design Critique

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 12. Libertarian Paternalism

5 / 32

Defaults

People often have to make important decisions while having little information on the subject. This makes the default action extremely important, i.e., the option that is implemented if an individual does not make an active choice. Consider a …rm that o¤ers a savings plan where employees are automatically enrolled unless they explicitly opt out; and a …rm that

• ¤ers a savings plan without automatic enrollment (the default action

is not to join). Madrian and Shea (2001) found out that automatic enrollment can increase participation rates from 49% to 86%.1 Another important application of defaults are organ donations.

1Madrian, Brigitte, and Dennis Shea (2001): “The Power of Suggestion: Inertia in

401(k) Participation and Savings Behavior,” Quarterly Journal of Economics 116 (4), 1149 - 1187.

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 12. Libertarian Paternalism

6 / 32

Defaults

Demand for organs greatly exceeds supply. By the end of 2008 there were more than 97000 persons on the waiting lists for organs (mostly kidneys) in the United States. Most of them (around 60%) will die because there are not su¢ciently many donors. The major obstacle to increasing donations is the need to get the consent of the family members of the (potential) donor. We will see that the share of donors depends on how consent is elicited.

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 12. Libertarian Paternalism

7 / 32

Defaults

Explicit Consent. Concrete steps are necessary to register as an

• rgan donor.

Routine Removal. The state owns the rights to body parts of people who are dead and can remove their organs without asking anyone’s permission (this rule is not used by any state). Presumed Consent. All citizens would be presumed to be consenting donors, but they have the opportunity to register their unwillingness to donate. The following graph displays the performance of these rules in a number of countries. Interpret the results!

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 12. Libertarian Paternalism

8 / 32

Defaults

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 12. Libertarian Paternalism

9 / 32

Defaults

Johnson and Goldstein (2003) analyzed this issue in an online survey.2 They asked people whether they would be willing to be donors. There were three conditions: explicit consent (one click to become a donor), presumed consent (one click to become a non-donor) and mandated choice (one click to become either a donor or a non-donor). So essentially, only the default action is varied between the conditions. The next graph shows the experimental outcome. Again, interpret the results!

2Johnson, Eric, and Daniel Goldstein (2003): “Do Defaults Save Lives?,” Science

302(5649), 1338 - 1339.

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 12. Libertarian Paternalism

10 / 32

Defaults

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 12. Libertarian Paternalism

11 / 32

Defaults

Presumed consent is not without problems. First, it is often overruled by family members if they are asked by doctors whether organs can be used or not. Second, it is hard to sell politically (one cannot just presume anything he wants). Hence, the best option in this case is mandated choice (e.g., a requirement that you check a box stating your organ donation preferences).

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 12. Libertarian Paternalism

12 / 32

Framing and Information Provision

One of the most cost-e¤ective ways to change behavior is to provide easy-to-understand information on complex issues. A good example is fuel e¢ciency of cars. In Europe, customers compare fuel e¢ciency using the number “liters per kilometer”. In the US, customers compare “miles per gallon” (MPG). These two number essentially measure the same thing. However, note that the relationship between gas used and MPG is

• curvilinear. Consequently, the two number may produce di¤erent

behavioral patterns.

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 12. Libertarian Paternalism

13 / 32

Framing and Information Provision

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 12. Libertarian Paternalism

14 / 32

Framing and Information Provision

Consider the following experiment by Larrick and Soll (2008).3 In an online survey, they presented subjects the following scenario: “A town maintains a ‡eet of vehicles for town employee use. It has two types of vehicles. Type A gets 15 MPG. Type B gets 34 MPG. Each car in the ‡eet is driven 10,000 miles. There are two plans for replacing the original cars with corresponding hybrid models. The goal is to reduce the gas consumption of the ‡eet. Option 1 is to replace the 100 vehicles that get 15 MPG with vehicles that get 19

• MPG. Option 2 is to replace the 100 vehicles that get 34 MPG with

vehicles that get 44 MPG.” In a control treatment, the scenario was the same but additionally all numbers were presented as GPM (gallons-per-100 miles), which is easier to understand. What options did subjects choose?

3Larrick, Richard, and Jack B. Soll (2008): “The MPG Illusion,” Science 320(5883),

1593 - 1594.

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 12. Libertarian Paternalism

15 / 32

Framing and Information Provision

In the MPG-treatment, 75% of subjects chose Option 2 (which of course is wrong). In the GPM-treatment, 64% of subjects chose Option 1. So the percentage choosing the more fuel-e¢cient option increases from 25% to 64%! The lesson here is that even the way in which numbers are presented may make a huge di¤erence!

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 12. Libertarian Paternalism

16 / 32

Social Norms

When uncertain what to do, many individuals follow the behavior of

• thers (“social norms”).

Making this behavior salient can change behavior. Allcott (2011) conducts a …eld experiment with customers of a utility company in the USA.4 They received letters in which their energy use was compared with that of their neighbors. The letters also contained energy conservation tips. This intervention resultet in energy savings of 2 percent.

4Allcott, Hunt (2011): “Social Norms and Energy Conservation,” Journal of Public

Economics 95(9–10), 1082–1095.

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 12. Libertarian Paternalism

17 / 32

Social Norms

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 12. Libertarian Paternalism

18 / 32

Product Design

In many countries, people with low incomes save too little for retirement (e.g., due to self-control problems or low numeracy skills). Thaler and Benartzi (2004) used the lessons from behavioral economics to develop a …nancial product (called “Save More Tomorrow”, SMarT) that helps people to increase savings.5 This product was available to blue-collar workers in various companies that o¤er de…ned-contribution plans. Since participation is voluntary, it is an example for “libertarian paternalism.” We discuss from which principles the product was developed and how successful it was.

5Thaler, Richard, and Shlomo Benartzi (2004): “Save More Tomorrow: Using

Behavioral Economics to Increase Employee Saving,” Journal of Political Economy 112(S1), S164-S187.

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 12. Libertarian Paternalism

19 / 32

Product Design

What do we have to take into account if we develop a product that is supposed to help people saving more for retirement? Bounded rationality (low numeracy skills) Self-control problems Loss aversion Procrastination (inertia)

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 12. Libertarian Paternalism

20 / 32

Product Design

Procrastination People who procrastinate constantly delay important decisions. Hence, they never take an action and the status quo remains (“status quo bias”). By using defaults, we can implement a certain enrollment rate at a savings plan. However, due to procrastination, this may actually lower the savings rate (once people join a plan, they do not change it anymore and stay with the same savings rate).

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 12. Libertarian Paternalism

21 / 32

Product Design

Loss Aversion It is therefore painful for an individual to cut spending once she is accustomed to certain spending level. The current reduction in consumption is perceived as a loss, the future increase in consumption is perceived as a gain. Hence, households may be reluctant to increase their contributions to the savings plan. Pay increases may provide the right time to try to get employees to save more.

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 12. Libertarian Paternalism

22 / 32

Product Design

The SMarT Program The savings plan developed by Thaler and Benartzi has the following features. Employees are approached about increasing their contribution rates a considerable time before their scheduled pay increase. If employees join, their contribution to the plan is increased beginning with the …rst paycheck after a raise. The contribution rate continues to increase on each scheduled raise until the contribution rate reaches a preset maximum. The employee can opt out the plan at any time.

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 12. Libertarian Paternalism

23 / 32

Product Design

Thaler and Benartzi o¤ered the SMarT plan at a number of companies, starting in 1998. We will have a look at the results of the …rst implementation, a midsize manufacturing company with very low participation rates as well as low saving rates. In addition to SMarT, the company hired an investment consultant and o¤ered his services to every employee.

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 12. Libertarian Paternalism

24 / 32

Product Design

Some more information on what the …nancial consultant actually did. On the basis of information that the employee provided, the consultant used commercial software to compute a desired saving rate (think of the life cycle model). He then made a recommendation to the employee about by how much he/she should increase contributions to the current saving plan (usually around 5%). Only 28% were willing to accept the consultant’s advice. For the rest of the employees, the SMarT plan was o¤ered.

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 12. Libertarian Paternalism

25 / 32

Product Design

78% of those employees who were unwilling to accept the consultant’s advice agreed to join the SMarT plan. 80% of those who joined the SMarT plan remained in the plan through four pay raises. Those who dropped out did not reduce their contributions, but remained with the actual level. The following table summarizes the consultant’s and SMarT’s success in increasing saving rates.

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 12. Libertarian Paternalism

26 / 32

Product Design

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 12. Libertarian Paternalism

27 / 32

Product Design

Those who joined the SMarT plan, increased savings from 3,5% to 13,6%. Notably, these were those people who refused the immediate increase in savings suggested by the …nancial consultant, i.e., those people for whom saving is rather di¢cult. Overall, saving rates in the company increased from 4,4% to 10,6%.

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 12. Libertarian Paternalism

28 / 32

Critique

The Slippery Slope “Once we accept modest paternalism for savings or cafeteria lines or environmental protection, highly intrusive interventions will surely follow.” Libertarian paternalism requires low-cost, opt-out rights (freedom of choice). This distinguishes it from real paternalism. Some kind of in‡uence is inevitable, because one has to present choices in some order (recall the example from the start). Staying neutral sometimes is impossible.

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 12. Libertarian Paternalism

29 / 32

Critique

Bad In‡uence “Decision makers in the public sector may use the tools from libertarian paternalism for their own interests.” The very same is obviously true for the private sector. However, the interests of decision makers in the public sector are more in line the

• nes of their “customers” (voters) than it is the case for private

sector decision makers (shareholders). Freedom of choice guarantees some protection against bad in‡uence.

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 12. Libertarian Paternalism

30 / 32

Critique

Redistribution “The measures of libertarian paternalism require some funding and therefore constitute a redistribution from the sophisticated ones (who need no help) to the unsophisticated ones.” The measures described so far only impose minimal costs on society. The sophisticated ones are protected by freedom of choice. If people who need help impose costs on society, then measures taken to help these people also increase welfare for society.

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 12. Libertarian Paternalism

31 / 32

Critique

Neutrality “Why should public policy makers know better than the people what is the right thing to do?” Sometimes it is quite clear that preferences are very di¤erent from what people actually do (recall organ donations). Economic theory and psychology can inform about optimal behaviors (recall saving behavior). Policy can make good guesses if it consults experts who know much more about a problem than the average decision maker. This is likely to be the case if people have few opportunities to learn.

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 12. Libertarian Paternalism

32 / 32