Hyperbolic Discounting How we misread the future Reading

Hyperbolic Discounting How we misread the future

Reading • Predictably Irrational, Chapter 6, The Influence of Arousal • Predictably Irrational, Chapter 7, The Problem of Procrastination and Self-Control • Nudge, Chapter 2, Resisting Temptation

Discounting • Being happy now (today, this week, this year) is typically more desirable than the prospect of being happy in the future (tomorrow, next week, next year) • Suppose happiness during a time period (day, week, year) can be measured • The discount rate is the additional future happiness that can compensate for the loss of one unit of happiness in the present

Discounting • Example: – For Alice, suppose the loss of one unit of present happiness can be compensated by the gain of 1. 07 units of future happiness – In this case, Alice’s discount rate is 0. 07 – For Bob, suppose the loss of one unit of present happiness can be compensated by the gain of 1. 02 units of future happiness – Bob’s discount rate is 0. 02 • Who is more patient, Alice or Bob?

Discounting • The higher your discount rate, the more important the present is to you (relative to the future) • The higher your discount rate, the more impatient you are

Dynamic Inconsistency • Typically, when we are making plans about what we’d do in the future, we are patient – our discount rates are low • But when the future finally arrives, we become impatient and succumb to temptation – our discount rates spike • This phenomenon is called hyperbolic discounting or dynamic inconsistency

Dynamic Inconsistency • When we think about what we’d do in the future, we tend to make well thought out plans • But when the future finally arrives, we give in to temptation and abandon the plans we had made

Does the future me want different things? Choose among 24 movie videos • Some are “low brow” • Some are “high brow” Does my choice depend on whether I am picking for tonight, next Thursday, or the following Thursday? Reed, Lowenstein & Kalyanaraman (1999) Mixing virtue and vice: Combining the immediacy effect and the diversification heuristic. Journal of Behavioral Decision Making, 12, 257 -273.

Does the future me want different things? Did people choose… a) More “low brow” movies now, more “high brow” movies for later b) More “high brow” movies now, more “low brow” movies for later c) About the same regardless of whether picking for now or later Reed, Lowenstein & Kalyanaraman (1999) Mixing virtue and vice: Combining the immediacy effect and the diversification heuristic. Journal of Behavioral Decision Making, 12, 257 -273.

Next week I will want things that are good for me… High Brow ; 44% Low Brow ; 56% Choosing for tonight High Brow; 63% Low Brow; 29% Low Brow; 37% Choosing for next Thursday High Brow; 71% Choosing for second Thursday Reed, Lowenstein & Kalyanaraman (1999) Mixing virtue and vice: Combining the immediacy effect and the diversification heuristic. Journal of Behavioral Decision Making, 12, 257 -273.

Does the future me want different things? When choosing between a healthy and unhealthy snack for delivery in one week, do people systematically misproject future desires? ← Next Week→ Predicted preference ← Right Now→ Actual preference D. Read (Leeds U. ) & B. van Leeuwen (Leeds U. ), 1998, Predicting hunger: The effects of appetite and delay on choice. Organizational behavior and human decision processes, 76, 189 -205.

Does the future me want different things? Are people a) More likely to choose the unhealthy snack for next week b) More likely to choose the unhealthy snack for right now c) About as likely either way ← Next Week→ Predicted preference ← Right Now→ Actual preference D. Read (Leeds U. ) & B. van Leeuwen (Leeds U. ), 1998, Predicting hunger: The effects of appetite and delay on choice. Organizational behavior and human decision processes, 76, 189 -205.

Does the future me want different things? 26% chose the unhealthy snack for delivery in one week right after lunch. 74% ← Next Week→ Predicted preference 26% D. Read (Leeds U. ) & B. van Leeuwen (Leeds U. ), 1998, Predicting hunger: The effects of appetite and delay on choice. Organizational behavior and human decision processes, 76, 189 -205.

Does the future me want different things? One week later, when allowed to change their choice at the delivery, 70% chose the unhealthy snack for immediate consumption 74% ← Next Week→ 26% 30% ← Right Now→ 70% Predicted preference Actual preference D. Read (Leeds U. ) & B. van Leeuwen (Leeds U. ), 1998, Predicting hunger: The effects of appetite and delay on choice. Organizational behavior and human decision processes, 76, 189 -205.

Hyperbolic discounting This is an example of under-rating the intensity of future desires

Hyperbolic Discounting, Measured • Evidence suggests that discounting is steeper in the immediate future than in the further future. • In one study, the median subject is indifferent between – $15 now and $20 in one month (for an annual discount rate of 345 percent) and between – $15 now and $100 in ten years (for an annual discount rate of 19 percent).

Hyperbolic Discounting • When evaluating outcomes in the distant future, individuals are patient and make plans to exercise, stop smoking, and look for a better job. • As the future gets near, the discounting gets steep, and the individuals engage in binge eating, light another (last) cigarette, and stay put on their job. • Preferences with these features therefore induce dynamic inconsistency.

Gym Memberships • In one study, three U. S. health clubs offered a choice between – a monthly contract with lump sum fee of approximately $80 per month and no payment per visit, and – a pay-per-visit contract with fee of $10. • Health club users that choose the monthly contract attend only 4. 4 times per month. • These users pay $17 per visit even though they could pay $10 per visit

Homework and Deadlines • The subjects are fifty-one professionals enrolled in a section of a semester-long executive education class at Sloan (MIT), with three homeworks as a requirement. • At the beginning of the semester, they set binding deadlines (with a cost of lower grades for delay) for each of the homeworks.

Homework and Deadlines • According to the standard model, they should set deadlines for the last day of the semester: – There is no benefit to setting early deadlines, since they do not receive feedback on the homeworks, and there is a cost of lower flexibility. – A maximization without constraints is always preferable to one with constraints.

Homework and Deadlines • According to a model of self-control, instead, the deadlines provide a useful commitment device. • Since homework completion is an investment good, individuals spend less time on it than they wish to ex ante. • A deadline forces the future self to spend more time on the assignment. • The results support the self-control model: 68 percent of the deadlines are set for weeks prior to the last week, indicating a demand for commitment.

Homework and Deadlines • This result leaves open two issues. – First, do the self-set deadlines improve performance relative to a setting with no deadlines? – Second, is the deadline setting optimal? • If the individuals are partially naive about the self-control, they will underestimate the demand for commitment.

Homework and Deadlines • In a second (laboratory) experiment, sixty students complete three proofreading assignments within twenty-one days. • The control group can turn in each assignment at any time within the twenty-one days • A first treatment group can choose three deadlines (as in the class-room setting described above), and • A second treatment group faces equal-spaced deadlines.

Homework and Deadlines • The first result is that self-set deadlines indeed improve performance: – The first treatment group does significantly better than the control group, detecting 50 percent more errors (on average, 105 versus 70) and earning substantially more as a result (on average, $13 versus $5).

Homework and Deadlines • The second result is that the deadline setting is not optimal: – the group with equal-spaced deadlines does significantly better than the other groups, on average detecting 130 errors and earning $20. • This provides evidence of partial naiveté about the self-control problems.

Credit Card Offers • Borrowers were offered two types of credit card contracts: – Low teaser interest rate for first six months, followed by high post-teaser interest rate after the first six months – Higher teaser rate, followed by lower post-teaser rate • Borrowers went for the first contract even though, given their borrowing behavior, they would have been better off with the second

In the future, I expect to prefer the long-term, rational choice. Later v. Now When the future becomes “right now, ” I prefer the pleasurable choice! [hyperbolic discounting]

In the future, I expect to prefer the long-term, rational choice and quit smoking. When the future becomes “right now, ” I keep smoking. Hyperbolic discounting with smoking?

Self-predicted future behavior in teenage smoking 15% of light smokers (less than one cigarette per day) in high school predicted they “might” be smoking in 5 years. 15% predicted 5 years later, what percentage were still smoking? a) Less than 5% b) 5% to 15% c) About 15% d) 15% to 30% e) Greater than 30% L. Johnston (U. Michigan), P. O’Malley (U. Michigan), J. Bachman (U. Michigan), 1993, Illicit drug use, smoking, and drinking by America’s high school students, college students, and young adults, 1975 -1987. National Institute on Drug Abuse: Rockville, Maryland.

Self-predicted future behavior in teenage smoking 15% of light smokers (less than one cigarette per day) in high school predicted they “might” be smoking in 5 years. 15% predicted 5 years later, 43% were still smoking. 43% actual L. Johnston (U. Michigan), P. O’Malley (U. Michigan), J. Bachman (U. Michigan), 1993, Illicit drug use, smoking, and drinking by America’s high school students, college students, and young adults, 1975 -1987. National Institute on Drug Abuse: Rockville, Maryland.

Emotions • Hyperbolic discounting may be more likely when people find themselves in an emotionally aroused state

Economic Consequences of Hyperbolic Discounting • Excessive spending, inadequate saving (especially for retirement), bankruptcy, and debt default • Inadequate personal investment in education • National neglect of investment on infrastructure, public education, scientific research • Excessive national spending on wars and other trivial issues that have immediate appeal

PROJECTION BIAS

Projection Bias • Hyperbolic discounting is not the only kind of systematically incorrect prediction that we are prone to • There’s also projection bias

Projection Bias: our current state (hot v. cold) influences projections of our future desires. Cold v. Hot

Does the current state of hunger change which snack people will order for delivery next week? a) Yes, hungry people choose the unhealthy snack b) Yes, hungry people choose the healthy snack c) No

Projection of future preferences depends on our current state People asked right after lunch (not hungry), chose the unhealthy snack for delivery in one week right after lunch 42% of the time. People asked four hours after lunch (hungry), chose the unhealthy snack for delivery in one week right after lunch 78% of the time I’m not hungry right now, so next week I will prefer the healthy snack. I’m hungry, so next week I will prefer the candy bar. Cold v. Hot

Projection bias Cold v. Hot Cold state projects to future cold state I’m not hungry right now, so next week I will prefer the healthy snack just like I do right now. Hot state projects to future hot state I’m hungry, so next week I will prefer the unhealthy choice just like I do right now.

Return rates of cold weather items • Conlin, O’Donoghue, and Vogelsang (2007) find that a reduction in the order date temperature of 30°F—corresponding to a decrease, for example, from 40°F to 10°F— increases the average return rate of a coldweather item by 3. 96 percent, consistent with projection bias.

An experiment with heroin addicts Heroin addicts in treatment receive a replacement drug. It produces a mild high designed to ward off heroin cravings. A double-dose produces a longer, more intense high. Choice between money or a double-dose, either to be received in five days. G. Badger (U. Vermont), W. Bickel (U. Arkansas), L. Giordano (Duke), E. Jacobs (S. Illinois U. ), G. Loewenstein (Carnegie Mellon), L. Marsch (St. Luke’s Hospital, NY), 2007, Altered states: The impact of immediate craving on the valuation of current and future opiods. The Journal of Health Economics, 26, 865 -876.

Projection bias predicts that heroin addicts… a) Would forego more money for the promise of an extra dose in 5 days if they were currently craving. b) Would forego less money for the promise of an extra dose in 5 days if they were currently craving. c) Would forego more money for an immediate extra dose than for one in 5 days. d) The state of craving would make no impact because they don’t get the extra dose for 5 more days.

$70 $60 $50 Amount of future money willing to give up for double-dose in 5 $40 days $30 $20 $10 When in a HOT state, I act as if I will always be in a HOT state. After regular dose Before regular dose $0 When in a COLD state, I act as if I will always be in a COLD state.

I need a hit now, so next week I will prefer an extra dose to $50. rr te Af Be fo re re gu la eg r do ul ar se do se $70 $60 $50 Amount of future money willing to give up for$40 $30 double-dose in 5 days $20 $10 $0 I don’t need drugs now, so next week I will prefer $50 to an extra dose.