What Endowment Effect A Public Good Experiment

1
What Endowment Effect?A Public Good Experiment

• Edward J. López
• Assistant Professor
• Department of Economics
• University of North Texas
• elopez_at_unt.edu

• William Robert Nelson, Jr.
• Assistant Professor
• School of Management
• University at Buffalo
• wrnelson_at_buffalo.edu

2
The Endowment Effect

• Individuals with loss aversion or status quo bias
  may value a resource more highly once the
  property right is established.
• Seminal cite Thaler 1980 JEBO

3
Evidence of the Endowment Effect

• Traditionally tested using CVM or lab/market
  auctions, showing WTA WTP
• Knetsch 1989 AER
• Kahnemann, Knetsch and Thaler 1991 JEP

4
Endowment Effect as Anomaly

• Modeled as some manipulation (pivot/kink/rotation)
  of standard indifference curves at point of
  initial endowment
• Morrison 1997 AER
• Implies possibly intersecting indifference curves
• Kahnemann, Knetsch and Thaler 1991 JEP

5
Value Disparities

• Commonly observed
• Smaller for more substitutable goods
• Hanemann 1991, 2003 AER (theory)
• Shogren et al. 1994 AER (evidence)
• Diminish with market discipline and market
  experience
• Shogren et al. 1994, 1997 AER (lab auction
  evidence)
• List 2003 QJE (field auction evidence)

6
Endowment vs. Substitution

• Controversy has recently centered on what
  explains observed value disparities
• Endowment and Substitution alternative
  explanations
• not necessarily mutually exclusive
• not necessarily mutually inclusive

7
Summarizing the Literature

• Some degree of value disparity is commonly
  observed
• Endowment and Substitution are alternative
  explanations
• Market experience and discipline tend to reduce
  value disparities
• EE is some pivot/kink/rotation of indifference
  curves at point of endowment

8
Our contribution

• The question of whether value disparities are
  preference-based (i.e., fundamental) remains.
• Are preferences anomalous yet constrained in the
  usual manner by market institutions?
• There may be value in testing for the presence of
  an endowment effect in a non-market situation
  that controls for the substitution effect as a
  viable alternative explanation.

9
Use a Public Good Game (VCM)

• Our design allows us simultaneously to
• Eliminate market discipline (uses public good)
• Eliminate market experience (one shot)
• Control for substitution effects (cash based)
• Observe treatments that one would expect to
  elicit the endowment effect (account and duration
  framing)

10
The Experiment

• 2 person public good game (VCM)
• MPCR0.75
• Each participant given 10 endowment
• Each participant makes one decision
• Number of dollars to contribute to public account
• Two types of treatment variation

11
Treatment Variation 1

• Account Framing (AF)
• Private Group told money starts in private
  account
• Property right is established
• Resembles WTA decision
• Public Group told money starts in public account
• Initial endowment is shared
• Resembles WTP decision

12
Treatment Variation 2

• Duration Framing
• Long Group holds money for some time before
  making contribution decision
• Short Group makes decision immediately
• Laboratory treatments
• Short holds cash for 1 minute
• Long holds cash for 25 minutes
• Classroom treatments
• Short-hold cash for 1 minute
• Long-hold cash for 1 week

13
Why an Effect in AF?

• Originating account is initial property right
• Focal point for prospective reciprocators to
  reach fairness equilibrium
• Shelling 1960 Rabin 1993 AER
• Warm glow and cold prickle might differ in
  magnitude
• Andreoni 1995 QJE

14
Why an Effect in DF?

• Temporal componentEE may take time to bind
• Knetch and Sinden 1984 QJE
• Kahneman, Knetch and Thaler 1991 JEP
• More readily part with windfall gains than earned
  wealth
• Thaler and Johnson 1991
• PIH
• Friedman 1957

15
Hypotheses

• H1 AF imparts an endowment effect
• Private group contributions less than Public
  group
• money will tend to stay where it is said to start
• H2 More time holding the endowment increases
  strength of endowment effect
• Long group contributions less than Short group

16
Results Overview

• 284 undergraduate participants
• From U. at Buffalo and U. of North Texas
• No difference in responses that affect tests
• The mean contributions
• classroom treatments 4.72
• laboratory treatments 4.61

17
Treatments and Hypotheses
18
H1 is rejected

• Public Group deposited about 1 more on average
  than Private Group
• two tailed p0.07
• Opposite of what endowment effect suggests
• Consistent with Andreonis (1995) result that the
  warm glow of giving provides a greater influence
  than the cold prickle of taking

19
H2 receives no support

• Long and Short groups contribute no differently
• Laboratory Results
• Long mean 5.20
• Short mean 5.14 (two tailed p0.95)
• Classroom Results
• Long mean 5.06
• Short mean 5.40 (two tailed p0.68)

20
Extra Result

• DF treatments physically held cash
• AF treatments only wrote down their decisions
• No evidence that AF decision differed from DF
• DF were 0.47 higher (4.96 vs. 4.50
• two tailed p0.22
• Not a clean test because lacks control

21
Conclusion

• The one-shot, cash-denominated VCM creates
  favorable circumstances for the endowment effect
  to emerge
• eliminates market discipline
• eliminates experience
• holds the substitution effect constant

22
Conclusion (cont.)

• Endowment effect is not apparent in our data
• In AF treatments, significant treatment effect in
  the opposite direction that the endowment effect
  would imply
• In DF treatments, no significant difference
  between treatments is observed

23
Bonus Track

• Supports standard lab experiment methods
• Participants treat new endowments as money they
  have held
• Not playing with other peoples money

24
Acknowledgements

• Funding
• Russell Sage Foundation Behavioral Roundtable
• NT Office of Research Services
• Comments
• Dan Houser, John List, Charles Plott, Vernon
  Smith, Bart Wilson
• Classroom time
• K. Battaglia, M. Strazicich, D. Molina
• Experiment help
• Todd Jewell, Ife Isiekew, Sangkyoo Kang, JaeHoon
  Kim, Kaunyoung Lee, Nathan Roseberry, Diego
  Segatore, Shanhong Wo, Oksana Zhuk