Experimental and Behavioral Economics

Transcription

1 Experimental and Behavioral Economics Doing Good or Doing Well? Image Motivation and Monetary Incentives in Behaving Prosocially (By Dan Ariely, Anat Bracha, Stephan Meier) Anna Vahminceva Summer term

2 Agenda I. Motives to behave prosocially II. Behavioral Hypotheses III. Laboratory experiment Click for Charity IV. Field study Bike for Charity VI. Conclusions 2

3 Motives to behave prosocially Intrinsic motivation: value of giving per se (pure altruism) Extrinsic motivation: material reward or benefit Image motivation (signaling motivation): the desire to be liked and respected by others and by one s self 3

4 Motives to behave prosocially extrinsic motivations can have detrimental effects on prosocial behavior The addition of extrinsic incentives makes it more difficult to answer the question: Is the individual behaving prosocially to do good or to do well? 4

5 Behavioral Hypotheses Individual s image is a function of how prosocial or greedy other people think the individual is. A crucial property of image is its dependency on visibility Assume image value from a prosocial activity is positive => increasing number of observers increases the image value of prosocial behavior and leads to higher effort 5

6 Behavioral Hypotheses Image-Motivation Hypothesis: changing visibility changes the level of prosocial activity For a positive image, increasing visibility increases the level of prosocial activity Effectiveness Hypothesis: extrinsic rewards are less effective the greater is the visibility of the prosocial act detrimental effect of extrinsic incentives is more likely to occur for a visible prosocial effort than for a private one 6

7 Experiment Click for Charity Subjects could donate to a charitable organization by clicking two keys (X and Z) on a computer keyboard for up to five minutes. decreasing payment schedule: 1 cent for each of the first 200 pairs of X-Z presses 0.5 cents for each of the next 200 pairs 0.25 cents for each of the next 200 pairs,, 0.01 cents for each pair above 1,200 7

8 Experiment Click for Charity The visibility of the task: subjects are randomly assigned to make their effort choice in public or in private The payment scheme: participants could, in addition to the donation, earn money for themselves The nature of the cause: subjects are randomly assigned to one of two charities (American Red Cross or the NRA) Charities are classified as good (American Red Cross) or as bad (NRA) 8

9 Results of the Study Source: Ariely et al. (2009) Monetary incentives crowd out the image motivation to behave prosocially 9

10 Results of the Study Source: Ariely et al. (2009) 10

11 Results of the Study Analyzing good and bad causes together (OLS regression) shows that: in the private condition monetary incentives increase effort significantly in the public condition monetary incentives do not increase individual effort the nature of the cause is significantly more important in the public than in the private setting The results support both the Image-Motivation Hypothesis and the Effectiveness Hypothesis 11

12 Results of the Study May be the effect due to a physical limit? Limit-treatment subjects could earn the money for themselves only by pressing the keys flat rate of 0.11 cents per pair + bonus ($2 by pressing more than 1,000 pairs, $20 by pressing more than 1,500 pairs The average of key pressed is 1,290 pairs (900 pairs for the Red Cross in the public sphere) Ineffectiveness of monetary incentives in the public condition cannot be attributed to a physical limit. 12

13 Field Study Bike for Charity Cycling for up to ten minutes Donation rate of $1 per mile The same conditions as in Click for Charity Results private monetary incentives are more effective in the private than in the public condition extrinsic incentives increase effort in the private condition but not in the public condition 13

14 Conclusion Monetary incentives depend crucially on visibility Public cotribution: with extrinsic incentives, the signal of a prosocial act gets diluted Private contribution: extrinsic incentives are very likely to increase prosocial behavior 14