Armin Falk

1
Intrinsic motivation experiments and
applicationsPart II Economic Implications

• Armin Falk
• Economics and Psychology
• Universität Bonn, WS 07/08

2
Economic Importance of the crowding out hypothesis

• Fundamental incentives are essence of
  economics
• Potential applications
• Environmental economics
• Pricing instruments, i.e., pollution charges
• Social policy
• Crowding out of own responsibility for life with
  monetary incentives
• Contract theory
• Contract enforcement and economic vs.
  psychological contracts

3
Important economic application Motivation in
principal agent relations

• Among all management objectives, motivating
  employees is one of the most important ones
• Traditional principal-agent theory offers a clear
  advice Principals should motivate their agents
  with the help of explicit incentive contracts,
  promotions and other forms of material rewards or
  sanctions
• Assumption
• All humans are selfish
• Only material incentives motivate

4
But

• In the presence of social preferences or
  intrinsic motivation, the introduction of
  explicit incentives may increase effort to a
  lesser degree than predicted by the standard
  model
• It may even lead to lower efforts

5
Literature

• Falk, A., Kosfeld, M., (2006), The Hidden Cost of
  Control, American Economic Review, 96(5),
  1611-1630.
• Fehr, E. and Falk, A., (2002), Psychological
  foundations of incentives, European Economic
  Review, 46(4-5), 687-724.
• Fehr, E. and S. Gächter (2000), Do Incentive
  Contracts Crowd Out Voluntary Cooperation?,
  Institute for Empirical Research in Economics,
  University of Zürich, Working Paper No. 34
• Berg, Dickhaut and Mc Cabe, (1995), Trust,
  reciprocity and Social History, Games and
  Economic Behavior, 10(1), 122-142.
• Fehr, E., Rockenbach, B., (2003), Detrimental
  effects of sanctions on human altruism, Nature,
  422, 137-140
• Gneezy, U., Rustichini, A., (2000a), A Fine is a
  price, The Journal of Legal Studies, 29, 117
• Gneezy, U., Rustichini, A., (2000b), Pay enough
  or dont pay at all, The Quarterly Journal of
  Economics, 115(2), 791810
• Gneezy, U. (2003), The W effect of incentives,
  The University of Chicago Graduate School of
  Business

6
The hidden cost of control

• Falk, A., Kosfeld, M., (2006), The Hidden Cost of
  Control, American Economic Review, 96(5),
  1611-1630

7
An example The HP-way (1)

• In the late 1930s, when I was working for
  General Electric ., the company was making a big
  thing of plant security. GE was especially
  zealous about guarding its tool and parts bins to
  make sure employees didnt steal anything. Faced
  with this obvious display of distrust, many
  employees set out to prove it justified, walking
  off with tools and parts whenever they could.
  When HP got under way, the GE memories were still
  strong and I determined that our parts bins and
  storerooms should always be open.

8
The HP-way (2)

• Keeping storerooms and parts bins open was
  advantageous to HP in two important ways. From a
  practical standpoint, . A second reason, less
  tangible but important, is that the open bins and
  storerooms were a symbol of trust, a trust that
  is central to the way HP does business.
• David Packard (1995)

9
Research questions

• How does the principals decision to control
  affect employees motivation?
• Does the use of control entail hidden costs by
  signaling distrust?
• Do principals anticipate these costs?
• Do the costs depend on the level of control?
• Answers can be given only empirically, need for
  experiment to get clean evidence and causal
  interpretation
• tight control (repeated games, controlling
  payoffs, eliciting beliefs etc.)
• individual differences and types

10
Experimental design (1)

• Principal-agent game
• Agent chooses productive activity x
• working time
• number/quality of produced items
• x is costly to the agent c(x)x
• x is beneficial to the principal b(x)2x
• Endowments 120 for agent and 0 for principal
• Payoffs
• ?p 2x
• ?a 120-x

11
Experimental design (2)

• Before the agent chooses x, the principal
  determines the agents choice set
• principal can enforce a minimum level of x equal
  to x10
• or he can leave the decision about x to the
  agents discretion

12
Principal
restrict
dont restrict
Agent
Agent
x
x
10
0
120
120
13
Interpretation of the principals decision

• restrict/control
• minimum presence requirement
• minimum output/quality
• complete incentive contract, which makes choice
  of xlt10 undesirable
• locked doors

• dont restrict/trust
• working times in responsibility of agent
• appeal w.r.t. output choice
• incomplete incentive contract, which leaves
  considerable freedom of choice
• open doors

14
Experimental procedures

• Strategy method for agent
• One-shot and anonymous interaction
• Main treatment
• 45 minutes, CHF 24.10
• z-Tree (Fischbacher 1999)
• Three further control treatments
• Total of 562 subjects
• Questionnaire study

15
Behavioral predictions (1)

• Standard economic model
• agent chooses minimal x
• xcontrol 10 gt xtrust 0
• principal controls
• Inequity aversion (e.g., Fehr/Schmidt 1999)
• agent chooses positive x independently of the
  decision of the principal
• xcontrol ? xtrust gt 0
• principal controls

16
Behavioral predictions (2)

• Distrust aversion
• agent is intrinsically motivated (e.g., by
  inequity aversion) and reacts negatively to
  signals of distrust
• xcontrol lt xtrust
• principal trusts

17
Results agents choices of x
18
Results agents choice of x
19
Results agents choice of x
20
Results agents choice of x
21
Distribution of x dependent on principals choice
trust
control
x
22
Cumulative distribution
control
trust
x
23
Costs and benefits of control
xcontrol - xtrust
agents
24
Three types of agents
25
Types of agents
26
Three types of agents
27
Three types of agents
28
Principals choice
29
Principals choices
30
Principals choices and beliefs
self-fulfilling prophecy of (dis)trust
31
Yes, but what if control devices are more
effective?

• Message is not that trust is always better than
  control
• Hidden costs depend on
• Frequency of different types (see above)
• Strength of incentives/degree of control

32

• Two further treatments
• Instead of 10, principal can enforce a minimum
  level of
• x equal to x20 treatment x_20
• x equal to x5 treatment x_5
• or let agent freely decide about x

33
Results agents choice of x (x20)
34
Distribution of x dependent on principals choice
(x20)
trust
control
35
Principals choices and beliefs (x20)
self-fulfilling prophecy of (dis)trust
36
Results agents choice of x (x5)
37
Distribution of x dependent on principals choice
(x5)
trust
control
38
Principals choices and beliefs (x5)
self-fulfilling prophecy of (dis)trust
39
Cost of control depending on degree of control
40
Cost of control depending on degree of control
Negative signal, low enforcement power
Negative signal, higher enforcement power
Positive signal, no enforcement power
41
Distrust aversion as a motive control experiment
Principal
restrict
dont restrict
Agent
Agent
Agent
x
x
10
0
120
120
x
10
120
Main treatment
Control treatment (n72)
42
(No Transcript)
43
Distrust aversion as a motive (2)
What do you feel if the principal forces you to
transfer at least 10 points?
44
Distrust aversion as a motive (2)
What do you feel if the principal forces you to
transfer at least 10 points?
45
Distrust aversion as a motive (3)
The restriction signals that the principal does
not trust me to transfer a positive amount.
I transfer less if the principal thinks I am not
trustworthy compared to if he thinks I am
trustworthy.
all agents (n72)
72
86
46
Distrust aversion at the workplace (questionnaire
data)

• Six hypothetical situations
• cash register in supermarket (employer checks
  correctness)
• working times at new job (formal agreement vs.
  appeal)
• job interview (references contacted vs. trusted)
• call center (quality feedback vs. no feedback)
• copy machine not for private use (doors closed
  vs. open) HP!
• internet access not for private use (special
  software to control private use)

47

• Two conditions
• Trust vs. control condition (no subject answers
  for
• both)
• Dep. variable
• Work motivation
• very low, low, middle, high, very high
• N 405, Ss did not take part in any experiment

48

• Situation 1 (Supermarkt)
• Sie haben als Ferienjob eine neue Arbeitsstelle
  in einem Supermarkt angetreten. Ihre Aufgabe
  besteht darin, abends die Kassenstände zu
  überprüfen, d.h. zu schauen, ob die Geldbeträge
  in den Kassen mit den Buchungen übereinstimmen.
• Im Prinzip könnten Sie sich leicht auf Kosten
  des Supermarkts bereichern, indem Sie einfach
  Geld aus der Kasse entnehmen. Sie haben aber die
  Kassen gewissenhaft und ohne eigenen Vorteil
  überprüft und die Ergebnisse ehrlich berichtet.

49
Situation 1 Supermarket

• You recently started a new side job in a
• supermarket. Your task is to check the cash
• register in the afternoon, which means to control
  if
• the money in the cash register matches the
• accounting.
• You could easily cheat on the supermarket in just
• taking some money out of the cash register.
• However you had the chance of cheating you are
• honest and diligently checked the cash box and
• reported the correct amounts.

50
Conditions

• Trust condition
• The manager of the supermarket trusts you and
• does not double-check the cash register.
• Whats your work motivation?
• Control Condition
• On the way back home you notice that you forgot
• your umbrella. When you got back to the
• supermarket you see the manager counting the
• money a second time
• Whats your work motivation?

51

• Vertrauensbedingung
• Der Filialleiter glaubt Ihren Angaben und prüft
  die Kassenstände nicht noch einmal nach.
• Wie hoch ist Ihre Arbeitsmotivation?
• Kontrollbedingung
• Auf dem Heimweg stellen Sie fest, dass Sie
  Ihren Schirm vergessen haben. Als Sie den
  Supermarkt betreten, sehen Sie, dass der
  Filialleiter noch einmal alle Kassenstände
  nachprüft.
• Wie hoch ist Ihre Arbeitsmotivation?

52
Whats your work motivation?(Supermarket)

53
Situation 2 Working time at new job

• Control condition
• Commit yourself by signing a declaration.
• Trust condition
• Appeal from the employer

54
Whats your work motivation?(working times)
55
Situation 3 Interview, references

• Control condition
• Reference is being verified
• Whats your work motivation?
• Trust condition
• Applicant is hired without checking the
  references.
• Whats your work motivation?

56
Whats your work motivation?(References)
57
Situation 4 Copy machine for private use

• Control condition
• Room for copy machine is locked, you have to pick
• up the key.
• Trust condition
• Room is open, appeal from the employer.

58
Whats your work motivation?(copy machine)
59
Situation 5 Internet access not for private use

• Control condition
• Specific software that enables control
• Whats your work motivation?
• Trust condition
• Appeal to follow the rules.
• Whats your work motivation?

60
Whats your work motivation?(Internet)
61
Distrust aversion in the field questionnaire
evidence
62
Questionnaire evidence ctd.
63
Conclusions

• Message there are hidden costs of control
• Offers a behavioral explanation for the
  deliberate incompleteness of many contracts
• Message is not that trust is always better than
  control
• Hidden costs depend on
• Frequency of different types
• Strength of incentives relation is
  non-monotonous
• Framing
• Ultimate management goal develop contracts that
  motivate opportunistic agents without
  de-motivating intrinsically motivated ones.

64
Reciprocity driven voluntary cooperation and
explicit incentives

• Fehr/Gächter (2000)Do Incentive Contracts Crowd
  Out Voluntary Cooperation?
• Principal offers contract (w, ê, f)
• w ? 0, 100 fixed wage
• ê ? 0.1, 1 desired effort
• 0 ? f ? 13 wage reduction in case of detected
  shirking e lt ê (detection probability 1/3)
• Agent can accept or reject the contract
• If accepted Agent chooses actual effort e ?
  0.1, 1
• Cost of effort c(e), with c(0.1) 0 c(e) gt 0
  c(e) gt 0

65
Payoffs

• Principal
• Agent

66
Predictions

• Agent chooses e ê only if (1/3)f ? c(ê)
• In general
• Agent accepts the contract only if w ? c(e)
• In the experiment 0 ? f ? 13 (limited sanction
  possibility)? Maximal enforceable effort is ê
  0.4
• Efficiency-enhancing e gt 0.4 only possible with
  voluntary cooperation

67
Hypotheses

• Hypothesis I
• Incentive contracts do not influence reciprocity
  based voluntary cooperation
• Hypothesis II
• Incentive contracts undermine voluntary
  cooperation

68
Experimental Procedures

• 2 treatments
• Trust treatment (TT) (w, ê)
• Incentive treatment (IT) (w, ê, f)
• 126 subjects
• 6 Principals, 8 Agents in two rooms
• Manually conducted at University of Zurich
• posted bid market more agents than principals
  (competitive pressure)
• 12 Periods

69

• Effort is lower in the IT compared with the TT.
  This
• is due to the following reasons
• A fraction of agents shirks in the IT even when
  the No-Shirking Condition (NSC) is met (ITIC)
• Voluntary cooperation (e e gt 0) vanishes
  almost completely for incentive compatible
  contracts in the IT
• If the NSC is violated in the IT agents choose
  the minimum effort in the vast majority of cases

70

• In the TT effort levels above the minimum are
  provided in the majority of cases
• In the TT voluntary cooperation responds strongly
  to the wage level while in the IT voluntary
  cooperation does no longer respond to the wage
  level

71
Voluntary cooperation
0.7
period 9-12
period 1-4
period 5-8
0.6
0.5
0.4
Voluntary cooperation (e - e)
0.3
0.2
0.1
0
-0.1
50
50
50
0-10
0-10
0-10
11-20
21-30
31-40
41-50
11-20
21-30
31-40
41-50
11-20
21-30
31-40
41-50
Intervals of offered wages
72
Evolution of wages
73
Voluntary cooperation in period 1
Trust
Incentive
0.9
0.9
0.8
0.8
0.7
0.7
0.6
0.6
0.5
0.5
0.4
0.4
0.3
0.3
Voluntary cooperation (e - e)
Voluntary cooperation (e - e)
0.2
0.2
0.1
0.1
0
0
-0.1
-0.1
-0.2
-0.2
-0.3
-0.3
0
20
40
60
80
0
20
40
60
80
Offered wage
Offered wage
74
Why is there an undermining of voluntary
cooperation?

• The redistributive effect of the fine may
  undermine voluntary cooperation?
• The explicit ex-ante threat to punish is
  perceived as a hostile action and reciprocal
  workers respond with hostility (shirk a lot) to
  hostile actions.
• Does negative framing (fine) induce hostility?

75
Inequity aversion and effort choice(for a given
wage)
 
payoffs

100e w f
100e w
w c(e)
w c(e) f
ef
ee
effort
76
Does the framing of incentives matter? The Bonus
contract

• Bonus contract If e ? ê a bonus of 0 ? b ? 13
  will be paid if e lt ê the bonus will be paid
  with probability 2/3
• identical incentive structure to the sanction
  contract
• only a different framing

77
Framing of incentives and voluntary cooperation
0.7
Trust
Incentive
Bonus
Treatment
Treatment
Treatment
0.6
0.5
0.4
0.3
Voluntary cooperation (e - e)
0.2
0.1
0
-0.1
50
50
50
0-10
0-10
0-10
11-20
21-30
31-40
41-50
11-20
21-30
31-40
41-50
11-20
21-30
31-40
41-50
Interval of offered compensation (w(b))
78
Offered contracts, Profits and Efficiency

• Incentive compatible contracts

• Mean Payoffs and Efficiency
• NIC IC
• Trust
• Agents 25.7
• Principals 7.1
• Sanctions
• Agents 19.6 9.1
• Principals 1.8 18.2
• Bonus
• Agents 17.6 11.6
• Principals 7.3 8.8

70
60
50
40
Percent
30
20
10
0
1
2
3
4
5
6
7
8
9
10
11
12
Period
79

• If the use of a hostile incentive destroys
  voluntary cooperation, then the deliberate
  non-use of a hostile incentive may be perceived
  as a kind action and may, hence, increase
  voluntary cooperation.
• Experimental test
• Trust Experiment by Fehr and Rockenbach, (2003),
  Detrimental effects of sanctions on human
  altruism, Nature

80
Design

• Investor and a trustee receive 10 DM
• Investor can send any amount between 0 and 10 to
  the trustee
• Each DM sent is tripled by the experimenter
• Investor also announces a desired back-transfer
• Trustee is free to send back any amount
• Back-transfer is not tripled
• Investment game by Berg, Dickhaut and McCabe
  (1997)

81

• Two treatments
• Trust treatment and punishment treatment
• In the punishment treatment the trustees payoff
  is reduced by 4 DM if she sends back less than
  desired
• Fine can be implemented or not be implemented

82
Your decisions in trust game
83
Decisions of experimental economics class
84
Table 1. Average behaviour and average payoffs of
investors and responders


Trust condition
Incentive condition
Incentive condition

fine chosen
no fine chosen



Investment (x)

6.5

6.8

8.7

Desired back
-
transfer in
59.
9

67.4

63.7

percent of tripled investment
?
x)
(
/3

Actual back
-
transfer (
y
)

7.8

6.0

12.5

Actual back
-
transfer in percent
40.6

30.3

47.6

of tripled investment (
y
/3
x
)

Actual back
-
transfer in percent
74.38

54.50

74.08

of desired back
-
transfer (
y
/

?
)

Investors payoff

11.3

9.2

13.8

Responders
payoff

21.8

22.4

23.5

Number of observations

24 pairs

30 pairs

15 pairs


85

• The average back-transfer of the responders
• as a function of the investors transfers
• In the trust condition and the incentive
  condition the back-transfers of the responders
  increase in the investors transfers
  irrespective of whether the fine is imposed or
  not
• If the investors impose a fine in the incentive
  condition responders reduce their back-transfers
  indicating a detrimental effect of the incentive
  on cooperative behavior
• Responders back-transfers are highest if the
  investor deliberately refrains from imposing the
  fine

86
The average back-transfer of the responders as a
function of the investors transfers
16
Incentive condition - fine imposed
14
Trust condition - no fine possible
Incentive condition - fine not imposed
12
10
Average back-transfer of trustee
8
6
4
2
0
0-1
2-4
5-7
8-10
Transfer of investor
87

• Actual back-transfer in percent of tripled
• investment as a function of the investors
• desired back-transfers
• The desired back-transfer is categorized as low
  if, in case that the responder meets this
  back-transfer, it implies that the investor earns
  the same or less than the responder
• The desired back transfer is categorized as high
  if the investor earns more than the responder

88

• At low desired back-transfers the sanctioning
  threat reduces actual back transfers in the
  incentive treatment but the effect is not
  significant.
• At high desired back-transfers the sanctioning
  threat has a large negative impact on the actual
  back-transfers in the incentive condition
  indicating that if sanctions are used to achieve
  a distributional advantage they strongly
  undermine cooperation

89
Actual back-transfer in percent of tripled
investment as a function of the investors
desired back-transfers
70
60
50
40
Actual back-transfer in percent of tripled
investment
30
20
Trust condition - fine not possible
10
0
low
high
Desired back-transfer
90
Students vs. CEOs

• Fehr and List (JEEA) performed a similar study as
  Fehr and Rockenbach (2003)
• Both principal and agent receive endowment of 10
• Trust treatment
• Principal offers any amount x between 0 and 10,
  which is tripled
• Principal states a desired pay back
• Agent decides on payback between 0 and 3 x
• Punishment treatment
• Principal can threaten to punish agent if payback
  lt desired payback (fine either 0 or 4)
• Fine not given to the principal

91

• Standard prediction
• Principal can enforce x 3 or 4 in the
  punishment
• treatment and 0 in the trust treatment
• Comparison of students and CEOs from Costa Rica
• Important to study behavior of higher level
  decision-makers
• 126 students, 76 CEOs from the coffee mill sector
  who attended THE COSTA RICA COFFEE INSTITUTES
  annual conference in March 2001
• All anonymous interaction (single blind)
• Earnings CEOs 65, Students 5.65

92
Results

• CEOs transfer more money than students
• For a given transfer, CEOs pay back more than
  students
• Not using the fine-option in the punishment
  treatment leads to higher back transfers than in
  the trust treatment (where the option is not
  available)
• CEOs use the punishment option less often than
  students

93
Back-transfers of CEOs and Students in the Trust
Treatment
14
12
10
8
Agent's back-transfer
6
4
2
0
0 to 5
6 to 8
9 and 10
Transfer of the principal
Students
CEOs
94
Back-transfers of CEOs and Students in the
Punishment Treatment
Students
CEOs
95
The Impact of the Punishment Threat on CEOs
Back-transfers
96
The Impact of the Punishment Threat on Students
Back-transfers
97
Two Field studies

• I. Gneezy and Rustichini Pay Enough or Dont
• Pay At All, QJE, 2000
• Test of monetary incentives on performance
• Experiment 1
• 160 students received a show-up fee of NIS 60
• To solve 50 quiz problems (IQ-test) in 45 minutes
• Subjects divided into four groups (treatments)
• Treatments differ with respect to the marginal
  payoff for a correct answer
• Nothing (asked to answer as many questions as
  they can)
• 10 cents of a Shekel, 1 or 3 Shekels

98
Summary statistics for the number of correct
answers by treatment
99
Hypothesis testing

• Mann-U-Tests for significance of results (pair
  wise comparisons), p-values

100

• How well could people predict the behavior of the
  subjects in the IQ test, depending on incentives?
• Participants were promised a payment
  proportional to the performance of their agent,
  and had to choose the incentive scheme for the
  agent
• 10 cents vs. nothing
• 87 chose to be matched with agents who were
  paid 10 cents over unpaid agents
• Wrong anticipation of detrimental effects of
  incentives

101

• Experiment 2
• 180 high school students
• Go from house to house to collect donations for
  charitable organizations (typical volunteer work
  on a particular day each year)
• Divided into three groups (treatments)
• No pay Small speech recalling the importance of
  the activity
• Speech Promise of 1 percent of collected money
• Speech Promise of 10 percent of collected money
• Money paid by experimentalists

102
Hypothesis testing results

• Significant differences (10 percent) between
  No/1percent and 1percent/10percent
• As before, principals chose the wrong incentive
  (76 percent chose the 1 percent incentive)

103

• II. Gneezy and Rustichini (2000),
• A Fine is a Price, Journal of Legal Studies
• Suppose you are a daycare manager
• The problem Parents come late to collect their
  child
• The solution A fine
• To test this solution
• Experimental study in Israeli day-care centers
• The contract signed at the beginning of the year
  states that the day-care center closes at 4.00
  p.m
• No mention of sanction
• Late coming is a problem

104
Organization of the field study

• 20 weeks
• 10 day-care centers
• 6 treatment, 4 control groups

105

• Announcement of the introduction of the fine one
  week before
• The number of late coming parents was registered
• The fine was 10 NIS
• For a delay of 10 minutes or more
• Per kid
• Fine was removed in the treatment group without
  any explanation

106
Average number of late-arriving parents each
week, by group type
25
20
15
No .of late arrivals
10
5
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
week number
Control group
107
Results

• In first four weeks no significant difference
  between treatment and control group
• In treatment group increase in late-comings after
  fine is introduced
• Rate finally settled (after 2 weeks) at an almost
  twice as high level as the initial one
• Removing the fine did not affect the number of
  late-coming final number of late coming parents
  in the treatment group higher than in control
  group

108
Two further questions

• Is the detrimental effect of sanctions or
  incentives monotonous?
• What if the incentives get stronger?
• What if positive and negative incentives are used
  in one experiment (only one word is changed in
  the experiment)
• Study by Gneezy (2004)

109
Economic Applications

• Frey, Jegen (2001) Motivational Interactions
  Effects of Behavior
• Frey (2001), Not just for the money.
• Labor market
• impact of higher wages on working effort
  particularly performance wages
• Natural environment
• Pricing instruments, i.e. pollution charges
• Social policy
• crowding out of own responsibility for life with
  monetary incentives
• Organization theory
• limits of the firm when only considering
  extrinsic incentives
• Contract theory
• avoid crowding out by psychological contracts

110
Labor market

• Distinguishing between crowding-out and
  crowding-in of work morale.
• Emphasis on Crowding-Out
• Is the case if an extrinsic reward is perceived
  to be controlling.
• Concerns
• what about performance based wages (partially),
  options for managers, higher extrinsic monetary
  rewards might initially increase extrinsic
  motivation, thus work morale but after better
  performance better job Perspectives thus more
  interesting tasks which implies an increase in
  intrinsic motivation.

111
Constitutional Design

• Frey applies the crowding-out theory to the
  effects of constitutional and legal rules on
  citizens
• He states that distrustful public laws undermine
  civic virtues, doubtful argumentation.

112
Environmental Policy (1)

• Important question in this setting
• Do environmental incentive instruments influence
  the environmental morale and if so in which
  direction?
• The lower costs for keeping restrictions the
  larger is the environmental morale.
• Defining the costs for different groups
• From a firms point of view does competition
  increase costs of environmental norms.
• From a citizen point of view distinguish
  industrial countries, where following
  environmental ethics is not costly and developing
  countries, where people have to exploit all
  opportunities maybe even for their own survival).

113
Environmental Policy (2)

• Effectivity of environmental incentives
  determined by
• How environmental morale affects the behavior
  (determined by the individual costs).
• Intensity to which the instruments are used.
• Frey argues that for example, self-organized
  groups or ecological activity groups initially
  cared for environmental problems, an governmental
  extrinsic intervention could destroy the morale
  to keep on protecting the environment, thus
  destroys the intrinsic motivation.
• WHY? The other direction makes more sense!

114
Environmental Policy (3)

• Failure of the policy instrument to support the
  persons intrinsic motivation to protect the
  environment i.e. tradable licenses as a license
  to pollute.
• BUT as if a firm had an intrinsic motivation to
  protect the environment per-se and the license
  destroys this motivation! It is more an extrinsic
  motivation to protect the environment, because
  there DOES NOT exist any or almost no intrinsic
  motivation to do so)!

115
Environmental Policy (4)

• Rewards or punishment contingent on behavior
• Pricing of household refuse based on quantity,
  which makes intrinsic motivation unnecessary.
• No differentiation between persons with high and
  low environmental morale, i.e. environmental tax
  for sectors average pollution can lead to an
  adjustment of the level of pollution to the
  sectors average by the managers, thus may be a
  crowding out of intrinsic motivated firms

116
Social and organizational policy

• Regulatory policy
• Voluntary work
• Crime Prevention
• Price of Deterrence Effect is expected
• Blood Donation

117
Possible reasons for negative effect of
incentives (see discussion in Gneezy 2003 and
Fehr/Falk EER 2002)

• Information (Learning about negative
  implications)
• Distrust
• Insult (Sex for money)
• Fairness payoff consequences
• Cooperation and social norms vs. exchange
• Signaling (Benabou/Tirole-story)
• Crowding out of intrinsic motivation