brigitte madrian harvard university
TRANSCRIPT
BEHAVIORAL ECONOMICS AND
FINANCIAL REGULATION
Brigitte Madrian
Harvard University
December 7, 2017
Behavioral Economics and Public Policy 2
What is Behavioral Economics?
Behavioral economics incorporates insights from economics and other behaviorally oriented disciplines including psychology, sociology, anthropology, and cognitive neuroscience, to enrich standard economic models in ways that improve our ability to understand and predict human behavior, market outcomes, and public policy.
3
What is Behavioral Economics?
Traditional Economics Behavioral Economics
4
Neurology of Decision Making:
Multiple Systems Hypothesis
The brain integrates signals from multiple systems in making decisions
These systems process information differently
System 1 System 2
Affective Analytic
Fast Slow
Reflexive Reflective
Effortless Effortful
Impulsive Deliberative
Myopic Patient
5
INFORMATION
FAILURES
Traditional Economic Motivations for
Policy Intervention 6
MARKET
POWER
Traditional Economic Motivations for
Policy Intervention 7
EXTERNALITIES
Traditional Economic Motivations for
Policy Intervention 8
PUBLIC
GOODS
Traditional Economic Motivations for
Policy Intervention 9
Behavioral Economics and Public Policy
What does an understanding of
behavioral economics bring to the policy table?
Additional motives for policy intervention
New policy tools
Ways in increase the effectiveness of
traditional policy tools
10
Behavioral Economics and Household Financial
Outcomes 11
Where Do Traditional Economic Models
Fall Short?
Limited attention
Selective attention
Salience
Limited computational capacity
Choice overload
Mental accounting
Use of heuristics
Biased reasoning
Errors in probabilistic
thinking
Motivational biases
People are Imperfect Optimizers:
They Make Mistakes!
12
Salience: Consumers Pay More
Attention…
…to up front fees …than to shrouded fees
Air travel ticket price Air travel add-on fees
Product prices Sales tax or shipping costs
Bank interest rates Other bank fees
Cell phone monthly fee Cell phone overage charges
Mental Accounting
241
414
0
100
200
300
400
500
One envelope Two envelopes
Rup
ees
saved o
ver
14
weeks
One vs. Two Envelopes
Source: Soman and Cheema (2011). “Earmarking and Partitioning”
14
How Much to Save: Economics…
15
0
1
2
3
1% 2% 3% 4% 5% 6% 7% 8% 9% 10% 11% 12% 13% 14% 15%
Frac
tio
n o
f e
mp
loye
es
Year-End Contribution Rate
Source: Choi, Laibson, Madrian and Metrick (2006)
…vs. Psychology
Heuristic thinking: people disproportionately choose to
Save at contribution rates that are multiples of 5
16
Future Value of $1000 Invested Today
0
2.500
5.000
7.500
10.000
12.500
15.000
17.500
20.000
0 5 10 15 20 25 30 35
At a 10% interest rate
compounded annually
At a 10% interest rate
with no compounding
Heuristic thinking: people anchor on linear
compounding, and adjust upward, but not completely
17
Procrastination
Overconfidence
Addiction
Affect
People Have Self Control Problems:
They Cannot Always Follow Through with Their Plans
Where Do Traditional Economic Models
Fall Short? 18
Mood and Tipping
19% 24% 27% 29%
Source: Rind (1996). “Effect of Beliefs About Weather Conditions on Tipping”
Average tip given to hotel room service delivery person
by the weather conditions reported to hotel guests.
19
Where Do Traditional Economics
Models Fall Short?
Reference-dependent preferences
Endowment effect
Loss aversion
Status quo bias
Menu effect
Framing
Other-regarding preferences
Altruism
Fairness
Social norms
Interpersonal preferences
Non-standard preferences: what individuals want is not
what traditional economic models presume they want
20
Loss Aversion and Investor Decisions
Individuals treat
investments that have gone
up in value differently
from investments that have
gone down in value
More likely to hold onto
stocks that have fallen in
value
Set a higher list price
(riskier strategy) for real
estate
21
Default/Endowment Effects and Asset
Allocation Outcomes
23% 20%
95%
27%
0%
25%
50%
75%
100%
Match default: employer stock
Match default: none
Alloca
tion
to E
mplo
yer
Sto
ck
Own
Match
Employer Stock Allocation in Employee Savings Accounts
IPO Lotteries in India
Source: Choi, Laibson and Madrian (2009)
63,9%
45,8%
36,6%
1,2% 1,7% 1,5% 0%
20%
40%
60%
80%
At IPO 12 months later
24 months later
Hold
s IP
O S
tock
Won IPO lottery
Lost IPO lottery
Source: Anagol, Balasubramanian and Ramadorai (2016)
22
Implications for Financial Market Regulation 23
The Interaction between Individual
Behavioral Biases and Firm Motives
Behavioral
Fallability
Market neutral and/or
wants to overcome
consumer fallability
Market exploits
consumer fallability
Consumers
misunderstand
compounding
Savings context:
• Banks would like to help
reduce this bias
Borrowing context:
• Banks would like to
exploit this bias
Consumers
procrastinate
Claiming the EITC:
• Tax preparation companies
would like to reduce this
bias
Claiming rebates:
• Retailers would like
to exploit this bias
Barr, Mullainathan and Shafir (2013), “The Case for Behaviorally Informed Regulation”
24
The New Yorker, February 1, 2010 25
Behavioral Economics and Public Policy
What does an understanding of
behavioral economics bring to the policy table?
Additional motives for policy intervention
New policy tools
Ways in increase the effectiveness of
traditional policy tools
26
Behaviorally Informed Regulation:
The Case of Disclosure
Reduce Information Asymmetries
Reduce Search Costs
Traditional Rationale for Disclosure Requirements
27
S&P 500 Index Fund Experiment
Subjects allocate a
hypothetical $10K across
four real S&P 500 Index
Funds
All subject groups are more
financially literate than the
population of American
investors
Harvard undergraduates
Wharton MBAs
Harvard non-faculty staff
28
Experimental Conditions
Control
Subjects receive only four prospectuses
Prospectuses are often the only information investors receive from companies
Fees transparency treatment
Eliminate search costs by also distributing fee summary sheet (repeats information in prospectus)
Returns treatment
Highlight extraneous information by distributing summary of funds’ annualized returns since inception (repeats information in prospectus)
29
Portfolio Fees of Control Group:
Student Experiment
$421 $431
$309
$349
$389
$429
$469
$509
$549
$589
MBA College
Control
Minimum
Possible
Fee
Maximum
Possible
Fee
$443: average fee with random fund allocation
0% of College Controls
put all funds in
minimum-fee fund
6% of MBA Controls
put all funds in
minimum-fee fund
Source: Choi, Laibson and Madrian (2010)
30
Fee Treatment Effect on Portfolio Fees:
Student Experiment
$421 $431
$366
$410
$309
$349
$389
$429
$469
$509
$549
$589
MBA College
Control
Fee treatment
19% of MBA Controls
put all funds in
minimum-fee fund
10% of College Controls
put all funds in
minimum-fee fund
Source: Choi, Laibson and Madrian (2010)
31
Returns Treatment Effect on Fees:
Student Experiment
$421 $431
$440
$486
$309
$349
$389
$429
$469
$509
$549
$589
MBA College
Control
Returns treatment
Source: Choi, Laibson and Madrian (2010)
32
Lessons on Disclosure
Lesson 1: Consumers are inattentive to relevant
information when it is not salient
Lesson 2: Consumers do respond to relevant
information when it is salient and easy to find
Although doesn’t necessarily result in optimal decision-
making
Lesson 3: Consumers also respond to irrelevant
information when it is salient
33
Making Disclosure More Effective
• Do the calculations C
• Translate information to personal objectives O
• Provide relative comparisons R
• Expand important outcomes E
Source: “Designing Better Energy Metrics for Consumers” (Larrick, Soll and Keeney, 2016)
34
C: Do the Calculations 35
O: Translate Information to Personal
Objectives 36
R: Provide Relative Comparison
Source: https://www.canada.ca/en/financial-consumer-agency/services/loans/payday-loans.html
37
E: Expand the Outcome 38
Evidence on the Impact of Behaviorally
Informed Disclosure 39
Behaviorally Informed Disclosure on
Future Payday Loan Utilization
Source: Bertrand and Morse (2011)
-5,1% -5,0%
-6%
-5%
-4%
-3%
-2%
-1%
0%
Relative comparison
Dollar calculation over time
Fu
ture
pay
day
loan
utiliza
tion
re
lative to c
ontr
ol gro
up
Disclosure Treatment
40
Increase in the Likelihood of Choosing the
Lowest Cost Option
(Relative to Control Group)
Loan
Choice
Savings Account
Choice
Comparison Table
5 options +25 pp +7-18 pp
10 options No effect +12 pp
Cost in pesos vs.
percentages +8 pp +4 pp
The Effect of Behaviorally Informed
Disclosure: A Study in Peru and Mexico
Source: Gine, Cueller and Mazer, “Information Disclosure and Demand Elasticity of Financial Products,” 2017
41
Behavioral Economics and Public Policy
What does an understanding of
behavioral economics bring to the policy table?
Additional motives for policy intervention
New policy tools
Ways in increase the effectiveness of
traditional policy tools
42
Behaviorally Informed Regulation:
Products
Standardized “plain
vanilla” products
Reduces complexity of
product comparison
Facilitates disclosure of
relevant
characteristics/prices
Facilitate price competition
Prohibitions on risky
products/product features
Fee caps
43
Behaviorally Informed Regulation:
Process
Choice architecture
The design of
environment and
context in which
choices are made
44
Behaviorally Informed Regulation:
Process
“Smart” defaults
“Smart” choice menus
Fewer options
Tiered options
Facilitate comparison
Decision support tools
Decision timing
Decision verification
45
Behaviorally Informed Regulation:
People
Consumers:
Require demonstration of
understanding before
allowing certain
transactions involving risky
products/large stakes
Financial services industry
employees:
Conduct standards (e.g.,
suitability requirements,
fiduciary standard)
46
A lot of our policy models
traditionally are based on a
rather naïve understanding of
what drives behavior. But if you
have a more intelligent, nuanced
account of how people make
decisions, you can design policy
that is more effective, less costly,
and makes life easier for most
citizens.
—David Halpern, Director of the
UK Behavioural Insights Team
Can We Improve Government? 47