Topic 15

Asymmetric Information and

Agency Problem
(Chapter 15)
Ratna K. Shrestha
 Introduction
⚫ Frequently a seller or producer knows more
about the quality of the product (hidden
characteristics) than the buyer does.
⚫ Managers know more about firm’s competitive
position, costs, and investment opportunities
than firm owners or shareholders.
⚫ This situation when one party knows more
than others is called Asymmetric
Information.
2
 Introduction

⚫ Many markets such as insurance, financial

credit and employment are characterized by
asymmetric information about product quality.
⚫ For example in the case of employment, a
potential candidate knows more about her
skills than the prospective employer.
⚫ Such asymmetric information leads to
opportunistic behavior: (a) adverse selection
or lemons problem and (b) moral hazard.
3
 Asymmetric Information

⚫ With asymmetric information, informed party

benefits at the expense of uninformed or less
informed party.
⚫ (1) Adverse Selection or Lemons Problem:
When consumers lack information on a
product quality, say used cars, they are likely
to offer an average price equal to their
average value of the good and bad cars.

4
 15.1 Adverse Selection
⚫ Since the price offered is less than the
value of the good cars, sellers of good cars
will not sell their products.
⚫ Similar situations also arises in many other
markets: Health Insurance, Auto Insurance
Financial Credit etc.

5
 Adverse Selection
⚫ Good drivers will find the premium too high
and choose not to insure.
⚫ This results in a higher proportion of risky
drivers in the pool of insured, raising the
accident probability among the insured.
⚫ In response to this (higher accident
probability), insurance company raises the
premiums, driving low-risk drivers out of the
market—adverse selection.

6
 Lemons Problem: Example
⚫ Assume risk neutral potential buyers of used
cars who value high quality cars at $2,000 and
low quality at $1,000. If they cannot distinguish
between high and low quality sellers, they will
offer $1,500 on the average.
⚫ If high quality sellers are willing to accept $1250
and low quality $750, all cars will be sold.
⚫ In this case asymmetric information causes no
inefficiency as all cars are bought by buyers
who value them more than the sellers.
7
 Lemons Problem
⚫ On the other hand, if high quality sellers’ need
at least $1,750, then high quality cars will not
be sold; only low quality cars will be sold.
⚫ Knowing that only low quality are being sold,
at equilibrium, buyers pay $1,000, the price
they are willing to pay.
⚫ This is inefficient because high quality cars
remain at the hands of the sellers who value
them less (= $1,750) than potential buyers
(who value them at $2,000).
8
 Lemons Problem

⚫ The deadweight loss due to high quality cars

not being sold (adverse selection)= $2,000 –
1,750 = $250/car.
⚫ If both buyers are sellers know the quality of
the cars, then good quality will be sold for
$2,000 and bad quality for $1,000.
⚫ In this case of NO asymmetric information,
there is no adverse selection problem.

9
 Example: Adverse Selection
⚫ Suppose 50% of the population is healthy
and the other half is unhealthy. If a healthy
gets sick, the medical cost is $1,000; for
unhealthy this cost is $10,000. In a given
year, the probability that anyone gets sick is
0.4. Although each person knows whether he
or she is healthy, the insurance company
does not. Both insurance company and the
people are risk neutral.

10
 Example: Adverse Selection
⚫ a) If the insurance company offers complete,
actuarially fair insurance, what is the
premium?
EV of medical coverage for both sick and
healthy people combined = 0.4*(0.5*10,000 +
0.5*1,000) = $2,200.
So actuarially fair premium (that yields zero
profits to the insurance company) = 2,200 per
head irrespective of health status.

11
 Example: Adverse Selection
⚫ b) At the price you determined in part a, do
healthy people purchase the insurance?
Each healthy person's expected willingness to
pay = 0.4*1,000 = $400/year, which is less
than the acturially fair premium of $2,200/year.
Therefore healthy people will not buy the
medical coverage—adverse selection
problem.

12
 Example: Adverse Selection
⚫ c) If only unhealthy people purchase
insurance, what is the price of the insurance?
Each unhealthy person’s willingness to pay or
medical coverage=0.4*10,000 = $4,000 per
year, which is higher than premium charged.
So they will buy the insurance.
Given that insurance company knows only
unhealthy people buy insurance, at equilibrium
it will charge premium = $4,000 per head/year.

13
15.2 Reducing Adverse Selection
Ways to reduce Adverse Selection:
⚫ 1) Universal Coverage/Risk Pooling
 By mandating insurance coverage to all,
informed people are prevented from
behaving opportunistically.
Example: Mandatory Auto Insurance.
Drivers (good or bad) have no choice but
to buy the basic auto insurance.

14
 Reducing Adverse Selection

⚫ 2. Screening: Uninformed people can

screen or gather information on the hidden
characteristics of informed people.
⚫ Examples: a) Insurance companies gather
health history (and habits) of potential
customers. b) Employers can screen
candidates (CV, interviews, etc.)

15
 Reducing Adverse Selection
⚫ 3. Signaling is used primarily by informed
party to reduce or eliminate adverse selection.
⚫ Examples: producers can convey information
to buyers about their product quality.
⚫ However such signal must be credible. One
way to make a credible signal is by providing
product warranties that cannot be matched by
low quality producers.

16
 Quiz 1
Buyers: Good fridges = $600 and poor fridges = $300.
Sellers: good >= $400, poor fridges >= $200. Both are
risk neutral.
a) If the proportion of good and bad are 50-50, then
all fridges are sold at $450.
b) If only 10% of available fridges are good, then
there is an adverse selection problem..
c) Both a and b.
d) None.
 Quiz 2
If buyers are uncertain but sellers know the
quality. This kind of asymmetric information
a) Can cause buyers to be risk averse.
b) can cause adverse selection.
c) can cause inefficiency in that potential
gains from trade are wasted.
d) b and c.

18
15.3 Moral Hazard/Agency
Problem
⚫ Moral Hazard: When one party has more
information, then that party may take
advantage of the uninformed party through
unobserved hidden actions. Examples:
(a) a worker may shirk if not monitored by the
manager.
(b) when you have auto insurance, you may be
more reckless while driving.

19
 Principal-Agent Problem
❑The principal agent problem arises under
asymmetric information of the “hidden action”
type called Moral Hazard.
❑ A principal hires someone – an “agent” but
cannot fully observe the action of the agent.
In this case the agency has an incentive to
shirk or cheat, against the interest of the
principal—“agency” problem.

20
Principal-Agent Problem
❑Thus the agency problem arises when
agent’s objective does not coincide with
the principal’s objective.

21
 Reducing Moral Hazard
⚫ Efficient Contract:
 If both parties are risk neutral, efficiency
requires that the combined profit/benefit be
maximized.
 If one party is more risk averse, then
efficiency requires the less risk-averse
party bear more risk.

22
 Reducing Moral Hazard: Example
⚫ Efficient Contracts: Ice Cream Shop
 Paul (principal) owns many ice cream
parlors. He contracts with Amy (agent) to
manage his Miami shop.
 Paul’s profits (before paying to Amy) is
given in Table 15.1.
 Assume Paul is risk neutral and Amy risk
averse.

23
Table 15.1 Ice Cream Shop Profits
Demand low or high with 50-50% chance.

Amy’s cost of normal (low) effort = 0 but that

of extra (Hard) effort is $40/day.
24
Reducing Moral Hazard
Inefficient Contract
 Paul cannot observe Amy’s effort.

 Contract: Amy gets a fixed-fee/wage of

$100/day. With this contract, Amy’s bears
zero risk.
 If Amy works normally, she gets $100. But,
if she works hard, she only nets $60 (100-
40). So she works normally—moral hazard
problem.

25
 Reducing Moral Hazard
Inefficient Contract
 With Amy’s normal effort, Paul’s EV = expected
profits – payment to Amy = 0.5(100 – 100) +
0.5(300 – 100) = $100; and σ2 = 10,000
 Total surplus = Paul’s+ Amy’s =100+100 = 200.

 Inefficient as max possible = 360.

 But efficient from risk standpoint. Risk averse

Amy bears no risk and risk neutral Paul bears
all risk.

26
 15.4 Efficient Contracts
⚫ Fixed-Fee Contract
 Assume Amy pays Paul fixed fee = $200
(with certainty) to operate the shop.
 Paul bears no risk as he receives a fixed
fee.
 Amy bears all the risk and gets the residual
profit.

27
Contracts to Reduce Moral Hazard
⚫ Fixed-Fee Contract
 Amy’s EV.

EVHARD = 0.5 (300 – 200 – 40) + 0.5(500 – 200 –

40) = 0.5*60+0.5*260 = $160
EVNORMAL =0.5(100 – 200)+ 0.5(300 – 200)
= 0.5*(-100) + 0.5*100 = $0.

EV(HARD) > EV(NORMAL). So. Amy chooses

HIGH effort.

28
Contracts to Reduce Moral Hazard

⚫ Joint Profit = Amy + Paul = 160 + 200 = $360

⚫ Joint profits is maximized
⚫ But there is no efficient risk sharing as Amy,
risk averse agent, bears all the risk.
⚫ Example: This contract is practiced in
agricultural sector, where tenants rent land
from landlords by paying a fixed fee and
keeping all the residual profits from farming.

29
 Contingent Contracts
⚫ 1. State Contingent Contract
 Contract: Assume Amy pays Paul $100 if
demand is low and $300 if demand is high.
Amy’s (agent)
EVHARD = 0.5(300 -100 - 40) + 0.5(500 - 300 -
40) = $160
EVNORMAL= 0.5(100-100) +0.5(300-300) = 0
EVHARD > EVNORMAL
With Amy’s hard effort, Paul’s EV = $200

30
 Contracts to Reduce Moral
Hazard
⚫ 2. Profit Sharing
 Paul and Amy agree to split the earnings of
the ice cream shop equally.
 Amy’s

EVHARD = 50% of (0.5*300+0.5*500) – 40=$160

EVNORMAL= 50%(0.5*100+0.5*300)= $100
She prefers to work hard.
 With Amy’s hard effort, Paul’s EV = $200

31
 Contracts to Reduce Moral
Hazard
⚫ Example: For what minimum %, say x, of
profit sharing will Amy put hard effort?
Amy’s EVHARD = x(0.5*300+0.5*500) – 40
= 400 x – 40.
EVNORMAL= x(0.5*100+0.5*300) – 0 = 200 x.
Amy will put hard effort if
EVHARD > EVNORMAL
400x – 40 > 200x
200x > 40 or X > 1/5 = 20%.
32
 Contracts to Reduce Moral
Hazard
⚫ 3. Bonuses
 Paul offers Amy a base wage of $100 and a
bonus of $200 if the shop’s earnings (before
paying Amy) > $300.
Amy’s EV
EVNormal = 100 + 0 =$100. (risk free)
EVHARD = 100 + (0.5*0 + 0.5*200) – 40=$160.

33
 Contracts to Reduce Moral
Hazard
In Bonus example,
From EV criteria, Amy chooses Hard effort.
From risk criteria, she chooses Normal.

⚫ If Amy is risk averse: inconclusive

⚫ Risk taker: hard effort.
⚫ Risk neutral: hard effort.

34
 Contracts to Reduce Moral
Hazard
4. Options:
⚫ An option gives the holder the right to buy
a certain number of shares at a given price
(the exercise price).
⚫ The holder can sell the stocks if the stock
price > exercise price.
⚫ Options thus provides agents incentives to
work hard.

35
 Contracts to Reduce Moral
Hazard
⚫ 5.
Commissions: agent receives a payment per
unit of output produced. This provides
incentives to work hard (but the reward may not
be good enough to offset extra effort)
Piece Contract: Amy is paid for every serving of
ice cream she sells.
Revenue Sharing: agent receives some share
of revenues earned, say 5%.

36
 Using Monitoring to Reduce
Moral Hazard
⚫ Due to difficulties in monitoring, employers
may use several techniques:
1. Hostages:
⚫ Examples:
a) bonding,
b) deferred payments, and
c) efficiency wages.

37
 Using Monitoring to Reduce
Moral Hazard
A) Bonding to Reduce Monitoring Efforts
 Example: an employer may require an
employee (agent) to provide a performance
bond, which can be forfeited if the agent fails
to complete certain duties.
B) Deferred Payments: Example--Pension
plan rewards only workers who stay longer
and who avoid being fired by working hard.

38
 Using Monitoring to Reduce
Moral Hazard
C) Efficiency Wages: pay higher wages than
worker’s opportunity cost.
 This reduces shirking (hence monitoring) as
employee cannot get higher wage
elsewhere.

39
 Example
❑ The manager’s effort is given in the left
column. Each cell shows the net income or
revenue to the firm (without subtracting wage
payments):
Bad Luck Good Luck
Low Effort 20 40
Med. Effort 40 80
High Effort 80 100

40
 Example

❑ The manager’s cost of low effort is $0,

medium effort is $10, and high effort is $30.
❑Bad luck and good luck each have a 50%
probability.
❑The manager and the owner are both risk-
neutral.

41
 Possible Contracts
Consider three possible contracts.
1) Fixed Wage: The manager receives a fixed
wage of $15.
2) Profit Sharing: The manager receives 50%
of net income and no fixed wage.
3) Bonus: The manager gets 80% of net
income exceeding $40 and nothing else.
Thus, if revenues are $80, the manager would
get 80% of (80 – 40) = $32.

42
 Fixed Wage Contract
❑ 1) For the fixed wage contract, what is the
expected return (net of the cost of effort) to
the manager for each effort level and which
effort level will be provided?
❑ Manager’s EV = Fixed wage – cost of
efforts.
Low effort, EVL = 15 - 0 = $15
Medium effort, EVM = 15 – 10 = $5
High effort, EVH = 15 – 30 = - $15

43
 Fixed Wage Contract
➢ Manager will choose Low Effort as this effort
level gives him the highest EV of income.
➢Given that manager chooses Low Efforts,
the firm’s expected profits,
EV = E(I) from low efforts – wage payments
= (0.5*20 + 0.5*40) – 15 = $15.
The expected total surplus for both
managers and firm together = 15 + 15 = $30

44
 Profit-sharing or Bonus
Contract?
(2) With 50% Profit sharing contract, which effort
level will be chosen by the manager?

➢ With profit sharing, manger’s

EV = 0.5(E(I)) - cost of effort.
o Low effort, EVL = 0.5*(0.5*20 + 0.5*40)–0=$15.
o Medium effort,
EVM = 0.5*(0.5*40 + 0.5*80) – 10 = $20
o High effort,
EVH = 0.5*(0.5*80 + 0.5*100) – 30 = $15
45
 Profit-sharing or Bonus
Contract?
 Manager will choose: Medium Effort
Given that manager chooses Medium Efforts,
the firm’s EV of profits,
EV = E(I) – share paid to the managers
= 0.5 *E(I)
= 0.5*(0.5*40 + 0.5*80) = $30
Thus with profit sharing total expected
surplus for both (managers and firm)
combined = 20 + 30 = $50
46
 Profit-sharing or Bonus
Contract?
(3) For the bonus contract, manager’s
EV = 0.8(the firm’s income above 40) - the
cost of effort.

o Low effort, EVL = 0.

o Medium effort, EVM = 0.8(0.5(40-40) +
0.5(80 – 40)) – 10 = 16 - 10 = $6
o High effort, EVH= 0.8(0.5(80 – 40) + 0.5(100
- 40)) – 30 = 40 – 30 = $10

47
 Profit-sharing or Bonus
Contract?
 Manager will choose: High Effort.
Given that manager chooses High Effort, firm
EV of profits = E(I) – bonus payments to
managers = (0.5*80 + 0.5*100) – 40 = $50
Note that bonus payment = 0.8[0.5(80-40) +
0.5(100-40)] = $40.
➢ With this, total surplus = 10 + 50 = $60.

48
 The Solution
(1) With a fixed wage: The manager chooses low
effort and gets $15 and the firm make $15.
Total surplus = $30.
(2) With profit-sharing: The manager puts
medium effort and gets $20. The expected
profits for the firm = $30. Total = $50.
(3) With bonus contract: The manager puts high
effort and gets $10 and the expected profits
for the firm = $50. Total = $60

49
 Comments
❑ Since bonus contract gives the highest
expected profits for the firm, it will choose
bonus contract.
❑ More generally, even high powered contracts
might not be able to generate full efficiency.

50
 Quiz 3
Assume the fixed wage contract provides a
salary of $10. The manager’s net return, Y, is
this salary minus the cost of effort.
a) With high effort the manager gets the
highest net return.
b) If the manager maximizes the net return,
we expect low effort to be chosen.
c) Medium effort would generate losses for
the manager.

51
 Quiz 4
Assume the bonus contract pays the manager
50% of the firm’s income above 40. If the firm
revenue is $50 with p = 0.6 and $70 with p = 0.4
under the manager’s chosen level of effort, the
firms expected profits is:
a) $58
b) $40.
c) $49
d) 9

52
 Quiz 5
Bad Luck (40%) Good Luck (60%)
Low Effort (0) 10 40
High Effort (10) 40 80
a) With a fixed wage of 10 the expected
return to the owner exceeds 20.
b) If net income is shared 50-50, the owner’s
expected return exceeds 30.
c) All of the above.
d) None
53