Вы находитесь на странице: 1из 38

Eco 525: Financial Economics I

Lecture 02: Risk Preferences and


Savings/Portfolio Choice
Prof. Markus K. Brunnermeier

21:58 Lecture 02 Risk Preferences Portfolio Choice Slide 2-1


Eco 525: Financial Economics I

State-by-state Dominance
- State-by-state dominance incomplete ranking
- riskier
Table 2.1 Asset Payoffs ($)

t=0 t=1
Cost at t=0 Value at t=1
1 = 2 =
s=1 s=2
investment 1 - 1000 1050 1200
investment 2 - 1000 500 1600
investment 3 - 1000 1050 1600

- investment 3 state by state dominates 1.


21:58 Lecture 02 Risk Preferences Portfolio Choice Slide 2-2
Eco 525: Financial Economics I

State-by-state Dominance (ctd.)

Table 2.2 State Contingent ROR (r )

State Contingent ROR (r )


s=1 s=2 Er
Investment 1 5% 20% 12.5% 7.5%
Investment 2 -50% 60% 5% 55%
Investment 3 5% 60% 32.5% 27.5%

- Investment 1 mean-variance dominates 2


- BUT investment 3 does not m-v dominate 1!

21:58 Lecture 02 Risk Preferences Portfolio Choice Slide 2-3


Eco 525: Financial Economics I

State-by-state Dominance (ctd.)


Table 2.3 State Contingent Rates of Return

State Contingent Rates of Return


s=1 s=2
investment 4 3% 5%
investment 5 3% 8%
1 = 2 =
E[r4] = 4%; 4 = 1%
E[r5] = 5.5%; 5 = 2.5%

- What is the trade-off between risk and expected return?


- Investment 4 has a higher Sharpe ratio (E[r]-rf)/ than investment 5
for rf = 0.
21:58 Lecture 02 Risk Preferences Portfolio Choice Slide 2-4
Eco 525: Financial Economics I

Stochastic Dominance

Stochastic dominance can be defined


independently of the specific trade-offs (between
return, risk and other characteristics of probability
distributions) represented by an agent's utility
function. (risk-preference-free)
Less demanding than state-by-state dominance

21:58 Lecture 02 Risk Preferences Portfolio Choice Slide 2-5


Eco 525: Financial Economics I

Stochastic Dominance
Still incomplete ordering
More complete than state-by-state ordering
State-by-state dominance stochastic dominance
Risk preference not needed for ranking!
independently of the specific trade-offs (between return, risk and other
characteristics of probability distributions) represented by an agent's
utility function. (risk-preference-free)
Next Section:
Complete preference ordering and utility
representations
Homework: Provide an example which can be ranked
according to FSD , but not according to state dominance.
21:58 Lecture 02 Risk Preferences Portfolio Choice Slide 2-6
Eco 525: Financial Economics I
Table 3-1 Sample Investment Alternatives

States of nature 1 2 3
Payoffs 10 100 2000
Proba Z 1 .4 .6 0
Proba Z 2 .4 .4 .2
EZ 1 = 64, z 1 = 44
EZ 2 = 444, z 2 = 779
Pr obability
F1
1.0

0.9
F2
0.8

0.7

0.6

0.5
F 1 and F 2
0.4
0.3

0.2

0.1
Payoff
0 10 100 2000
21:58 Lecture 02 Risk Preferences Portfolio Choice Slide 2-7
Eco 525: Financial Economics I

First Order Stochastic Dominance


Definition 3.1 : Let FA(x) and FB(x) , respectively,
represent the cumulative distribution functions of two
random variables (cash payoffs) that, without loss of
generality assume values in the interval [a,b]. We say
that FA(x) first order stochastically dominates (FSD)
FB(x) if and only if for all x [a,b]
FA(x) FB(x)

Homework: Provide an example which can be ranked


according to FSD , but not according to state dominance.
21:58 Lecture 02 Risk Preferences Portfolio Choice Slide 2-8
Eco 525: Financial Economics I

First Order Stochastic Dominance


1

0.9

0.8 FB

0.7 FA

0.6

0.5

0.4

0.3

0.2

0.1

0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
X

21:58 Lecture 02 Risk Preferences Portfolio Choice Slide 2-9


Eco 525: Financial Economics I

Table 3-2 Two Independent Investments

Investment 3 Investment 4
Payoff Prob. Payoff Prob.
4 0.25 1 0.33
5 0.50 6 0.33
12 0.25 8 0.33
1

0.9

0.8

0.7
0.6 investment 4
0.5

0.4

0.3
0.2
investment 3
0.1

0
0 1 2 3 4 5 6 7 8 9 10 11 12 13

Figure 3-6 Second Order Stochastic Dominance Illustrated

21:58 Lecture 02 Risk Preferences Portfolio Choice Slide 2-10


Eco 525: Financial Economics I

Second Order Stochastic Dominance


Definition 3.2: Let FA ( ~x ) , FB ( ~x ) , be two
cumulative probability distribution for
random payoffs in [a, b]. We say that FA ( ~x )
second order stochastically dominates
(SSD) FB ( ~x ) if and only if for any x :

[ FB (t) - FA (t) ] dt 0
x

-
(with strict inequality for some meaningful
interval of values of t).
21:58 Lecture 02 Risk Preferences Portfolio Choice Slide 2-11
Eco 525: Financial Economics I

Mean Preserving Spread


xB = xA + z (3.8)
where z is independent of xA and has zero mean

for normal distributions fA (x)

f B (x)

~
x , Payoff
= x fA (x)dx = x f B (x)dx

Figure 3-7 Mean Preserving Spread


21:58 Lecture 02 Risk Preferences Portfolio Choice Slide 2-12
Eco 525: Financial Economics I

Mean Preserving Spread & SSD


Theorem 3.4 : Let FA() and FB () be two distribution
functions defined on the same state space with identical
means. Then the follow statements are equivalent :

FA ( ~
x ) SSD FB ( ~ x)
x ) is a mean preserving spread of FA ( ~
FB ( ~ x)
in the sense of Equation (3.8) above.

21:58 Lecture 02 Risk Preferences Portfolio Choice Slide 2-13


Eco 525: Financial Economics I

Expected Utility & Stochastic Dominance

Theorem 3. 2 : Let FA ( ~x ) , FB ( ~x ) , be two cumulative


probability distribution for random payoffs ~ x [a , b].
Then FA ( ~x ) FSD FB ( ~x ) if and only if
for all non decreasing utility functions U().
E A U(~
x) E B U(~
x)

21:58 Lecture 02 Risk Preferences Portfolio Choice Slide 2-14


Eco 525: Financial Economics I

Expected Utility & Stochastic Dominance


Theorem 3. 3 : Let FA ( ~x ) , FB ( ~x ) , be two cumulative
probability distribution
for random payoffs ~ x defined on [a , b] .
Then, FA ( ~x ) SSD FB ( ~x ) if and only if E A U(~x) E B U(~x)
for all non decreasing and concave U.

21:58 Lecture 02 Risk Preferences Portfolio Choice Slide 2-15


Eco 525: Financial Economics I

Arrow-Pratt measures of risk aversion


and their interpretations

absolute risk aversion = - U" (Y) R A (Y)


U' (Y)

Y U" (Y)
relative risk aversion =- R R (Y)
U' (Y)

risk tolerance =
21:58 Lecture 02 Risk Preferences Portfolio Choice Slide 2-16
Eco 525: Financial Economics I

Absolute risk aversion coefficient

Y+h
Y

1 Y-h

21:58 Lecture 02 Risk Preferences Portfolio Choice Slide 2-17


Eco 525: Financial Economics I

Relative risk aversion coefficient

Y(1+)
Y

1 Y(1-)

Homework: Derive this result.

21:58 Lecture 02 Risk Preferences Portfolio Choice Slide 2-18


Eco 525: Financial Economics I

CARA and CRRA-utility functions


Constant Absolute RA utility function

Constant Relative RA utility function

21:58 Lecture 02 Risk Preferences Portfolio Choice Slide 2-19


Eco 525: Financial Economics I

Investor s Level of Relative Risk Aversion

1
( Y + CE) ( Y + 50,000 )1 12 ( Y + 100,000 )1
1
2
= +
1- 1- 1-

=0 CE = 75,000 (risk neutrality)


=1 CE = 70,711
Y=0 =2 CE = 66,246
=5 CE = 58,566
= 10 CE = 53,991
= 20 CE = 51,858
= 30 CE = 51,209

Y=100,000 =5 CE = 66,530
21:58 Lecture 02 Risk Preferences Portfolio Choice Slide 2-20
Eco 525: Financial Economics I

Risk aversion and Portfolio Allocation


No savings decision (consumption occurs only at t=1)
Asset structure
One risk free bond with net return rf
One risky asset with random net return r (a =quantity of risky assets)

21:58 Lecture 02 Risk Preferences Portfolio Choice Slide 2-21


Eco 525: Financial Economics I

Theorem 4.1: Assume U'( ) > 0, and U"( ) < 0 and let
denote the solution to above problem. Then
a > 0 if and only if E~r > rf
a = 0 if and only if E~r = r f

a < 0 if and only if E~r < rf .

Define W(a ) = E{U (Y0 (1 + rf ) + a (~r rf ))}. The FOC can


then be written W(a ) = E[U(Y0 (1 + rf ) + a (~r rf ))(~r rf )] = 0 .
By risk aversion (U''<0), W(a ) = E[U(Y0 (1 + rf ) + a (~r rf ))(~r rf )2 ]
< 0, that is, W'(a) is everywhere decreasing. It follows that
will be positive if and only if W(0) = U(Y0 (1 + rf ))E(~r rf ) > 0
(since then a will have to be increased from its value of 0 to
achieve equality in the FOC). Since U' is always strictly
positive, this implies a > 0 if and only if E(~r rf ) > 0 . W(a)
The other assertion follows similarly. a
21:58 Lecture 02 Risk Preferences Portfolio Choice 0 Slide 2-22
Eco 525: Financial Economics I

Portfolio as wealth changes

Theorem 4.4 (Arrow, 1971): Let be the


solution to max-problem above; then:

(i)

(ii)

(iii) .

21:58 Lecture 02 Risk Preferences Portfolio Choice Slide 2-23


Eco 525: Financial Economics I

Portfolio as wealth changes


Theorem 4.5 (Arrow 1971): If, for all wealth levels Y,

(i)

(ii)

(iii)

where O = da/a / dY/Y (elasticity)

21:58 Lecture 02 Risk Preferences Portfolio Choice Slide 2-24


Eco 525: Financial Economics I

Log utility & Portfolio Allocation


U(Y) = ln Y.

2 states, where r2 > rf > r1

Constant fraction of wealth is invested in risky asset!

21:58 Lecture 02 Risk Preferences Portfolio Choice Slide 2-25


Eco 525: Financial Economics I

Portfolio of risky assets as wealth changes


Now -- many risky assets
Theorem 4.6 (Cass and Stiglitz,1970). Let the vector
1 ( Y0 )
. denote the amount optimally invested in the J risky assets if

.
1 ( Y0 ) a1
J 0
( Y ) . .
the wealth level is Y0.. Then = f ( Y0 )
. .

J ( Y0 ) a J
if and only if either

(i) U ' ( Y0 ) = ( Y0 + ) or
(ii) U ' ( Y0 ) = e Y0 .

In words, it is sufficient to offer a mutual fund.


21:58 Lecture 02 Risk Preferences Portfolio Choice Slide 2-26
Eco 525: Financial Economics I

LRT/HARA-utility functions
Linear Risk Tolerance/hyperbolic absolute risk aversion

Special Cases
B=0, A>0 CARA
B 0, 1 Generalized Power
B=1 Log utility u(c) =ln (A+Bc)
B=-1 Quadratic Utility u(c)=-(A-c)2
B 1 A=0 CRRA Utility function
21:58 Lecture 02 Risk Preferences Portfolio Choice Slide 2-27
Eco 525: Financial Economics I

Prudence and Pre-cautionary Savings


Introduce savings decision
Consumption at t=0 and t=1
Asset structure
NO risk free bond
One risky asset with random gross return R

21:58 Lecture 02 Risk Preferences Portfolio Choice Slide 2-28


Eco 525: Financial Economics I

Prudence and Savings Behavior


Risk aversion is about the willingness to insure
but not about its comparative statics.
How does the behavior of an agent change when
we marginally increase his exposure to risk?
An old hypothesis (going back at least to
J.M.Keynes) is that people should save more now
when they face greater uncertainty in the future.
The idea is called precautionary saving and has
intuitive appeal.
21:58 Lecture 02 Risk Preferences Portfolio Choice Slide 2-29
Eco 525: Financial Economics I

Prudence and Pre-cautionary Savings


Does not directly follow from risk aversion alone.
Involves the third derivative of the utility function.
Kimball (1990) defines absolute prudence as
P(w) := u'''(w)/u''(w).
Precautionary saving if any only if they are prudent.
This finding is important when one does comparative
statics of interest rates.
Prudence seems uncontroversial, because it is weaker
than DARA.

21:58 Lecture 02 Risk Preferences Portfolio Choice Slide 2-30


Eco 525: Financial Economics I

Pre-cautionary Saving

(+) (-) in s

Is saving s increasing/decreasing in risk of R?


Is RHS increasing/decreasing is riskiness of R?
Is U() convex/concave?
Depends on third derivative of U()!

N.B: For U(c)=ln c, U(sR)R=1/s does not depend on R.


21:58 Lecture 02 Risk Preferences Portfolio Choice Slide 2-31
Eco 525: Financial Economics I

Pre-cautionary Saving
2 effects: Tomorrow consumption is more volatile
consume more today, since its not risky
save more for precautionary reasons
~ ~
R R
Theorem 4.7 (Rothschild and Stiglitz,1971) : Let A , B
be
~ two~ return distributions with identical means such that
RB = RA + e, (where e is white noise) and let s and s be,
A B
respectively, the savings out of Y0 corresponding to the
~ ~
return distributions R A
and R .B
If R ' R ( Y ) 0 and RR(Y) > 1, then sA < sB ;
If R ' R ( Y ) 0 and RR(Y) < 1, then sA > sB

21:58 Lecture 02 Risk Preferences Portfolio Choice Slide 2-32


Eco 525: Financial Economics I

Prudence & Pre-cautionary Saving


U ' ' ' (c)
P(c) =
U ' ' (c)

cU ' ' ' (c)


P(c)c =
U ' ' (c)
~ ~
Theorem 4.8 : Let A ,R B be two return distributions such
R
~ ~
that R A SSD R B , and let sA and sB be, respectively, the
savings out of Y0 corresponding to the return distributions
~ ~
R A and R B . Then,
sA sB iff cP(c) 2, and conversely,
sA < sB iff cP(c) > 2
21:58 Lecture 02 Risk Preferences Portfolio Choice Slide 2-33
Eco 525: Financial Economics I

Joint saving-portfolio problem


Consumption at t=0 and t=1. (savings decision)
Asset structure
One risk free bond with net return rf
One risky asset (a = quantity of risky assets)

max U ( Y0 s ) + EU ( s (1 + rf ) + a ( ~r rf )) (4.7)
{ a ,s }
FOC:
s: U(ct) = E[U(ct+1)(1+rf)]
a: E[U(ct+1)(r-rf)] = 0

21:58 Lecture 02 Risk Preferences Portfolio Choice Slide 2-34


Eco 525: Financial Economics I

for CRRA utility functions

s: ( Y0 s) ( 1) + E ([s(1 + rf ) + a ( ~r rf )] (1 + rf ) ) = 0
a: [
E (s(1 + rf ) + a ( ~r rf )) ( ~r rf ) = 0 ]
Where s is total saving and a is amount invested in risky asset.

21:58 Lecture 02 Risk Preferences Portfolio Choice Slide 2-35


Eco 525: Financial Economics I

Multi-period Setting
Canonical framework (exponential discounting)
U(c) = E[ t u(ct)]
prefers earlier uncertainty resolution if it affect action
indifferent, if it does not affect action
Time-inconsistent (hyperbolic discounting)
Special case: formulation
U(c) = E[u(c0) + t u(ct)]
Preference for the timing of uncertainty resolution
recursive utility formulation (Kreps-Porteus 1978)

21:58 Lecture 02 Risk Preferences Portfolio Choice Slide 2-36


Eco 525: Financial Economics I

Multi-period Portfolio Choice

Theorem 4.10 (Merton, 1971): Consider the above canonical


multi-period consumption-saving-portfolio allocation problem.
Suppose U() displays CRRA, rf is constant and {r} is i.i.d.
Then a/st is time invariant.

21:58 Lecture 02 Risk Preferences Portfolio Choice Slide 2-37


Eco 525: Financial Economics I

Preference for the


Digression:
timing of uncertainty resolution
$100 $150

$100 $ 25
0
$150
$100
Kreps-Porteus $ 25

Early (late) resolution if W(P1,) is convex (concave)

Marginal rate of temporal substitution risk aversion


21:58 Lecture 02 Risk Preferences Portfolio Choice Slide 2-38