• Intertemporal Choice

(Gravelle Chap. 15, Pp. 406-436; Varian Chap. 19)

A. Introduction
• The theory of the consumer developed previously was related to
decisions made by economic units (consumers) in a single time
period
• It therefore took no account of saving, disaving or lending and
borrowing, which we would normally expect to be important
aspects of consumer behaviour
• Moreover, the operation of the capital market-the market for
borrowing and lending-influence an economy in important ways, so
it is useful to develop the theory of the operation of that market.
• In the theory of the firm the firm changes its scale by investing in
new capacity, and so we could view the problem of determining
long-run equilibrium output as the problem of choosing the most
profitable amount of investment to produce and so it is important to
study how such investment decisions are taken
 B. Optimal consumption over time
• We begin with the theory of consumer
• Assume time is divided into two equal discrete intervals-say into
years
• Given the consumer’s annual income, we assume that the
atemporal theory applies, and the consumer allocates this income
optimally over goods. However, he also has a further choice,
which is either to lend some of his income, or to borrow

• Lending will result in a reduction in current consumption, but an

increase in future consumption and conversely for borrowing

• Thus an analysis of lending and borrowing decisions is an analysis

of the consumer’s choice of a consumption pattern over time.
• For this analysis we make the following additional assumptions

a. Prices are given in each period and the sum of the values of the
goods could be used to get total sum of consumption expenditure
in each period (composite commodity)., rather than quantities of
particular goods and services
b.Only two time periods exist (so we can use two dimensional
analysis):
0 = present time (period)
1 = next time (period)
M0 = Present time total consumption expenditure

M1 = next time total consumption expenditure

C. There is competitive capital market, so there is a given price for
borrowing and lending, which is usually expressed as an interest
rate, r. Thus $100 borrowed at time 0 implies that $100(1+r) must
be repaid at time 1 (r is annual interest rate). So 1+r can be thought
of as the price paid for borrowing or received for lending $1.
D. There is no constraint on borrowing or lending since we assume
perfect capital market, implying that demand equals supply in the
capital market
• Given these assumptions, we can construct a model of choice of
consumption over time
• The elements of consumer’s optimisation problem are
a. choice variables M0 and M1;
b. Consumers’ have preference ordering over combinations of
current and future consumption expenditure (M0, M1),
satisfying all the axioms of consumer preferences. Therefore
u(.) and indifference curves have the usual shape, U0, U1>0;
• C. Feasible set is defined as follows
 Initial endowment with an income time-stream

M 0 , M 1 >0.

We assume at least one of them is positive.

 Let A be the amount the consumer lends or borrows
in year 0, with A>0 for borrowing and
A<0 for lending. Thus the consumer’s
feasible consumptions will be constrained by
0 < M0 < M0 +A……(B.1)
0 < M1 < M 1 +A……(b.2)
• The optimal point will be on a boundary of the feasible set. Then, solving for A in
B.2 and substituting in B.1 gives

(M1  M1)
( M0  M0 )+ = 0………………………………………………(B.3)
(1 r)
Or
M M
M0  1  M0  1 V0 ………………………………………………(B.4)
1 r 1 r
• Equation B.3 should be compared with consumer’s budget constraint in the static
model of utility maximization and
M0  M0

And

M1  M1
Represent the demand for consumption in the two periods, and 1/(1+r) represents
relative price.
• Equation B.4 is wealth constraint. These values are expressed in terms of
income at time 0 (V0 is the present value of wealth-the present value of his
endowed income time-stream).
• The economic interpretation of B.4 is that by borrowing or lending the
consumer may achieve consumption time-stream which differs from his
endowed time-stream, but in doing this he is constrained by his wealth
• Since r is constant (perfect capital market), the wealth constraint can be graphed
as a straight line (V0 on in Fig. 11.1)
• The slope of V is
0
dM
1
=
(
1
+r
) (
B
.
5
)
dM
0

since [B.4] implies the equation M1 = (1+r)V0 - (1+r)M0.

• V0 must pass through initial endowment point

M  (M 0 , M 1 )

since this point always satisfies B.4

• The wealth line V0 represents the set of market exchange
(borrowing or lending) opportunities to the consumer;
– by lending he can move leftwards from along the line;
– by borrowing he moves rightward
• Each point on V0 represents simultaneously a consumption
time-stream, an amount of borrowing or lending in year 0,
and a corresponding repayment in year 1.
• The absolute value of the slope of V0 is determined by r.
• A reduction in r leads to a flatter line, such as V′0 in the
figure.
M
• Note that this line must continue to pass through , since
the initial endowment point continues to satisfy [B.4].
• So change in r causes V0 to rotate through M
A
c
h
a
n
ge
i
na
n
i
n
it
i
a
l
en
d
o
wm
e
nt
(
r
re
m
a
in
i
n
g
c
on
s
t
an
t
)
w
il
l
c
h
an
g
et
h
e
i
nt
e
r
c
ep
t
o
fV
b
u
0t
1 1
n
o
t
t
h
e
sl
o
p
eo
f
t
h
el
i
n
e,
a
n
dt
h
el
i
n
e
sh
i
f
t
sp
a
r
a
l
l
el
t
o
V,
e
.
g
0.
,
t
oV0

1 1
T
h
i
sl
i
n
ec
o
r
r
es
p
o
n
ds
t
o
,
s
ay
,
an
i
n
c
re
a
s
eM
i
nt
o
0M
,
0o
r
a
n
i
nc
r
e
aM
s
e
i
nt
o
1M
o
1r
i
n
t
e
r
me
d
ia
t
e
i
nc
r
e
a
se
s
i
n
bo
t
h
•The new wealth line must pass through the new initial endowment
point.
•Given our assumptions about preferences, we expect to obtain a
tangency solution to the consumer’s optimization problem such as
that in Fig. 11.2 at M*.
•(Why is it reasonable to assume that we would not have a corner
solution?)
•The consumer’s chosen consumption time-stream at (M*0, M*1) is
achieved by borrowing an amount
• It would be possible for the indifference curves, initial endowments,
or interest rate to be such that the optimal point implies lending

(
M
*
t
ot
h
e
le
f
t
o
fM
o
nV
)
,
o
0r
n
e
i
t
he
r
b
o
rr
o
w
i
n
gn
o
r
*
L
e
n
d
i
nM
g
(c
o
in
c
i
d
e
sw
i
t
hM
)

• As before the Slope of the indifference curve is the negative of the

ratio of marginal utilities
U
0

,
U
1

where U0 is the marginal utility of period 0 consumption

• At the optimum, therefore, using [B.5]

U
0
1
r
…
………
…
…
…
……
…
…
……
…
…
…
……
…
…
….
(
B.
6
)
U
1
• Reducing M0 by $1 reduces U by U0
• There will exist an increase in M1 which will make the consumer just
as well off as before the $1 reduction in M0
• This compensating increase in M1 is $(1+p) and is defined by

U
0(
M 0,M
1)
U1(
M 0,M
1)(
1
p)…
………
………
………
………
………
….(B
.7
)

• where the notation shows that U0 and U1 depend on M0 and M1

• In words, raising M1 by $1 increases U by the marginal utility U1 so
that increasing M1by $(1+p) will raise U by U1(1+p) and this just
offsets the effect of the $1 reduction in M0
• Since U0 and U1>0, we must have 1+p >0 so that p>-1, p can be
interpreted as the consumer’s subjective rate of interest since it shows
how much extra consumption in period 1 is required to compensate
for the loss of $1 of the current consumption. P may be negative if
less than $1 extra of M1 is required.
• It is subjective because it depends on the consumer’s preferences, not on
observable market phenomena. Since u1 and u0 depend on M1 and M0 so
must p: p = p(M0,M1).
• Rearranging B.7, we get
U
0
1
…
…
p…
……
…
…
…
……
…
…
……
…
…
…
……
…
…
(B
.
8
)
U
1

• As we move along indifference curve from left to right substituting M 0 for

M1, p will decline since the slope of indifference curve gets flatter, and
current consumption relatively less valuable
• P is also known as the consumer’s rate of time preference
• We can use B.8 to write the optimum condition in B.6 simply
as
P = p(M0, M1) = r……………………………(B.9)
• B.9 or B.6 together with the constraint B.4 provide two
equations to determine the optimal M1* and M0*.

• The consumer is in equilibrium when p equals r(the

market rate of interest).
• In other words, the consumer lends up to the point
at which r is just sufficient to compensate for the
marginal reduction in current consumption.

• Alternatively, he borrows until he reaches the point

at which the price he must pay (in terms of reduced
consumption next period) is just sufficient to offset
the value to him of the additional consumption this
period
 c. The optimal investment decision
To analyse the optimal investment decision we assume the
following
• A firm is a single decision taker with available
specific set of productive investment opportunities;
• The firm has access to capital market for borrowing
or lending;
• There are two periods, 0 and 1;
• The firm is the only income source for the owner;
• The firm owner solves three problems: (1)
investment problem (r= i); (2) production problem
(MVPz =wz); and (3) consumption problem (ux = px)
• Since there is borrowing or lending, consumption in each period need no equal the
cash income generated by the firm;
• Consumption time-stream is given by M0 and M1;
• Feasible set in this case differs from that of section B in that it is now determined by
the technological possibilities of production and investment as well as by r (not by
endowment);
Production and investment possibilities
Let
• D0 = cash flow received by the owner from the firm (revenue
minus expenditure) in period 0;
•  D1 = Revenue minus expenditure in period 1;
•  L0 and L1(labour) is the only variable input in period 0 and period 1,
respectively;
• K0 = stock of capital input in period 0 inherited at the beginning of
period 0;
• K1 = Capital input in period 1;
 Cash flow equations
• D0 = Pf(L0,k0)-wL0-Pk(K1-K0) = Pf(L0,k0)-wL0-I……(C.1)
• D1 = Pf(L1,k1)-wL1………………………………………(C.2)
• where P, w, and Pk are the competitive market prices of the firm’s output, labour and
capital, respectively; f(Lt,Kt) is production function of the firm.
• I =Pk(K1-K0) is the expenditure on investment in period 0
• Assume also all prices are constant over time and there is no depreciation of the
capital stock
• If K1<K0, the firm is disinvesting, i.e., selling off its capital to increase its cash flow
in the first period;
• The firm will choose the labour input to maximize the cash flow in each period for
given levels of the capital stock, since choice of Lt affects the cash flow of that
period and the owner always prefers higher cash flow to lower in a period given that
the cash flow of the other period is unaffected;
• Given the optimal choice of the variable input Lt (by separately maximizing C.1 and
C.2) in each period and the fixed initial capital stock K0, the cash flows in each
period depend only on the optimal stock K1 chosen for period 1
 * *
D
0=Pf(L,K
0 0)wL
0Pk(
K 1K0)D0(
K 1)…
………
………
………
…..(C
.3
)
* *
D
=
1 P
f
(L,K)
1 1wL
1D
(
1K)
1……
……
……
……
……
……
……
……
….
… (
C.4
)

• Since both periods’ cash flows depend on K1 we can derive a

relationship between D0 and D1 by varying K1 (i.e., by investing or
disinvesting).
• Increasing K1 reduces D0 by the rate Pk. L1*will depend on K1 (L0*
does not since the marginal cash flow from variations in L0 is

• ) since the marginal period 1 cash flow from

variations in L1 is
Pf ( L1 , K 1 )
 w
L1

• which is affected by K .
1
• Hence

dD1 f ( L*1 , K 1 ) f ( L*1 , K 1 ) dL*1 f ( L*1 , K 1 )

P  (P  w) P
dK 1 K 1 L1 dK 1 K 1
• where we have used the fact that L*1 maximizes D1 for a given
K1 and so the marginal period 1 cash flow from L1 is zero at
L1*(example of Envelope theorem).
• Increasing K1 will therefore reduce D0 and increase D1 (as long
as the marginal product of capital is positive in period 1).
• Figure 11.3 plots feasible combinations of cash flows that the
owner of the firm can receive by varying his investment
decisions, i.e., by altering K1 so moving along the curve PP
• By investing or disinvesting he moves along PP (investment possibility curve).

D  ( D0 , D1 )
is assumed to be the cash flow time-stream the firm will generate if it neither
invests nor disinvests, that is, K1 = K0 and I = 0
• By increasing K1, say to K11 through investing I1 =Pk(k11-k0) (the cash flow of the
first period is reduced to D01 = D0-I* and the next period’s cash flow increased to
D1 1

Borrowing and investing

• The owner of the firm has access to a capital market on which he
can borrow or lend at the interest rate r-so his consumption
expenditure time-stream (M0,M1) may differ from the cash flow
time-stream (D0.D1) he receives from the firm
• Figure 1.4 combines wealth line with the curve PP from Figure 11.3
• It shows the feasible set when there are possibilities of
altering the time pattern of consumption by both
production and capital market activities, i.e., by moving
along PP (investing or disinvesting) and along wealth
line (borrowing or lending).
• On Figure 1.4 by investing or disinvesting, he can move
along PP and achieve different combinations of cash
flows
• Given his investment decision (choice of D0 and D1) he
can enter the capital market and trade (borrow or lend)
to any point along the wealth line through his cash flow
combination
• For example, if he neither invests nor disinvests, he will
be at point
D
o
n
P
P
• He could then lend out some of his period 0 cash flow and move
along the wealth line V to a point such as M1 where he is better
off (he is now on a higher indifference curve than

U
D
t
h
r
o
u
g
h)

• Higher wealth line is associated with higher consumption.

• Optimal investment decision (choice of K1 or equivalently choice
of D0, D1) will be that which maximizes the owner’s wealth
• On Fig. 1.4 this is the cash flow combination D 0*, D1* achieved by
choosing a second period optimal stock of capital k1* and investing
I* = Pk(K1*-K0)in the first period. V* is the highest possible wealth
line attainable by investment along PP and so by investing I* the
owner is put in the best possible position for engaging in the capital
market.
• Given the optimal wealth-maximizing investment and
production decisions, he chooses some combinations of
consumption expenditures in the two periods (M0, M1) along
the maximum wealth line V*.
• Fig.11.4 shows two possible final equilibrium positions
• Given the pattern of preferences given by U*, overall optimal
position is at M* = (M0*, M1*).
• M* is reached in two steps: first investing I* to get D*; and
second borrowing M0*-D0* to repay D1*-M1* = (1+r)(M0*-
D0*) out of the cash flow in period 1
• Generally the first step is a solution to the investment decision
problem and the second step is a solution to the consumption
decision problem
• On the other hand, given U** the overall optimal solution is M**
= (M0**, M1**), again reached in two steps: First the firm invests I*
and second the owner lends D0*-M0** to receive M1**-D1* = (1+r)
(D0*-M0**) in period 1 to add to period 1 cash flow
• There is a third kind of solution possibility, which would arise if
an indifference curve happened to be tangent to V* at D*, in
which case I* would still solve the investment decision problem,
with no lending or borrowing on the capital market and D*=M*.
• This implies three types of solutions to the firm’s problem of
choosing an optimal consumption time-stream: (1) borrowing,
(2) lending, and (3) neither.
• In each case the solution to the investment decision problem is the
same and is independent of the pattern of preferences (separation
of investment decision from consumption decision because of the
existence of capital market) with the investment decision
requiring information only on investment opportunities (the curve
PP) and interest rate, r.
• Note that the negative of 1+r is the slope of wealth line (e.g., V*)
and the negative of 1+i is the slope of PP. At D* 1+r = 1+i, i.e.,
the firm invests up to the point where the marginal rate of
return on investment (i) equals the market rate of interest (r).
• Up to D*, i>r but after D* i< r.
• D* can be obtained by solving
• Max V = (D1/(1+r))+D0
(D1,D0)
 
Subject to D1 = P(D0)
 
Where p is a function representing the curve PP
• The first order conditions are (for D0>0)
dD1/dD0 = -(1+i) = -(1+r)
at the optimal point (D0*,D1*)

• Thus we have the important result that the optimal investment

decision is solved by maximizing the present value of the
income-stream generated by investment, or equivalently the
owner’s wealth, over the available set of income streams

• For this no information on preferences is necessary

• Note also that this solution will determine the capital stock,
labour and output produced by the firm in period 1
Intertemporal optimisation with several periods (T>2) 
• a. Consumption

M
ax
U(
x) s
.
t.
1
P(
tjX 
tj X)
tjR
0
j t

Where
• X is a vector of consumption goods, X = (x01,x02,…,xTj).
X01 = consumption of good 1 in year 0
• P1tj is the present value of the price of good j (at period 0) in
year t and
1
P
tj 1
P=
tj , t 1,.....
T, P0j P
0j
(1r0)(
1r1)...(
1rt1)
• R0 is a given sum of money at t = 0 determined by past
decisions;

i
s
xi
n
i
t
i
a
l
e
n
d
tjo
w
m
en
t
o
f
g
o
o
d
ji
n
y
e
a
r
t
;

r
t

1i
si
n
t
e
r
e
s
tr
a
t
e
e
xp
e
c
t
e
d
t
op
r
e
v
a
i
li
n
p
e
r
i
od
t
-
1

• This maximization problem uses the same method as

maximizing utility within a time period.

• Equilibrium solution determines not only the quantities of

consumption goods but also a pattern of lending and
borrowing over time, given we know Ptj, r0, rt and rT-1
 • b. Firm’s maximization of wealth by maximizing present
value of its cash flow (This is the same as maximizing the
present value of firm’s profit)

( PI )
t tk t
Max V= t
 P1
Y
tj tj P1
tkIt)
(1r0)(
1r1...(
1rt1) t j

S.t. gt(yt,kt) = 0 (production constraint) and

Kt-1(1-)+It-1=Kt (capital constraint)
• The Lagrangean function is
L=V+tgt (Yt,Kt)t (Kt1(1)It1 Kt)
t t
 Where
• gt represents the implicit production function facing the firm;
• Ytj is the firm’s net output of good j in year t with Ytj<0 for non
capital input and Ytj>0 for output;
• It is the firm’s acquisition of the capital good in year t;
• Kt is the firm’s capital stock in year t and  is the capital stock
depreciation rate;
• Yt is the vector of non-capital goods in year t. and
 
 t  P Y
tj =
tj fi
rm ’
s to
talp
rofi
t i
n y
ea r t
.
j
• The first-order conditions are:


 Ptj1   g tj  0 , t  1,..., T , j  1,... j
 Y tj

  g tk   t   t  1 (1   )  0 , t  1,..., T
K t

 Ptk1   t  1  0 , t  0 ,1,..., T  1
I t
where

g t g t
g tj  , and g tk 
Ytj k t
• Notice there is no investment in year T since there is no production after
• These conditions together with the constraints determine the firm’s production plan
from year 0 to year T, including its purchase of investment goods and the evolution
of its capital.

D. Capital Market Equilibrium

• In short capital market equilibrium is said exist if demand for capital equals the
supply of capital, both of which are functions of interest rate
• The equilibrium interest rate is given by r*.
• If r is greater than r*, demand for capital will be less than its supply.
• If r is less than r*, demand for capital exceeds its supply..
• Let Aj = Aj(r) be individual j’s well behaved net demand function for capital in
period 0 as a function of interest rate, r, where dAj(r)/dr<0.
• It follows that an equilibrium interest rate (r*) satisfies

J
A
(
r
*)
=A
(
jr*)
0
,
j
j

1
=1
,
2
,…,
Jin
d
i
vi
d
ua
ls
• In other words, an equilibrium interest rate is found at the point where aggregate supply
(lending) equals demand (borrowing)

• A(r) can be regarded as the horizontal sum of individual curves showing lending and
borrowing as functions of interest rate
• The relationship between r and borrowing or lending is derived from optimisation
problems of firms and consumer’s demand for current consumption, M0* and future
consumption, M1*, leads to net demand for current funds
M
(
M
*
-
)
0
0
• The problem of the consumer is given by
M a x u ( M 0 ,M 1 )

S .T . v 0  M 0  M 1  M 0  M 1 (w e a lth c o n s tr a in t)

L a g ra n g e

L = u ( M 0 ,M 1 ) +  ( M 0  M 1  M 0  M 1)

FO C

M t* = M t* (  , v 0 )

• It is possible to derive the Marshalian demand (Mt) and the

Hicksian demand (ht) for consumption in period t and the
impacts of r and  (discount factor=1/(1+r) on these demands
 Choice Under Uncertainty
• Until now, we have been concerned with the behavior of a consumer
under conditions of certainty.
• However, many choices made by consumers take
• place under conditions of uncertainty.
• In this section we explore how the theory of consumer choice can be
used to describe such behavior.
• Distinctions between certainty and risk
• Uncertainty: the lack of complete certainty, that is, the
existence of more than one possibility. The true outcome
/state/result/value is not known.
• Risk: A state of uncertainty where some of the possibilities
involve a loss, catastrophe or other undesirable outcome
with some subjective/objective probabilities (known).E.g.,
there is a 40% chance that it will rain today.
• Before we apply consumer theory to describe behaviour under
uncertainty, we first describe the set of choices facing the consumer
unde uncertainty.
• We shall imagine that the choices facing the consumer take the form
of lotteries.

• This notation means: "the consumer receives prize x with

probability p and prize y with probability (1 - p)."
• The prizes may be money, bundles of goods, or even further
• lotteries.
• Most situations involving behavior under risk can be put into this
lottery framework.
• We will make several assumptions about the consumer‘
perception of the lotteries open to him.
• Assumption (L3), sometimes called "reduction of compound lotteries," is somewhat
suspect since there is some evidence to suggest that consumers treat compound
lotteries different than one-shot lotteries.
• However, we do not pursue this point here.
• (Note that the left hand side of (L3) is compound lottery while the right hand side is
its reduced version
• Under these assumptions we can define L, the space of lotteries available
• to the consumer.
• The consumer is assumed to have preferences on this lottery space: given any two
lotteries, he can choose between them.
• As usual we will assume the preferences are complete, reflexive, and
transitive.
• The fact that lotteries have only two outcomes is not restrictive since
we have allowed the outcomes to be further lotteries.
• This allows us to construct lotteries with arbitrary numbers of prizes
by compounding two prize lotteries.
• For example, suppose we want to represent a situation with three
prizes x, y, and z where the probability of getting each prize is one
third.
• By the reduction of compound lotteries, this lottery is equivalent to the
lottery

• According to assumption L3 above, the consumer only cares about the

net probabilities involved, so this is indeed equivalent to the original
lottery.
 Expected utility

• Under some additional hypotheses, we can find a particular

monotonic transformation of the utility function that has a very
convenient property, the expected utility property:
• The expected utility property says that the utility of a lottery is the
expectation of the utility from its prizes.
• We can compute the utility of any lottery by taking the utility that
would result from each outcome, multiplying that utility times the
probability of occurrence of that outcome, and then summing over the
outcomes.
• Utility is additively separable over the outcomes and linear in the
probabilities.
• What is of interest here is the existence of a utility function with the
above convenient property, not the mere existence of utility function.
• For that we need these additional axioms:
• Assumption (Ul) is an assumption of continuity
• Assumption (U2) says that lotteries with indifferent prizes are
indifferent.
• In order to avoid some technical details we will make two further
assumptions.

• Assumption (U3) is purely for convenience.

• Assumption (U4) can be derived from the other axioms. It just says
that if one lottery between the best prize and the worst prize is
preferred to another it must be because it gives higher probability of
getting the best prize.
• Under these assumptions we can state the main theorem.

Expected utility theorem. If (L, ) satisfy the above

axioms, there is a utility function u defined on L that satisfies the
expected utility property:

• Uniqueness of the expected utility function

• We have now shown that there exists an expected utility function u:
L R.
• Of course, any monotonic transformation of u will also be a utility
function that describes the consumer's choice behavior.
• But will such a monotonic transform preserve the expected utility
property?
• Does the construction described above characterize expected utility
functions in any way?
 Uniqueness of expected utility function.
• An expected utility function is unique up to an affine transformation.
Other notations for expected utility
• We have proved the expected utility theorem for the case where there
are two outcomes to the lotteries.
• As indicated earlier, it is straightforward to extend this proof to the
case of a finite number of outcomes by using compound lotteries.
• If outcome xi is received with probability p, for i =
• I , . . . , n, then the expected utility of this lottery is simply

• Subject to some minor technical details, the expected utility theorem

also holds for continuous probability distributions.
• If p(x) is a probability density function defined on outcomes x, then
the expected utility of this gamble can be written as
• We can subsume both of these cases by using the expectation
operator.
• Let X be a random variable that takes on values denoted by x.
• Then the utility of X is also a random variable, u(X).
• The expectation of this random variable, Eu(X) is simply the expected
utility associated with the lottery X.
• In the case of a discrete random variable, Eu(X) is given by (11.2),
and in the case of a continuous random variable Eu(X) is given by
(11.3).

Risk aversion
• Let us consider the case where the lottery space consists solely of
gambles with money prizes.
• We know that if the consumer's choice behavior satisfies the various
required axioms, we can find a representation of utility that has the
expected utility property.
• For example, to compute the consumer's expected utility of a gamble

we just look at

• This construction is illustrated in Figure 11.1 for p = 1/2.

• Notice that in this example the consumer prefers to get the expected
value of the lottery.
• That is, the utility of the lottery

is less than the utility of the expected value of the lottery, px + (1 - p)y.
• Such behavior is called risk aversion.
• A consumer may also be risk loving; in such a case, the consumer
prefers a lottery to its expected value.
• If a consumer is risk averse over some region, the chord drawn
between any two points of the graph of his utility function in this
region must lie below the function.

• This is equivalent to the mathematical definition of a concave

function.

• Hence, concavity of the expected utility function is equivalent to risk

aversion.

• It is often convenient to have a measure of risk aversion.

• Intuitively, the more concave the expected utility function, the more
risk averse the consumer.

• Thus, we might think we could measure risk aversion by the second

derivative of the expected utility function.
• However, this definition is not invariant to changes in the expected
utility function: if we multiply the expected utility function by 2, the
consumer's behavior doesn't change, but our proposed measure of
risk aversion does.
• However, if we normalize the second derivative by dividing by the
first, we get a reasonable measure, known as the Arrow-Pratt
measure of (absolute) risk aversion

risk aversion depends on the the magnitude of the first

and second derivatives.
• The following analysis gives further rationale for this measure.
• Let us represent a gamble now by a pair of numbers (xl, x2) where the
consumer gets X1 if some event E occurs and x2 if not-E occurs.
• Then we define the consumer's acceptance set to be the set of all gambles the
consumer would accept at an initial wealth level w.

• If the consumer is risk averse, the acceptance set will be a convex set.
• The boundary of this set-the set of indifferent gambles-can be given by an implicit
function x2(x1), as depicted in Figure 11.2.

• Suppose that the consumer's behavior can be described by the maximization of

expected utility. Then x2(x1) must satisfy the identity:

• The slope of the acceptance set boundary at (0,0) can be found by differentiating
this identity with respect to xl and evaluating this derivative at x1 = 0:
• That is, the slope of the acceptance set at (0,O) gives us the odds.
• This gives us a nice way of eliciting probabilities-find the odds at
which a consumer is just willing to accept a small bet on the event in
question.
• Now suppose that we have two consumers who have identical
probabilities on the event E.
• It is natural to say that consumer i is more risk averse than consumer j
if consumer i's acceptance set is contained in consumer j's acceptance
set.
• This is a global statement about risk aversion for it says that j will
accept any gamble that i will accept.
• If we limit ourselves to small gambles, we get a more useful measure.
• It is natural to say that consumer i is locally more risk averse than
consumer j if i's acceptance set is contained in j's acceptance set in a
neighborhood of the point (0'0).
• This means that j will accept any small gamble that i will accept.
• If the containment is strict, then i will accept strictly fewer small
gambles than j will accept.
• It is not hard to see that consumer i is locally more risk averse
than consumer j if consumer i's acceptance set is "more
curved" than consumer j's acceptance set near the point (0,0).
• This is useful since we can check the curvature of the
acceptance set by calculating the second derivative of x2(xl).
• Differentiating the identity (11.4) once more with respect to
xl, and evaluating the resulting derivative at zero, we find

Larger x’’2(0) implies taking less small gamblesmore risk

aversion
• EXAMPLE: The demand for insurance
• Suppose a consumer initially has monetary wealth W
• There is some probability p that he will lose an amount L-for
example, there is some probability his house will burn down.
• The consumer can purchase insurance that will pay him q dollars in
the event that he incurs this loss.
• The amount of money that he has to pay for q dollars of insurance
coverage is q; here  is the premium per dollar of coverage.
• If  is equal to the probability of loss, the insurance is said to be fair
insurance
• How much coverage will the consumer purchase? We look at the
utility maximization problem

• Taking the derivative with respect to q and setting it equal to zero, we

find
• If the event occurs, the insurance company receives q - q dollars.
• If the event doesn't occur, the insurance company receives  q dollars.
• Hence, the expected profit of the company is

• Let us suppose that competition in the insurance industry forces these profits to
zero. This means that

• Under the zero-profit assumption the insurance firm charges an actuarially fair
premium: the cost of a policy is precisely its expected value, so that p = .
• Inserting this into the first-order conditions for utility maximization. we find
• If the consumer is strictly risk averse so that u’’(W<) 0 (this is to avoid division by
zero in the second-order condition-see first-order condition), then the above
equation implies

from which it follows that L = q*.

• Thus, the consumer will completely insure himself against the loss L. (generally,
this is true if the consumer is risk averse and the insurance premium is fair)
• This result depends crucially on the assumption that the consumer
cannot influence the probability of loss.
• If the consumer's actions do affect the probability of loss, the insurance firms may
only want to offer partial insurance (rationing), so that the consumer will still have
an incentive to be careful.
• a model of this sort is investigated in Chapter 25, page 455. (Moral hazard)
 Global risk aversion
• The Arrow-Pratt measure seems to be a sensible interpretation of
local risk aversion: one agent is more risk averse than another if he is
willing to accept fewer small gambles.
• However, in many circumstances we want a global measure of risk
aversion-that is, we want to say that one agent is more risk averse than
another for all levels of wealth.
• What are natural ways to express this condition?
• The first plausible way is to formalize the notion that an agent with
utility function A(w) is more risk averse than an agent with utility
function B(w) is to require that

for all levels of wealth w.

• This simply means that agent A has a higher degree of risk aversion
than agent B everywhere.

• Another sensible way to formalize the notion that agent A is more risk
averse than agent B is to say that agent A's utility function is "more
concave" than agent B's.

• More precisely, we say that agent A's utility function is a concave

transformation of agent B's; that is, there exists some increasing,
strictly concave function G(.) such that

• A third way to capture the idea that A is more risk averse than B is to
say that A would be willing to pay more to avoid a given risk than B
would.
• It may seem difficult to choose among these three plausible sounding
interpretations of what it might mean for one agent to be "globally
more risk averse" than another.
• Luckily, it is not necessary to do so: all three definitions turn out to
be equivalent!
• As one step in the demonstration of this fact we need the following
result, which is of great use in dealing with expected utility functions.

• Jensen's inequality. Let X be a nondegenerate random variable and f

(X) be a strictly concave function of this random variable. Then E f
(X) < f (EX).
 Risk aversion and investment
•the investment in the risky asset will be increasing,
constant, or decreasing in wealth as risk aversion is
decreasing, constant, or increasing in wealth.