Lecture Notes Handout

Microeconomics 2 ITMI 1017

Prof. Swagata Bhattacharjee
Topic: Exchange Economy
Ref: Serrano and Feldman Ch 15; Varian Ch 32.

General Equilibrium Framework

So far, we have discussed the demand side and production side of the market in partial equilibrium framework.
A general equilibrium framework combines all the elements of an economy and analyzes the economy as a
whole. In this lecture we will study an economy with "pure exchange," where we will not consider the
production decisions. Instead, in this framework, we will characterize the idea of e¢ ciency and illustrate the
e¢ ciency of the competitive equilibria.

Two Consumer- Two Goods Economy

Let us assume there are two economic agents (Robinson Crusoe (R) and Friday (F )) and two goods: x
and y, in the economy. There is a …xed quantity of both these commodities available in the economy and
no production takes place. The two consumers each have an initial endowment of the goods: denoted by
0
wR = (x0R ; yR ) and wF = (x0F ; yF0 ); and then if they want they can trade among themselves. Let the total
quantity of x and y be respectively X and Y:

X = x0R + x0F
0
Y = yR + yF0

Starting with the initial endowments, after trade the consumers end up at an allocation (xR ; yR ) and
(xF ; yF ): Note that, at every feasible allocation in the exchange economy, the consumers consume non-
negative quantities and the supply of each good must equate the demand of that good, i.e., the allocation
(xR ; yR ) and (xF ; yF ) are feasible if:
xR ; yR ; xF ; yF 0
and

X = xR + xF
Y = y R + yF

Agents’preferences are represented by the following utility functions respectively:

f or R : uR (xR ; yR )
f or F : uF (xF ; yF )

Note that every agent’s utility depends on his own consumption and not what the other person consumes.
The fact that the consumers are inherently di¤erent and their preferences are di¤erent makes mutually
bene…cial trade possible.

Edgeworth Box:
To depict the allocations in this exchange economy, Edgeworth and Bowley devised a diagram, called an
Edgeworth box, to show all possible allocations of the available quantities of goods x and y between the two
consumers.

The dimensions of the Edgeworth Box are the total quantities of x and y:

There are two origins: the lower left corner of the box is for R; the upper right corner of the box is for
F:

1
 Every point inside the box (a; b) depicts a "feasible" allocation in the economy where R consumes:
xR = a; yR = b and F consumes: xF = X a:yF = Y b:

Thus, Edgeworth box shows four quantities: (xR ; yR ) and (xF ; yF ) in two-dimensional space.
In this Edgeworth box, now let us draw representative indi¤erence curves for the two agents. Under the
standard assumptions of continuity, monotonicity and convexity, the rational preferences are represented by
indi¤erence curves in Figure . Notice that while R0 s indi¤erence curve, convex to the lower left origin OR
looks familiar, F 0 s ICs, convex to OF look upside down. In this Figure , The endowment point is denoted
as W: At initial endowment, R0 s utility level is associated with IR and F 0 s utility is the level of IC IF :

Figure 1: Edgeworth Box

Pareto E¢ ciency
So far, in partial equilibrium analysis, we focussed on one good and de…ned the "social surplus" (the sum
of consumers’surplus and producers’surplus) as the total net bene…t created in the market for that good.
Now we need to extend the same notion in general equilibrium analysis. To do so, we use the concept of
"Pareto E¢ ciency," following the work of Vilfredo Pareto.
If A and B are two feasible allocations (i.e. each contains non-negative quantities of both the goods and
the total consumption of each good sum up to the total amount of that good available in the economy), we
say A Pareto-dominates B if both the consumers R and F like A at least as well as B: In that case, A is
Pareto-improving compared to B:
Pareto E¢ ciency: A feasible allocation is Pareto optimal or Pareto E¢ cient if there is no other feasible
allocation that both consumers like at least as well, and that is preferred by at least one of them. In other
words, an allocation is Pareto optimal if there is no Pareto-improving allocation compared to that, i.e. no
other feasible allocation makes at least one of the consumers better o¤, without making any of them worse
o¤.
Now let us see how we can …nd out Pareto optimal allocations. In Figure 1, the initial endowment point is
not Pareto optimal, because any of the points in the area enclosed by the two ICs IR and IF pareto dominates
this point. At any of these points, both R and F are better o¤ (at a higher indi¤erence curve). In the same
logic, any point in the Edgeworth box where the indi¤erence curves of the two consumers intersect can never

2
be Pareto optimal, because a move to the interior of the lens-shaped area between those two indi¤erence
curves will make both the consumers better o¤. So, assuming that the indi¤erence curves are smooth and
convex, the Pareto optimal points must be the ones where the indi¤erence curves of the consumers touch
each other, that is, tangent to each other. At this tangency point, moving towards any direction will make
one of the consumers worse o¤, and no other feasible allocation will be Pareto-improving compared to this
point.
So, at the Pareto optimal points, the slopes of the ICs are same, i.e.:

M RS R = M RS F

i.e.,
M Ux R M Ux F
=
M Uy M Uy

Mathematical Derivation of Contract Curve

By de…nition, a Pareto e¢ cient allocation makes each consumer as well o¤ as possible, given the utility
of the other agent. So, let us pick an arbitrary utility level u0 as the utility level for agent F: Then, the
maximization problem is:

max uR (xR ; yR )
s:t: uF (xF ; yF ) = u0
X = xR + xF
Y = yR + y F

The Lagrangian is:

$ = uR (xR ; yR ) + [ u0F uF (xF ; yF )] + 1 [X (xR + xF )] + 2 [Y (yR + yF )]

FOCs are obtained by di¤erentiating w.r.t. xr ; xF ; yR ; yF :

@uR
1 = 0
@xR
@uR
2 = 0
@yR
@uF
1 = 0
@xF
@uF
2 = 0
@xF
Dividing the …rst equation by the second and the third by the fourth one:
@uR
@xR 1
M RS R = @uR
=
@yR 2
@uF
@xF 1
M RS F = @uF
=
@yF 2

Hence, the Pareto e¢ cient points must have

M RS R = M RS F

Contract Curve

The contract curve is the set of all Pareto-optimal allocations in the Edgeworth Box. In the Figure , the
curve joining all the yellow-dotted points is the contract curve.

3
 But there remains a question: to which of the many allocations on the contract curve will consumers end
up after trade?
In a perfectly competitive market, each consumer will only agree to trade if it is bene…cial to him. So,
the …nal equilibrium will be on the part of the contract curve where neither of the consumers is worse o¤
compared to the initial endowment point. This is called the Core. In Figure , the allocations on the contract
curve which can not be sustained as an equilibrium are marked.
Core: The core is the set of all Pareto-optimal allocations that are welfare-improving for both consumers
relative to their own endowments (as illustrated in Figure 2).
Rational trade should achieve a core allocation.

Figure 2: Core

Example 1: Cobb-Douglus Utility Function

Suppose there are two consumers. The utility functions for consumer i is given by:

ui (xi ; yi ) = xai i yi1 ai

ai 2 (0; 1)

Endowments are (x0i ; yi0 ) = (1; 1):

The ICs are smooth and convex, so the Pareto optimal points are points of tangency between the ICs of
the two consumers. Tangency condition is:

M RS 1 = M RS 2

Now,
M U1i ai yi
M RS i = =
M U2i 1 ai xi
So, Pareto e¢ ciency requires:
a1 y1 a2 y2
=
1 a1 x1 1 a2 x2
Using Feasibility condition:

x1 + x2 = 1
y1 + y2 = 1

4
 We have,
y1 a2 (1 a1 ) 1 y1
= (1)
x1 a1 (1 a2 ) 1 x1
This is the equation of the contract curve, i.e., the set of all Pareto optimal points.
If a1 = a2 = 21 ; this equation will simplify to:

x1 = y1

which is just a straight line through the origin.

Example 2: Perfect Complements

Suppose there are two consumers. The utility functions for consumer 1 is given by:

u1 (x1 ; y1 ) = minfx1 ; y1 g

and for consumer 2 is:

u2 (x2 ; y2 ) = minf2x2 ; y2 g
Endowments are (x0i ; yi0 ) = (1; 1):

Show the situation in an Edgeworth Box diagram.

Find the Contract Curve. What is Core? Point it in the diagram.
Answer: Please refer to "Contract Curve_Eg."

Example 3: Perfect Substitutes

Suppose there are two consumers. The utility functions for consumer 1 is given by:

u1 (x1 ; y1 ) = x1 + y1

and for consumer 2 is:

u2 (x2 ; y2 ) = 2x2 + y2
Endowments are (x0i ; yi0 ) = (1; 1):

Show the situation in an Edgeworth Box diagram.

Find the Contract Curve. What is Core? Point it in the diagram.
Answer: Please refer to "Contract Curve_Eg."

Competitive Equilibrium: De…nition

Now, let us turn to the de…nition of Competitive Equilibrium in an Exchange Economy, also known as the
Walrasian Equilibrium.
A Walrasian Equilibrium is a vector of allocation and price ((xR ; yR ),(xF ; yF );(px ; py )) such that:

1. (xR ; yR ) is optimal for R and (xF ; yF ) is optimal for F: i.e.,

(xR ; yR ) solves:

max uR (xR ; yR )
Such that px xR + py yR = px x0R + py yR
0

and (xF ; yF ) solves:

max uF (xF ; yF )
Such that px xF + py yF = px x0F + py yF0

5
 2. (xR ; yR ),(xF ; yF ) is a feasible allocation, i.e.

xR + xF = X
yR + y F = Y

Fundamental Theorems of Welfare Economics

First Fundamental Theorem of Welfare Economics:
Given that consumers’ preferences are well-behaved, trading in perfectly competitive markets imple-
ments a Pareto-optimal allocation of the economy’s endowment. In other words, every competitive
equilibrium is Pareto optimal.

Second Fundamental Theorem of Welfare Economics:

Given that consumers’preferences are well-behaved, for any Pareto-optimal allocation there are prices
and an allocation of the total endowment that makes the Pareto-optimal allocation implementable by
trading in competitive markets. In other words, any Pareto optimal points can be implemented as a
competitive equilibrium of the economy using transfers among the agents.