Transcript Measuring the deadweight loss due to monopoly

Estimating Dead Weight Loss Due to
Imperfectly Competitive Market Structures
Economists are naturally interested in
estimating the size of dead weight losses
(DWLs) resulting from allocative inefficiency.
Estimating DWL is difficult because the
investigator will not normally know the true
value of marginal cost. Hence, DWL must be
estimated indirectly.
Harburger’s Approach1
The ABH dead weight triangle is approximated by the
following equation (equation 4.1 in the text):
1
DWL  ( PM  PC )(QC  QM )
2
By algebraic manipulation it can be shown that:
1
DWL  d 2 P * Q *
2
1Arnold
[1]
Harburger. "Monopoly and Resource Allocation," American
Economic Review, Dec. 1965: 77-87
Explanation of equation [1]
 is price elasticity of demand
d is the price cost margin, that is:
d
P  MC
P
P* is the monopoly price
Q* is the monopoly output
•To estimate d, Harburger measured the difference between rate
of return for the industry and the average rate of return for all
industries.
•Harburger assumed that, for all industries,  = 1
Harburger’s estimates
Based on data for U.S. industries in the 1920s,
Harburger estimated the DWL due to monopoly
to be equal to 1/10 of 1 percent of GNP. Hence,
the welfare loss due to pricing above marginal
cost is very small and would hardly justify the
allocation of substantial resources for antitrust
enforcement.
Cowling and Mueller’s Approach2
•Equation [1] above reveals that estimates of DWL are sensitive
to assumptions made about elasticity of demand ( )Cowling and
Mueller made adjustments to the methodology used by Harburger
and , using a sample of 734 U.S. firms for 1963-66, reached
radically different conclusions as regards the magnitude of
welfare losses.
•Cowling and Mueller changed a key assumption of Harburger;
namely, that for all industries,  = 1.
2 Keith
Cowling and Dennis Mueller. "The Social Costs of
Monopoly Power," Economic Journal, December 1978: 727-48.
To estimate industry-level price elasticities (), Cowling
and Mueller took advantage of the fact that the firm’s profit
maximizing price (P*) satisfies the following condition:
P*

P * MC
[2]
Recall that d is the price-cost margin . Thus we can say:
1
P*

d P *  MC
[3]
Thus by equation [2], we can say:
1

d
Thus if you can
estimate d, you
can estimate 
Thus substitute 1/d for  in equation [1] and you get:
11 2
1
DWL   d P * Q*  dP * Q *
2d 
2
[4]
Substituting (P*- MC)/P* for d in equation [4] gives us:
DWL 
1  P *  MC 
1
1
P
*
Q
*

(
P
*

MC
)
Q
*

*


2
P
2
2

[5]
Thus, Cowling and Mueller showed that DWL for an industry
was equal to ½ of the economic profits () of firms in the
industry.
Figure 1
A
Price
PM
PC
C
MC
B
D
MR
0
QM
QC
Quantity
1  P *  MC 
1
1
DWL  
 P * Q*  ( P *  MC )Q*   *
2
P
2
2

•DWL is given by the black
–shaded triangle.
• is given by the greenshaded rectangle
Price
P*
MC
MR
0
Q*
Measuring Dead Weight Loss
D
Quantity
Cowling and Mueller’s results
Back to lesson 3
Assuming that 12 percent is a "normal" rate of return on capital,
Cowling and Mueller produced 2 estimates of DWL in the U.S.
economy:
• The low estimate, which does not include advertising
expenditures as a component of the dead weight loss, was 4
percent of GNP (about $403 billion in 2001).
• The high estimate, which reckoned advertising expenditures as
"wasted resources," was 13 percent of GNP (about $1.394 trillion
in 2001).
Moral of the story: If these estimates are in the ballpark,
then antitrust enforcement may be an extremely cost
effective way to improve social welfare