VIEWS: 4 PAGES: 14

• pg 1
```									B/C – A and distributional issues

(Cost Benefit Analysis DEC 51304)
Zerbe & Dively Ch.11
R. Jongeneel
Lecture Plan
 Basics
 FullCompensation Criterion
 Kaldor-Hicks criterion
 Identify gainers and losers
 No undesirable transfers
 Opportunity cost rule
 Use distributional weights
Basics

                                
H    N
dW   Pi dX  (ai  1) Pi dX
i
h
i
h

h 1 i 1
efficiency   distribution

 h: marginal social utility of income of
individual h
 h: h - h MUYh times MSUh
 Problem: uncertainty w.r.t. MSUYh
Full compensation criterion
 Assumes   nothing is known about
MSUYh
 Accepts Pareto Principle as decisive
criterion
 Requires that losers from a policy
change are fully compensated from
the gains of the winners
 Conclusion: PPI is actually realized!
Example:
Full compensation criterion (before)
Group MSUY (ah)    Net    Distributional    Change in
benefit     effect     economic welfare

Poor     1.2       -70         -14             -84

Rich     0.8       +110        -22             +88

Total              +40         -36             +4
Example:
Full compensation criterion (after)
Group MSUY (ah)     Net    Distributional    Change in
benefit     effect     economic welfare

Poor       1.2        0           0              0

Rich       0.8       +40          -8            +32

Total                +40          -8            +32

Assumption: zero transfer costs
The Kaldor-Hicks criterion
H     N
 Basic   rule:   dW   Pi dX        i
h

h 1 i 1
 Ignores   distributional effects / assures
same MSUY for all h’s

 Simply   calculates the present value of costs
and benefits and evaluates whether
NPVbenefit exceeds NPVcosts
The Kaldor-Hicks criterion
Defense:
 Distributional effects easy to identify and
remedy
 Existing distribution (chosen by government)
and already optimal (implying equal MSUYh
for all h)
Criterion:
 Potential Pareto Improvement
 Compensation possible, but not actually
carried out
Identify gainers and losers
 Avoid   problem of distributional
assumptions
 Simply   identifies who gains (and how
much) and who loses (and how much)
 Let   the policy maker decide
No undesirable transfers
(Willig & Baily)
       Accepts widely used assumptions:
i)      MSUY decreases as Y increases
ii)     MSUY is non-negative for all h
iii)    It is undesirable to transfer money from a
poorer to a richer individual
       Rank all individuals from poorest to
richest
j            j
A>B if    NB
h 1
A
h     NB ; j  1,2,.....N
h 1
B
h
No undesirable transfers
 Net  benefits must be superior at each stage
of the summation process from poorest to
richest income group
 Weaknesses:
- Some policy changes with net social welfare
benefits might be rejected (no regressive
transfers)
- Results are sensitive to group definition
Opportunity cost rule (Harberger)
   The benefit of a transfer cannot be greater than the
cost of making the transfer by the most efficient
alternative means (opportunity costs)
Example I
Poor      Middle      Rich    Transfer   NB
B or C
NB K-H           +40        +29       -70         0      -1

NB Opp.Cst       +40        +29       -70        +8      +7

Harberger: Net transfer cost only 1 (cf.-1) instead of
20% of +40 (=8)
Opportunity cost rule (Harberger)

Example II
Poor    Middle     Rich    Transfer    NB
B or C

NB K-H         -100      +50      +58        0        +8

NB Opp.Cst     -100      +50      +58       -20       -12

Harberger: Taking into account the costs of actually
compensating the poor (20% of 100=20) makes the project
undesirable
Distributional weights
 Attach explicit weights to costs and benefits
accruing to different groups

K      N
dW   wk  PdX
i           i
k

k 1   i 1

with weight of class k equal to wk

```
To top