Lecture Notes in Public Economics

Roozbeh Hosseini
October 30, 2008

In preparing these notes I have beneted from: Chari and Kehoe (1998), V.V. Chari and Larry Jones
rst year Macro lectures notes and V.V. Chari, Narayana Kocherlakota and Mike Golosovs lectures on
Public Economics. These notes are prepared for the Public Economics course at the W.P. Carey School
of Business, Arizona State University. They contain errors. Please use with caution and do not circulate.
YOU, the reader, are the only person responsible for the errors in the notes! Comments are welcome.
1
ECN 741: Public Economics Fall 2008
Contents
1 Ramsey Taxation - Primal Approach 3
1.1 Ramsey problem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.2 Elasticities and optimal taxes . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.2.1 Additive separable utility functions . . . . . . . . . . . . . . . . . . . 7
1.2.2 Quasi-linear utility function . . . . . . . . . . . . . . . . . . . . . . . 7
1.2.3 Complementarity with leisure . . . . . . . . . . . . . . . . . . . . . . 8
1.3 Uniform commodity taxation . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.4 Intermediate good taxation . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2 Optimal Fiscal Policy-Dynamic Ramsey Taxation 12
2.1 Ramsey problem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.2 Chamley-Judd result . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.2.1 Heterogeneous consumers . . . . . . . . . . . . . . . . . . . . . . . . 16
2.2.2 Non-Steady State . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.2.3 Werning (QJE, 2007) . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.3 Taxing Capital in Life Cycle Economies (Erosa and Gervais (2002)) . . . . . 26
3 Mirrleesian Approach to Optimal Taxation 30
3.1 The New Dynamic Public Finance . . . . . . . . . . . . . . . . . . . . . . . . 39
3.1.1 Inverse Euler Equation . . . . . . . . . . . . . . . . . . . . . . . . . . 40
3.1.2 Long-run properties of ecient allocations . . . . . . . . . . . . . . . 44
3.1.3 Aggregate Risk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
3.1.4 Inter-temporal optimality with balanced growth preferences . . . . . 58
3.2 Implementing Ecient Allocations . . . . . . . . . . . . . . . . . . . . . . . 59
3.2.1 Wedges and Taxes . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
References 64
2
ECN 741: Public Economics Fall 2008
1 Ramsey Taxation - Primal Approach
Consider an economy with n types of consumption good that are produced using labor input:
F(c
1
+ g
1
, . . . , c
n
+ g
n
, l) = 0 (1)
c
i
is private and g
i
is public consumption of good i and l is the labor input. F is a constant
return to scale technology. Consumers face the following maximization problem
max
c
1
,...,cn,l
U(c
1
, . . . , c
n
, l)
subject to
n

i=1
p
i
(1 +
i
)c
i
= l
in which
i
is the taxed levied on consumption of good i (wage is normalized to 1).
There is a representative rm that produces goods using technology F:
max
x
1
,...,xn,l
n

i=1
p
i
x
i
l
F(x
1
, . . . , x
n
, l) = 0
Government has to nance its purchase g = (g
1
, . . . , g
n
) using linear taxes
i
n

i=1
p
i
g
i
=
n

i=1
p
i

i
c
i
(2)
Lets take government purchase as given. A Competitive Equilibrium is
Consumers and producers allocations: (c, x, l)
prices: p = (p
1
, . . . , p
n
)
policy: = (
1
, . . . ,
n
)
such that
1. Given policy and prices p, (c, l) solve consumers problem.
2. Given prices, p, (x, l) solves producers problem.
3
ECN 741: Public Economics Fall 2008
3. Government budget (equation (2)) holds
4. Allocations are feasible (or market clearing if you like!)
c
i
+ g
i
= x
i
for i = 1, . . . , n (3)
Proposition 1 Any competitive equilibrium allocations must satisfy the resource feasibility
constraint
F(c
1
+ g
1
, . . . , c
n
+ g
n
, l) = 0 (4)
and an implementability constraint
n

i=1
U
i
c
i
+ U
l
l = 0. (5)
Furthermore, any allocations that satisfy (4) and (6) can be supported as a competitive
equilibrium for appropriately constructed polices and prices.
Proof.
Suppose (c, x, l) is a competitive equilibrium allocation. Then the following FOC must hold
U
i
U
l
= (1 +
i
)p
i
for i = 1, . . . , n
together with the following budget constraint
n

i=1
p
i
(1 +
i
)c
i
= l.
Replacing out for prices (and taxes) from FOC into budget constraint gives the imple-
mentability constraint. The feasibility follows by denition of equilibrium.
Now consider allocations (c, x, l) that are feasible (given vector of g) and satisfy (5). Con-
struct prices from the FOC of the rm
p
i
=
F
i
F
l
for i = 1, . . . , n
set policy as
1 +
i
=
U
i
U
l
F
l
F
i
for i = 1, . . . , n
4
ECN 741: Public Economics Fall 2008
You can verify that the policy and prices (as constructed above) together with the allocation
(c, x, l) is a competitive equilibrium.
We are interested in the problem of choosing the best policy to maximize the welfare of
consumers. One restriction on such a problem is that the resulting allocation be a competitive
equilibrium allocation for each given policy. The timing is the following: First, government
chooses a policy, Second, private agents makes decision. We are interested in nding the
equilibrium of this game.
1.1 Ramsey problem
Suppose the set of feasible policy for government in .
Denition 1 A Ramsey equilibrium is a policy = (
1
,
n
) , allocation rules c(),
x() and l() and price function p() such that
arg max

U(c(

), l(

))
subject to
n

i=1
p
i
g
i
=
n

i=1
p
i

i
c
i
and (c(

), x(

), l(

)) together with p(

) is a competitive equilibrium for every

.
Suppose , (c(), x(), l()) and p() is a Ramsey equilibrium. Then we call (c(), x(), l())
a Ramsey allocation.
Proposition 2 Suppose c

and l

are part of a Ramsey allocation. Then

(c

, l

) arg max
c,l
U(c, l)
subject to (5) and (4).
Proof.
Follows from the denition of Ramsey allocation.
5
ECN 741: Public Economics Fall 2008
1.2 Elasticities and optimal taxes
Suppose n = 2. Consider the following Ramsey problem
max
c
1
,c
2
,l
U(c
1
, c
2
, l)
subject to
1. Implementability constraint
U
1
c
1
+ U
2
c
2
+ U
l
l = 0 (6)
2. Feasibility
F(c
1
+ g
1
, c
2
+ g
2
, l) = 0 (7)
Let and be multipliers on implementability constraint (equation (6)) and feasibility
(equation (7)). First order conditions are
U
i
+ (U
i
+ U
1i
c
1
+ U
2i
c
2
+ U
li
l) = F
i
i = 1, 2
U
l
+ (U
l
+ U
1l
c
1
+ U
2l
c
2
+ U
ll
l) = F
l
We can write these equations as
1 + H
l
=
F
l
U
l
in which, H
i
=
(U
1i
c
1
+U
2i
c
2
+U
li
l)
U
i
and H
l
=
(U
1l
c
1
+U
2l
c
2
+U
ll
l)
U
l
.
Note that from individual problem we have
1 +
i
=
U
i
U
l
F
l
F
i
in other words the optimal wedge must satisfy
1 +
i
=
1 + H
l
1 + H
i
There you go! If H
i
> H
j
, then it is optimal to tax good i more than good j.
The problem is that, it is not very helpful. Unfortunately, without imposing assumption on
U we cannot say much more. Next we consider some special (yet, interesting) cases.
6
ECN 741: Public Economics Fall 2008
1.2.1 Additive separable utility functions
Suppose U is of the form
U(c
1
, c
2
, l) = u
1
(c
1
) + u
2
(c
2
) v(l)
then
H
i
=
U
ii
c
i
U
i
Our goal to relate H
i
to income elasticity of demand for good i. In order to do that, suppose
there is a non-wage income m, such that p
1
c
1
+ p
2
c
2
= l + m. Consider FOC of consumer
(notice that I have ignored taxes for this part)
U
i
(c
i
(p, m)) = p
i
(p, m)
in which (p, m) is the lagrange multiplier on budget constrain. Lets take derivative w.r.t
m
U
ii
c
i
m
= p
i

m
=
U
i

m
or
U
ii
c
i
U
i
m
c
i
c
i
m
=
m

m
.
Let
i
=
m
c
i
c
i
m
. Then
H
i
=
m

m
1

i
Therefore, H
i
> H
j
if and only if
j
>
i
. Combine this with the above and we get the
following:
Result 1 If preferences are additive separable, necessities should be taxed more than luxuries.
Example : U(c
1
, c
2
, l) = log(c
1
) + log(c
2
c) v(l)
1.2.2 Quasi-linear utility function
Consider the utility function in the previous section and assume that v(l) = l. Then there
is no income eect and using income elasticities for guiding us about optimal taxes is not
useful. However, we use price elasticities. Consider again the FOC of consumer
U
i
(c
i
) = p
i

7
ECN 741: Public Economics Fall 2008
Note that in this case = 1 (independent of prices). Take derivative w.r.t p
i
U
ii
c
i
p
i
= =
U
i
p
i
and
H
i
=
1

i
Result 2 If preferences are additive separable and quasi-linear, price-inelastic goods should
be taxed more.
1.2.3 Complementarity with leisure
Sandmo (1987) and Corlett and Hauge (1953-54) argue that goods that are more complement
with leisure should be taxed more heavily. The next example shows this
Example : U(c
1
, c
2
, l) = c

1
+ c

2
(1 l)

1.3 Uniform commodity taxation

One of the most useful and interesting result in optimal taxation is the uniform commodity
taxation result. Suppose the preferences are weakly separable in consumption and leisure
U(c
1
, . . . , c
n
, l) = W(G(c
1
, . . . , c
n
), l) (8)
furthermore, G() is homothetic.
Proposition 3 Suppose preferences satisfy (8), then it is optimal to tax all goods at the
same rate, i.e.
i
=
j
for all i and j.
Proof.
Note that the fact that G() is homothetic implies that
U
i
(c, l)
U
j
(c, l)
=
U
i
(c, l)
U
j
(c, l)
or
U
i
(c, l) =
U
i
(c, l)
U
j
(c, l)
U
j
(c, l).
8
ECN 741: Public Economics Fall 2008
Dierentiate w.r.t and set = 1 we get

n
k=1
U
ik
c
k
U
i
=

n
k=1
U
jk
c
k
U
j
Also, note that U
l
= W
l
, U
li
= W
lg
G
i
and U
i
= W
g
G
i
. Therefore,
H
i
=

n
k=1
U
ik
c
k
U
i

U
il
l
U
i
=

n
k=1
U
ik
c
k
U
i

W
lg
l
W
g
= H
j
This can be generalized to utility functions of the form
u(c
1
, . . . , c
k
, G(c
k+1
, . . . , c
n
), l)
in which, G() is homothetic. Then the result is that commodities (c
k+1
, . . . , c
n
) should be
taxed at uniform rate.
Exercise: Suppose consumer is endowed with y unit of good one that cannot be taxed
away. Does the uniform commodity taxation still hold? what if the utility function is
additive separable?
Exercise: Suppose government is restricted to setting taxed on c
1
to zero. How would
modify the Ramsey problem? Does the uniform commodity taxation hold?
1.4 Intermediate good taxation
Another powerful and important result in Ramsey taxation is that intermediate good shall
not be taxed.
Suppose there are two sectors. One sector produces commodity x
1
that is consumed by
private agent, c
1
and by government, g. Commodity x
1
is produced using intermediate good
z and labor l
1
as input according to the following production function
f(x
1
, z, l
1
) = 0.
The other sector, uses labor l
2
as input to produce good x
2
that can be used as input in
production of good x
1
(that is z) or it can be consumed (c
2
and g
2
). The technology is the
following
9
ECN 741: Public Economics Fall 2008
h(x
2
, l
2
) = 0.
Private agents solves
max
c,l
U(c
1
, c
2
, l
1
+ l
2
)
subject to
p
1
(1 +
1
)c
1
+ p
2
(1 +
2
)c
2
l
1
+ l
2
.
Producer of good x
1
solves
max
x
1
,z,l
1
p
1
x l
1
p
2
(1 +
z
)z
subject to
f(x
1
, z, l
1
) = 0.
The FOC for this problem implies
f
z
f
l
= p
2
(1 +
z
).
Producer of good x
2
solves
max
x
2
,l
2
p
2
x
2
l
2
subject to
h(x
2
, l
2
) = 0.
and FOC implies
h
x
h
l
= p
2
.
Combining the FOC condition for two sector we get
h
x
h
l
(1 +
z
) =
f
z
f
l
.
Government budget constraint is

1
p
1
c
1
+
2
p
2
c
2
+
z
p
2
z = p
1
g
1
+ p
2
g
2
10
ECN 741: Public Economics Fall 2008
Finally, feasibility and market clearing
c
1
+ g
1
= x
1
c
2
+ g
2
+ z = x
2
f(x
1
, z, l
1
) = 0
h(x
2
, l
2
) = 0
The Ramsey problem is
max U(c
1
, c
2
, l
1
+ l
2
)
subject to
U
1
c
1
+ U
2
c
2
+ U
l
(l
1
+ l
2
) = 0
f(c
1
+ g
1
, z, l
1
) = 0
1
h(c
2
+ g
2
+ z, l
2
) = 0
2
FOC w.r.t z

1
f
z
=
2
h
x
FOC w.r.t l
1
and l
2
U
l
+ (U
ll
(l
1
+ l
2
) + U
l
+ U
cl
c) = f
l

1
U
l
+ (U
ll
(l
1
+ l
2
) + U
l
+ U
cl
c) = h
l

2
and therefore,
f
l

1
= h
l

2
.
This implies that
h
x
h
l
=
f
z
f
l
It means that it is optimal to set
z
= 0 and not distort production eciency. For more on
intermediate good taxation and production eciency see Diamond and Mirrlees (1971).
11
ECN 741: Public Economics Fall 2008
2 Optimal Fiscal Policy-Dynamic Ramsey Taxation
The main focus of this section is the derivation of Chamley-Judd result (Chamley (1986)
and Judd (1985)). We are only going to consider deterministic environment. See Chari and
Kehoe (1994) and Chari and Kehoe (1998) for stochastic environment and optimal policy
over business cycle.
The environment is the following. There are innitely lived identical consumers. Government
has to nance expenditure g
t
every period and levies distortionary taxes (or subsidies) on
consumption, investment, labor and capital income. It can also issue debt.
Consumers problem: consumers are endowed with k
0
unit of capital and b
0
unit of govern-
ment debt
max
ct,lt,xt,k
t+1
,b
t+1

t=0

t
U(c
t
, l
t
)
subject to
(1 +
ct
)c
t
+ (1 +
xt
)x
t
+ b
t+1
(1
lt
)w
t
l
t
+ (1
kt
)r
t
k
t
+ R
bt
b
t
;
t
k
t+1
(1 )k
t
+ x
t
b
t+1
M
k
0
, b
0
given
in which M is some large positive number.
The FOCs are

t
U
ct
=
t
(1 +
ct
) (9)

t
U
lt
=
t
w
t
(1
lt
) (10)
(1 +
xt
)
t
=
t+1
[(1
xt+1
)(1 ) + (1
kt+1
)r
t+1
] (11)

t
=
t+1
R
bt+1
(12)
Government Budget:
g
t
+ R
bt
b
t
= b
t+1
+
xt
x
t
+
ct
c
t
+
lt
w
t
l
t
+
kt
r
t
k
t
(13)
Feasibility:
c
t
+ g
t
+ k
t+1
= F(k
t
, l
t
) + (1 )k
t
(14)
12
ECN 741: Public Economics Fall 2008
Competitive pricing implies that
r
t
= F
k
(k
t
, l
t
) (15)
w
t
= F
l
(k
t
, l
t
)
A competitive equilibrium is: the sequence of allocations x = c
t
, l
t
, b
t+1
, k
t+1
, x
t

t=0
, prices
r
t
, w
t
, R
bt

t=0
, policy =
ct
,
lt
,
xt
,
kt+1

t=0
such that, the allocations solve consumer
problem, given prices and policy, prices are competitive, government budget holds and allo-
cations are feasible.
A Ramsey Equilibrium is a policy , an allocation rule x() and price rules r(), w() and
R
b
() such that:
arg max

t=0

t
U(c
t
, l
t
)
subject to 12 and x() be a competitive equilibrium, and
for any policy

, allocation x(

) and prices (r(

), w(

), R
b
(

)) be a competitive equilib-
rium.
We next derive the implementability condition. Note that if conditions of Ekeland and
Scheinkman (1986) and/or Weitzman (1973) are satised, then the equilibrium allocations
should also satisfy the following Transversality conditions
lim
t

t
b
t+1
= 0 (16)
lim
t

t
k
t+1
= 0 (17)
Now multiply consumers budget constraint by
t
and sum over t and use (16)-(17)

t=0

t
[(1 +
ct
)c
t
+ (1 +
xt
)(k
t+1
(1 )k
t
) + b
t+1
] =

t=0

t
[(1
lt
)w
t
l
t
+ (1
kt
)r
t
k
t
+ R
bt
b
t
] .
Now use (9)-(12) and we get

t=0

t
[(1 +
ct
)c
t
(1
lt
)w
t
l
t
] =
0
[(1 +
x0
)(1 ) + (1
k
0
)r
0
] k
0
+ R
b0
b
0
.
13
ECN 741: Public Economics Fall 2008
Now replace (9)-(10) and we arrive at the implementability constraint

t=0

t
[U
ct
c
t
+ U
lt
l
t
] = U
0
[(1 +
x0
)(1 ) + (1
k
0
)r
0
] k
0
+ R
b0
b
0
(18)
Proposition 4 A feasible allocation x = c
t
, l
t
, b
t+1
, k
t+1
, x
t

t=0
is a competitive equilibrium
allocation if and only it satises the implementability constraint (18) (for some period zero
policies).
Proof.
Suppose x is the competitive equilibrium allocation, then following the steps outlines above
we can show that it should satisfy the implementability constraint (18). Now suppose an
allocation x

is feasible and satisfy (18) for some proof zero policies.

Note that in any competitive equilibrium, the bond holding must satisfy
b
t+1
=

s=t+1

ts
[U
cs
c
s
+ U
ls
l
s
]
U
ct
k
t+1
(19)
in other words, any sequence of c

t
, l

t
and k

t+1
uniquely identies a sequence of b
t
that is a
part of competitive equilibrium. Candidate wage and rate of return on capital is given by
(15). Therefore, from the FOC (9)-(12) we have
1
lt
1 +
ct
=
U

lt
F

lt
U

ct
(1 +
xt
)
U

ct
1 +
ct
=
U

ct+1
1 +
ct+1

(1
xt+1
)(1 ) + (1
kt+1
)F

kt+1

(20)
U

ct
1 +
ct
=
U

ct+1
1 +
ct+1
R
bt+1
any two of the four taxes can be chosen such that the above conditions hold.
2.1 Ramsey problem
The Ramsey problem is the following
max
ct,k
t+1
,lt

t=0

t
U(c
t
, l
t
)
14
ECN 741: Public Economics Fall 2008
subject to

t=0

t
[U
ct
c
t
+ U
lt
l
t
] = U
0
[(1 +
x0
)(1 ) + (1
k
0
)r
0
] k
0
+ R
b0
b
0
;
c
t
+ g
t
+ k
t+1
= F(k
t
, l
t
) + (1 )k
t
;
t
Dene function W(, , ) as
W(c, l, ) = U(c, l) + [U
c
c + U
l
l] .
Now we can rewrite the Ramsey problem as
max
ct,k
t+1
,lt

t=0

t
W(c
t
, l
t
, )
subject to
c
t
+ g
t
+ k
t+1
= F(k
t
, l
t
) + (1 )k
t
;
t
Take rst order conditions
W
lt
W
ct
= F
lt
(21)
W
ct
W
ct+1
= (1 + F
kt
) for t 1 (22)
2.2 Chamley-Judd result
Proposition 5 If the solution to the Ramsey problem converges to a steady state, then at
the steady state, the tax rate on capital income is zero.
Proof.
In (22) at the steady state we have
(1 + F
kt+1
) = 1.
This implies that at the steady state there is no inter-temporal distortion. Compare with
(20) we have
(1 +
xt
)(1 +
ct+1
)
(1 +
ct
)(1 +
xt+1
)
=

1 +

1
kt+1
1 +
xt+1

F
kt+1

15
ECN 741: Public Economics Fall 2008
Note that any feasible allocation that satises (18) can be implemented by two of the four
taxes (that is we only need two of the
c
,
l
,
x
and
k
to implement the same allocations).
This in turn implies that

kt
= 0
1 +
ct
1 +
xt
= constant
2.2.1 Heterogeneous consumers
Suppose there are two type of consumers i = 1, 2 with preferences

t=0

t
U
i
(c
it
, l
it
)
The resources constraint for the economy is
c
1t
+ c
2t
+ k
t+1
= F(k
t
, l
1t
, l
2t
) + (1 )k
t
(23)
implementability constraint for consumer i is

t=0

U
i
ct
c
it
+ U
i
lt
l
it

= U
i
0

[(1 +
x0
)(1 ) + (1
k
0
)r
0
] k
i
0
+ R
b0
b
i
0

(24)
Suppose government puts welfare weights
i
on consumers of type i. The Ramsey problem
is
max
1

t=0

t
U
1
(c
1t
, l
1t
) +
2

t=0

t
U
2
(c
2t
, l
2t
)
subject to (23) and (24).
Attached multiplier
i
to implementability constraint of type i and write
W(c
1
, c
2
, l
1
, l
2
,
1
,
2
) =

i=1,2

i
U
i
(c
i
, l
i
) +
i

U
i
c
c
i
+ U
i
l
l
i

16
ECN 741: Public Economics Fall 2008
max

t=0

t
W(c
1t
, c
2t
, l
1t
, l
2t
,
1
,
2
)
subject to
c
1t
+ c
2t
+ k
t+1
= F(k
t
, l
1t
, l
2t
) + (1 )k
t
;
t
where W
i
is dened the obvious way.
First order conditions imply
W
c
it
= W
c
it+1
(1 + F
kt+1
)
and in the steady state
1 = (1 + F
kt+1
)
and, therefore, tax on capital should be zero in the steady state.
Capitalists vs Workers (Judd 1985)
Suppose consumer of type 1 does not hold any asset and cannot save, borrow or invest. We
call these Worker. Also, assume that all the capital is held by consumer 2 who do not
supply any labor. We call these Capitalists. The implementability constraint for Worker
is
U
1
ct
c
1t
+ U
1
lt
l
1t
= 0 t
and for Capitalist

t=0

U
2
ct
c
2t

= U
2
0

[(1 +
x0
)(1 ) + (1
k
0
)r
0
] k
2
0
+ R
b0
b
2
0

(25)
Suppose the welfare weight on Worker utility is 1 and on Capitalist utility is zero.
max

t=0

t
U
1
(c
1t
, l
1t
)
subject to
U
1
ct
c
1t
+ U
1
lt
l
1t
= 0 t

t=0

t
U
2
ct
c
2t
= U
2
0

[(1 +
x0
)(1 ) + (1
k
0
)r
0
] k
2
0
+ R
b0
b
2
0

(26)
c
1t
+ c
2t
+ k
t+1
= F(k
t
, l
1t
, l
2t
) + (1 )k
t
;
t
17
ECN 741: Public Economics Fall 2008
Dene
W(c
1
, c
2
, l
1
, l
2
,
1
,
2
) = U
1
(c
1
, l
1
) +
i

U
i
c
c
i
+ U
i
l
l
i

First order conditions

U
2
cct
c
2t
+ U
2
ct

+
t
= 0

t
=
t+1
(1 + F
kt+1
)
in steady state
t+1
=
t
and therefore
1 = (1 + F
kt+1
)
and again, tax of capital is zero in the steady state.
Exercise: In the above set up we have implicitly assumed that government can levy dierent
taxes on dierent consumer types. How would you add the following restrictions to the
problem
1. Tax on capital income has to be uniform across dierent types. Does the result hold
with this restriction? Under what assumptions?
2. Tax on labor income has to be uniform across dierent types. Does the result hold?
Under what assumptions?
3. Tax on capital income cannot be more than 100 percent. Does the result hold? Under
what assumptions?
Dividend Taxes?!!! (an interesting example)
Suppose we write the environment as in McGrattan and Prescott (2005) with corporate taxes
and dividend taxes. Consumers can trade share of corporations, s
t
, at price v
t
. Let d
t
be
dividend and
dt
be dividend tax. Consumers solve
max
ct,s
t+1
,lt

t=0

t
U(c
t
, l
t
)
subject to

t=0
p
t
[c
t
+ v
t
(s
t+1
s
t
)]

t=0
p
t
[(1
dt
)d
t
s
t
+ (1
lt
)w
t
l
t
]
s
0
= 1
18
ECN 741: Public Economics Fall 2008
FOC implies
U
ct
U
lt
= (1
lt
)w
t
p
t
v
t
= p
t+1
v
t+1
+ p
t+1
(1
dt+1
)d
t+1
And therefore implementability constraint is

t=0

t
[U
ct
c
t
+ U
lt
l
t
] = U
c0
[v
0
+ (1
d0
)d
0
] s
0
(27)
There is a corporation that maximizes the present discounted value of owners dividends and
pays taxes
t
on corporate income.
max

t=0
p
t
(1
dt
)d
t
subject to
d
t
= f(k
t
, l
t
) x
t
w
t
l
t

t
(f(k
t
, l
t
) k
t
w
t
l
t
)
k
t+1
= (1 )k
t
+ x
t
First order conditions for the corporation is
f
lt
= w
t
p
t
(1
dt
)
p
t+1
(1
dt+1
)
= 1 (1
t+1
)(f
kt+1
)
For this economy the feasibility is
c
t
+ k
t+1
+ g
t
= f(k
t
, l
t
) + (1 )k
t
s
t
= 1
and there is also a government budget constraint

t=0
p
t
g
t
=

t=0
p
t
[
dt
d
t
s
t
+
t
(f(k
t
, l
t
) k
t
w
t
l
t
) +
lt
w
t
l
t
]
Question: What is the appropriate implementability constraint? Is constraint (27) su-
cient? In other words, is it true that any feasible allocation that satisfy (27) can be supported
in a competitive equilibrium? If not, what other constraints should be added?
19
ECN 741: Public Economics Fall 2008
2.2.2 Non-Steady State
Proposition 6 Suppose the utility function is of the form
U(c, l) =
c
1
1
v(l),
Then Ramsey taxes on capital income is zero for t 2.
Proof.
Do it as an exercise.
Exercise: Can you establish any connection between this result and uniform commodity
taxation?
2.2.3 Werning (QJE, 2007)
Werning (2007) studies a dynamic environment in which individuals are heterogeneous in
their skills. Instead of ruling out lump-sum taxation, he allows them. However, he does not
allow government to condition the lump-sum tax on individual skill. Instead he allows for a
distortionary labor income (and capital income) tax that government can use to redistribute
income across people with dierent skill. In some sense, it is one step away from traditional
Ramsey setup, towards rationalizing distortionary taxes.
The environment is the following: let c be consumption and l be the hours worked. Individual
with skill who works l hours produce y = l eciency labor unit. If period utility over
hours worked and consumption is U(c, l), then we can write it in terms of consumption and
eciency labor unit as U
i
(c, y) = U(c, y/
i
).
Suppose there are =

i
, . . . ,
N

. We call the individual of type

i
, person i or type
i. The fraction of type i is
i
. Assume

i

i
= 1.
Aggregate state of economy is s
t
S (nite set) and is publicly observable. Denote the
history of aggregate shocks by s
t
= (s
0
, . . . , s
t
). Probability of history s
t
is Pr(s
t
).
Consumer problem
Individual of type i solves
max

t,s
t

t
Pr(s
t
)U
i
(c
t
(s
t
), y(s
t
)) (28)
20
ECN 741: Public Economics Fall 2008
sub. to

t,s
t
p(s
t
)

c(s
t
) + k(s
t
)

t,s
t
p(s
t
)

w
t
(s
t
)(1 (s
t
))y(s
t
) + R(s
t
)k(s
t1
)

T
k
i
(s
0
) = k
i
0
is given
in which R(s
t
) = 1 + (1 (s
t
))(r
t
(s
t
) ) and T =

t,s
t
p(s
t
)T(s
t
) is present value of
lump-sum taxes. Note that there is heterogeneity in skills
i
as well as initial capital holding
k
i
(s
0
).
Feasibility
Let L(s
t
) =

i
y
i
(s
t
), C(s
t
) =

i
c
i
(s
t
), K(s
t
) =

i
k
i
(s
t
). Then feasibility is
C(s
t
) + K(s
t
) + g(s
t
) = F(K(s
t1
), L(s
t
), s
t
, t) + (1 )K(s
t1
) (29)
Govern met
Government has exogenously given sequence of expenditure g(s
t
) to nance. It can levy
linear tax on capital income (s
t
). It can also levy the following tax on income
(s
t
)w
t
(s
t
)y
i
(s
t
) + T(s
t
)
Government budget constraint is

t,s
t
p(s
t
)g(s
t
) T +

t,s
t
p(s
t
)

(s
t
)w
t
(s
t
)L(s
t
) + (s
t
)(r
t
(s
t
) )K(s
t1
)

(30)
Firms
As usual the rms problem is static and implies marginal product pricing
r
t
(s
t
) = F
k
(K(s
t1
), L(s
t
), s
t
, t) (31)
w
t
(s
t
) = F
L
(K(s
t1
), L(s
t
), s
t
, t)
Equilibrium is dened the usual way.
Next we derive the implementability constraints. Werning (2007) develops a methodology
that incorporates the fact that labor income taxes are uniform across types (so no extra con-
straint needs to be added to the optimal taxation problem). Also, he shows implementability
constraints can be written only in terms of aggregates.
21
ECN 741: Public Economics Fall 2008
First observe that in any equilibrium
U
i
y
(s
t
)
U
i
c
(s
t
)
=
U
j
y
(s
t
)
U
j
c
(s
t
)
= w(s
t
)(1 (s
t
)) (32)
U
i
c
(s
t
)
U
i
c
(s
0
)
=
U
j
c
(s
t
)
U
j
c
(s
0
)
=
p(s
t
)

t
Pr(s
t
)p(s
0
)
i, j
Therefore, given the aggregate consumption and labor output (C(s
t
), L(s
t
)), the assignment
of allocation of consumption and labor output c
i
(s
t
), y
i
(s
t
) are ecient. In other words,
given any sequence of aggregate output (C(s
t
), L(s
t
)), there are weights =

1
, . . . ,
N

such that

i

i
= 1 and c
i
(s
t
), y
i
(s
t
) is the solution to
U
m
(C(s
t
), L(s
t
); ) max
{c
i
,y
i
}

i
U
i
(c
i
, y
i
)
sub. to

i
c
i
= C(s
t
),

i
y
i
= L(s
t
)
Denote the solution by
c
i
= h
i
c
(C, L; ), y
i
= h
i
y
(C, L; ) (33)
therefore
(c
i
(s
t
), y
i
(s
t
)) = h
i
(C, L; )
in which h
i
= (h
i
c
, h
i
y
).
Note also that
U
m
C
(C(s
t
), L(s
t
); ) =
i
U
i
c
(c
i
, y
i
) (34)
U
m
L
(C(s
t
), L(s
t
); ) =
i
U
i
y
(c
i
, y
i
)
and therefore, in any equilibrium
U
m
L
(s
t
)
U
m
C
(s
t
)
= w(s
t
)(1 (s
t
)) (35)
U
m
c
(s
t
)
U
m
c
(s
0
)
=
p(s
t
)

t
Pr(s
t
)p(s
0
)
i, j
Now lets look at individual is implementability constraint

t,s
t

U
i
c
(c
i
(s
t
), y
i
(s
t
))c
i
(s
t
) + U
i
y
(c
i
(s
t
), y
i
(s
t
))y
i
(s
t
)

= U
i
c
(c
i
(s
0
), y
i
(s
0
))

R
0
k
i
0
T

22
ECN 741: Public Economics Fall 2008
Now we can replace individual is allocations in terms of aggregate allocations using (33)
and (34)

t,s
t

U
m
C
(C(s
t
), L(s
t
); )h
i
c
(C(s
t
), L(s
t
); ) = (36)
+U
m
L
(C(s
t
), L(s
t
); )h
i
y
(C(s
t
), L(s
t
); )

= U
i
c
(C(s
0
), L(s
0
); )

R
0
k
i
0
T

i(37)
Note that (36) is expressed entirely in terms of aggregate allocations, weights and initial
endowments.
Proposition 7 Given initial wealth R
0
k
i
0
, an aggregate allocation C(s
t
), L(s
t
), K(s
t
) can
be implemented in a competitive equilibrium if and only if
1. It is feasible
2. There exists weights and lump-sum T such that implementability constraint (36)
holds for all i = 1, . . . , N
Proof.
Any equilibrium allocation is feasible and we just showed that it satisfy (36) . Suppose there
is a feasible aggregate allocation that satises (36) for sum weights and lump-sum taxes.
Then individual allocations and prices can be constructed using (33) and (35). Then it is
immediate that (32) (consumer optimality) holds. The individual allocations constructed as
such are also feasible since they satisfy (36) .
A Panning Problem
Suppose
i
is planers weight on type i.

i

i
= 1. Consider the following planning
problem
max

t,s
t
,i

t
Pr(s
t
)U
i
(h
i
(C(s
t
), L(s
t
); ))
sub. to

t,s
t

U
m
C
(C(s
t
), L(s
t
); )h
i
c
(C(s
t
), L(s
t
); )
+U
m
L
(C(s
t
), L(s
t
); )h
i
y
(C(s
t
), L(s
t
); )

= U
i
c
(C(s
0
), L(s
0
); )

R
0
k
i
0
T

i ;
i

i
23
ECN 741: Public Economics Fall 2008
C(s
t
) + K(s
t
) + g(s
t
) = F(K(s
t1
), L(s
t
), s
t
, t) + (1 )K(s
t1
)
Make our usual change of variable
W(C, L; , , )

i
U
i
(h
i
(C, L; ))
+
i

U
m
C
(C, L; )h
i
c
(C, L; ) + U
m
L
(C, L; )h
i
y
(C, L; )

and rewrite the problem as

max

t,s
t
,i

t
Pr(s
t
)W(C(s
t
), L(s
t
); , , ) U
i
c
(C(s
0
), L(s
0
); )

R
0
k
i
0
T

sub. to
C(s
t
) + K(s
t
) + g(s
t
) = F(K(s
t1
), L(s
t
), s
t
, t) + (1 )K(s
t1
)
First order conditions are
F
L
(K(s
t1
), L(s
t
), s
t
, t) =
W
C
(C(s
t
), L(s
t
); , , )
W
L
(C(s
t
), L(s
t
); , , )
W
C
(C(s
t
), L(s
t
); , , ) =

s
t+1
|s
t
W
C
(C(s
t+1
), L(s
t+1
); , , )R

(s
t+1
)Pr(s
t+1
)
in which R

(s
t+1
) = 1 + + F
K
(K(s
t
), L(s
t+1
), s
t+1
, t + 1).
FOC with respect to tax on initial capital

i
k
i
0
= 0 or R
0
= 0
Optimal Taxes

(s
t
) = 1
U
m
L
(C, L; )
W
L
(C, L; , , )
W
C
(C, L; , , )
U
m
C
(C, L; )
Consumer inter-temporal optimality in equilibrium implies
U
m
C
(C(s
t
), L(s
t
); ) =

s
t+1
|s
t
U
m
C
(C(s
t+1
), L(s
t+1
); )R(s
t+1
)Pr(s
t+1
)
24
ECN 741: Public Economics Fall 2008
One way to get this is to set the capital income taxes such that
R(s
t+1
) = R

(s
t+1
)
U
m
C
(C(s
t
), L(s
t
); )
W
C
(C(s
t
), L(s
t
); , , )
W
C
(C(s
t+1
), L(s
t+1
); , , )
U
m
C
(C(s
t+1
), L(s
t+1
); )
Note that FOC with respect to initial capital implies
N

i=1

i
k
i
0

i
= 0
Example: Consider the following preferences
U
i
(c, y) =
c
1
1

(y/
i
)

Note that we have h

i
c
(C, L; ) =
i
c
C and h
i
y
(C, L; ) =
i
y
L, with

i
c
=
(
i
)
1/

i
(
i
)
1/
and
i
y
=
(
i
)

1
(
i
)
1
1

i
(
i
)

1
(
i
)
1
1
and therefore
U
m
=
m
u
c
1
1

m
v

(y/
i
)

and W =
W
u
c
1
1

W
v

(y/
i
)

in which
m
u
,
m
v
,
W
u
and
W
v
are some constant. Note that this implies

(C, L) = 1

m
v

m
u

W
u

W
v
Note also that
U
m
C
(C(s
t
), L(s
t
); )
W
C
(C(s
t
), L(s
t
); , , )
W
C
(C(s
t+1
), L(s
t+1
); , , )
U
m
C
(C(s
t+1
), L(s
t+1
); )
= 1
and therefore
R(s
t+1
) = R

(s
t+1
)
which implies
(s
t
) = 0 for all t 1
This implies that the result for optimal taxes on capital income holds from date zero (not
just for t 1 as it was the case before). When k
i
0
= k
0
for all i, taxing initial capital is
25
ECN 741: Public Economics Fall 2008
like a lump-sum tax. But since lump-tax is allowed here, it is not necessary. However, when
individuals are heterogeneous in their initial wealth, then taxing wealth for redistribution is
desirable.
Example: Now consider the following preferences
U
i
(c, y) = log(c) + (1 ) log(1
y

i
)
then h
i
c
(C, L; ) =
i
C and h
i
y
(C, L; ) =
i

i
(1 L) and

i
=

i

i
therefore,
U
m
(C, L; ) = log(C) + (1 ) log(1 L) +

log

+ (1 ) log

i
/
i

.
Also we can we can verify that
W(C, L) =
W
U
(log(C) + (1 ) log(1 L)) +
W
U
L
(1 )
1 L
and therefore

(L) =
1
(1 L)
W
U
/
W
U
L
+ 1
also
(s
t
) = 0 for all t 1
2.3 Taxing Capital in Life Cycle Economies (Erosa and Gervais
(2002))
Here, I present a 2 period version of Erosa and Gervais (2002). Individuals live 2 periods
(born at age 0, die at age 1). Each generation is indexed by its date of birth. For example
in period t, the generations alive are t 1, t. Assume no population growth.
Each individual is endowed with one unit of time at each age j and can transform one unit
of time into z
j
unit of ecient labor. Let c
t,j
be the consumption of generation t at age j.
Other variables follow the same notation.
26
ECN 741: Public Economics Fall 2008
Consumers problem is the following (for generation t > 0)
max U(c
t,0
, l
t,0
) + U(c
t,1
, l
t,1
)
subject to
(1
c
t,0
)c
t,0
+ a
t,1
(1
l
t,0
)w
t
z
0
l
t,0
(1
c
t,1
)c
t,1
(1
l
t,1
)w
t
z
1
l
t,1
+ (1 + (1
k
t,1
)(r
t
))a
t,1
There is a constant return to scale technology and
r
t
= f
k
(k
t
, l
t
)
w
t
= f
l
(k
t
, l
t
)
and feasibility requires that
c
t
+ k
t+1
= f(k
t
, l
t
) + (1 )k
t
c
t
= c
t,0
+ c
t1,1
l
t
= l
t,0
+ l
t1,1
k
t
= a
t1,1
Government budget constrain is

t=0
p
t
g
t
=

t=0
p
t

j=0,1

c
tj,j
c
tj,j
+

j=0,1

l
tj,j
w
t
z
j
l
tj,j
+
k
t1,1
(r
t
)a
t1,1

Let U
t
= U(c
t,0
, l
t,0
) + U(c
t,1
, l
t,1
) be the lifetime utility of generation t for a given se-
quence of consumption and leisure and let 0 < < 1 be governments discount factor across
generations. Government objective is to maximize

t=0

t
U
t
Exercise: Show that, in this environment, implementability constraint for generation t is
the following
U
c
t,0
c
t,0
+ U
l
t,0
l
t,0
+

U
c
t,1
c
t,1
+ U
l
t,1
l
t,1

= 0 (38)
27
ECN 741: Public Economics Fall 2008
Exercise: Show that a feasible allocation is implementable if and only if it satisfy (38).
Ramsey problem
Ramsey problem is the following
max

t=0

t
[U(c
t,0
, l
t,0
) + U(c
t,1
, l
t,1
)]
subject to
U
c
t,0
c
t,0
+ U
l
t,0
l
t,0
+

U
c
t,1
c
t,1
+ U
l
t,1
l
t,1

= 0 ;
t

t
c
t
+ k
t+1
= f(k
t
, l
t
) + (1 )k
t
;
t

t
c
t
= c
t,0
+ c
t1,1
l
t
= l
t,0
+ l
t1,1
k
t
= a
t1,1
First order conditions are

t
U
c
t,0
+
t

U
c
t,0
+ U
cc
t,0
c
t,0
+ U
lc
t,0
l
t,0

=
t

t
(39)

t
U
c
t,1
+
t

U
c
t,1
+ U
cc
t,1
c
t,1
+ U
lc
t,1
l
t,1

=
t+1

t+1

t
U
l
t,0
+
t

U
l
t,0
+ U
ll
t,0
l
t,0
+ U
lc
t,0
c
t,0

=
t

t
f
lt
(40)

t
U
l
t,0
+
t

U
l
t,1
+ U
ll
t,1
l
t,1
+ U
cl
t,1
c
t,1

=
t+1

t+1
f
lt+1

t
=
t+1

t+1
(1 + f
kt+1
) (41)
Combine (39) and (41)
U
c
t,0
+
t

U
c
t,0
+ U
cc
t,0
c
t,0
+ U
lc
t,0
l
t,0

U
c
t,1
+
t

U
c
t,1
+ U
cc
t,1
c
t,1
+ U
lc
t,1
l
t,1
= (1 + f
kt+1
) (42)
Steady State: In the steady state (c
t,0
, c
t,1
, l
t,0
, l
t,1
, a
t,1
) = (c
0
, c
1
, l
0
, l
1
, a
1
) and
t
= .
Therefore,
28
ECN 741: Public Economics Fall 2008
U
c
0
+ (U
c
0
+ U
cc
0
c
0
+ U
lc
0
l
0
)
U
c
1
+ (U
c
1
+ U
cc
1
c
1
+ U
lc
1
l
1
)
= (1 + f
kt+1
)
Note that this in general does not imply zero tax on capital. When prole of labor produc-
tivity, z
j
, is not at over lifetime, in general consumption and leisure allocations over lifetime
is not at.
Question: Intuitively, why is it optimal to distort inter-temporal decision in this environ-
ment?
We can impose assumptions on preferences (both for government and individuals) to arrive
at zero capital taxation result again.
Proposition 8 Suppose period utility function is of the following form
u(c, l) =
c
1
1
v(l)
Then the Ramsey problem prescribes no inter-temporal distortions for periods t 1, provided
that labor income taxes can be age-dependent.
Proof.
Note that equation (42) becomes
U
c
0,t
U
c
1,t
= (1 + f
kt+1
)
Individual problems Euler equation is
U
c
0,t
U
c
1,t
= (1 + (1
t,1
)(f
kt+1
))
This result should be viewed as a consequence of uniform commodity taxation.
Question: Note that we get this result independent of . Isnt that surprising? Why is
that?
29
ECN 741: Public Economics Fall 2008
3 Mirrleesian Approach to Optimal Taxation
One criticism of the Ramsey approach is the ad hoc assumption of linear distortionary taxes
exclusion of Lump-sum taxation. At the same time without imposing these restrictions or
without including informational frictions the Lump-taxes are very desirable in these models.
We saw Werning (2007) as an attempt to move away from this limitations and expand
the set of instruments available to government, i.e., Lump-sum taxes together with uniform
distortionary taxes across individuals. In doing that he appeals to informational friction, i.e.,
the fact that individuals type (skill) is not observable and hence cannot be taxes. But is it
the best government can do? Is it possible for government to implement more sophisticated
instruments and achieve better outcomes (lets agree for now that better means, higher
welfare given a welfare function)? What are these instruments? How we possibly restrict
the set instruments available to government?
In Mirrleesian approach set of instruments is pinned down by information/enforcement lim-
itation of the government. The optimal tax policy is found among those policies that are
incentive compatible, given the information/enforcement restrictions. In doing that we pro-
ceed in two steps
1. Find a socially optimal allocation given information/enforcement restrictions
2. Devise a tax system that implements this allocation
Environment with nite number of agents
There are N agents, indexed by n = 1, . . . , N, who live T < period. The preferences are
T

t=1

t1
(u(c
t
) v(l
t
)) , 1 > > 0, u

, u

, v

, v

> 0
in which c
t
is consumption and is observed. l
t
is hours worked (or eort) and it is not
observed.
Let be a nite set of skills or ability. Nature makes a draw
T
n
= (
n1
, . . . ,
nt
)
T
=
for each agent n. The
T
n
draws are i.i.d across agents. Let () be probability
density function over
T
draws. Each agent privately learns
nt
at the beginning of period
t. Individual who has skill
nt
and works l
nt
hours in period t can produce y
nt
unit of output
according to
y
nt
= l
nt

nt
30
ECN 741: Public Economics Fall 2008
l
nt
and
nt
are both private information but y
nt
is observable. In what follows we make a
change of variable l
nt
=
ynt
nt
.
Denition 2 An allocation is a sequence of functions (c
n
, y
n
)
N
n=1
c
n
:

N
R
T
+
y
n
:

N
R
T
+
such that c
nt
and y
nt
are (
t
1
, . . . ,
t
N
) measurable. Denote the set of feasible allocation by
FA.
Denition 3 An allocation (c
n
, y
n
)
N
n=1
is feasible if
T

t=1
N

n=1
c
nt
(
T
1
, . . . ,
T
n
)R
t

t=1
N

n=1
y
nt
(
T
1
, . . . ,
T
n
)R
t
, R > 1
for all (
T
1
, . . . ,
T
n
) such that (
T
1
, . . . ,
T
n
) > 0.
So far, we know the information structure, we know what allocations are and we know what
allocations are feasible. But how does this guide towards a set of tax policies? In other
words the remaining question is, given the private information, what allocations can be
implemented?
One period economy
Suppose for now T = 1.
Denition 4 A Game (or a Mechanism) is a set of actions (A
1
, . . . , A
N
) and outcome
functions
(g
c
, g
y
) :
N

n=1
A
n
FA
The timing is as follows:
Nature makes a draw for each agent n
Agents privately observe and then simultaneously choose an action a
n
A
n
31
ECN 741: Public Economics Fall 2008
The outcome is determined according to outcome function
Denition 5 Let (A, g
c
, g
y
) be a Mechanism. A Bayesian Nash Equilibrium (BNE) is a
collection of strategies

N
n=1
,

n
: A
n
such that

n
(
n
) arg max
An

n
(
n
)

u(g
c
n
(,

n
(
n
))) v

g
y
n
(,

n
(
n
))

We call g
c
n
(

1
(
1
), . . . ,

N
(
N
)) and g
y
n
(

1
(
1
), . . . ,

N
(
N
)) (
1
, . . . ,
N
) equilibrium out-
come.
Denition 6 A feasible allocation (c
n
, y
n
)
N
n=1
is implementable if there is a mechanism
(A, g
c
, g
y
) and a BNE

N
n=1
of that mechanism such that
c
n
= g
c
n
(

1
(
1
), . . . ,

N
(
N
)), y
n
= g
y
n
(

1
(
1
), . . . ,

N
(
N
))
So far we have made it clear what exactly do we mean by implementability. But is it helpful?
Notice that our setup so far does not impose any restriction on the type of games (mecha-
nisms) considered. Any equilibrium outcome of some game is implementable. Think for a
moment about the following problem: we want to nd the best implementable allocation.
That means we need to search in the space of games, nd a game that has a BNE that
implements that best allocation as its outcome. This is a very complicated problem.
Good news is that there is a very powerful result that allows us to restrict attention to a
very particular game without lose of generality. We need couple of more denitions.
Denition 7 A Direct Mechanism is a a game such that A
n
= for all n.
Denition 8 A truth-telling BNE of a direct mechanism (, g
c
, g
y
) is

n
(
n
) =
n
for all n
such that

n
arg max

n
(
n
)

u(g
c
n
(,
n
)) v

g
y
n
(,
n
))

A feasible allocation is truthfully implementable if

c
n
= g
c
n
(
1
, . . . ,
N
), y
n
= g
y
n
(
1
, . . . ,
N
)
In a direct mechanism, players are basically asked to report their skill type. We are interested
in equilibria in which type is revealed truthfully. It turns out there is not loss of generality
in doing that.
32
ECN 741: Public Economics Fall 2008
Proposition 9 (Revelation Principle)
A allocation (c
n
, y
n
)
N
n=1
is implementable if and only if it is truthfully implementable in a
direct mechanism.
Proof.
Suppose allocation (c
n
, y
n
)
N
n=1
is implementable as outcome of some mechanism (A, g
c
, g
y
).
We construct a truth-full mechanism (, g
c
, g
y
) as the following
g
c
(
1
, . . . ,
N
) = g
c
(

1
(
1
), . . . ,

N
(
N
)), g
y
(
1
, . . . ,
N
) = g
y
(

1
(
1
), . . . ,

N
(
N
))
In which

N
n=1
is the BNE of the mechanism (A, g
c
, g
y
). We only need to show that
truth-telling is a BNE of (, g
c
, g
y
). Suppose not, i.e., suppose there is a type
n
and a
report such that

n
(
n
)

u( g
c
n
(,
n
)) v

g
y
n
(,
n
))

>

n
(
n
)

u( g
c
n
(
n
,
n
)) v

g
y
n
(
n
,
n
))

n
(
n
)

u(g
c
n
(

n
(
n
),

n
(
n
))) v

g
y
n
(

n
(
n
),

n
(
n
))

in which the last inequality follows from denition of ( g

c
, g
y
). This implies that there must
exist =
1
n
() A
n
such that

n
(
n
)

u(g
c
n
(,

n
(
n
))) v

g
y
n
(,

n
(
n
))

>

n
(
n
)

u(g
c
n
(

n
(
n
),

n
(
n
))) v

g
y
n
(

n
(
n
),

n
(
n
))

this is a contradiction. Therefore, (c

n
, y
n
)
N
n=1
can be implemented by
c
n
= g
c
(
1
, . . . ,
N
), y
n
= g
y
(
1
, . . . ,
N
)
Using Revelation Principle we can restrict attention to direct mechanism and allocations
that are truthfully revealing. This means that the set of implementable allocations are the
33
ECN 741: Public Economics Fall 2008
the ones that satisfy the following incentive compatibility constraints

n
(
n
)

u(c
n
(
n
,
n
)) + v

y
n
(
n
,
n
)

n
(
n
)

u(c
n
(

,
n
)) + v

y
n
(

,
n
)

n,
n
,

.
We are going to primarily focus on environment with unit measure of agents.
Environment with innite number of agents
Consider the same environment as before (for general T < ) except that now there are
unit mass of agents. Nature makes a draw
T
= (
1
, . . . ,
t
)
T
= for each
agent. The
T
draws are i.i.d across agents. Let () is probability density function over
T
draws. There is no aggregate uncertainty, therefore (
T
) is also the mass of people who
have the draw
T
. Let D

T
[(
T
) > 0

.
Dene allocation as
t
measurable functions
c
t
: D R
T
+
y
t
: D R
T
+
Allocation is feasible if

D
T

t=1
R
t
c
t
(
T
)(
T
)

D
T

t=1
R
t
y
t
(
T
)(
T
)
Dene a mechanism as set of actions A

T
t=1
X
t
and outcome functions
g : A (A) R
2T
+
and g
t
(a, ) = g
t
(a

) if a
t
= a
t
and

(a
t+1
,...,a
T
)
( a
t
, a
t+1
, . . . , a
T
) =

(a

t+1
,...,a

T
)

( a
t
, a

t+1
, . . . , a

T
) a
t
in which ,

(A) are measure of actions chosen.

34
ECN 741: Public Economics Fall 2008
A BNE is a strategy

: D A (
t
is
t
measurable) such that
T

t=1

t1
(
T
)

u(g
c
t
(

t
(
T
),

)) v

g
c
t
(

t
(
T
),

t=1

t1
(
T
)

u(g
c
t
(

t
(
T
),

)) v

g
c
t
(

t
(
T
),

for all

: D A (

t
is
t
measurable) and

(a) =

{
T
|(
T
)=a}
(
T
). An allocation
is implementable if there exist a mechanism (A, g
c
, g
y
) and a BNE

such that
(c
t
(
T
), y
t
(
T
)) = g
t
(

(
T
),

)
A mechanism is a direct mechanism if A =
T
. A strategy is truth-full if
t
(
T
) =
t
for all
T

T
. An allocation is truthfully implementable if there exists a BNE of a direct
mechanism, i.e., if
T

t=1

t1
(
T
)

u(g
c
t
(
T
,

)) v

g
y
t
(
T
,

t=1

t1
(
T
)

u(g
c
t
(

t
(
T
),

)) v

g
y
t
(

t
(
T
),

for all

: D
T
(

t
is
t
measurable)

(a) =

{
T
|
T
=a}
(
T
) = (a). And
(c
t
(
T
), y
t
(
T
)) = g
t
(
T
, )
Revelation Principle: An allocation (c, y) is implementable only if it is truthfully imple-
mentable.
Note then that implementable allocation can be characterized by the following incentive
compatibility constraints
T

t=1

T
D
(
T
)

u(c
t
(
T
)) v

y
t
(
T
)

t=1

T
D
(
T
)

u(c
t
(

t
(
T
))) v

y
t
(

t
(
T
))

for all

: D D (

t
is
t
measurable).
35
ECN 741: Public Economics Fall 2008
Example : one period problem
Suppose T = 1 and there are only two types,
H
and
L
with
H
>
L
. Consider the
utilitarian planers problem
max (
H
)

u(c(
H
)) v

y(
H
)

+ (
L
)

u(c(
L
)) v

y(
L
)

sub. to.
u(c(
H
)) v

y(
H
)

u(c(
L
)) v

y(
L
)

u(c(
L
)) v

y(
L
)

u(c(
H
)) v

y(
H
)

(
H
) [c(
H
) y(
H
)] + (
H
) [c(
H
) y(
H
)] = 0
For a moment suppose there is no private information. Then the optimal allocation must
satisfy (note that v() is convex):
u

(c(
H
)) = u

(c(
L
)) = c(
H
) = c(
L
)
1

H
v

y(
H
)

=
1

L
v

y(
L
)

= y(
H
) > y(
L
)
Note that there is no distortion
u

(c()) =
1

y()

We will show that this allocation does not satisfy I.C. constraints. Note that y(
H
) > y(
L
),
therefore
u(c(
L
)) v

y(
L
)

= u(c(
H
)) v

y(
L
)

> u(c(
H
)) v

y(
H
)

the I.C. for type H is violated.

So we know that when individuals have private information about their type, at least of the
the I.C. constraints is binding at the optimal solution.
36
ECN 741: Public Economics Fall 2008
Consider a relaxed planning problem with only type Hs I.C. constraint
max (
H
)

u(c(
H
)) v

y(
H
)

+ (
L
)

u(c(
L
)) v

y(
L
)

sub. to.
u(c(
H
)) v

y(
H
)

u(c(
L
)) v

y(
L
)

; (
H
)
(
H
) [c(
H
) y(
H
)] + (
H
) [c(
H
) y(
H
)] = 0 ;
we will characterize the solution to this problem and then we will verify that at the solution
the I.C. constraint of type L is slack.
(1 + )u

(c(
H
)) =

1
(
H
)
(
L
)

(c(
L
)) =
(1 + )
1

H
v

y(
H
)

=
1

L
v

y(
L
)

(
H
)
(
L
)
1

H
v

y(
L
)

=
First, note that there is no distortion for type H
u

(c(
H
)) =
1

H
v

y(
H
)

Also, observe that

c(
H
) > c(
L
)
note that incentive compatibility implies
v

y(
H
)

y(
L
)

= u(c(
H
)) u(c(
L
)) > 0
and therefore
y(
H
) > y(
L
)
37
ECN 741: Public Economics Fall 2008
Next we check that under these allocations, the I.C. constraints for type L is slack.
v

y(
H
)

y(
L
)

y(
H
)
y(
L
)
v

dy >

y(
H
)
y(
L
)
v

dy
= v

y(
H
)

y(
L
)

= u(c(
H
)) u(c(
L
))
rearrange terms
u(c(
L
)) v

y(
L
)

> u(c(
H
)) v

y(
H
)

So, we know that our characterization is valid. Note that under this characterization
1

L
v

y(
L
)

(c(
L
)) =
(
H
)
(
L
)

H
v

y(
L
)

(c(
L
))

<
(
H
)
(
L
)

H
v

y(
H
)

(c(
H
))

< 0
and hence
1

L
v

y(
L
)

< u

(c(
L
))
there is distortion for the low type.
Suppose we want to implement this allocation with a nonlinear tax function T(y). Let
T(y) = y c if y y(
H
), y(
L
) and T(y) = y otherwise. Consider the consumers
problem
max u(c) v

sub. to.
c = y T(y)
FOC
u

(c)(1 T

(y)) =
1

38
ECN 741: Public Economics Fall 2008
discussion above implies that
T

(y(
L
)) > 0, T

(y(
H
)) = 0
3.1 The New Dynamic Public Finance
So far we have characterized the set of achievable allocations by any mechanism. The goal
of the planner is to nd the best achievable allocation.
T

t=1

t1

T
D
(
T
)(
1
)

u(c
t
(
T
)) v

y
t
(
T
)

(43)
sub. to.
T

t=1

T
(
T
)

u(c
t
(
T
)) v

y
t
(
T
)

t=1

T
(
T
)

u(c
t
(

t
(
T
))) v

y
t
(

t
(
T
))

for all

: D D (

t
is
t
measurable).

D
T

t=1
c
t
(
T
)(
T
)/R
t1

D
T

t=1
y
t
(
T
)(
T
)/R
t1
(c
t
, y
t
) are
T
-measurable
c
t
(
T
), y
t
(
T
) 0 t,
T
D
Note that we allow for allocation to depend on date on realization of .
The goal of this section is to characterize the properties of constraint ecient allocation (i.e.
the solution to the above planning problem). In particular we are interested in identifying
the optimal inter-temporal distortions.
To start we look at the full information problem.
Full information optima
Suppose
t
is public information. Then the planning problem is the same, except that there
will no incentive compatibility constraint. Let be multiplier on feasibility.

T
(
T
)(
1
)u

(c
t
(
T
))
t1
= /R
t1

T
(
T
)
39
ECN 741: Public Economics Fall 2008
Note that this is the FOC with respect to a c
t
(
T
) at a particular draw
T
. But we know
that c
t
(
T
) is
t
measurable, therefore we dont need to sum over all
T
D, but only
those that contain the particular history
t

T
|
t
(
T
)(
1
)u

(c
t
(
T
))
t1
= /R
t1

T
|
t
(
T
)
by measurability of c
t
(
T
).
u

(c
t
(
T
))
t1
(
1
) = /R
t1
Note: This implies that optimal c
t
(
T
) is actually
1
-measurable, i.e., it is independent from

t
for t > 1. In other words there is full insurance.
Note: The following Euler equation hold
u

(c
t
(
T
)) = RE

(c
t+1
(
T
))[
t

planner is happy to allow access to outside trade.

Note: Another Euler equation also holds.
1
u

(c
t
(
T
))
=
1

t1
R
t1
(
1
)
=
1
R
E

1
u

(c
t+1
(
T
))
[
t

3.1.1 Inverse Euler Equation

Consider again the original planning problem (43)(with incentive constraints). Let (c

, y

)
be the solution to this problem. Now consider the following perturbation around (c

, y

)
y

= y

s
= c

s
for all s = t, t + 1(for xed t)
for all histories
t
u(c

t
(
T
)) + u(c

t+1
(
T
)) = k + u(c

t
(
T
)) + u(c

t+1
(
T
)) for all
t+1
such that (
t+1
[
t
) > 0
40
ECN 741: Public Economics Fall 2008

T
|
t
(
T
)

t
(
T
) + c

t+1
(
T
)/R

T
|
t
(
T
)

t
(
T
) + c

t+1
(
T
)/R

Note: (c

, y

) is feasible and incentive compatible.

Note: What we are doing is perturbing u(c
t
(
T
)) by some amount and then make an appro-
priate perturbation in every immediate history following
t
so that incentive compatibility is
preserved. If (c

, y

) is the solution to (43), this perturbation cannot improve welfare. One

implication of this is that (c

, y

) solves the following maximization problem and k = 0 at

the optimal solution.
max
k,c

t
(
T
),c

t+1
(
T
)
k
sub. to
u(c

t
(
T
)) + u(c

t+1
(
T
)) = k + u(c

t
(
T
)) + u(c

t+1
(
T
)) for all
t
,
t+1
such that (
t+1
[
t
) > 0

T
|
t
(
T
)

t
(
T
) + c

t+1
(
T
)/R

T
|
t
(
T
)

t
(
T
) + c

t+1
(
T
)/R

let (
t+1
) and be multipliers.
Lets write the FOC

t+1
|
t
(
t+1
)u

(c

t
(
T
)) =

T
|
t
(
T
) = (
t
)
for all
t
,
t+1
such that (
t+1
[
t
) > 0.
1
u

(c

t
(
t+1
))(
t+1
) = /R

T
|
t+1
(
T
) = (
t+1
)/R
Substitute for (
t+1
)

t+1
|
t
(
t+1
)/R
u

(c

t
(
t+1
))
u

(c

t
(
T
)) = (
t
)
Cancel terms and evaluate this the solution c

= c

R
u

(c

t
(
T
))
=

t+1
|
t
(
t+1
)
(
t
)
1
u

(c

t+1
(
T
))
(44)
1
Note that I am abusing notation here. (
t
) means the probability of history
t
.
41
ECN 741: Public Economics Fall 2008
Note: The Intertemporal condition only depends on consumption which is observable.
Note: The additive separability assumption was key in deriving this result. Look at Farhi
and Werning (2008) for a version of this result that is derived for more general class of utility
functions.
Note: This result does not hold if private information aect the marginal utility of consump-
tion (for example in Atkeson and Lucas (1992) taste shock model).
This result implies that it is not desirable for planer to allow access to saving. To see this
look at the following Euler equation (which has to hold if there is access to saving)
u

(c
t
(
T
)) = R

t+1
|
t
(
t+1
)
(
t
)
u

(c
t+1
(
T
)) (45)
But lets look back at Inverse Euler Equation (44)
u

(c
t
(
T
)) = R
1

t+1
|
t
(
t+1
)
(
t
)
1
u

(c

t+1
(
T
))
> R
1
1
P

t+1
|
t
(
t+1
)
(
t
)
u

(c

t+1
(
T
))
= R

t+1
|
t
(
t+1
)
(
t
)
u

(c

t+1
(
T
))
Note that at the ecient allocation the individuals are saving constrained. In other words,
if individuals can privately save, they will choose to do so and it is desirable for planer to
prevent them from doing that.
Another way of seeing this is the following: suppose (45) holds. Then we must have
R
u

(c

t
(
T
))
<

t+1
|
t
(
t+1
)
(
t
)
1
u

(c

t+1
(
T
))
Now suppose the planer wants to increase utility at time t by and decrease it at time t +1
by
1
. The cost of increase of utility in period t is u

(c
t
(
T
))/ . On the other hand planer
hands in u

(c
t+1
(
T
))/ less at each
t+1
that follows
t
. Therefore it can free up resources.
42
ECN 741: Public Economics Fall 2008
On dynamics of consumption
Consider again the full information optimal allocation
u

(c
t
(
T
))
t1
(
1
) = /R
t1
Suppose for simplicity R = 1. Then
1. Allocation is independent of history (expect possibility
1
)
2. There is no mobility in short-run or long-run
3. Inequality is constant
Now consider private information optimal allocations. Assume
t
is i.i.d. Consider two
dierent history
t
and

t
u

(c
t
(
T
[
t
)) = R

t+1
|
t
(
t+1
)
(
t
)
u

(c
t+1
(
T
[
t
))
u

(c
t
(
T
[

t
)) = R

t+1
|
t
(
t+1
)
(
t
)
u

(c
t+1
(
T
[

t
))
note that (
T
[

t
) = (
T
[
t
). Now suppose u

(c
t
(
T
[
t
)) > u

(c
t
(
T
[

t
)), then there exist a
history
t+1
such that (
t+1
[
t
) = (
t+1
[

t
) and
u

(c
t+1
(
T
[
t
)) > u

(c
t+1
(
T
[

t
))
good shocks up to period t has persistent eect on period t + 1 allocations.
Next we consider inequality. Assume, u(c) = log(c), then
1
u

(c)
= c. Start from inverse Euler
equation (R = 1)
1
u

(c
t
(
T
))
= E

1
u

(c
t+1
(
T
))
[
t

We want to know what happens to variance of consumption over time

V ar

1
u

(c
t
(
T
))

= V ar

1
u

(c
t+1
(
T
))
[
t

= V ar

1
u

(c
t+1
(
T
))

V ar

1
u

(c
t+1
(
T
))
[
t

43
ECN 741: Public Economics Fall 2008
If V ar

1
u

(c
t+1
(
T
))
[
t

> 0 for some

t
, then
V ar

1
u

(c
t
(
T
))

< V ar

1
u

(c
t+1
(
T
))

and therefore
V ar

c
t
(
T
)

< V ar

c
t+1
(
T
)

So inequality grows. And it is ecient.

How about mobility? In short-run there is mobility. What about long-run?
Note that
1
u

(ct)
is a martingale (R = 1). We also know that (by feasibility) E

1
u

(c
t+1
)

<
Martingale Convergence Theorem: If x
t

t=1
is stochastic process adapted to lteration
T
t

t=1
such that x
t
= E[x
t+1
[T
t
] and E[x
t
] < for all t, then
lim
t
x
t
a.s.
= x

<
where x

is a random variable with E[x

] < .
Therefore,
1
u

(ct)
converges to a nite number and hence there is no mobility in the long-run.
3.1.2 Long-run properties of ecient allocations
This part is mostly based on Farhi and Werning (2006, 2007, 2005), PHELAN (2006) and
Atkeson and Lucas (1992).
In what follows we maintain the following assumptions:
T =
=
H
,
L

t
i.i.d over time
Very important note: In what follows I skip some detailed steps. Most of the arguments
are heuristic and have loose ends. For more rigorous proofs please look at the references
above.
44
ECN 741: Public Economics Fall 2008
Immiseration result
For this part assume R = 1. Consider the following planning problem
w
0
= max
T

t=1

t1

T
D
(
t
)

u(c
t
(
t
)) v

y
t
(
t
)

(46)
sub. to.
T

t=1

t
(
t
)

u(c
t
(
t
)) v

y
t
(
t
)

t=1

t
(
t
)

u(c
t
(

t
(
t
))) v

y
t
(

t
(
t
))

for all

: D D (

t
is
t
measurable).

t
T

t=1
(
t
)

c
t
(
t
) y
t
(
t
)

/R
t1
0
The solution to this problem must also be the solution to the following dual problem
K(w
0
) = min

t
T

t=1
(
t
)

c
t
(
t
) y
t
(
t
)

/R
t1
(47)
sub. to
T

t=1

t
(
t
)

u(c
t
(
t
)) v

y
t
(
t
)

t=1

t
(
t
)

u(c
t
(

t
(
t
))) v

y
t
(

t
(
t
))

t=1

t1

T
D
(
t
)

u(c
t
(
t
)) v

y
t
(
t
)

w
0
In which K(U
0
) is the cost of delivering ex-ante utility U
0
to everyone.
We want to write this recursively. Consider an allocation sequence (c
t
(
t
), y
t
(
t
)). Consider
a history

t
. Then dene the ex ante utility of an agent with history

t
under this plan as
w
t
(

t
) =

s=t

s
|

t
(
s
)
st

u(c
s
(
s
)) v

y
t
(
s
)

we call w
t
(

t
) the promised utility after history

t
.
Let (c

t
(
t
), y

t
(
t
)) be the solution to problem (47) w

t
(

t
) be promised utility after history

t
.
We can show that (c

t
(

t
,
t+1
), y

t
(

t
,
t+1
), w

t
(

t
,
t+1
)) solve the following Bellman Equation
45
ECN 741: Public Economics Fall 2008
at w = w

t
(

t
) (note: I imposed the R = 1 assumption here)
K(w) = min
c,y,W

() [c(, w) y(, w) + K(w

(, w))]
sub. to
u(c(, w)) v

y(, w)

+ w

(, w) u(c(

, w)) v

y(

, w)

+ w

, w) ,

()

u(c(, w)) v

y(, w)

+ w

(, w)

w
The second constraint is called promise keeping constraint.
Proposition 10 K(w) is strictly increasing and strictly convex (assumptions on v() is
needed). Also. Let w and w be the lowest and highest possible values for promised utility.
Then, lim
ww
K

(w) = 0 and lim

w w
K

(w) = lim
w w
K(w) = 0.
Let (,

) be multiplier on incentive constraint and the multiplier on promise keeping.

First order condition with respect to c(, U) is
u

(c(, w))

(,

, ) + ()

= ()
and with respect to w

(, U)

(,

, ) + ()

= ()K

(w

(, w)) (48)
and therefore
K

(w

(, w)) =
1
u

(c(, w))
this implies the following lemma
Lemma 1 Given any w [w, w], if w

(, w) = w

, w) for some ,

, then c(, w) =
c

(, w)
Sum the equation (48) over all

()K

(w

(, w)) =

(,

, ) + =
46
ECN 741: Public Economics Fall 2008
also from envelope condition we have
K

(w) =
therefore
K

(w) =

()K

(w

(, w))
Start from a given w
0
, construct a stochastic process w
t
as
w
t+1
= w

(
t
, w
t
)
then
K

(w
t
) = E
t
[K

(w
t+1
)]
hence w
t
is a martingale. By martingale convergence theorem there must exist a w

such
that w
t
a.s.

. Suppose K

(w

) > 0. Note that convergence implies that

w

(, w

) = w

, w

) ,

and therefore
c(, w

) = c(

, w

) ,

and then incentive compatibility implies

y(, w

) = y(

, w

) ,

but we know from our two type static example that the planer can do better by dierentiating
various types. Therefore, this is a contradiction. Hence K

(w

) = 0 and w

= w.
No Immiseration result
Consider the following planning problem in which planer values future consumption more
than agent (

> )
w
0
= max
T

t=1

t1

T
D
(
t
)

u(c
t
(
t
)) v

y
t
(
t
)

(49)
47
ECN 741: Public Economics Fall 2008
sub. to.
T

t=1

t
(
t
)

u(c
t
(
t
)) v

y
t
(
t
)

t=1

t
(
t
)

u(c
t
(

t
(
t
))) v

y
t
(

t
(
t
))

for all

: D D (

t
is
t
measurable).

t
T

t=1
(
t
)

c
t
(
t
) y
t
(
t
)

/R
t1
0
assume the following (in addition to the assumption mentioned at the beginning of the
discussion)

R = 1
v

=
v(y)

u() is unbounded below, hence w =

E

= 1.
Suppose the above problem has a solution and let

be the multiplier on resources constraint.
Then the solution to the above problem must also solve the following
P(w
0
) = max
T

t=1

t1

T
D
(
t
)

u(c
t
(
t
))
v(y
t
(
t
))

c
t
(
t
) +

y
t
(
t
)

(50)
sub. to.
T

t=1

t
(
t
)

u(c
t
(
t
))
v(y
t
(
t
))

t=1

t
(
t
)

u(c
t
(

t
(
t
)))
v(y
t
(

t
(
t
)))

for all

: D D (

t
is
t
measurable).
T

t=1

t1

T
D
(
t
)

u(c
t
(
t
))
v(y
t
(
t
))

w
0
Again, we can show that the solution to the above problem also solves the following Bellman
equation (after any history)
48
ECN 741: Public Economics Fall 2008
P(w) = max
c,y,w

()

u(c(, w))
v(y(, w))

c(, w) +

y(, w) +

P(w

(, w))

(51)
sub. to
u(c(, w))
v(y(, w))

+ w

(, w) u(c(

, w))
v(y(

, w))

+ w

, w) ,

()

u(c(, w))
v(y(, w))

+ w

(, w)

w
We want to show that in this problem in the long-run the promised utility cannot be at
misery (w

> ). We use two lemmas to show this

Lemma 2 The value function P(w) is strictly concave and continuously dierentiable on
(, w). Moreover,
lim
v
P(w) = lim
v w
P(w) = lim
v w
P

(w) =
and
lim
v
P

(w) = 1
Proof.
We are not going to prove concavity and dierentiability. We take them as given.
Warning: This proof is not complete! There are some steps that needs to be lled in or
reformulated. I present it to provide the core idea of the proof as I understand it.
Warning:The proof that I presented in class is wrong! The mistake is in constructing the
upper and lower bound value function (P
FI
and P
m
). In particular, I mentioned in class
that P
i
(w) = w K
i
(w) (for i = FI, m, etc), in which K
i
(w) is the cost of delivering utility
w. This is correct only if

= . When

> individual utilities are discounted at dierent
rate in the planners objective and expected present discounted value of an allocation in the
planners objective is not the same as the one promised to the agent.
Correct proof has the same idea but those functions are constructed correctly.
Dene value function P
FI
(w) as
P
FI
(w) = max
T

t=1

t1

T
D
(
t
)

u(c
t
(
t
))
v(y
t
(
t
))

c
t
(
t
) +

y
t
(
t
)

49
ECN 741: Public Economics Fall 2008
sub. to.
T

t=1

t1

T
D
(
t
)

u(c
t
(
t
))
v(y
t
(
t
))

w
P
FI
(w) is the value to the planner from delivering utility w to individual, if we ignore
incentive constraint. Note that P
FI
(w) > P(w). Also, P
FI
(w) is strictly concave and
dierentiable.
Next consider the following maximization problem
m = max
c,y,
u(c)
v(y)

c +

y
The above problem has a solution (u

(c) =

,v

(y) =

, and belongs to a compact set).
Next, note that
P
FI
(w) =
T

t=1

t1

T
D
(
t
)

u(c
t
(
t
))
v(y
t
(
t
))

c
t
(
t
) +

y
t
(
t
)

=
T

t=1

t1

T
D
(
t
)

u(c
t
(
t
))
v(y
t
(
t
))

c
t
(
t
) +

y
t
(
t
)

+
T

t=1

t1

T
D
(
t
)

u(c
t
(
t
))
v(y
t
(
t
))

c
t
(
t
) +

y
t
(
t
)

t=1

t1

T
D
(
t
)

u(c
t
(
t
))
v(y
t
(
t
))

c
t
(
t
) +

y
t
(
t
)

= w +
T

t=1

t1

T
D
(
t
)

c
t
(
t
) +

y
t
(
t
)

+
T

t=1
(

t1

t1
)

T
D
(
t
)

u(c
t
(
t
))
v(y
t
(
t
))

c
t
(
t
) +

y
t
(
t
)

w +
T

t=1

t1

T
D
(
t
)

c
t
(
t
) +

y
t
(
t
)

+ m

1
1

1
1

K(w) + m

1
1

1
1

in which

K(w) = min
T

t=1

t1

T
D
(
t
)

c
t
(
t
) y
t
(
t
)

50
ECN 741: Public Economics Fall 2008
sub. to.
T

t=1

t1

T
D
(
t
)

u(c
t
(
t
))
v(y
t
(
t
))

w
Note that

K(w) is strictly convex and dierentiable and lim
w

K

(w) = 0.
Let P
max
(w) = w

K(w) + m. Then P
max
(w) P
FI
(w) and both are strictly con-
cave. Also, lim
w
P
FI
(w) lim
w
P
max
(w) = . Therefore, lim
w
P

FI
(w)
lim
w
P

max
(w) = 1.
Also, Note that lim
w
P(w) lim
w
P
FI
(w) = and therefore (since both are
strictly concave)
lim
w
P

(w) lim
w
P

FI
(w) = 1.
lim
w
P

(w) 1
Next, consider allocations (c(w
0
,
t
), y(w
0
,
t
)) that solve the original problem. Suppose they
attain the value P(w
0
). Dene new allocations ( c(w,
t
), y(w,
t
)) for w w
0
as
c(w,
t
) = c(w
0
,
t
)
t
t
y(w,
t
) = y(w
0
,
t
)
t
t > 1
y(w,
1
) = v
1
(v(y(w
0
,
1
)) + w
0
w)
Now dene P
m
(w) for w w
0
as
P
m
(w) =
T

t=1

t1

T
D
(
t
)

u( c(w,
t
))
v( y(w,
t
))

c(w,
t
) +

y(w,
t
)

1
(
1
)

u(c(w
0
,
1
))
v(y(w
0
,
1
)) + w
0
w

c(w
0
,
1
) +

y(w
0
,
1
)

+
T

t=2

t1

T
D
(
t
)

u(c(w
0
,
t
))
v(c(w
0
,
t
))

c(w
0
,
t
) +

y(w
0
,
t
)

1
(
1
)

w w
0

1
+

y(w,
1
)

1
(
1
)

u(c(w
0
,
1
))
v(y(w
0
,
1
))

c(w
0
,
1
)

+
T

t=2

t1

T
D
(
t
)

u(c(w
0
,
t
))
v(c(w
0
,
t
))

c(w
0
,
t
) +

y(w
0
,
t
)

51
ECN 741: Public Economics Fall 2008
Note that P
m
(w) is strictly concave, P
m
(w) P(w). Also, note that only the rst term
depends on w. Therefore
P

m
(w) = 1

1
v

(v(y(w,
0
)) + w
0
w)

and lim
w
P

m
(w) = 1.
Therefore, lim
w
P

(w) lim
w
P

m
(w) = 1. Hence, we proved that lim
w
P

(w) =
1.
In the next lemma we show that 1 P(w

(w, )) can be bounded above and bellow for all .

Lets rewrite the problem (51) again
P(w) = max
c,y,w

()

u(c(, w))
v(y(, w))

c(, w) +

y(, w) +

P(w

(, w))

(52)
sub. to
u(c(, w))
v(y(, w))

+ w

(, w) u(c(

, w))
v(y(

, w))

+ w

, w) ,

()

u(c(, w))
v(y(, w))

+ w

(, w)

w
Suppose only high types incentive constraint binds. Let and be multipliers on the IC
and promise keeping (and let (
H
) = )
FOC w.r.t c(w, )

1
u

(c(w,
H
))

+ + = 0
(1 )

1
u

(c(w,
L
))

+ = 0
FOC w.r.t w

(w, )

(w,
H
) + + = 0
(1 )

(w,
L
) + (1 ) = 0
52
ECN 741: Public Economics Fall 2008
FOC w.r.t y(w, )

H
+

1
v

(y(w,
H
))

H
= 0

(1 )

L
+ (1 )

1
v

(y(w,
L
))

(1 )

L
+

H
= 0
combining these FOCs together with envelope condition P

(w) = we get
E[1 P

(w

(w, ))] =

(1 P

(w)) + 1

and

1
v

(y(w, ))

= (1 + )E

= 1 +
Assume that we know y(w,
H
) > y(w,
L
) and w

(w,
H
) > w

(w,
L
). Now we can prove
the next lemma.
Lemma 3 The following inequalities hold
(1 P

(w))

1 +

H

+ 1

1 P

(w

(w, )) (1 P

(w))

H
+ 1

Proof.
From the FOC for y(w,
H
) we get
(1 + )
H
= E

(y(w, ))

(y(w, ))
= 1 + +

from this we can get the following bound on

(1 + )(
H
1)
Now we use this in the FOC for w

(w,
H
) and we get
P

(w

(w,
H
))

(
H
1)
= P

(w)

(
H
1)
53
ECN 741: Public Economics Fall 2008
after rearranging terms
1 P

(w

(w,
L
)) < 1 P

(w

(w,
H
)) (1 P

(w))

H
+ 1

Using similar argument we can show that

1
(1 + )(1
L
)

L
and
P

(w

(w,
L
)

1 +

H

= P

(w)

1 +

H

and hence
1 P

(w

(w,
H
)) > 1 P

(w

(w,
L
)) (1 P

(w))

1 +

H

+ 1

Now let w , then P

(w) 1 and
lim
w
P

(w

(w, )) =

< 1
So w

(w, ) can never stay at misery level.

For more detailed arguments and complete proof of the existence of stationary distribution
see Farhi and Werning (2006, 2007, 2005).
3.1.3 Aggregate Risk
In this section we derive inverse Euler equation in an environment in which there is aggregate
risk. We need to introduce new notations. Let be the space of individual shocks and Z
be the space of aggregate shock. The timing is the following
1. Nature draws z
T
Z
T
according to p.d.f
z
(z
T
).
54
ECN 741: Public Economics Fall 2008
2. Nature draw individual shocks
T

T
according to

(
T
[z
T
). These draws are i.i.d
across individuals conditional on z
T
.
By law of large number, given z
T
, the fraction of population with shocks
T
is

(
T
[z
T
).
We impose the following restriction:
Assumption 1 For all
T

T
,

(
t
[z
T
) =

(
t+1
,...,
T
)

(
t
,
t+1
, . . . ,
T
[z
T
) is indepen-
dent of z
t+1
, . . . , z
T
.
This assumption implies that conditional on z
t
, (
t+1
, . . . ,
T
) and (z
t+1
, . . . , z
T
) are inde-
pendent.
Remarks: Note that in this setup we are not imposing any restriction on time series prop-
erties of
t
and z
t
. However, our assumption implies that by observing history of private
shocks up to date t the individuals cannot infer anything about future aggregate shocks.
As before, we assume that agents learn z
t
and
t
at the beginning of date t and
t
is private
information.
There is an initial capital stock

K
1
in the economy.
Denition 9 An allocation is a sequence of functions (c, y, K) such that
c :
T
Z
T
R
T
+
, c
t
: (
t
, z
t
)-measurable
y :
T
Z
T
R
T
+
, y
t
: (
t
, z
t
)-measurable
K : Z
T
R
T
+
, K
t+1
: z
t
-measurable
Denition 10 An allocation is feasible if
C
t
(z
T
) + K
t+1
(z
T
) = (1 )K
t
(z
T
) + F(K
t
(z
T
), Y
t
(z
T
), z
T
) (53)
C
t
(z
T
) =

(
T
[z
T
)c
t
(
T
, z
T
)
Y
t
(z
T
) =

(
T
[z
T
)y
t
(
T
, z
T
)
K
1
=

K
1
55
ECN 741: Public Economics Fall 2008
Denition 11 An allocation is incentive compatible if

z
T
Z
T

z
(z
T
)
T

t=1

t1

(
T
[z
T
)

u(c
t
(
T
, z
T
)) v

y
t
(
T
, z
T
)

(54)

z
T
Z
T

z
(z
T
)
T

t=1

t1

(
T
[z
T
)

u(c
t
(

t
(
T
, z
T
))) v

y
t
(

t
(
T
, z
T
))

for all

: (

t
is (
t
, z
t
) measurable).
The planing problem
max

z
T
Z
T

z
(z
T
)
T

t=1

t1

(
T
[z
T
)

u(c
t
(
T
, z
T
)) v

y
t
(
T
, z
T
)

sub. to (53) and (54).

The inverse Euler equation
Suppose (c

, y

, K

) is the solution to the above problem. Fix a public history z

t
. We perturb
the solution to (c

, y

, K

) such that
u(c

t
(
t
, z
t
)) = u(c

t
(
t
, z
t
)) +

z
t+1

z
( z
t
, z
t+1
)
t+1
(
t
, z
t
, z
t+1
) +
t
(55)
u(c

t+1
(
t+1
, z
t
, z
t+1
)) = u(c

t+1
(
t+1
, z
t
, z
t+1
))
t+1
(
t
, z
t
, z
t+1
) (56)

t
(
t
[ z
t
)c

t
(
t
, z
t
) + K

t+1
( z
t
)

t
(
t
[ z
t
)c

t
(
t
, z
t
) + K

t+1
( z
t
) (57)

t+1

(
t+1
[ z
t
, z
t+1
)c

t+1
(
t+1
, z
t
, z
t+1
) K

t+1
( z
t
)(1 ) F(K

t+1
( z
t
), Y

t
( z
t
, z
t+1
), z
t
, z
t+1
)(58)

t+1

(
t+1
[ z
t
, z
t+1
)c

t+1
(
t+1
, z
t
, z
t+1
) K

t+1
( z
t
)(1 ) F(K

t+1
( z
t
), Y

t
( z
t
, z
t+1
), z
t
, z
t+1
)
The idea is that (c

, y

, K

) must be the solution to the following maximization problem

(with = 0)
0 = max

t+1
,c

t
,c

t+1
,,K

t+1

56
ECN 741: Public Economics Fall 2008
sub. to (55)-(58).
Let (
t
, z
t
),(
t+1
, z
t
, z
t+1
), ( z
t
) and ( z
t
, z
t+1
) be multipliers on (55), (56), (57) and (58).
Write the FOCs
u

(c

t
(
t
, z
t
))(
t
, z
t
) = ( z
t
)

(
t
[ z
t
) (59)
u

(c

t+1
(
t+1
, z
t
, z
t+1
))(
t+1
, z
t
, z
t+1
) = ( z
t
, z
t+1
)

(
t+1
[ z
t
, z
t+1
) (60)

z
( z
t
, z
t+1
)(
t
, z
t
) =

t+1
,z
t+1
|
t
, z
t
(
t+1
, z
t
, z
t+1
) (61)
( z
t
) =

z
t+1
( z
t
, z
t+1
)(1 + F
K
(K

t+1
( z
t
), Y

t
( z
t
, z
t+1
), z
t
, z
t+1
)) (62)
substitute (
t
) and (
t+1
, z
t
, z
t+1
) from (59) and (60) into (61)

z
( z
t
, z
t+1
)
( z
t
)

(
t
[ z
t
)
u

(c

t
(
t
, z
t
))
=

t+1
,z
t+1
|
t
, z
t
( z
t
, z
t+1
)

(
t+1
[ z
t
, z
t+1
)
u

(c

t+1
(
t+1
, z
t
, z
t+1
))
Take ( z
t
, z
t+1
) out of summation and rearrange terms

z
( z
t
, z
t+1
)

(
t
[ z
t
)
u

(c

t
(
t
, z
t
))

t+1
,z
t+1
|
t
, z
t

(
t+1
[ z
t
, z
t+1
)
u

(c

t+1
(
t+1
, z
t
, z
t+1
))

1
=
( z
t
, z
t+1
)
( z
t
)
Note that by the Independence assumption we had

(
t+1
[ z
t
, z
t+1
)

(
t
[ z
t
)
=

(
t+1
[ z
t
, z
t+1
)

(
t
[ z
t
, z
t+1
)
=

(
t+1
[ z
t
, z
t+1
,
t
)
Let
( z
t
, z
t+1
)
( z
t
, z
t+1
)
( z
t
)
z
( z
t
, z
t+1
)
=

u

(c

t
(
t
, z
t
))

t+1
| z
t
,z
t+1
,
t

(
t+1
[ z
t
, z
t+1
)
u

(c

t+1
(
t+1
, z
t
, z
t+1
))

1
then
1 =

z
t+1

z
( z
t
, z
t+1
)( z
t
, z
t+1
)(1 + F
K
(K

t+1
( z
t
), Y

t
( z
t
, z
t+1
), z
t
, z
t+1
))
or
( z
t
, z
t+1
) =

u

(c

t
(
t
, z
t
))

1
u

(c

t+1
(
t+1
, z
t
, z
t+1
))

t+1
, z
t
, z
t+1

1
57
ECN 741: Public Economics Fall 2008
1 = E

( z
t
, z
t+1
)(1 + F
K
(K

t+1
( z
t
), Y

t
( z
t
, z
t+1
), z
t
, z
t+1
))[ z
t

Note that (using Jensens inequality)

( z
t
, z
t+1
) =

u

(c

t
(
t
, z
t
))

1
u

(c

t+1
(
t+1
, z
t
, z
t+1
))

t+1
, z
t
, z
t+1

1
<
E

(c

t+1
(
t+1
, z
t
, z
t+1
))

t+1
, z
t
, z
t+1

(c

t
(
t
, z
t
))
and therefore
u

(c

t
(
t
, z
t
)) < E

(c

t+1
(
t+1
, z
t
, z
t+1
))(1 + F
K
(K

t+1
( z
t
), Y

t
( z
t
, z
t+1
), z
t
, z
t+1
))

t+1
, z
t
, z
t+1

Example 1: suppose is singleton. Then

( z
t
, z
t+1
) =
u

(c

t+1
( z
t
, z
t+1
))
u

(c

t
( z
t
))
and therefore
1 = E

(c

t+1
( z
t
, z
t+1
))
u

(c

t
( z
t
))
(1 + F
K
(K

t+1
( z
t
), Y

t
( z
t
, z
t+1
), z
t
, z
t+1
))[ z
t

Example 2: suppose Z is singleton. Then

t+1
=

u

(c

t
(
t
))

1
u

(c

t+1
(
t+1
))

t+1

1
1 =
t+1
(1 + F
K
(K

t+1
, Y

t
))
and therefore
(1 + F
K
(K

t+1
, Y

t
))
u

(c

t
(
t
))
= E

1
u

(c

t+1
(
t+1
))

t+1

and therefore (using Jensens inequality)

u

(c

t
(
t
)) < (1 + F
K
(K

t+1
, Y

t
))E

(c

t+1
(
t+1
))

t+1

3.1.4 Inter-temporal optimality with balanced growth preferences

The additive separable preferences that we have assumed so far are not consistent with
balanced growth fact (except if u(c) = log(c)). However, the perturbation that we used
58
ECN 741: Public Economics Fall 2008
in deriving the inverse Euler equation depends crucially on additive separability. If the
preferences are of the form
u(c, l) =
c
1
1
v(l)
then the proof presented will not work. See Farhi and Werning (2008) for derivation of the
inter-temporal optimality condition for larger class of preferences (including those that are
consistent with balanced growth).
3.2 Implementing Ecient Allocations
3.2.1 Wedges and Taxes
So far we have shown that ecient allocations must satisfy a condition like the following (if
there is no aggregate shock)
(1 + F
K
(K

t+1
, Y

t
))
u

(c

t
(
t
))
= E

1
u

(c

t+1
(
t+1
))

t+1

.
We also showed that this implies the following
u

(c

t
(
t
)) < E

(c

t+1
(
t+1
))(1 + F
K
(K

t+1
, Y

t
))

t+1

.
In other words there is an inter-temporal wedge
1
t
=
u

(c

t
(
t
))
E

(c

t+1
(
t+1
))(1 + F
K
(K

t+1
, Y

t
))

t+1
< 1.
But does this mean that the ecient allocations can be implemented by a positive tax on
capital?
Two period example
Consider the following example:
T = 2.
= 0, 1, (
1
= 1) = 1, (
2
= (1, 1)) = 0.5 and (
2
= (1, 0)) = 0.5. (Note that
this implies y
2h
= y
2
((1, 1)) = l
2h
(1, 1) and y
2l
= y
2
((1, 0)) = 0)
59
ECN 741: Public Economics Fall 2008
u(c, l) = log(c)
l
2
2
, = 1.
F(K, Y ) = RK + wY , = 1.
There is endowment of K
1
in period 1.
The planners problem is
max
c
1
,c
2h
,c
2l
,y
1
,y
2h
,K
2
log(c
1
)
y
2
1
2
+ 0.5

log(c
2h
)
y
2
2h
2

+ 0.5 [log(c
2l
)]
sub. to
c
1
+ K
2
= RK
1
+ wy
1
0.5c
2h
+ 0.5c
2l
= RK
2
+ 0.5wy
2h
log(c
2h
)
y
2
2h
2
log(c
2l
)
c
1
, c
2h
, c
2l
, y
1
, y
2h
, K
2
0
Let (c

1
, c

2h
, c

2l
, y

1
, y

2h
, K

2
) be the solution. Then it should satisfy the following FOC
c

1
+ K

2
= RK
1
+ wy

1
0.5c

2h
+ 0.5c

2l
= RK

2
+ 0.5wy

2h
log(c

2h
)
y
2
2h
2
= log(c

2l
)
Rc

1
= 0.5c

2h
+ 0.5c

2l
w
c

2h
= y

2h
w
c

1
= y

1
We want to implement these allocations in a competitive equilibrium. We assume the fol-
lowing market structure
There is a single rm that rents capital and labor.
60
ECN 741: Public Economics Fall 2008
Individuals have the same endowment of k
1
= K
1
and choose how much to work and
how much to save.
Government taxes savings at rate
k
.
Government makes transfers T
2h
and T
2l
to individual who produce y
2h
and zero output
in the second period. Transfers can be negative.
Competitive equilibrium is allocations (c
1
, y
1
, c
2h
, y
2h
, c
2l
, k
2
) and policy (
k
, T
2h
, T
2l
) such
that:
1. Given the policy, allocations solve consumer problem
max
c
1
,c
2h
,c
2l
,y
1
,y
2h
,k
2
log(c
1
)
y
2
1
2
+ 0.5

log(c
2h
)
y
2
2h
2

+ 0.5 [log(c
2l
)]
sub. to
c
1
+ k
2
= Rk
1
+ wy
1
c
2h
= R(1
k
)k
2
+ wy
2h
+ T
2h
, if y
2h
> 0
c
2h
= R(1
k
)k
2
+ T
2l
, if y
2h
> 0
c
2l
= R(1
k
)k
2
+ T
2l
c
1
, c
2h
, c
2l
, y
1
, y
2h
, k
2
0
2. Markets clear
c
1
+ K
2
= RK
1
+ wy
1
0.5c
2h
+ 0.5c
2l
= RK
2
+ 0.5wy
2h
k
1
= K
1
k
2
= K
2
Note that the government budget constraint is satised in equilibrium
R
k
K
2
= 0.5T
2h
+ 0.5T
2l
We want the tax system be such that y
2h
> 0. Then it is necessary that
log(c
2h
)
y
2
2h
2
log(c
2l
)
61
ECN 741: Public Economics Fall 2008
Lets assume we have such a tax system. Then the consumer FOCs are
1
c
1
= R(1
k
)

.5
c
2h
+
.5
c
2l

w
c
1
= y
1
w
c
2h
= y
2h
Also note that
c
2h
= R(1
k
)k
2
+ wy
2h
+ T
2h
c
2l
= R(1
k
)k
2
+ T
2l
Question: How do we pick taxes to implement ecient allocations (c

1
, c

2h
, c

2l
, y

1
, y

2h
, K

2
)
in the equilibrium?
Why dont we try the simplest way to do this. And choose the following taxes
1
k
=
[0.5c

2h
+ 0.5c

2l
]
1

0.5
c

2h
+
0.5
c

2l

and
T
2h
= c

2h
R(1
k
)K

2
wy

2h
T
2l
= c

2l
R(1
k
)K

2
This way the equilibrium allocations satisfy all of the optimality conditions listed above.
They are also feasible and hence should implement the ecient allocations.
What is missing?
The taxes above guarantee that the individual saves the correct amount if he plans to tell
the truth about his ability in the second period. Also, they guarantee that the individual
tells the truth if he saves the correct amount.
Question: What if the individual saves more than K

2
and plan to lie about their ability if
they happen to be the able type?
We rst show that this double deviation is feasible and improves individuals welfare. Consider
three dierent plan by the agent
62
ECN 741: Public Economics Fall 2008
PLAN 1:
Save K

2
( K
1
+ wy

1
c

1
)
Produce y

2h
if
2
= 1 and zero if
2
= 0
PLAN 2:
Save K

2
( K
1
+ wy

1
)
Produce zero in the second period
Note rst that, because incentive compatibility in planners problem binds, individual is
indierent between PLAN 1 and PLAN 2. Also,
1
c

1
< R(1
k
)

.5
c

2h
+
.5
c

2l

< R(1
k
)
1
c

2l
PLAN 3:
Save K

2
+ .
Consume c

2l
+ R(1
k
). and produce zero.
Dene function g()
g() = log(c

1
) + log(c

2l
+ R(1
k
)) [log(c

1
) + log(c

2l
)]
and note that
g

(0) =
1
c

1
+
R(1
k
)
c

2l
> 0
and therefore
PLAN 3 ~ PLAN 2 PLAN 1
Therefore under the proposed tax system individuals prefer PLAN 3 to PLAN 1. Hence this
implementation fails. In other words the ecient allocation cannot be implemented by a
separable tax system in which capital taxes do not depend on individuals income ex-post.
63
ECN 741: Public Economics Fall 2008
References
Atkeson, A. and J. Lucas, Robert E (1992): On Ecient Distribution with Private
Information, Review of Economic Studies, 59, 42753. 3.1.1, 3.1.2
Chamley, C. (1986): Optimal Taxation of Capital Income in General Equilibrium with
Innite Lives, Econometrica, 54, 60722. 2
Chari, V.V., C. L. and P. Kehoe (1994): Optimal scal policy in a business cycle
model, Journal of Political Economy, 102, 61752. 2
Chari, V. and P. J. Kehoe (1998): Optimal scal and monetary policy, Minneapolis
Fed Sta Report, SR 251. , 2
Corlett, W. and D. Hauge (1953-54): Complementarity and the excess burden of
taxation, Review of Economic Studies, 21, 2130. 1.2.3
Diamond, P. and J. Mirrlees (1971): Optimal taxation and public production I: pro-
duction eciency, American Economic Review, 61, 827. 1.4
Ekeland, I. and J. A. Scheinkman (1986): Transversality Conditions for some innite
horizon discrete time optimization problems, Mathematics of operation research, 11. 2
Erosa, A. and M. Gervais (2002): Optimal Taxation in Life-Cycle Economies, Journal
of Economic Theory, 105, 338369. 2.3
Farhi, E. and I. Werning (2005): Inequality, Social Discounting and Estate Taxation,
. 3.1.2, 3.1.2
(2006): Progressive Estate Taxation, . 3.1.2, 3.1.2
(2007): Inequality and Social Discounting, Journal of Political Economy, 115, 365
402. 3.1.2, 3.1.2
(2008): Optimal Savings Distortions with Recursive Preferences, The Journal of
Monetary Economics, 55, 2142. 3.1.1, 3.1.4
Judd, K. L. (1985): Redistributive Taxation in a Simple Perfect Foresight Model, Journal
of Public Economics, 28, 5983. 2
McGrattan, E. R. and E. C. Prescott (2005): Taxes, Regulations, and the Value of
U.S. and U.K. Corporations, Review of Economic Studies, 72, 767796. 2.2.1
64
ECN 741: Public Economics Fall 2008
PHELAN, C. (2006): Opportunity and Social Mobility, Review of Economic Studies, 73,
487504. 3.1.2
Sandmo, A. (1987): A Reinterpretation of Elasticity Formulae in Optimum Tax Theory,
Economica, 54, 8996. 1.2.3
Weitzman, M. L. (1973): Duality Theory of Convex Programming for Innite Horizon
Economic Models, Management Science, 19. 2
Werning, I. (2007): Tax Smoothing with Redistribution, Quarterly Journal of Eco-
nomics, 122, 92567. 2.2.3, 2.2.3, 3
65