Professor Mark Gertler

Macroeconomic Theory II
Spring 2003
Lecture 7
Sticky Price Models
1 A Basic Sticky Price Model without Capital
1.1 Households
The representative household chooses
n
!
!+"
" !
!+"
(#)" $
!+"
"
#
!+"
$
!+"
"
%
!+"
$
!+"
o
!
"=0
to max
%
!
(
!
X
"=0
&
"
"
1
1 !'
!
1"&
!+"
+
(
'
1 !'
'

)
!+"
*
!+"

1"&
#
!
(
(
1 +'
(
$
1+&
$
!+"
#)
(1)
subject to
!
!
=
+
!
*
!
$
!
+!
!
+,-
!
!
)
!
!)
!"1
*
!
!
1
1+"
!
.
!
!.
!"1
*
!
" (2)
where '" '
'
/ 1 and '
(
0 01
Setting up the Bellman equation
2

)
!"1
*
!
"
.
!"1
*
!

= max
"
1
1 !'
!
1"&
!+"
+
(
'
1 !'
'

)
!+"
*
!+"

1"&
#
!
(
(
1 +'
(
$
1+&
$
!+"
+%
!
&2

)
!
*
!+1
"
.
!
*
!+1

#
(3)
subject to (2).
FOC
The rst order conditions for the consumers problem are:
$
!
:
+
!
*
!
=
(
(
$
&
$
!
!
"&
!
(4)
.
!+1
*
!+1
:
!
"&
!
1
1 +3
!
= &2
2

)
!
*
!+1
"
.
!
*
!+1

*
!
*
!+1
(5)
where the envelope condition is
2
2

)
!"1
*
!
"
.
!"1
*
!

= !
"&
!
(6)
1
updating the envelope condition one period forward and plugging it back into (5) we
obtain
1 = %
!
(
(1 +3
!
)

*
!
*
!+1

&

!
!
!
!+1

"&
)
(7)
or
!
"&
!
= -
!+1
&!
"&
!+1
" (8)
where
-
!+1
= (1 +3
!
)
*
!
*
!+1
1 (9)
.
!+1
*
!+1
:
!
"&
!
= (
'

)
!
*
!

"&
#
+&2
1

)
!
*
!+1
"
.
!
*
!+1

*
!
*
!+1
1 (10)
where the envelope condition implies
2
1

)
!"1
*
!
"
.
!"1
*
!

= !
"&
!
1 (11)
updating one period forward and plugging it into (10)
(
'

#
1
$
!

"&
#
!
"&
!
= 1 !
1
1 +3
!
(12)
Notice that if (
'
= 0, for 3
!
0 0 the return from holding bonds dominates the return
from holding real money balances. Equation (8) is a standard Euler equation. Equation
(4) is an intratemporal condition capturing the consumption/leisure trade-o!. It has the
interpretation that the marginal rate of substitution between consumption and leisure be
equal to the real wage. Equation (12) is the dynamic condition for the choice of money
holdings. The marginal cost of foregoing one unit of consumption today must be equal to
the pecuniary benet of being able to buy an extra unit of consumption tomorrow, plus the
nonpecuniary benet measured by the current utility ow of an extra unit of money. Noting
that from (8)
*
!
*
!+1
&!
"&
!+1
=
!
"&
!
1 +3
!
"
we can write (12) as follows:
!
"&
!

3
!
1 +3
!

= (
'

)
!
*
!

")
1 (13)
2
1.2 Firms
1.2.1 Final Goods Producers
The economy is composed of a continuum of wholesale producers, whose total is normalized
to unity. Final good producers ensemble the intermediate goods according to the following
production function
4
*
!
=
!
"
1
Z
0
4
*
!
(#)
%!1
%
5#
#
$
%
&!1
(14)
where 6 0 1 is the price elasticity of demand. We can see how this is a CES production
function, which also exhibits diminishing marginal product, property that will drive the rms
to diversify and produce with all the intermediate goods available.
The nal good producer will minimize its costs. Therefore it will choose 4
*
!
(#) to
min
1
Z
0
*
!
(#) 4
*
!
(#) 5#
st
!
"
1
Z
0
4
!
(#)
%!1
%
5#
#
$
%
%!1
"

4
Writting the langrangian
7 =
1
Z
0
*
!
(#) 4
*
!
(#) 5# !8
%
&
'
&
(
!
"
1
Z
0
4
!
(#)
%!1
%
5#
#
$
%
%!1
!

4
)
&
*
&
+
FOC
The rst order condition with respect to 4
*
!
(#) is
*
!
(#) !8
6
6 !1
!
"
1
Z
0
4
!
(#)
%!1
%
5#
#
$
%
%!1
"1
6 !1
6
4
*
!
(#)
"
1
%
= 0 (15)
*
!
(#) = 8

1
R
0
4
!
(#)
%!1
&
5#

%
%!1

1
R
0
4
!
(#)
%!1
%
5#
4
*
!
(#)
"
1
%
*
!
(#) = 8
4
*
!
4
*
%!1
%
!
4
*
!
(#)
"
1
%
3
*
!
(#) = 8

4
*
!
(#)
4
*
!
!
"
1
%
(16)
Now we need to solve for the lagrange multiplier. To do so, we write the rst order
condition in a di!erent way
*
!
(#) = 8
94
*
!
94
*
!
(#)
(17)
1
4
*
!
*
!
(#) 4
*
!
(#) = 8
94
*
!
94
*
!
(#)
4
*
!
(#)
4
*
!
1
4
*
!
1
Z
0
*
!
(#) 4
*
!
(#) 5#
| {z }
+!#Total cost of prod.
= 8
1
Z
0
94
*
!
94
*
!
(#)
4
*
!
(#)
4
*
!
5#
| {z }
1
%
!
4
*
!
= 8 (18)
Since the nal goods producers operate in perfect competition, total cost of production must
be equal to the total value of the goods sold. Hence
%
!
= *
!
4
*
!
(19)
Combining (18) and (19) we obtain
8 = *
!
(20)
Using this result into (16)
*
!
(#) =

4
*
!
(#)
4
*
!
!
"
1
%
*
!
(21)
or
4
*
!
(#) =

*
!
(#)
*
!

")
4
*
!
(22)
Market Demand of Intermediate Good z
Integrating over all nal good rms we can obtain the total demand of intermediate
good z.
4
!
(#) =
1
Z
0
4
*
!
(#) 5:
4
4
!
(#) =
1
Z
0

*
!
(#)
*
!

")
4
*
!
5:
4
!
(#) =

*
!
(#)
*
!

")
1
Z
0
4
*
!
5:
| {z }
,!
4
!
(#) =

*
!
(#)
*
!

")
4
!
(23)
Price Index
If we plug the demand of good # by rm : into the production function of rm : :
4
*
!
=
%
'
(
1
Z
0
"

*
!
(#)
*
!

")
4
*
!
#%!1
&
5#
)
*
+
%
%!1
4
*
!
=
%
'
(

4
*
!
%!1
&
1
Z
0

*
!
(#)
*
!

"()"1)
5#
)
*
+
%
%!1
4
*
!
= 4
*
!
%
'
(
1
Z
0

*
!
(#)
*
!

"()"1)
5#
)
*
+
%
%!1
1 =
!
"

1
*
!

"()"1)
1
Z
0
[*
!
(#)]
"()"1)
5#
#
$
%
%!1
1 =

1
*
!

")
!
"
1
Z
0
[*
!
(#)]
"()"1)
5#
#
$
%
%!1
*
)
!
=
!
"
1
Z
0
[*
!
(#)]
1")
5#
#
$
%
%!1
*
!
=
!
"
1
Z
0
[*
!
(#)]
1")
5#
#
$
1
%!1
(24)
5
1.2.2 Intermediate Good Producers
As we have seen, the intermediate good producer faces a downward sloping demand. This
is due to the fact that the intermediate good maket is monopolistically competitive.
The production funcion of an intermediate good rm # is
4
!
(#) = ;
!
$
!
(#) (25)
The market demand for rm z is
4
!
(#) =

*
!
(#)
*
!

")
4
!
(26a)
The rm will choose *
!
(#)" 4
!
(#) and $
!
(#) to
max %
!"1

*
!
(#)
*
!
4
!
(#) !
+
!
*
!
$
!
(#)

st (25) and (26a)

*
!
(#)
*
!

")
4
!
!)!
!

*
!
(#)
*
!

")
4
!
)
(27)
FOC
The rst order condition is
%
!"1
(
(1 !6)

*
!
(#)
*
!

")
4
!
*
!
+6)!
!

*
!
(#)
*
!

"(1+))
4
!
*
!
)
= 0
%
!"1
(

*
!
(#)
*
!

")
4
!
*
!
"
(1 !6) +6)!
!

*
!
(#)
*
!

"1
#)
= 0 (28)
if we devide both sides by (1 !6) and dene (1 +<) #
6
1 !6
=
1
1"
1
%
, we obtain
%
!"1
(

*
!
(#)
*
!

")
4
!
*
!
"
1 +
6
1 !6
)!
!

*
!
(#)
*
!

"1
#)
= 0
or
%
!"1
(

*
!
(#)
*
!

")
4
!
*
!
"
1 + (1 +<) )!
!

*
!
(#)
*
!

"1
#)
= 0
1
This condition comes from the rm minimizing cost.
6
where
6
1 !6
=
1
1 !
1
)
= 1 + < is the mark-up of the rm, which is inversely related to
the elasticity of demand, 6. In the case of perfect competition, when 6 = $"
1
1"
1
%
= 1, and
hence <, the net mark-up over the marginal cost, is zero.
We can rewrite the FOC as
%
!"1
(

*
!
(#)
*
!

")
4
!
*
!
"
1 + (1 +<) )!
!

*
!
(#)
*
!

"1
#)
= 0 (29)
1.2.3 One period Sticky Prices
If rms set prices before the beginning fo the period, there exists a di!erence between ex-
ante and ex-post mark-up. In the absence of shocks or under exible prices, the mark-up is
constant and equal to the desired mark-up.
Ex-Ante
The condition that holds ex-ante and which determines the desired mark-up by the
rm(<) is
%
!"1

*
!
(#)
*
!

= %
!"1
[(1 +<) )!
!
] (30)
Ex-Post
Ex-post, there exist a realization of *
!
and +
!
, and we will have an ex-post mark-up
*
!
(#)
*
!
| {z }
*"./0 1! !
= (1 +<
!
)
| {z }
/."234! '156"72
)!
!
(31)
Unanticipated increases in demand will drive up the marginal cost, through an incrase
in the demand of labor, and down the ex-post mark-up. Therefore we have counter-cyclical
mark-up.
1.3 Government and Money Creation Process
Government
)
!
!)
!"1
*
!
= ,-
!
(32)
Money Creation Process
)
!+1
)
!
= 1 +=
'
1 (33)
7
1.4 Symmetric Monopolistic Competitive Equilibrium
A symmetric equilibrium is characterized by the following conditions
*
!
(#) = *
!
%#" (34)
4
!
(#) = 4
!
%#" (35)
$
!
(#) = $
!
%#" (36)
!
!
= 4
!
" (37)
and
.
!
= 01 (38)
Equation (38) is a bonds market clearing condition. In equilibrium the supply of bonds is
zero. Everybody is indi!erent between borrowing and lending. Since every individual is the
same in this economy, aggregate saving is equal to zero in equilibrium. Equation (37) is a
goods market clearing condition.
1.4.1 Aggregate Demand
The aggregate demand side of the economy is characterized by the following equations:
(1) Wide Economy Resource Constraint
!
!
= 4
!
(39)
(2) Euler Equation for Bonds
!
!
= %
!

(1 +3
!
)

*
!
*
!+1

& (!
!+1
)
"
1
)

"8
(40)
where > =
1
&
.
1.4.2 Aggregate Supply
The aggregate supply side of the economy is characterized by the following equations:
(3) Aggregate Production Function
4
!
= ;
!
$
!
(41)
(4) Labor Market Equilibrium
;
!
= (1 +<
!
) (
(
$
&
$
!
!
"&
!
(42)
8
which come from combining the ex-post FOC for the intermediate rm and the FOC for
labor supply for the consumer:
;
!
= (1 +<
!
)
+
!
*
!
+
!
*
!
= (
(
$
&
$
!
!
"&
!
(5) Ex-ante Condition for Price Setting

*
!
= (1 +<) %
!"1
()!
(
!
) (43)
which comes from the ex-ante FOC for the intermediate rm, which expects
$
!
(9)
$!
= 1, and
hence
%
!"1

*
!
(#)
*
!

= %
!"1
[(1 +<) )!
!
]
1 = %
!"1
[(1 +<) )!
!
]

*
!
= (1 +<) %
!"1
()!
(
!
)
Surprises in the nominal wages (+
!
) or in technology (;
!
) will cause the actual mark-up (<
!
)
to deviate from the desired mark-up (<).
(6) Euler Equation for Money

)
!

*
!
= !
"
*
*
#
!

1 !
1
1 +3
!

"
1
*
#
(
"
1
*
#
'
(44)
where the nominal money stock ()
!
) is xed by policy and the level of price is xed due to
its sticky nature. We can see how, in this case, an unticipated shock in the money supply
does not guarantee that the prices are going to move to adjust, since they are sticky. Hence,
there exist an interaction between nominal money and real variables.
If we use the wide economy resource constraint (39) in both euler equations (40) and
(44), we will obtain the IS and LM relations. Also combining the aggregate production
function (41) and the labor market equilibrium (42) we obtain the Agreggate Supply.
IS
4
!
= %
!

(1 +3
!
)

*
!
*
!+1

& (4
!+1
)
"
1
)

"8
(45)
LM

)
!

*
!
= 4
"
*
*
#
!

1 !
1
1 +3
!

"
1
*
#
(
"
1
*
#
'
(46)
AS
1 +<
!
=
1
(
(
4
"(&+&
$
)
!
;
1+&
$
!
(47)
which shows how unanticipated increases in demand decrease the mark-up, pushing the
economy temporarily closer to the competitive equilibrium.
9
1.5 Log-Linearization
The log-linearization of the model is done in the same manner as in Lecture 6, and therefore
the derivations will be omitted.
IS
?
!
= !> [@
!
!(%
!
A
!+1
! A
!
)] +%
!
?
!+1
(48)
where A
!
denotes the log-deviation of the price at period t, which is xed at t since prices
are set at period t-1.
Since prices are xed only for one period, the best forecast that economic agents can
make about the future is the exible price scenario. If we denote ?
$
!
as the log-deviation of
output from the steady state for the exible price model, the best forecast will be
%
!
?
!+1
= %
!
?
$
!+1
Using this last equation and the Fisher parity condition
@
!
= B
!
+%
!
A
!+1
! A
!
(49)
we can re-write the IS as follows
?
!
= !> B
!
+%
!
?
$
!+1
(50)
LM
C
!
! A
!
= ( ?
!
!D@
!
(51)
where ( =
&
&
#
and D =
1
&
#
1
AS
<
!
= !(' +'
(
) ?
!
+ (1 +'
(
) (
!
(52)
In the exible price equilibrium the ex-post mark-up does not deviate from the desired
mark-up, and therefore the AS becomes
0 = !(' +'
(
) ?
$
!
+ (1 +'
(
) (
!
?
$
!
=

1 +'
(
' +'
(

(
!
(53)
where ?
$
!
is the log-deviation of output from the steady state in the case of exible prices. If
you we plug (53) into (52) we obtain
<
!
= !(' +'
(
) ( ?
!
! ?
$
!
) (54)
We know, by equation (31), that the ex-post mark-up is equal to the inverse of the marginal
cost. Hence, combining the log-linearized version of (31)
2
and (54)
CE
!
= (' +'
(
) ( ?
!
! ?
$
!
) (55)
Money Growth
C
!
! C
!"1
= =
'
(56)
2
!"
!
=! #
!
10
1.5.1 Absence of Technology Shocks
In the absence of technology shocks, (
!
= 0 %F" we obtain
%
!
?
$
!+1
=

1 +'
(
' +'
(

%
!
(
!+1
| {z }
=0
= 0
and the IS becomes
?
!
= !> B
!
= !>[@
!
!(%
!
A
!+1
! A
!
)] (57)
If there is only unexpected shocks in the economy, agents will create expectations con-
sidering that the future is like the exible price model. Hence, updating the LM one period
forward
C
!+1
! A
$
!+1
= ( ?
$
!+1
!D@
$
!+1
where
@
$
!+1
= B
$
!+1
+%
!+1
A
$
!+2
! A
$
!+1
but, if (
!
= 0 %F, then ?
$
!+1
= 0, and updating (57) one period forward, B
$
!+1
= 0. Hence
C
!+1
! A
$
!+1
= !D@
$
!+1
and @
$
!+1
= %
!+1
A
$
!+2
! A
$
!+1
therefore
A
$
!+1
= C
!+1
+D

%
!+1
A
$
!+2
! A
$
!+1

(58)
If money growth is constant
%
!+1
A
$
!+2
! A
$
!+1
= A
$
!+1
!A
!
= C
!+1
! C
!
= =
'
(59)
using (59) into (58) we obtain
A
$
!+1
= C
!+1
+D=
'
= C
!
+ (1 +D) =
'
Finally, since the best forecast of future prices is the exible price case, %
!
A
!+1
= A
$
!+1
"equation
(57) becomes
?
!
= !>[@
!
! C
!
!(1 +D) =
'
+ A
!
)] (60)
Consider an increase in C
!
:
1. the standard IS-LM analysis applies;
2. as C
!
increases, agents form expectations about future ination. As a result, the real
interest rate decreases to make money more attractive for given real money balances;
3. beliefs about future demand matter as well (through ?
!+1
). Notice that this e!ect will
be clearer when prices will be xed for more than one period.
From (60) when C
!
increases, ?
!
increases as well. With sticky prices, money a!ects
real output. Furthermore, given C
!
" if < increases, expected ination increases as well. This
puts a downward pressure on the real interest rate, so that real output increases.
11
1.6 Staggered Long-Term Pricing
Following the Calvo [1] setup, rms adjust their prices infrequently. The opportunity to
adjust follows a Bernoulli distribution. Dene G the probability of keeping prices constant
and (1!G) the probability of changing prices. In other words, each period there is a constant
probability (1!G) that the rm will be able to adjust its price, independently of past history.
This implies that the fraction of retailers setting prices at F is (1 !G). Thus, only a fraction
of rms is setting prices at a certain period of time. The draw is independent of history
and we do not need to keep track of rms changing prices. The time that elapses between
price adjustments follows a geometric distribution. The expected time over which the price
is xed, i.e., the expected waiting time for the next price adjustment is therefore
1
1":
1 The
problem of the rm changing price at time F consists of choosing *
!
(#) to max
%
!
!
X
"=0
(G&)
"
"
!;!+"
"
*
!
(#)
*
!+"
4
!;!+"
(#) !
<
!+"
-
!+"
*
!+"
4
!;!+"
(#)
#
(61)
subject to
4
!;!+"
(#) =

*
!
(#)
*
!+"

")
4
!+"
" (62)
where
"
!;!+"
=

!
!+"
!
!

"&
1 (63)
In (62), 4
!;!+"
(#) is the rms demand function for its output at time F +3" conditional on the
price set 3 periods before at time F" i.e., *
!
(#)1 &"
!;!+"
is the relevant discount factor between
F and F +31 Using the cost-minimization condition )!
(
!+"
=
<
!+"
-
!+"
, and substituting (62) into
(61), the objective function can be written as:
%
!
!
X
"=0
(G&)
"
"
!;!+"
(

*
!
(#)
*
!+"

1")
4
!+"
!
)!
(
!+"
*
!+"

*
!
(#)
*
!+"

")
4
!+"
)
1 (64)
The FOCs write as
%
!
!
X
"=0
(G&)
"
"
!;!+"
4
!+"
(
1 !6
*
!+"

*
$
!
(#)
*
!+"

")
+
6
*
!+"
)!
(
!+"
*
!+"

*
$
!
(#)
*
!+"

")"1
)
= 01 (65)
Multiplying (65) by *
!
(#)" dividing by (1 !6)" and then simplifying:
%
!
!
X
"=0
(G&)
"
"
!;!+"
4
!+"
(
*
$
!
(#)
*
!+"

*
$
!
(#)
*
!+"

")
!
6
6 !1
)!
(
!+"
*
!+"

*
$
!
(#)
*
!+"

")
)
= 01 (66)
Then
%
!
!
X
"=0
(G&)
"
"
!;!+"

*
$
!
(#)
*
!+"

1")
4
!+"
= (1 +<) %
!
!
X
"=0
(G&)
"
"
!;!+"
)!
(
!+"
*
!+"

*
$
!
(#)
*
!+"

")
4
!+"
" (67)
12
where
1 +< =
6
6 !1
1 (68)
Now, multiply and divide the RHS of (67) by
$
!
(9)
$
!+"
:
%
!
!
X
"=0
(G&)
"
"
!;!+"

*
$
!
(#)
*
!+"

1")
4
!+"
= (1 +<) %
!
!
X
"=0
(G&)
"
"
!;!+"
1
*
$
!
(#)
)!
(
!+"

*
$
!
(#)
*
!+"

1")
4
!+"
1
(69)
Since *
!
(#) does not depend on 3" we can take it out of the summation and bring it to the
LHS of (69):
*
$
!
(#)%
!
!
X
"=0
(G&)
"
"
!;!+"

*
$
!
(#)
*
!+"

1")
4
!+"
= (1 +<) %
!
!
X
"=0
(G&)
"
"
!;!+"
)!
(
!+"

*
$
!
(#)
*
!+"

1")
4
!+"
1
(70)
We can simplify the terms *
!
(#)
1")
which appears in both sides and write
*
$
!
(#)%
!
!
X
"=0
(G&)
"
"
!;!+"

1
*
!+"

1")
4
!+"
1 = (1 +<) %
!
!
X
"=0
(G&)
"
"
!;!+"

1
*
!+"

1")
4
!+"
1)!
(
!+"
1
(71)
Thus, the FOC can be written as
*
$
!
(#) = (1 +<)
%
!
P
!
"=0
(G&)
"
"
!;!+"

1
$
!+"

1")
4
!+"
1)!
(
!+"
%
!
P
!
"=0
(G&)
"
"
!;!+"

1
$
!+"

1")
4
!+"
1
" (72)
or also
*
$
!
(#) = (1 +<) %
!
!
X
"=0
H
!;!+"
)!
(
!+"
1 (73)
The optimal price *
!
(#) is a markup over a weighted average of expected future nominal
marginal costs (nominal wages), where the markup is (1 +<) =
)
)"1
(which corresponds to
the desired markup under exible prices) and where the weight H
!;!+"
is
H
!;!+"
=
(G&)
"
"
!;!+"

1
$
!+"

1")
4
!+"
1
%
!
P
!
"=0
(G&)
"
"
!;!+"

1
$
!+"

1")
4
!+"
1
1 (74)
Note that:
1. for G = 0 the pricing equation becomes
*
!
=
6
6 !1
+
!
(75)
as in the exible price case;
13
2. the optimal price depends on forecasted future values of aggregate variables (4
!
" *
!
) as
well as future marginal costs.
Equivalently, one can express the prices set by forward-looking rms relative to the
overall price index *
!
1 One can divide both sides of (71) by *
!
:
*
$
!
(#)
*
!
%
!
!
X
"=0
(G&)
"
"
!;!+"

1
*
!+"

1")
4
!+"
1 = (1 +<) %
!
!
X
"=0
(G&)
"
"
!;!+"

1
*
!+"

1")
4
!+"
1)!
(
!+"
1
(76)
We could then multiply and divide each term of the RHS of (76) by *
!+"
for 3 = 1" 2" 111:
*
$
!
(#)
*
!
%
!
!
X
"=0
(G&)
"
"
!;!+"

1
*
!+"

1")
4
!+"
1 = (1 +<) %
!
!
X
"=0
(G&)
"
"
!;!+"

1
*
!+"

1")
4
!+"
1)!
!+"
*
!+"
*
!
1
(77)
We can note that
$
!+"
$!
is the cumulative gross ination rate between F and F + 31 We can
write:
*
!+"
*
!
=
""1
Y
==0
*
!+1+=
*
!+=
=
"
Y
==1
!
!+=
= !
!;!+"
1 (78)
We can then write the FOC as
*
$
!
(#)
*
!
= (1 +<)
%
!
P
!
"=0
(G&)
"
"
!;!+"

1
$
!+"

1")
4
!+"
1!
!;!+"
)!
!+"
%
!
P
!
"=0
(G&)
"
"
!;!+"

1
$
!+"

1")
4
!+"
1
" (79)
or also
*
$
!
(#)
*
!
= (1 +<) %
!
!
X
"=0
H
!;!+"
!
!;!+"
)!
!+"
1 (80)
The optimal relative price
$
!
(9)
$
!
is a markup over a weighted average of expected future real
marginal costs (real wages), and the weights are now H
!;!+"
!
!;!+"
1
In equilibrium each producer that chooses a new price *
!
(#) in period F will choose the
same new price *
!
(#) and the same level of output. Then the dynamics of the consumption-
based price index will obey
*
!
=

G*
1")
!"1
+ (1 !G)*
$
!
(#)
1")
1
1!%
1 (81)
Log-linearization
Lets log-linearized the FOC of the rm. From equation (71) we can write
*
$
!
(#)
*
!
%
!
(
!
X
"=0
(G&)
"
"
!;"
4
!+"
*
)"1
!+"
)
= (1 +<) %
!
(
!
X
"=0
(G&)
"
"
!;"
)!
(
!+"
*
!
4
!+"
*
)"1
!+"
)
1
14
Consider rst the left-hand side (LHS) of the above equation, and note that we drop any
product of two or more variables, since we are using a rst order approximation:
LHS ' (*
)
4 )
"
!
X
"=0
(G&)
"
#
( A
$
!
! A
!
) + (*
)
4 )
!
X
"=0
(G&)
"
h
?
!+"
+ (6 !1) A
!+"
+

8
!;"
i
1
Consider now the right-hand side (RHS). Applying the same logic
3
:
RHS ' (*
)
4 )
!
X
"=0
(G&)
"
h
( CE
(
!+"
!A
!
) + ?
!+"
+ (6 !1) A
!+"
+

8
!;"
i
1
Combining LHS and RHS, simplifying, and dividing through by *
)
4 we get
A
$
!
= (1 !G&)%
!
(
!
X
"=0
(G&)
"

CE
(
!+"

)
"
which gives the log-deviation of the newly set price as a discounted streamof the log-deviation
of the nominal marginal cost.
Log-linearizing the price index (81) yields
A
!
= G A
!"1
+ (1 !G) A
$
!
Combine these last two equations to obtain:
I
!
= J CE
!
+&%
!
I
!+1
" (82)
with
J #
(1 !G)(1 !&G)
G
1
and where I
!
= A
!
! A
!"1
1
Using equation (55), we can write the log-linearized aggregate supply (82) as
I
!
= K ( ?
!
! ?
$
!
) +&%
!
I
!+1
This equation represents the AS curve. It can be observed that:
1. as output increases, ination goes up;
2. it is a forward looking AS since
I
!
= K%
!
!
X
"=0
&
"

?
!+"
! ?
$
!+"

(83)
Ination depends on future beliefs about capacity, meaning that ination is a forward
looking variable;
3
Note that we switch to real marginal cost by dividing and multiplying by $
!+"
%
15
3. there is no trade-o! problem for the Central Bank. To control ination the central
bank does not need to generate a recession. Stabilizing output the central bank is also
stabilizing ination.
The complete log-linear system of the dynamic sticky price model is: the IS curve:
?
!
= !> B
!+1
+%
!
?
!+1
(84)
the LM curve:
C
!
! A
!
= ( ?
!
!D@
!
(85)
the Fisher parity condition:
@
!
= B
!+1
+%
!
A
!+1
! A
!
(86)
the forward-looking Phillips curve:
I
!
= K ?
!
+&%
!
I
!+1
(87)
the ination rate:
%
!
I
!+1
= %
!
A
!+1
! A
!
(88)
exogenous process for money growth:
C
!+1
! C
!
= =
'
1 (89)
References
[1] Calvo, G.A. (1983). Staggered Prices in a Utility-Maximizing Framework. J ournal
of Monetary Economics, 12 (3), September, 383-398.
[2] Gal, J. and M. Gertler (1999). Ination Dynamics: A Structural Econometric
Analysis. J ournal of Monetary Economics, 44 (2): 195-222, October.
[3] Rotemberg J.J. and M. Woodford (1998). An Optimization-Based Econometric
Framework for the Evaluation of Monetary Policy: Expanded Version. National Bureau
of Economic Research Technical Working Paper no. 233, May.
[4] Svensson, L.E.O. (1998). Open-Economy Ination Targeting. National Bureau of
Economic Research Working Paper no. 6545, May.
[5] Obstfeld, M. and K. Rogo! (1996). Foundations of International Macroeconomics.
Cambridge, MA: MIT Press.
16