slides ch5

Lectures on Monetary Policy, In‡ation and the Business Cycle Monetary Policy Tradeo¤s: Discretion vs. Commitment by Jor...

0 downloads 204 Views 147KB Size
Lectures on Monetary Policy, In‡ation and the Business Cycle Monetary Policy Tradeo¤s: Discretion vs. Commitment

by Jordi Galí

The Monetary Policy Problem 1 X t min E0f [

yet2 +

y

t=0

subject to:

2 t ]g

(1)

= Etf t+1g + yet + ut where futg evolves exogenously according to t

ut =

u

ut

1

+ "t

In addition: yet =

1

(it

Etf

yt+1g rtn) + Etfe

t+1 g

Note: utility based criterion requires

y

=

(2)

Optimal Policy with Discretion Each period CB chooses (xt;

t) y

subject to

yet2 +

2 t

= yet + vt Etf t+1g + ut is taken as given. t

where vt

to minimize

Optimality condition:

yet =

Equilibrium

(3)

t y

= y q ut yet = q ut it = rtn + q [ (1 u) + t

where q

2+

1 y (1

u)

y u]

ut

(4) (5) (6)

Implementation: it = rtn + [(1 uniqueness condition:

y

u)

+ y

u]

t

> 1 (likely if utility-based:

> 1)

Alternatively, it = rtn + q [ uniqueness condition:

(1

u)

> 1:

+

y u]

ut +

(

t

yq

ut )

Optimal Policy with Commitment State-contingent policy fe yt ;

1 E0 2

1 t gt=0 1 X

that maximizes t

(

y

t=0

subject to the sequence of constraints: t

=

Et f

Lagrangean: L=

1

X 1 E0 2 t=0

t

[

y

First order conditions:

yet2 + y

t

for t = 0; 1; 2; :::and where

t+1 g

yet

+

1

2 t

+2

= 0.

t

(

=0

t 1

2 t)

yet + ut

+

t t

yet2 +

=0

t

yet

t+1 )]

Eliminating multipliers: ye0 =

for t = 1; 2; 3; :::..

yet = yet

(7)

0 y 1

t

(8)

y

Alternative representation:

for t = 0; 1; 2; :::where pbt

yet =

pt

p

y 1

.

pbt

(9)

Equilibrium pbt = a pbt for t = 0; 1; 2; :::where a

1

+ a Etfb pt+1g + a ut y

y (1+

)+

2

Stationary solution:

pbt =

for t = 0; 1; 2; :::where

pbt 1 + (1 p 2

1

1 4 a 2a

u)

ut

(10)

2 (0; 1):

! price level targeting ! yet =

for t = 1; 2; 3; :::as well as

yet

ye0 =

1

y (1

y (1

u)

u)

u0

ut

(11)

Optimal Monetary Policy: Discretion vs. Commitment

Appendix: Sources of Cost Push Shocks Variations in desired price markups. t Assumption: time varying desired markup: nt t 1 Log-linearized optimal price setting rule: 1 X pt = (1 ) ( )k Etf nt+k + mct+k + pt+k g = (1

)

k=1 1 X

(

k=1

where mc ft

mct + t

= = =

n t.

Thus,

Etf Etf Etf

+ t+1 g + t+1 g +

t+1 g

)k Etfmc f t+k + pt+k g mc ft mc c t + ( nt ) (yt y t) + ( nt

)

where y t equilibrium output under a constant price markup .

Exogenous Variations in Wage Markups mct = wt at = w;t + mrst at = w;t + ( + ') yt

(1 + ') at

Thus, mc c t = ( + ') (yt

y t) + (

w)

w;t

where y t : equilibrium output under a constant price and wage markup. Implied in‡ation equation: t

= Etf

t+1 g

+ (yt

y t) + (

w;t

w)