STAGGERED PRICE SETTING AND PERSISTENCE

Daniel Ahelegbey
Blanca Ballesta
Ruhollah Eskandari

DATA EVIDENCE

Monetary policy shocks have a delayed but persistent effect on output

PROBLEM

Most of the models (Lucas Islands, CIA, rigidity on wages…) don’t reproduce this
feature of the data (effects on output but not persistent)

FRAMEWORK

Money supply shocks effects are share between both prices and output. If we want a
large effect on output, it must be we have rather small effect on prices. Hence:

• We want sticky prices

• Prices & wages highly correlated
Wages do not respond to monetary
shocks either

Assumption. - Labor supply equation ߱௧ = ߛ‫ݕ‬௧

Elasticity of real wage wrt output (ߛ)

STAGGERED PRICE SETTING

FIRM TYPE I

‫݌‬ҧଶ ‫݌‬ҧସ

t=1 t=2 t=3 t=4 t=5 t=6 t=7 …

‫݌‬ҧଵ ‫݌‬ҧଷ ‫݌‬ҧହ

FIRM TYPE II
Say ‫݌‬ҧ௧ = price set up by firm at period t (that will be the same for period t+1). Hence

‫݌‬ഥ௧ + ‫݌‬ҧ௧ିଵ
‫݌‬௧ =
2
Money supply shock
‫݌‬ҧ௧
‫݌‬௧
‫݌‬ҧ௧ାଵ
‫݌‬௧ାଶ …

Strong propagation mechanism!!!

THE ORIGINAL MODEL

• The economy has two sectors

o An intermediate goods sector, where a continuum of monopolists set the

price on their unique intermediate good. With production function:
భ
ଵ
ܻ௧ = ቂ‫׬‬଴ ܻ௧ ሺ݅ሻ௤ ݀݅ ቃ , 0<‫<ݍ‬1
೜

o Goods sector operating under perfect competition, using intermediate

goods as inputs. They maximize profits:
ଵ
ଵ ௤ ଵ
ܲ௧ ቈන ܻ௧ ሺ݅ሻ௤ ݀݅ ቉ − න ܲ௧ ሺ݅ሻܻ௧ ሺ݅ሻ݀݅
଴ ଴

Profit maximization

After some manipulation of the FOC and making a Taylor’s approximation one can get:

1 1
‫݌‬ҧ௧ = ൫‫݌‬௧ + ‫ܧ‬௧ ሺ‫݌‬௧ାଵ ሻ൯ + ሺ߱௧ + ‫ܧ‬௧ ሺ߱௧ାଵ ሻሻ
2 2
NOTE: In general, we have into account the overall cost of inputs (‫ݒ‬௧ in Walsh, but
since we assume no capital –details later on-- the only input is labor with cost ߱௧ –
wages--)

Combining the last result with

തതതା
௣೟ ௣ҧ೟షభ
‫݌‬௧ =
ଶ

We get the equilibrium condition:

p −t = 1
2
p −t−1 + E t p −t+1  + w t + E t w t+1
Assumptions

Aggregate demand is given by

yt = m t − pt

The optimal price set in t

p −t = 12 p −t−1 + E t p −t+1  + w t + E t w t+1

The aggregate price in t

p t = 12 p −t + p −t−1 

Nominal money supply follows a random walk

E t m t+1 = m t

Real marginal costs are linearly related to output

w t = γy t

CONSUMERS' PROBLEM

In the particular framework of our homework, the supply of labor is determined by the
tastes of workers in competitive labor markets, hence workers decide their labor supply
typically maximize their utility function subject to their budget constraint.

Max ln(ct ) + ψ ln(1 − l t )

s. t
c t = w t lt

Since there is no capital in the economy

ct = yt

Therefore the consumer problem becomes

Max lnw t lt  + ψ ln1 − lt 

FOC
1
lt = 1+ψ

From the FOC and c t = y t = w t lt

w t = 1 + ψy t

Comparing with the assumption of real marginal costs linearly related to output by
CKM,
γ = 1 + ψ
γ = Labor demand elasticity (exogenously given)
ψ = Weight of leisure time in the utility function (structural parameter)

Substituting the real marginal costs linearly related to output in

p −t = 1
2
−
p t−1 −
+ E t p t+1  + w t + E t w t+1

p −t = 1
2
−
p t−1 −
+ E t p t+1  + γy t + E t y t+1 

Substituting the aggregate demand equation

p −t = 1
2
−
p t−1 −
+ E t p t+1  + γm t − p t + E t m t+1 − E t p t+1 

The aggregate price in t

pt = 1
2
p −t + p t−1
−


Therefore

p −t = 1
2
−
p t−1 −
+ E t p t+1  + γm t − 1
2
−
p −t + p t−1  + E t m t+1 − 1
2
−
E t p t+1 + p −t 

p −t = 1
2
−
p t−1 −
+ E t p t+1  + γm t − 1
2
p −t + p t−1
−
 + E t m t+1 − 1
2
−
E t p t+1 − 1
2
p −t 

γ γ
p −t 1 + 2
+ 2
= 1
2
−
1 − γp t−1 + 1
2
−
1 − γE t p t+1 + γm t + E t m t+1 

p −t 1 + γ = 1
2
−
1 − γp t−1 −
+ E t p t+1  + γm t + E t m t+1 

1−γ γ
p −t = 1
2
−
 1+γ p t−1 + E t p −t+1  +  1+γ m t + E t m t+1 

Since nominal money supply follows a random walk given by

E t m t+1 = m t

1−γ 2γ
p −t = 1
2
−
 1+γ p t−1 −
+ E t p t+1 + 1+γ
mt
 −
Assuming p t follows a linear pattern given by

p −t = ap −t−1 + bm t

E t p −t+1 = aE t p −t + bE t m t+1 = ap −t + bm t

The firm will set up the price according to weighted sum of both, the price the other
type of firms set up yesterday for today the monetary base.

If a is high (lower b) then a monetary shock will not be transformed (will have little
effect) on an increase on expected prices for tomorrow and therefore the prices will not
be much affected today. The shock will have real effects during more time since prices
adapt slowly and hence the persistence of will be longer.

If b is high (lower a) then prices adapt quickly and stronger to monetary shocks, hence
the shock will not lead to a persistent real effect on output.

Therefore

1−γ 2γ
p −t = 1
2
−
 1+γ p t−1 −
+ E t p t+1 + 1+γ
mt

Becomes

1−γ 2γ
p −t = 1
2
 1+γ p −t−1 + ap −t + bm t  + 1+γ
mt

1−γ 1−γ 1−γb+4γ

p −t 1 − a2  1+γ  = 1
2
 1+γ p −t−1 +  21+γ
m t

21+γ−a1−γ 1−γ 1−γb+4γ

p −t  21+γ
 = 1
2
 1+γ p −t−1 +  21+γ
m t

1−γ 1−γb+4γ
p −t =  21+γ−a1−γ p −t−1 +  21+γ−a1−γ m t

1−γ
a =  21+γ−a1−γ 

1−γb+4γ
b =  21+γ−a1−γ 
Solving for a

a 2 1 − γ − 2a1 + γ + 1 − γ = 0

1+γ
a 2 − 2a 1−γ + 1 = 0

1− γ
a= 1+ γ
,

|a| < 1, b = 1−a

Substituting γ = 1 + ψ

(1+ψ )
a = 11+− (1+ψ )
, |a| < 1

−1 < a < 0 b = 1− a

The solution to the model is

p −t = ap −t−1 + 1 − am t

The associated solution for aggregate prices

p t = ap t−1 + 1
2
1 − am t − m t−1 

After money supply shock prices in period t will react strongly. Hence, an increase in
money will not lead to a persistent real effect on output. The result shows that wages
and prices adjust strongly to the shocks hence no persistence effect on output.

The higher the parameter (the more consumers value leisure) the lower the parameter a
which implies less persistency of the monetary shocks in the real variable. If consumer
put less value on leisure time and only care about consumption then the prices will react
slowly to the monetary shocks and therefore there will be more place for an effective
monetary policy.
References

Chari, Kehoe and McGrattan (2000) “Can Sticky Price Models Generate Volatile and
Persistent Real, Exchange Rates?, NBER working paper, no. 7869
Ellison and Scott (2000) “Sticky Prices and Volatile Output”, Journal of Monetary
Economics, 46,621-632
Taylor (1980) “Aggregate Dynamics and Staggered Contracts”, Journal of Political
Economy, 88, 1-24
Watson (1993) “Measures of Fit for Calibrated Models”, Journal of Political Economy,
101, 1011-1041