Transcript Slide 1

Comments on Mourougane and Vogel's
The Impact of Selected Structural Reforms:
Speed of Adjustment and Distributional Effects
Maroje Lang, CNB
Reform delay and fatigue

Consensus about positive (long-term) effects of
structural reforms

However, structural reforms often delayed or stopped
Why ?
Motivation
(authors try to answer the following):
1.
2.
why reforms are delayed: adjustment path is important
for political economy reasons (short term costs vs. longterm benefits)
why reforms work in some countries: how institutions
(rigidities in labor and product markets and characteristics of
financial markets (including monetary policy)) affect the pace of
3.
adjustment to selected reforms
how to deal with opposition to reform: analyse the
distributional effects and offer solutions
Methods used
1.
2.



descriptive analysis - illustrate the actual data on impact of reforms
simulations using two types of models: small macroeconomic neoKeynesian model and micro founded DSGE model
to illustrate problem and show robustness
the right approach for the OECD study/working paper
some methods are self-contained – can be presented separately
4. DGE model

main and most interesting/promising part of the paper

closed economy with a number of frictions

analyses the adjustment path of change in selected
variables – proxies for structural reforms
... under different institutional settings
+ distribution between agents
Policy simulations
Model with
income
tax
employer social
security
replacement rate
consumption tax
employment
adjustment cost
(+search&matching)
price adjustment
financial market
imperfections
Reform proxies
income
employer
replacement
↓
fiscal
Conclusion
different
and price
costs (EU and US)
Employment adjustment costs
have a MODERATE impact
real adjustment ... while price
adjustment costs affect
financial market
imperfections
(liquidity constrained
consumers)
Financial frictions have
STRONG real effects
analysis of
distributional effects
and compensation
schemes
Some reforms can entail
term distributional costs
... but budgetary schemes
reduce them
security
contributions

monetary policy
fiscal rule
tax
Subject to
Extensions/refinements





model structural reforms which reduce employment and
price adjustment costs
conduct alternative monetary policy experiments using
DGE model (instead of NK, perhaps different results)
sensitivity analysis?
model solution procedure?
possible problem with derivation of the pricing equation
(Phillips curve)

if so, does it change the results?
Typos ?
(6)



1
h

 
 Et  o
 o
c
o
o 
(1   t ) Pt  (Ct  hCt 1 )
 (Ct 1  hCt ) 
(14)
Yt  Ct  t Nt  tYt
o
t
likewise (21)
1
Phillips curve (15) (likewise 20)
 Pt j j
e Wt
j
j
j
j j
Max
E

Y

1


N


N


Yt 

0
0
,
t
t
t
t
t
t
t

Pt j
Pt
t 0
 Pt





s.t.
 0 ,t   t
 Pt j 
Yt  N t    Yt
 Pt 
j
t
0
j
is stohastic discount factor
  P
 Pt
 P
P Y 
 1 t  1   Et  t 1  t 1  1 t 1 t 1  
 Pt 1  Pt 1
 Pt Yt 
 t  Pt
 
2
2
2

 Y
  Y
 








W
P
Y
Y
Y

1
e

t
t
t
t
t
t

1
t

1
t

1

  
   1 t
 
 1   
 1
 
 1  Et 
 1
 Yt 1

Pt 2  Pt 1 
Yt 1 2  Yt 1 
Yt   
 t  Yt




 




Alternative derivation of Phillips curve
Ireland(2004) and Ali(2001)
 Pt j j
e Wt
j
j
j 
Max
E0  0,t  Yt  1   t
Nt  t Nt  t Yt 
j
Pt
Pt
t 0
 Pt




 t 1  Pt 1  Pt 1 Yt 1 
 Pt
 Pt

 
 1
 1   Et 
 1

 Pt 1  Pt 1
 Pt Yt 
 t  Pt
2

 Y
  Y
 




W
Y
Y
e

t
t
t
t

1
t

1
t

1

  
   1 t
 1   
 1
 Et 
 1
 Yt 1

Pt
Yt 1
t  Yt
Yt   






 




3. Small Neo-Keynesian model

Model difficult to understand

model not shown; only some estimated equations

some variables not named and relationships not described


ω*; p (producer prices?) vs pcore (consumer prices); what is (w - p)?;
typos
Estimation using general to specific approach (+ moving
average terms)

makes understanding the model even more difficult

additional equations defining moving averages?
Small Neo-Keynesian model (cont.)

Derived conclusions thus difficult to evaluate
1.
no international spillovers - international linkages not presented
2.
monetary policy can speed up the adjustment path ???

additional reason against joining the currency union?
 cause for common EU reforms vs individual country reforms
 especially interesting as also concluded that employment and price
frictions have no strong effect
 is it due to the difference in models used? (NK vs DGE)
2. Descriptive analysis II
Correlation between the change in institutions and
cumulative change in NAIRU
Strong conclusion : very gradual impact of reforms (5-10yrs)

But: weak statistics (rule of thumb 0,1 correlation, no Granger causality)
+ macromodels suggest different time of impact - which one to use?

DSGE 3-4years; NK 20+ years
lag 0 ?

lag 0 ?

1. Descriptive analysis I
1.

Contribution of frictions to structural unemployment (for OECD)
Use results from Bassanini and Duval (2006) to
ˆ  ˆ
1. Define structural unemployment: u*  u  gap

2.
structural unemployment drops in late 1990's
Define contributions from reforms:

contribution = constant * variable


rather show variables (if must) (+correlations)
Problem with aggregated countries data : (different directions of
reforms)


contribution( x)  ˆ  x
tax wedge seems the most important
More interesting is looking into individual countries (US)
2005
2006
9%
2001
2002
2003
2004
1997
1998
1999
2000
1993
1994
1995
1996
1989
1990
1991
1992
1985
1986
1987
1988
1983
1984
US: the case for Bush tax cuts
ACTUAL US UNEMPLOYMENT
11%
10%
w ithout tax cut?
8%
7%
6%
5%
4%
3%
2%
1%
0%