Transcript Bild 1 - finanspolitiskaradet

Comments on
Anders Forslund:
Den svenska
jämviktsarbetslösheten
– en översikt
Bertil Holmlund
Department of Economics
Uppsala University
1
Outline



Theory
The Swedish development
Recent Swedish reforms




Unemployment insurance
The income tax system
Other tax reforms
Conclusions
2
Equilibrium unemployment theory
3
Can fiscal policy affect the equilibrium
unemployment rate (NAIRU)?




Yes, in an open economy
NAIRU depends in general on the real
exchange rate
Fiscal policy can influence the real
exchange rate
A fiscal expansion can reduce NAIRU
through a real appreciation
4
Equilibrium unemployment in
an open economy
Real exch.
rate
AD
WS/PS
Employment
5
A fiscal expansion in
an open economy
Real exch.
rate
AD
WS/PS
Employment
6
Does the real exchange rate
matter for unemployment?


Direct evidence: Lindblad and Sellin
(2003)
Indirect evidence: taxes and
unemployment


The real exchange rate is part of the total
wedge between real consumer and producer
wages
A real appreciation is analogous to a tax cut
7
Long run considerations


Balanced trade?
Relevant for the very long run
8
The Swedish development
9
10
Why did NAIRU double
between 1990 and 1994?

Institutional changes?





Major tax reform 1991 (less progressive taxes)
Benefit cuts 1993
Coordinated wage bargaining 1991 (Rehnbergavtalet)
Hard to identify major changes in institutions
that can explain a sharp rise in NAIRU in the
early 1990s
But NAIRU probably increased from the 1960s
up to the early 1990s
11
Did NAIRU fall over the 1990s?

Benefit cuts took place
Imperfect wage-indexation of benefits
Temporary work agencies (1993)
Product market competition increased
Coordinated bargaining (Industriavtalet 1997)

So NAIRU probably fell over the 1990s




12
Recent reforms


Less generous unemployment insurance
In-work tax credits (Jobbskatteavdrag)
13
Compare benefit cuts and inwork tax credits (S)

Income after tax when unemployed: B



Income after tax when employed: W+S
Standard model:


Unemployment depends on the replacement rate, b
Model with in-work tax credits, S :


b=B/W is the replacement rate
Unemployment depends on (b – s); s=S/W
In-work tax credits and benefit cuts are
approximately equivalent
14
Swedish reforms
b  B /W, s  S /W
bˆ  b  s (effective repl. rate)
bˆ  b  s  0.10  0.05  0.15
The change in the effective replacement rate has
been substantial
Should reduce equilibrium unemployment
15
Effects on labor force participation


Benefit cuts should reduce labor force
participation
In-work tax credits should increase labor
force participation
16
Other tax reforms with
implications for employment

Property taxes:



Effects on labor supply?


Abolishment of the wealth tax
Cuts in real estate taxes
Reduced supply on the intensive margin
In-work tax credits have income effects
17
In-work tax credit 2007:
average tax rates by income
18
In-work tax credit 2007:
marginal tax rates by income
19
Income and substitution effects
H hours of work, W the wage, Y nonlabor income
The last term is the marginal propensity to earn
out of nonlabor income, mpe
20
Compute the hours response from
exogenous tax cuts (income effects)
H
 Y 
 mpe 

H
 WH 
Y is the income increase associated with the tax
reduction
WH is annual earnings
mpe = -0.10
(Imbens et al, AER Sept. 2001)
21
Hours responses to exogenous tax cuts (%)
Tax cut (1000 SEK)
Income
(1000)
10
20
30
40
50
60
70
300
-0.5
-1.0
-1.4
-1.9
-2.4
-2.9
-3.3
500
-0.4
-0.9
-1.3
-1.8
-2.2
-2.7
-3.1
700
-0.3
-0.6
-1.0
-1.3
-1.6
-1.9
-2.2
22
Conclusions




NAIRU in Sweden has increased gradually
up to the early 1990s
NAIRU has probably fallen over the 1990s
Benefit cuts and in-work tax credits have
probably reduced the NAIRU
In-work tax credits in combination with
other tax reforms have reduced work
hours
23