Empirical Evidence and Earnings Taxation:
L
Lessons
from
f
the
h Mirrlees
Mi l
R i
Review
Lecture 2: Taxation
i off Earnings
i
Munich Lectures in Economics 2010
CESifo
November 2010
Richard Blundell
Bl ndell
University College London and Institute for Fiscal Studies
© Institute for Fiscal Studies
Empirical Evidence and Earnings Taxation
• This lecture will analyse the context, the
impact and the design of earnings tax
reforms
• It will
ill ffocus on ttwo questions:
ti
– How should we measure the impact
p
of
taxation on work decisions and earnings?
– How should we assess the optimality of
tax reforms?
Empirical Evidence and Earnings Taxation
•
A discussion on the role of evidence loosely
organised under five headings:
1. Key margins of adjustment to tax reform
2. Measurement of effective tax rates
3. The importance of information and complexity
4 Evidence
4.
E id
on the
th size
i off responses
5. Implications
p
for tax design
g
Empirical Evidence and Earnings Taxation
•
S b h di ((and
Sub-heading
d subtext)
bt t) ffor th
the llecture:
t
pp y Responses
p
at the Extensive Margin:
g
Labor Supply
What Do We Know and Why Does It Matter?
•
Key chapter (in Mirrlees Review): Brewer, Saez
and Shephard
Shephard,
http://www.ifs.org.uk/mirrleesReview
•
+ commentaries by Moffitt, Laroque and Hoynes
The extensive – intensive distinction is important
for a number
n mber of reasons:
•
Understanding responses to tax and welfare reform
– Jim Heckman, David Wise, Ed Prescott, etc.. all highlight
the importance of extensive labour supply margin,
– a balance needs to be struck between the two margins….
•
The size of extensive and intensive responses are also key
parameters in the recent literature on earnings tax design
–
•
used heavily in the Mirrlees Review.
But the relative importance of the extensive margin is
specific to particular groups
–
I’ll examine a specific case of low earning families (from
Blundell and Shephard
Shephard, 2010) in more detail in what follows
Draw on new empirical evidence: – some examples
•
labour supply responses for individuals and families
– at the intensive and extensive margins
– by age and demographic structure
•
taxable income elasticities
– top of the income distribution using tax return
information
•
income uncertainty
– persistence and magnitude of earnings shocks over
the life-cycle
•
ability to (micro-)simulate marginal and average rates
– simulate reforms
•
So where are the key margins of response?
•
Evidence suggests they are not all the extensive
margin..
– intensive
i t i andd extensive
t i margins
i both
b th matter
tt
– they matter for tax policy evaluation and earnings tax
design
– and they matter in different ways by age and
demographic groups
•
Getting it right for men
Employment for men by age – FR, UK and US 2007
100%
90%
FR
UK
US
80%
70%
60%
50%
40%
30%
20%
10%
0%
16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 52 54 56 58 60 62 64 66 68 70 72 74
Blundell, Bozio and Laroque (2010)
Total Hours for men by age – FR, UK and US 2007
2000
FR
UK
1750
US
1500
1250
1000
750
500
250
0
16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 52 54 56 58 60 62 64 66 68 70 72 74
Blundell, Bozio and Laroque (2010)
Key Margins of Adjustment
•
and for women …..
Female Total Hours by age – US, FR and UK 2007
1600
FR
1400
UK
US
1200
1000
800
600
400
200
0
16
18
20
22
24
26
28
30
32
34
36
38
40
42
44
46
48
50
52
54
56
58
60
Blundell, Bozio and Laroque (2010)
62
64
66
68
70
72
74
Female Hours by age – US, FR and UK 1977
1600
FR
1400
UK
US
1200
1000
800
600
400
200
0
16
18
20
22
24
26
28
30
32
34
36
38
40
42
44
46
48
50
52
54
56
58
60
Blundell, Bozio and Laroque (2010)
62
64
66
68
70
72
74
Decomposition of change in annual hours worked (1977-2007)
1400
United-States
2007 1308 h
2007: 1308 hours
1300
1977: 1212 hours
1200
1977: 1148 h
hours
1977: 1124 hours
1100
2007: 1094 hours 1000
2007: 953 hours 900
France
800
© Institute for Fiscal Studies United-Kingdom
Change in structure
Women 55-74
Men 55-74
Women 30-54
M 30
Men
30-54
54
W
Women
16-29
16 29
Men 16-29
Blundell, Bozio and Laroque (2010)
Thinking about Responses at the Intensive and
E t i Margin
Extensive
M i
•
•
1+1/α
⎧
h
Write within period utility as
− β if h > 0
⎪c −
U = ⎨ 1 + 1/ α
⎪c
if h = 0
⎩
α is the intensive labour supply elasticity and she works when
the value of working at wage w exceeds the fixed cost β.
•
Convenient to describe the distribution of heterogeneity
through
g the conditional distribution of β g
given α,, F(β|
(β| α)) and
the marginal distribution of α.
•
The labour supply and employment rate for individuals of type
α, is
1+α
h( w, α ) = w
α
⎛w ⎞
andd p ( w, α ) = F ⎜
⎟
⎝ 1+ α ⎠
Thinking about Responses at the Intensive and
E t i Margin
Extensive
M i
•
•
The intensive and the employment rate elasticity are
(1+α )
⎞
⎛ w(1+α ) ⎞
(1+α ) ⎛ w
f⎜
ε I (α ) = α and ε E (α ) = w
⎟/F⎜
⎟
+
+
1
1
α
α
⎝
⎠
⎝
⎠
The aggregate hours elasticity is a weighted sum across the
intensive and extensive margins
⎛ w1+α
⎞
d ln H 1
α
| α ⎟ + wα w1+α
= ∫ [α w F ⎜
d ln w H α
⎝ 1+ α
⎠
•
⎛ w1+α
⎞
f⎜
| α ⎟]dG (α )
⎝ 1+ α
⎠
1
= ∫ p( w, α )h( w, α )[ε I (α ) + ε E (α )]dG (α )
Hα
Of course, quasi-linear utility is highly restrictive and we
expect income effects to matter
matter, at least for some types of
households – we use more general models with fixed costs
Measuring Responses at the Intensive and Extensive Margin
•
Suppose the population share at time t of type j is qjt, then
total hours
J
H t = ∑ q jt H jt andd H jt = p jt h jt
j =1
•
Changes in total hours per person written as the sum of
changes across all types of workers and the change in
structure of the population
H t − H t −1 = Δ t + St
where Δ t = ∑ j =1 Δ jt with Δ jt = q jt −1[ H jt − H jt −1 ]
J
•
•
We can also mirror the weighted elasticity decomposition
⎡
Δh j
Δp j ⎤
ΔH 1 J
∑ q j ⎢ p j hj
+ p j hj
⎥
H
H j =1 ⎢⎣
hj
p j ⎦⎥
And derive bounds on extensive and intensive responses for
finite changes
Bounds on Intensive and Extensive Responses (1977-2007)
FR
Year
Men
16-29
Women
16-29
Men
30-54
Women
30-54
Men
55-74
Women
55-74
I-P, I-L
[-37,-28]
[-23, -19]
[-59, -56]
[-49, -35]
[-11, -8]
[-10, -9]
E-L,, E-P
[[-54,, -45]]
[[-19,, -16]]
[[-27,, -23]]
[[71,, 85]]
[[-28,, -25]]
[[6,, 7]]
Δ
-82
-38
-82
36
-36
-3
[-42, -36]
[-26, -23]
[-48, -45]
[-3, -2]
[-22, -19]
[-8, -6]
E-L, E-P
[-35, -29]
[14, 17]
[-25, -22]
[41, 41]
[-23, -20]
[15, 17]
Δ
-71
-9
-70
39
-42
10
[-6, -6]
[1, 1]
[-5, -5]
[14, 19]
[3, 3]
[3, 5]
E-L, E-P
[-13, -13]
[21, 21]
[-14, -14]
[72, 77]
[3, 3]
[33, 35]
Δ
-19
22
-19
90
6
38
UK I-P, I-L
US I-P, I-L
© Institute for Fiscal Studies Blundell, Bozio and Laroque (2010)
Bounds on Intensive and Extensive Responses (1977-2007)
FR
Year
Men
16-29
Women
16-29
Men
30-54
Women
30-54
Men
55-74
Women
55-74
I-P, I-L
[-37,-28]
[-23, -19]
[-59, -56]
[-49, -35]
[-11, -8]
[-10, -9]
E-L,, E-P
[[-54,, -45]]
[[-19,, -16]]
[[-27,, -23]]
[[71,, 85]]
[[-28,, -25]]
[[6,, 7]]
Δ
-82
-38
-82
36
-36
-3
[-42, -36]
[-26, -23]
[-48, -45]
[-3, -2]
[-22, -19]
[-8, -6]
E-L, E-P
[-35, -29]
[14, 17]
[-25, -22]
[41, 41]
[-23, -20]
[15, 17]
Δ
-71
-9
-70
39
-42
10
[-6, -6]
[1, 1]
[-5, -5]
[14, 19]
[3, 3]
[3, 5]
E-L, E-P
[-13, -13]
[21, 21]
[-14, -14]
[72, 77]
[3, 3]
[33, 35]
Δ
-19
22
-19
90
6
38
UK I-P, I-L
US I-P, I-L
© Institute for Fiscal Studies Blundell, Bozio and Laroque (2010)
Why is this distinction important for tax design?
•
Some key lessons from recent tax design theory (Saez
(2002, Laroque (2005), ..)
large extensive elasticity at low earnings can ‘turn
turn
• A ‘large’
around’ the impact of declining social weights
– implying a higher optimal transfer to low earning workers
than to those out of work
– a role for earned income tax credits
• But how do individuals perceive the tax rates on earnings
implicit in the tax credit and benefit system - salience?
– are individuals more likely to ‘take-up’ if generosity
increases? – marginal rates become endogenous…
• Importance of margins other than labo
labourr ssupply/hours
ppl /ho rs
– use of taxable income elasticities to guide choice of top tax
rates
•
Importance of dynamics and frictions
An Empirical Analysis in Two Steps
• The first step (impact) is a positive analysis of household
decisions. There are two dominant empirical approaches
to the measurement of the impact of tax reform
reform…
– both prove useful:
•
1. A ‘quasi-experimental’ evaluation of the impact of
historic reforms /and randomised experiments
•
2. A ‘structural’ estimation based on a general discrete
choice model with (unobserved) heterogeneity
• The second step (optimality) is the normative analysis or
p
p
policy
y analysis
y
optimal
– Examines how to best design benefits, in-work tax
credits and earnings tax rates with (un)observed
heterogeneity and unobserved earnings ‘capacity’
Focus first on tax rates on lower incomes
Main defects in current welfare/benefit systems
• Participation tax rates at the bottom remain very high in
UK and elsewhere
• Marginal tax rates are well over 80% for some low
income working families because of phasing-out
phasing out of
means-tested benefits and tax credits
– Working Families Tax Credit + Housing Benefit in UK
– and interactions with the income tax system
– for example, we can examine a typical budget
constraint for a single mother in the UK…
Particular Features of the UK Working Tax Credit
• hours of work condition
– minimum hours rule - 16 hours per week
– an additional hours-contingent payment at 30 hours
• family eligibility
– children ((in full time education or younger)
y
g )
– adult credit plus amounts for each child
• income
i
eligibility
li ibilit
– family net income below a certain threshold
– credit is tapered away at 55% (previously 70% under
FC)
The UK Working Families Tax Credit
£6,000
£5,000
WFTC
£4,000
£3,000
£2,000
£1,000
£0
£0
£5,000
£10,000
£15,000
Gross income (£/year)
© Institute for Fiscal Studies
£20,000
£25,000
The US EITC and the UK WFTC compared
£6,000
£5,000
WFTC
£4 000
£4,000
EITC
£3,000
£2,000
£1,000
£0
£0
£5,000
£10,000
£15,000
£20,000
£25,000
Gross income (£/year)
• P
Puzzle:
l WFTC about
b t ttwice
i as generous as th
the US EITC bbutt
with about half the impact. Why?
© Institute for Fiscal Studies
The interaction of WFTC with other benefits in the UK
Low wage
lone parent
£300
£250
£200
WFTC
Net earnings
Other income
£
£150
£100
£50
hours of work
48
44
40
36
32
28
24
20
16
8
12
4
0
£0
The interaction of WFTC with other benefits in the UK
Low wage
lone parent
£300
£250
£200
WFTC
Income Support
Net earnings
Other income
£150
£100
£50
hours of work
48
44
40
36
32
28
24
20
16
8
12
4
0
£0
The interaction of WFTC with other benefits in the UK
Low wage
lone parent
£300
£250
Local tax rebate
Rent rebate
WFTC
Income Support
Net earnings
Other income
£200
£150
£100
£50
hours of work
48
44
40
36
32
28
24
20
16
8
12
4
0
£0
Strong
g
implications
for EMTRs,
PTRs and
labour supply
The interaction between taxes, tax credits and benefits
£350
Income support
£300
Council tax
benefit
Net we
eekly inco
ome
£250
Housing benefit
£200
Working tax
credit
£150
Child tax credit
£100
Child benefit
£50
£0
£0
£20 £40 £60 £80 £100 £120 £140 £160 £180 £200 £220
Gross weekly earnings
Net earnings
less council tax
Notes: Lone p
parent,, with one child aged
g between one and four,, earning
g the minimum wage
g ((£5.80
per hour), with no other private income and no childcare costs, paying £80 per week in rent to live
in a council tax Band B property in a local authority setting council tax rates at the national
average
But this is just an example….
• What does the tax and benefit system imply across
the distribution of earnings and different family
types?
– What do effective marginal tax rates look like? – the
proportion of a small increase in earnings taken in
tax and withdrawn benefits
– What do participation tax rates look like? – the
incentive to be in paid work at all – defined by the
proportion of total earnings taken in tax and
withdrawn benefits.
40%
50
0%
60%
%
70%
80%
Average EMTRs for different family types
0
100
200
300
400
500
600
700
800
900 1000 1100 1200
Emplo er cost (£/week)
Employer
(£/ eek)
Single, no children
Partner not working
working, no children
Partner working, no children
Lone parent
Partner not working
working, children
Partner working, children
30%
%
40%
50%
60%
70%
7
Average PTRs for different family types
0
100
200
300
400
500
600
700
800
900 1000 1100 1200
E
Employer
l
costt (£/
(£/week)
k)
Single, no children
P t
Partner
nott working,
ki
no children
hild
Partner working, no children
Lone parent
P t
Partner
nott working,
ki
children
hild
Partner working, children
Can the reforms explain weekly hours worked?
Single Women (aged 18-45) - 2002
Blundell and Shephard (2009)
Hours’ distribution for lone parents, before WFTC
Blundell and Shephard (2010)
Hours’ distribution for lone parents, after WFTC
Blundell and Shephard (2010)
Hours trend for low ed lone parents in UK
1600
1550
1500
1450
1400
1350
1300
1250
1200
1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
Employment trends for lone parents in UK
0.8
0.75
0.7
0.65
0.6
College
0.55
No College
No College
0.5
0.45
0.4
0.35
0.3
1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
WFTC Reform: Quasi-experimental Evaluation
Matched Difference-in-Differences
Average Impact on % Employment Rate of Single Mothers
Single Mothers Marginal
Effect
F il
Family
45
4.5
Resources
Survey
Labour Force
4.7
Survey
Standard
Error
1 55
1.55
Sample Size
0.55
233,208
25 163
25,163
Data: FRS, 45,000 adults per year, Spring 1996 – Spring 2002.
B
Base
employment
l
t llevel:l 45% iin S
Spring
i 1998.
1998
Matching Covariates: age, education, region, ethnicity,..
Alternative approaches to measuring the impact:
• Structural model
– Simulate effect of actual or hypothetical reforms
Simulate effect of actual or hypothetical reforms
– Useful for (optimal) design too, but, robust?
• Quasi‐experiment/Difference‐in‐differences
– Compares outcomes of eligibles
p
g
and non‐eligibles
g
and estimates ‘average’ impact of past reform
– Only indirectly
Only indirectly related to what is needed for optimal design
related to what is needed for optimal design
– Can use this quasi‐experimental evidence to (partially) validate the structural model
validate the structural model
• Randomised experiment? SSP?
Canadian Self Sufficiency Program
• Randomised‐Control experimental design
• Do financial incentives encourage work among low g
g
skilled lone parents?
• The aim was to encourage employment among single g
p y
g g
parents on welfare
– 50% earnings supplement –
g
pp
as a tax credit
– at least 30 hours per week job
– On earnings up to an annual limit
On earnings p to an ann al limit of $36000
of $36000
• provided to the individual, not the employer, as in EITCs
EITC
39
Canadian Self Sufficiency Program
Budget Constraint for a Single Parent on Minimum Wage
Incom
me per M
Month ($$1995)
2500
2000
SSP
1500
1000
IA
500
0
0
5
10
15
20
25
30
35
40
45
50
55
Weekly Hours of Work
Income Assistance
Self Sufficiency Program
40
60
.2
.15
.1
Employment rate
.25
.3
SSP: Employment Rate by months after RA
0
10
20
30
40
Months after random assignment
control
experimental
50
60
Key
y features of the structural model
U (ch , h, P; X , ε )
typically
yp
y approximated
pp
by
y shape
p constrained sieves
Preferences
• Structural model allows for
- unobserved work-related fixed costs
- childcare costs
- observed and unobserved heterogeneity
- pprogramme
g
pparticipation
p
‘take-up’
p costs
• See Blundell and Shephard (2010)
Importance of take-up and information/hassle costs
0
Pro
obability of take-up
.2
.4
.6
..8
1
Variation in take-up
pp
probability
y with entitlement to WFTC
0
50
100
Lone parents
© Institute for Fiscal Studies
150
200
WFTC entitlement (£/week, 2002 prices)
Couples
Net Income schedule :
Tax
P: take-up
yhP = wh + I − t ( wh, I ) − Ch + Ψ 0 ( w, h, I ) + PΨ1 ( w, h, I )
Transfers
or
yhP = y hP + PΨ1 ( w, h, I )
the tax-credit payment function Ψ1 ( w, h, I ) depends on:
h
hours
(th
(through
h th
the h
hours condition
diti off entitlement)
titl
t)
other income I
demographic characteristics X
44
Structural Model Elasticities – low education lone parents
(a) Youngest Child Aged 5-10
Weekly
Earnings
0
Density
Extensive
Intensive
0.280 (.020)
0.321 (.009)
0.152 (.005)
0.058 (.003)
0.820 (.042)
0.085 (.009)
0.219 (.025)
0.194 (.020)
0.132 (.010)
0 4327
0.4327
50
0.1575
150
0.1655
250
0.1298
350
0.028
Employment elasticity
Structural Model Elasticities – low education lone parents
(c) Youngest Child Aged 0-4
Weeklyy
Earnings
Densityy
Extensive
Intensive
0
0.5942
50
0.1694
0.168 (.017)
0.025 (.003)
150
0 0984
0.0984
0 128 ((.012)
0.128
012)
0 077 ((.012)
0.077
012)
250
0.0767
0.043 (.004)
0.066 (.010)
350
0 0613
0.0613
0 016 ((.002)
0.016
002)
0 035 ((.005)
0.035
005)
Participation elasticity
0.536 (.047)
• Differences in intensive and extensive margins by age and
demographics have strong implications for the design of the tax
schedule...
h d l
• But do we believe the structural model estimates?
Structural Simulation of the WFTC Reform:
WFTC Tax Credit Reform
All
Change in employment rate:
Average change in hours:
6.95
0 74
0.74
1.79
02
0.2
y-child
0 to 2
3.09
0 59
0.59
0.71
0 14
0.14
y-child
3 to 4
7.56
0 91
0.91
2.09
0 23
0.23
y-child
5 to 10
7.54
0 85
0.85
2.35
0 34
0.34
Notes: Simulated on FRS data; Standard errors in italics.
– relatively ‘large’ impact
Blundell and Shephard (2010)
y-child
11 to 18
4.96
0 68
0.68
1.65
02
0.2
Impact of WFTC reform on lone parent, 2 children
W eekly n
net inco
ome
£300
1997
2002
£250
£200
£150
£100
0
4
8
12 16 20 24 28 32 36 40 44 48
Hours/week
•
Notes: Two children under 5. Assumes hourly wage of £4.10, no housing costs or council tax
liability and no childcare costs.
Impact of WFTC and IS reforms on lone parent, 2 children
W
Weekly
net inco
ome
£300
1997
2002
8
16 20
£250
£200
£150
£100
0
4
12
24 28
32
36 40
44
Hours/week
•
Notes: Two children under 5. Assumes hourly wage of £4.10, no housing costs or council tax
liability and no childcare costs.
48
Structural Simulation of the WFTC Reform:
Impact of all Reforms (WFTC and IS)
All
Change in employment rate:
Average change in hours:
4.89
0.84
1.02
0 23
0.23
y-child
0 to 2
0.65
0.6
0.01
0 21
0.21
y-child
3 to 4
5.53
0.99
1.15
0 28
0.28
y-child y-child
5 to 10 11 to 18
6.83
4.03
0.94
0.71
1.41
1.24
0 28
0.28
0 22
0.22
• shows the importance of getting the effective tax rates right
especially when comparing with quasi-experiments.
•
compare with experiment or quasi-experiment.
Evaluation of the ‘ex-ante’ structural model
• The diff-in-diff impact parameter can be identified from the
structural evaluation model
• Simulated diff-in-diff parameter
• The structural model then defines the average impact of the
policy on the treated as:
α SEM ( X ) = Pr[h > 0 | X , D = 1] − Pr[h > 0 X , D = 0]
• Compare
C
simulated
i l t d diff-in-diff
diff i diff momentt with
ith diff-in-diff
diff i diff
DD
αSEM
=∫
X
∫
X
T =1,t =1
f
(
X
,
ε
,
D
=
1)
)
dF
dFX −∫
ε
∫
ε
⎡
− ⎢∫ f ( X , ε , D = 0)dFεT =0,,t =1dFX − ∫
X
⎣ε
X
T =1,t =0
f
(
X
,
ε
,
D
=
0)
)
dF
dFX
ε
∫
ε
∫ε f ( X , ε , D = 0)dFε
T =0,,t =0
⎤
dFX ⎥
⎦
Evaluation of the ex-ante model
• The simulated diff-in-diff parameter from the structural
evaluation model is precise and does not differ
significantly from the diff-in-diff estimate
• Compare
C
simulated
i l t d diff-in-diff
diff i diff momentt with
ith diff-in-diff
diff i diff
– .21 (.73), chi-square p-value .57
• Consider additional moments
– education:
d ti
low
l education:
d ti
0.33
0 33 (.41)
( 41)
– youngest child interaction
• Youngest child aged < 5: .59 (. 51)
• Youngest
Y
t child
hild agedd 5-10:
5 10 .31
31 (.35)
( 35)
How do we think about an optimal design?
• Assume we want to redistribute ‘£R’ to low ed. single parents,
what is the ‘optimal’
optimal way to do this?
• Recover optimal tax/credit schedule in terms of earnings
– use Diamond-Saez
Di
dS
approximation
i i in
i terms off extensive
i andd
intensive elasticities at different earnings
Ti − Ti −1
1
=
ci − ci −1 ei hi
⎡
T j − T0 ⎤
h j ⎢1 − g j − η j
⎥.
∑
c j − c0 ⎥⎦
j ≥i
⎢⎣
I
• also ‘complete’ Mirrlees optimal tax computation
A ‘microeconometric’ optimal tax design framework
• Assume earnings (and certain characteristics) are all that is
observable to the tax authority
– relax below to allow for ‘partial’ observability of hours
Social welfare,
welfare for individuals of type X
W=
∫ ∫ε Γ(U (wh
− T (w, h ; X ), h ; X , ε ))dF (ε )dG(w; X )
w, X
The tax structure T(.) is chosen to maximise W, subject
to:
∫ ∫ε T ( wh , h ; X )dF (ε )dG ( w; X ) ≥ T (= − R )
w, X
for a given R.
R
Control preference for equality by transformation function:
Γ (U | θ ) =
1
θ
θ
(exp
U
)
− 1}
{
when
h θ is
i negative,
ti the
th function
f ti favors
f
the
th equality
lit off
utilities. θ is the coefficient of absolute inequality aversion.
If θ < 0 then analytical solution to integral over (Type I
extreme-value) j state specific errors (BS, 2010)
1⎡
⎤
θ
Γ(1 − θ ) ⋅ (∑ exp u ( j )) − 1⎥
⎢
θ⎣
h
⎦
Objective: robust policies for fairly general social welfare
weights, document the weights in each case
Implied Optimal Schedule, Youngest Child Aged 5-10
Weekly earnings
March 2002 prices
Blundell and Shephard (2010)
Implied Optimal Schedule, Youngest Child Aged 5-10
• Results Suggests ‘dynamic’ tax incentives according to age of
(youngest)
(y
g ) child
• Redistributing towards early years (see Table 10 in Blundell and
Shephard, 2010)
Implications for Tax Reform
• Change transfer/tax rate structure to match lessons from
‘new’ optimal tax analysis and empirical evidence
– in the Review we use a similar design framework for family
labour supply and early retirement
• Key role off labour supply responses at the extensive and
intensive margins
• Both matter but differ by gender, age, education and family
composition
– lone parents, married parents, pre-retirement low earners.
• Results for lone parents suggest lower marginal rates at the
bottom
– means-testing
g should be less aggressive
gg
– at least for some key groups =>
Implications for Tax Reform
• ‘Life-cycle’
‘Lif
l ’ view
i
off ttaxation
ti
– distinguish by age of (youngest) child for mothers/parents
– pre-retirement ages
– effectivelyy redistributing
g across the life-cycle
y
– a ‘life-cycle’ rearrangement of tax incentives and welfare
payments to match elasticities and early years investments
– results in Tax by Design show significant employment and
earnings increases
• Hours rules? – at full time for older kids,
– welfare gains depend on ability to monitor hours
• Dynamics and frictions?
– some time
ti
to
t adjust
dj t but
b t little
littl iin th
the way off experience
i
effects
ff t
for low-skilled
Dynamic effects on wages for low income welfare
recipients?
6
H
Hourly
rea
al wages
6.5
7
7.5
8
8.5
SSP: Hourly wages by months after RA
0
10
20
30
40
Months after random assignment
control
© Institute for Fiscal Studies
experimental
50
60
10
00
M
Monthly
earnings
e
200
300
400
SSP: Monthly earnings by months after RA
0
10
20
30
40
Months after random assignment
control
t l
© Institute for Fiscal Studies
experimental
i
t l
50
60
Evidence on experience effects from the SSP
•
Little evidence of employment enhancement or wage
progression
•
Other evidence, Taber etc, show some progression
but quite small
•
Remains a key area of research
–
ERA Policy in UK.
© Institute for Fiscal Studies
At the top too… the income tax system lacks coherence
Income tax schedule for those aged under 65, 2010–11
70%
Marginal in
ncome tax ratte
60%
50%
40%
30%
20%
10%
0%
0
20
40
60
80
100
120
Gross annual income (£000s)
140
160
180
Top tax rates and taxable income elasticities
An ‘optimal’ top tax rate (Brewer, Saez and Shephard, MRI)
e – taxable income elasticity
t = 1 / (1 + a·e) where a is the Pareto parameter.
Estimate e from the evolution of top incomes in tax return
data following large top MTR reductions in the 1980s
Estimate a (≈ 1.8)
1 8) from the empirical distribution
Top incomes and taxable income elasticities
A . T o p 1 % In c o m e S h a re a n d M T R , 1 9 6 2 -2 0 0 3
80%
16%
70%
14%
T o p 1 % in c o m e s h a re
40%
10%
30%
8%
20%
6%
10%
Source: MR1, UK SPI (tax return data)
2002
2
1998
1
1994
1
1990
1
1986
1
1982
1
1978
1
1974
1
1970
1
4%
1966
1
0%
Income S
Share
12%
T o p 1% M T R
50%
1962
1
Marginal Ta
ax Rate
60%
B . T o p 5 -1 % In c o m e a n d M T R , 1 9 6 2 -2 0 0 3
80%
16%
70%
14%
40%
10%
30%
8%
20%
T o p 5 -1 % M T R
10%
T o p 5 -1 % in c o m e s h a re
6%
2002
1998
1994
1990
1986
1982
1978
1974
1970
4%
1966
0%
Income S
Share
12%
50%
1962
Marginal Ta
M
ax Rate
60%
Taxable Income Elasticities at the Top
Simple Difference (top 1%)
DD using top 5-1%
as control
1978 vs 1981
1986 vs 1989
1978 vs 1962
2003 vs 1978
0.32
0 38
0.38
0.63
0 89
0.89
0.08
0 41
0.41
0.86
0 64
0.64
Full time series
0.69
(0.12)
0.46
(0.13)
With updated data the estimate remains in the .35 - .55 range with a
central estimate of .46, but remain quite fragile
Note also the key relationship between the size of elasticity and the tax
base (Slemrod and Kopczuk, 2002)
Pareto distribution as an approximation to the income distribution
Pro
obability density ( log scale )
0.0100
0.0010
Pareto distribution
Actual income distribution
0.0001
0.0000
0.0000
£100 000 £150,000
£100,000
£150 000 £200,000
£200 000 £250,000
£250 000 £300,000
£300 000 £350,000
£350 000 £400,000
£400 000 £450,000
£450 000 £500,000
£500 000
Pareto parameter quite accurately estimated at 1.8
=> revenue maximising tax rate for top 1% of 55%.
Reforming Taxation of Earnings
•
Change transfer/tax rate structure to match lessons from ‘new’
optimal tax analysis
•
l
lower
marginal
i l rates
t att th
the b
bottom
tt
– means-testing should be less aggressive
– distinguish by age of youngest child
•
age-based taxation
– pre-retirement ages
•
limits to tax rises at the top,
p, but
– base reforms - anti-avoidance, domicile rules, revenue shifting
•
Integrate different benefits and tax credits
– improve administration, transparency, take-up, facilitate
coherent design
•
Undo distributional effects of the rest of the package…
htt //
http://www.ifs.org.uk/mirrleesReview
if
k/ i l R i
Richard Blundell
University College London and Institute for Fiscal Studies
© Institute for Fiscal Studies