Public Economics: Poverty and Inequality
Andrew Hood
© Institute for Fiscal Studies
Overview
• Why do we use income?
• Income Inequality
– The UK income distribution
– Measures of income inequality
– Explaining changes in income inequality
• Income Poverty
– Measuring income poverty
– Universal Credit and poverty
• Summary and conclusions
© Institute for Fiscal Studies
Why income?
• Economic analysis tends to focus on income inequality and income
poverty
–
not because income is the only thing that matters...
–
...but because it is arguably the best measure of living standards
we’ve got
• Consumption is conceptually a better indicator of living standards
–
Income snapshots can be misleading
© Institute for Fiscal Studies
Those with the lowest incomes do not have the
lowest consumption
£490
£420
Median Expenditure
£350
£280
£210
£140
£70
£0
£0
£100
£200
£300
Income
Source: Brewer and O’Dea (2012)
£400
£500
Why income?
• Economic analysis tends to focus on income inequality and income
poverty
–
not because income is the only thing that matters...
–
...but because it is arguably the best measure of living standards
we’ve got
• Consumption is conceptually a better indicator of living standards
–
Income snapshots can be misleading
–
but consumption is much harder to measure and the data is much
better (and more up-to-date) for income
© Institute for Fiscal Studies
Measurement of income
• Income as measured by government in “Households Below
Average Income” (HBAI)
• Income is measured net of direct taxes and benefits
• Measured at the household level (implicitly assumes income
sharing)
• Adjusted for household size (equivalised)
• Adjusted for inflation
• Based on Family Resources Survey (from 1994-5 onwards)
– 25,000 households across the UK
– Subject to sampling error
© Institute for Fiscal Studies
Income Inequality
© Institute for Fiscal Studies
The UK income distribution in 2011–12
Household income (£ per week)
2,500
2,000
1,500
1,000
500
0
10
20
30
Source: Cribb et. al. (2013)
© Institute for Fiscal Studies
40
50
60
Percentile point
70
80
90
The UK income distribution in 2011–12
Household income (£ per week)
2,500
2,000
1,500
50th percentile:
£427
1,000
500
0
10
20
30
Source: Cribb et. al. (2013)
© Institute for Fiscal Studies
40
50
60
Percentile point
70
80
90
The UK income distribution in 2011–12
Household income (£ per week)
2,500
2,000
1,500
1,000
50th percentile:
£427
10th percentile:
£221
500
0
10
20
30
Source: Cribb et. al. (2013)
© Institute for Fiscal Studies
40
50
60
Percentile point
70
80
90
The UK income distribution in 2011–12
Household income (£ per week)
2,500
2,000
90th percentile:
£865
1,500
1,000
50th percentile:
£427
10th percentile:
£221
500
0
10
20
30
Source: Cribb et. al. (2013)
© Institute for Fiscal Studies
40
50
60
Percentile point
70
80
90
The UK income distribution in 2011–12
Household income (£ per week)
2,500
2,000
90th percentile:
£865
1,500
1,000
50th percentile:
£427
10th percentile:
£221
500
0
10
20
30
Source: Cribb et. al. (2013)
© Institute for Fiscal Studies
40
50
60
Percentile point
70
80
90
Gross annual earnings required to reach certain
percentiles of the UK income distribution
Single
individual
One-earner
couple,
no children
Two-earner
couple,a
no children
50th
£18,000
£29,000
£26,000
One-earner
couple,
two children
under 14
£39,000
90th
£41,000
£66,000
£59,000
£94,000
99th
£125,000
£198,000
£174,000
£290,000
a
With each partner earning the same amount.
Source: Cribb et. al. (2013)
• Equivalisation makes a really big difference
• The gap between the 90th and the 99th percentiles is pretty
significant
© Institute for Fiscal Studies
Measuring income inequality
• Broadly 2 types of inequality measures
1. Ratio measures – compare incomes at different points of the
distribution
2. Summary measures – attempt to collapse the whole income
distribution into a single number
© Institute for Fiscal Studies
Measuring income inequality: ratio measures
6.0
5.5
5.0
Ratio
4.5
4.0
3.5
3.0
2.5
2.0
1961
1963
1965
1967
1969
1971
1973
1975
1977
1979
1981
1983
1985
1987
1989
1991
1993-94
1995-96
1997-98
1999-00
2001-02
2003-04
2005-06
2007-08
2009-10
2011-12
1.5
90/10 ratio
Source: Cribb et. al. (2013)
© Institute for Fiscal Studies
99/50 ratio
Measuring income inequality: the Gini coefficient
100
Perfect equality
Share of total income (%)
90
Gini =
80
70
A
A + B
60
50
A
40
30
UK Lorenz curve in
2011-12: Gini = 0.34
20
B
10
0
0
10
20
30
40
50
60
70
80
Percentage of population, ranked by income
90
100
Gini coefficient: 1979 to 2011–12
Gini coefficient
0.40
0.35
0.30
2011-12
2009–10
2007–08
2005-06
2003–04
2001-02
1997–98
1995–96
1993-94
1991
1999-2000
Source: Cribb et. al. (2013)
1989
1987
1985
1983
1981
1979
0.25
• Gini rose dramatically in the 1980s (0.25 in 1979 to 0.34 in 1990)
• Big fall in recent years (0.36 in 2007–08 to 0.34 in 2011–12)
© Institute for Fiscal Studies
Why has income inequality risen?
• Lots of explanations
–
Skills-biased technological changes [see Acemoglu (2002), Machin
(2001) and Goldin and Katz (2008)]
–
Labour market institutions: weaker trade unions and a decline of
collective bargaining (Goodman and Shephard 2002)
–
More inequality in employment status across households (Gregg and
Wadsworth, 2008)
–
Changes in the tax and benefit system
• How can we test them?
© Institute for Fiscal Studies
Example 1: decomposition of inequality by
household employment structure
• Take overall inequality as measured by the mean log deviation:
• If we divide the population into g subgroups (each containing ng
members) overall inequality can be decomposed into a “withingroups” and a “between-groups” term (Shorrocks, 1980):
© Institute for Fiscal Studies
Example 1: decomposition of inequality by
household employment structure
• Brewer, Muriel and Wren-Lewis (2009) use this decomposition to
examine the impact of changes in household employment
structure on inequality
– Groups defined according to number of adults, number of earners and
age of household head
• Conclude that the growing disparity between “work-rich” and
“work-poor” households contributed significantly to the increase
in inequality during the 1980s
© Institute for Fiscal Studies
0.05
0.40
0.04
0.03
0.30
0.02
0.25
0.01
Source: Adam and Browne (2010).
Note: Tax and benefit systems from previous years have been uprated in line with the Retail Prices Index. Years
up to and including 1992 are calendar years; thereafter, years refer to financial years.
2008
2006
2004
2002
2000
1998
1996
1994
1992
1990
1988
1986
1984
1982
0.20
1980
0.00
Gini coefficient
0.35
1978
Increase in Gini relative to 2009-10
Example 2: replacing tax/benefit system with those
from previous years (UK)
Example 2: replacing tax/benefit system with
those from previous years (UK)
• The tax and benefit system matters for the level of income
inequality
– if Labour had left the system they inherited unchanged, the Gini in
2009–10 would have been 0.39 rather than 0.36, higher than the US
(assuming no behavioural response)
• Other things matter more than the tax and benefit system for the
level of income inequality
– Inequality rose during the 2000s despite inequality-reducing changes
to the tax and benefit system
© Institute for Fiscal Studies
Poverty
© Institute for Fiscal Studies
What is poverty?
• Destitution, relative deprivation, capability or functioning in
society, livelihood sustainability?
– but what can we measure?
• Economists have tended to define poverty as having income below
a certain “poverty line”
•
One alternative is a “poverty gap” measure
– weights people according to how far they are below the poverty line
– but the data towards the bottom of the income distribution is not
good enough
© Institute for Fiscal Studies
Poverty lines
• 2 kinds of poverty lines are used
1. Absolute Poverty lines
– defined as a certain level of real-terms income
– egs. $1 a day poverty line (in 1990 prices) (Ravallion et al 1991),
US government basket of goods and services
© Institute for Fiscal Studies
Calculating absolute poverty
Count the proportion of
people below that
poverty line
Draw a line of real-terms
income
Lowest
Highest
Income
© Institute for Fiscal Studies
Absolute poverty over time
Count the proportion of
people below that
poverty line
Draw a line of real-terms
income
Lowest
Highest
Income
© Institute for Fiscal Studies
Absolute poverty over time
Count the proportion of
people below that
poverty line
Draw a line of real-terms
income
Lowest
Highest
Income
© Institute for Fiscal Studies
Poverty lines
• 2 kinds of poverty lines are used
1. Absolute Poverty lines
– defined as a certain level of real-terms income
– egs. $1 a day poverty line (in 1990 prices) (Ravallion et al 1991),
US government basket of goods and services
2. Relative Poverty lines
– defined as a certain percentage of median income in a country
– eg. UK government uses 60% of median income for child poverty
targets
© Institute for Fiscal Studies
Calculating relative poverty
Take (e.g.) 60% of that amount.
Everyone with income less than
this is in relative poverty.
Find the middle person’s income
(the median)
Lowest
Income
© Institute for Fiscal Studies
Highest
Relative poverty over time – a moving target
...then “60% of median income” –
the relative poverty line – grows
too...
If median income grows...
...even with no change to incomes
of low-income people, relative
poverty goes up
Lowest
© Institute for Fiscal Studies
Income
Highest
Why look at relative and absolute poverty?
• Relative poverty is really a measure of inequality between the
middle and the bottom
– particularly problematic when median income is falling
• Absolute poverty lines become irrelevant in the long run
– often moved on an ad hoc basis eg. 2010 baseline for 2020 child
poverty targets
• Changes in absolute poverty perhaps more significant in the
short run, with changes in relative poverty more significant
in the long run
© Institute for Fiscal Studies
Poverty and government policy: a case study
• Universal Credit is a major reform to the UK benefits system
aiming to:
– simplify the system
– improve work incentives
• How does it work?
– Universal Credit will replace 6 major working-age benefits and
tax credits with a single monthly payment
– So-called “legacy benefits” are Jobseeker’s allowance,
employment and support allowance, income support, housing
benefit, child tax credit and working tax credit
• Roughly revenue-neutral on an entitlements basis
© Institute for Fiscal Studies
Universal Credit: improving work incentives
• Universal Credit has larger earnings disregards...
– You can earn more before your benefit starts to be withdrawn
• ... and a lower maximum withdrawal rate
– Single rate of 65% on post-tax earned income ( maximum 76.2%
effective marginal tax rate)
© Institute for Fiscal Studies
Budget constraint for a lone parent with 2 children
£500
£450
Weekly net income
£400
£350
£300
£250
Existing system
£200
Universal Credit
£150
£100
£50
£0
0
10
20
30
40
50
Hours worked per week, at £6.50 per hour
Source: Browne and Roantree (2013)
© Institute for Fiscal Studies
60
Average participation tax rates by earnings
40%
35%
30%
25%
20%
Before
After
15%
10%
5%
0%
£0
£10,000
£20,000
Source: Browne and Roantree (2013)
© Institute for Fiscal Studies
£30,000
£40,000
£50,000
Universal Credit: improving work incentives
• Universal Credit has larger earnings disregards...
– You can earn more before your benefit starts to be withdrawn
• ... and a lower maximum withdrawal rate
– Single rate of 65% on post-tax earned income ( maximum 76.2%
effective marginal tax rate)
• Average participation tax rates are substantially reduced for
low earners
– this should increase labour supply and hence reduce poverty
© Institute for Fiscal Studies
Universal Credit: increasing take-up
• Take-up rates for benefits and tax credits are surprisingly low
– below 70% for Jobseeker’s allowance
– around 80% for Housing Benefit
• Universal Credit should increase take-up
– system will be easier to understand
– those currently only claiming one benefit but entitled to more
will automatically get their full entitlement
• All else equal, higher take-up rates will reduce poverty
© Institute for Fiscal Studies
Universal Credit: increasing take-up
• We can isolate the projected impact of introducing Universal
Credit on child poverty
– we assume no behavioural response ie. work incentives don’t
matter
– we assume everyone who currently claims any legacy benefit
claims their full Universal Credit entitlement
© Institute for Fiscal Studies
The effect of Universal Credit on relative child
poverty (UK)
26%
24%
22%
20%
18%
16%
14%
Universal Credit
no Universal Credit
12%
10%
2010
2011
2012
2013
2014
Notes: Poverty line is 60% of median before-housing-costs (BHC) income. Years refer to financial years.
Source: Browne, Hood and Joyce (2013)
© Institute for Fiscal Studies
2015
2016
Universal Credit: increasing take-up
• We can isolate the projected impact of introducing Universal
Credit on child poverty
– we assume no behavioural response ie. work incentives don’t
matter
– we assume everyone who currently claims any legacy benefit
claims their full Universal Credit entitlement
• Universal Credit is projected to reduce relative child poverty
by 2 percentage points in 2016-17
– this is basically just the result of increased take-up (as reform is
revenue-neutral and we don’t model behavioural response)
• Overall fiscal consolidation increases poverty substantially
© Institute for Fiscal Studies
The effect of tax and benefit reforms since April
2010 on relative child poverty (UK)
26%
24%
22%
20%
18%
16%
14%
12%
UK (actual)
10%
2010
2011
2012
2013
2014
UK (no reforms)
2015
2016
2017
2018
Notes: Poverty line is 60% of median before-housing-costs (BHC) income. Years refer to financial years.
Source: Browne, Hood and Joyce (2013)
© Institute for Fiscal Studies
2019
2020
Summary
• Income inequality rose quickly across the distribution in the
1980s, and has been increasing at the very top since
– decompositions and counterfactual analysis can help us to
understand why
• Poverty can be defined according to an absolute or relative
income measure
• The introduction of Universal Credit has the potential to
reduce poverty through improved work incentives and higher
take-up rates
– but the fiscal consolidation overall is likely substantially increase
poverty rates
© Institute for Fiscal Studies
References (1)

Acemoglu, D. (2002)“Technical Change, Inequality and the Labor Market”, Journal of Economic
Literature 40 (1)

Adam, S., and Browne ,J. (2010) “Redistribution, work incentives and thirty years of UK tax
and benefit reform”, IFS Working Paper 10/24

Brewer, M., Muriel, A. and Wren-Lewis, L. (2009) “Accounting for changes in inequality since
1968: decomposition analyses for Great Britain”, Government Equalities Office, London.

Brewer, M., and O’Dea, C. (2012) “Measuring Living Standards with income and consumption:
Evidence from the UK”, IFS Working Paper W12/12

Browne, J., Hood, A. and Joyce, R. (2013) “Child and working-age poverty in Northern Ireland
from 2010 to 2020”, IFS Report R78

Browne, J. and Roantree, B. (2013) “Universal Credit in Northern Ireland: what will its impact
be, and what are the challenges?”, IFS Report R77
© Institute for Fiscal Studies
References (2)

Cribb, J., Hood, A., Joyce, R., and Phillips, D. (2013) “Living Standards, Poverty and Ineqaulity
in the UK: 2013” IFS Report R81

Goldin, C., and Katz, L. (2008) “The Race Between Education and Technology”, Harvard
University Press, Cambridge MA

Goodman, A. and Shephard, A. (2002), Inequality and living standards in Great Britain: some
facts, IFS Briefing Note 19 , Institute for Fiscal Studies, London

Gregg , P. and Wadsworth ,J. (2008) “Two Sides to Every Story: Measuring Polarization and
Inequality in the Distribution of Work”, Journal of the Royal Statistical Society Series A

Machin, S. (2001) “The Changing Nature of Labour Demand in the New Economy and SkillBiased Technology Change”, Oxford Bulletin of Economics and Statistics 63 (S1)

Ravallion, M., Datt, G., and van de Walle, D. (1991) “Quantifying Absolute Poverty in the
Developing World,” Review of Income and Wealth no.37 pp 345-361

Shorrocks, A. (1980) “The Class of Additively Decomposable Inequality Measures”,
Econometrica 48 (3)
© Institute for Fiscal Studies