Income Shocks, Assets and Labour Supply: Understanding … · 2014. 9. 15. ·...

© Institute for Fiscal Studies

Income Shocks, Assets and Labour Supply: Understanding Inequality Dynamics

Richard Blundell University College London and Institute for Fiscal Studies

Thanks to Jonathan Cribb, Ben Etheridge, Andy Hood, Wenchao Jin, Robert

Joyce, Cormac O’Dea, Luigi Pistaferri, Ian Preston, and Itay Suporta

This research is funded by the ERSC Centre (CPP) at IFS.

MISOC Workshop September 15th 2014

Motivation •  Many of the key determinants in the trends in income inequality and in

overall living standards over the past 25 years, including since the financial crisis, have been driven by changes in the labour market, including, for the UK: –  huge increase in entry cohorts with at least a BA degree during the 1990s

and early 2000s, –  large relaHve rise in top ‘earnings’ percenHle since the early 1990s, –  dramaHc fall in real wages since 2008, ….

•  To which we can add changes in savings and asset prices, in parHcular housing; and reforms to taxes and welfare benefits.

•  Aim: To layout a framework for examining the role of interacHons between labour market shocks, asset markets shocks and changes to the tax and welfare system, in the evoluHon of inequality, –  allowing an interpretaHon of what has happened in the UK, and also in the US and elsewhere.

Inequality over the Working Life •  Begin by focussing on the labour market as a primary source of shocks

and examine the links between the various definiHons of economic inequality:

•  wages => earnings => joint earnings => income => consumpHon –  the link between the various measures of inequality is mediated by

mulHple `insurance' mechanisms including:

1.  (Family) labour supply (wages → earnings → joint earnings) 2.  Taxes and welfare benefits: (earnings → income) 3.  Assets: Saving and borrowing (income → consumpHon) 4.  Informal contracts, gi^s, etc. •  But how important are each of these mechanisms? •  How do they change over the life-‐cycle and the business cycle? •  How should we design policies to best insure these shocks? First look at income and consumpHon inequality by year and by cohort….

Income DistribuHon by Year -‐ UK

Source: Blundell and O’Dea (2014).

0

200

400

600

800

1000

1200

1978

1979

1980

1981

1982

1983

1984

1985

1986

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

Year

mean

p90

p75

p50

p25

p10

Note: something is happening around 2000, then 2008!

ConsumpHon DistribuHon by Year -‐ UK

0

100

200

300

400

500

600

700

800

900 1978

1979

1980

1981

1982

1983

1984

1985

1986

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

Year

mean

p90

p75

p50

p25

p10


Note: ConsumpHon started falling back earlier than income in the recent recession….

Income Inequality by Year and Birth Cohort -‐ UK

Source: Blundell and O’Dea, 2014

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

1979

1980

1981

1982

1983

1984

1985

1986

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

Year

Variance, Log Income

1930s

1940s

1950s

1960s

1970s

1980s

Notice something is happening just after 2000; … much more noticeable in the US.

ConsumpHon Inequality by Year and Birth Cohort -‐ UK


0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

0.45

1979

1980

1981

1982

1983

1984

1985

1986

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

Year

Variance, Log Consump4on

1930s

1940s

1950s

1960s

1970s

1980s

ConsumpHon Inequality by Age and Birth Cohort -‐ UK

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

0.45

21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 61 63 65 67 69 71 73 75 Age

1930s

1940s

1950s

1960s

1970s

1980s

Notes: younger cohorts entering with much larger inequality than previous cohorts….


Income Inequality by Age and Birth Cohort -‐ UK

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 61 63 65 67 69 71 73 75 Age

1930s

1940s

1950s

1960s

1970s

1980s


Source: Attanasio and Pistaferri (2014)

.1.1

5.2

.25

.3.3

5.4

var(l

ntot

cons

)

20 25 30 35 40 45 50 55 60 65 70age

Born 1930s Born 1940sBorn 1950s Born 1960s

ConsumpHon Inequality by Year and Birth Cohort -‐ US

Inequality dynamics •  Similar episodes of growth in income inequality can have very

different dynamic underpinnings, with different policy implicaHons. •  Studying the the relaHonship between inequality in income,

consumpHon, wages and earnings can bring some structure to the key features of inequality.

•  A ‘shock’ to the earnings of one family member may: –  generate reacHons in the labour supply of other family members, –  be compensated, in part, through the welfare benefit system, –  perhaps also through drawing on savings and net equity, –  and even through informal networks.

•  This is the moHvaHng idea for the Blundell, Pistaferri and Suporta (2013) study, which uses data from the new PSID, 1996 – 2010, –  new panel data measures of consumpHon, earnings and assets; –  and previous studies using mixtures of panel and c-‐s data.

Consumption inequality dynamics •  BP (QJE 1998) and BPP(AER 2008) approach relates transitory and persistent

log income shocks to log consumpHon. In a simplified form:

Δ lnYit ≈ ν it +Δξitpermanent income shocks

transitory income shocks

Impact of shocks to asset prices, etc.

Ratio of net assets to net assets plus human capital wealth – insurance value of net assets

Δ lnCit ≈ Γit +φ(1−π it−1)ν it +ψγ tL (1−π it−1)ξit +εit

•  Relates the evoluHon of the income distribuHon to consumpHon distribuHon –  Persistent shocks will, in general, have a larger impact on consumpHon; –  ϕ and ψ allow for excess smoothness and excess sensiHvity; and for net savings to be

used to ‘insure’ shocks.

•  Income inequality episodes differ in underlying persistence/durability –  Using variance of log for UK and US find strong growth in variance of persistent shocks

in early/mid 1980s for all cohorts and early/mid 1990s for certain key working age cohorts.

Born 1940s

0

0.1

0.2

0.3

0.4

0.5

1979

1980

1981

1982

1983

1984

1985

1986

1987

1988

1989

1990

1991

1992

Var(c) Var(y) Cov(c,y)

Born 1950s

0

0.1

0.2

0.3

0.4

0.5

1979

1980

1981

1982

1983

1984

1985

1986

1987

1988

1989

1990

1991

1992

Var(c) Var(y) Cov(c,y)

Source: Blundell and Preston (1998)

For example, birth cohort inequality:

Implied variance of permanent shocks by cohort – Britain The ‘spike’ in persistent shocks - very similar in US, BPP (AER, 2008)

Source: Blundell and Preston (1998)

Consumption inequality dynamics with family earnings •  Blundell, Pistaferri and Suporta (2013) generalise this framework to allow

for mulHple earnings in a family –  Wage/earnings shocks can be correlated across family members –  Labour supply can respond across family members –  Taxes and welfare benefits deliver net current period income –  Labour supplies and consumpHon can be non-‐separable

•  Some findings: –  a key role for family labour supply in providing insurance for adverse shocks,

especially for those with low/no net assets,

–  substanHal role for taxes and welfare benefits, and –  once family labour supply, taxes, welfare benefits and savings are taken account,

there is lirle role for any other insurance mechanisms.

•  Recent recession – assets and real wages suffer an adverse shock: –  consumpHon falls for non-‐durables/necessiHes, largely across the distribuHon,

implying strong persistent/uninsurable shocks;

–  labour supply of women and older workers is maintained, even enhanced; –  ‘double whammy’ implies strong role for welfare benefits and tax credits.

For the rest of the talk….

Focus mostly on the UK:

1.  Earnings: Wage Changes, Employment (and ProducHvity) 2.  Incomes: Working-‐age (and Pensioners) 3.  ConsumpHon: Durable and Non-‐durable Expenditures –  think about what they tell us about the process of inequality and living

standards since the recession in the context of the ‘insurance’ model,

–  briefly examine the prospects for the future. -‐ the key idea is to use the link between earnings, incomes and consumpHon to build a more detailed picture of the process and prospects for inequality.

© InsHtute for Fiscal Studies

1. Earnings: Wages and Employment

•  Average real hourly wages fallen since this recession began. –  even though workforce composiHon has shi^ed towards more

producHve types,

•  Real wage falls have occurred within individuals: –  unprecedentedly high proporHons of employees experienced nominal

wage freezes,

•  The educaHon premium survived the large increase in those with BAs –  but real wages have fallen for all groups since the recession.

•  ‘EffecHve’ and actual labour supply is higher now than during previous recessions

–  due partly welfare policy changes and partly to wealth/real wage shocks.


Real Median Wage by EducaHon, Men

Source: Blundell, Green and Jin (2014)

BA or above proporHon by cohort, all


ProporHon with degrees or above by age: all


ProporHon with degrees or above by age: women


RaHo of BA (S3) median wage to that of S2

Source: Blundell, Green and Jin (2014) Remarkably… no cohort effects! BA premium stayed constant, even through the recession.

Average real hourly wage by qualificaHon among young people

0.00

2.00

4.00

6.00

8.00

10.00

12.00

14.00

18-‐21, GCSEs or above

18-‐21, no GCSEs or equilvants

22-‐24, degree or above

22-‐24, GCSEs, no degrees

22-‐24, no GCSEs

Source: Blundell, Crawford and Jin (2014)

% change to real hourly wage since last year, by period


Employment and labour market parHcipaHon

•  Some increase of parHcipaHon can be arributed policy changes, e.g. : –  Labour supply has increased among lone parents, partly as a

result of job search condiHons arached to benefit claims

–  Older workers are reHring later, partly as a result of increased SPA for women

Change to lone mothers’ parHcipaHon rate since policy change

-‐8% -‐6% -‐4% -‐2% 0% 2% 4% 6% 8% 10% 12% 14%

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Quarter since the last quarter before policy change

youngest kid 12-‐15, 2008Q3 youngest kid10-‐11, 2009Q3

youngest kid 7-‐9, 2010Q3 youngest kid 5-‐6, 2011Q3

Impact of SPA increase for women on employment

30%

35%

40%

45%

50%

55%

60%

Counterfactual male 60-‐64 employment rate

LFS employment rate 60-‐64 men

Counterfactual female 60-‐64 employment rate

LFS employment rate 60-‐64 year old women

Note: counterfactual employment rates are estimated. See Cribb et al (2013) “Incentives, shocks or signals: labour supply effects of increasing the female state pension age in the UK”

Employment rate for older workers: women aged 60-‐64

0.0

5.0

10.0

15.0

20.0

25.0

30.0

35.0

40.0

45.0

2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

Euro area (13 countries)

Germany (unHl 1990 former territory of the FRG)

Spain

France

Italy

United Kingdom

Employment rate for older workers: men aged 65-‐69

0.0

5.0

10.0

15.0

20.0

25.0

30.0

2003 2004 2005 2006 2007 2008 2009 2010 2011 2012



Spain

France

Italy

United Kingdom

Employment rate for older workers: men aged 60-‐64

0.0

10.0

20.0

30.0

40.0

50.0

60.0

70.0

2003 2004 2005 2006 2007 2008 2009 2010 2011 2012



Spain

France

Italy

United Kingdom

For the young employment fell back.... Employment rate: men aged 25-29

60.0

65.0

70.0

75.0

80.0

85.0

90.0

2003 2004 2005 2006 2007 2008 2009 2010 2011 2012



Spain

France

Italy

United Kingdom

Male employment by age -‐ UK


Employment with at least BA: men


Employment, le^ school at 16: men


Employment, le^ school at 16: women


Hours, le^ school at 16: men


2. Incomes •  Leading up the recession:

•  income growth had slowed in early 2000s. •  pensioners/ working-‐age childless were doing relaHvely well/badly

•  During recession and immediately a^erwards: •  real earnings for those in work fell •  employment rates fell for low skilled young adults but not for older ones •  benefits/tax credit incomes were robust.

–  As a result (…..drawing from IFS Living Standards Report): a)  income inequality fell (despite rise in earnings inequality among workers) b)  low educated young adults did worst; pensioners did best


-‐1.0%

-‐0.5%

0.0%

0.5%

1.0%

1.5%

2.0%

2.5%

3.0%

3.5%

4.0%

10 20 30 40 50 60 70 80 90 Average an

nual re

al income chan

ge Parents and children

Working-age without children

Pensioners

Percentile point

1996-‐97 to 2011-‐12: pensioners did relaHvely well; working-‐age childless relaHvely badly

Notes and source: see Figure 5.2 of Living Standards, Poverty and Inequality in the UK: 2013

Income trends not uniform: slower growth from early 2000s

100

105

110

115

120

125

1997–98=100

Mean income Median Income


Source: Family Resources Survey, various years

100

105

110

115

120

125

1997–98=100




Income trends not uniform: slower growth from early 2000s

... and large falls a^er the recession

100

105

110

115

120

125

1997–98=100





Source: Table 2.3 of Living Standards, Poverty and Inequality: 2014 Notes: This is a very slightly different sample to the overall income staHsHcs. Households with negaHve incomes are dropped. This makes a small difference to falls in income

Income sources, 2007–08 to 2009–10: steady income growth due to benefits/tax credits

-‐2 -‐1.5 -‐1 -‐0.5 0 0.5 1 1.5 2 2.5 3

Total income

Taxes and other deducHons

Other

Self-‐employment income

Savings and private pension income

Benefits and tax credits

Earnings

ContribuHon to income growth between 2007–08 to 2009–10 (in percentage points)

0.1

0.4

0.2

0.2

2.4

2.2

-‐0.7



Income sources, 2009–10 to 2012–13: large income falls due to falling earnings

-‐10 -‐8 -‐6 -‐4 -‐2 0 2 4

Total income


Other




Earnings


-‐8.1

-‐1.7

-‐0.2

3.2

-‐8.6

-‐1.0

-‐0.8



Income sources: 2007–08 to 2012–13

-‐10 -‐8 -‐6 -‐4 -‐2 0 2 4 6

Total income


Other




Earnings


-‐8.2

-‐1.4

0.1

3.5

-‐6.4

1.2

-‐1.5

-‐14.0% -‐13.0% -‐12.0% -‐11.0% -‐10.0% -‐9.0% -‐8.0% -‐7.0% -‐6.0% -‐5.0% -‐4.0% -‐3.0% -‐2.0% -‐1.0% 0.0%

10 20 30 40 50 60 70 80 90

Cumula4

ve re

al cha

nge Percentile point

Weekly earnings inequality (among workers) rose between 2007-‐08 to 2012-‐13...

Note: Excludes self-employment income Source: Family Resources Survey, various years

Note: towards bottom, much driven by falls in hours worked (not relatively larger fall in hourly wages)

...but net result was sHll a fall in income inequality


-12%

-8%

-4%

0%

4%

8%

10 20 30 40 50 60 70 80 90

Inco

me

chan

ge

2007–08 to 2009–10 2009–10 to 2012–13 2007–08 to 2012–13

Percentile point


3. Consumption

•  Expenditure falls have been deeper than in previous recessions. –  note that the start of the fall is coincident with the fall in GDP (not

income).

•  Unusually expenditure on consumer nondurables has fallen most –  especially among the young families with children and to some extent

among the middle aged. Less for the old.

•  This points to an expectaHon of a permanent/persistent fall in living standards, –  especially among the young and middle-‐aged.

•  The very recent growth is dominated by durable expenditure increases.


ConsumpHon Growth


80

85

90

95

100

105

2005Q1

2005Q3

2006Q1

2006Q3

2007Q1

2007Q3

2008Q1

2008Q3

2009Q1

2009Q3

2010Q1

2010Q3

2011Q1

2011Q3

2012Q1

2012Q3

2013Q1

Private consum

p4on

(2008Q

1 = 100) EU28

Spain

Italy

United Kingdom

Non-‐ (and semi) durables; UK recessions


90

95

100

105

110

115

120

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

Qua

rter before recession = 100

Quarters since start of recession

1980

1990

2008

ConsumpHon DistribuHon by Year -‐ UK

Notes: Non-durables and Services; FES Source: Blundell and O’Dea (2014).

0

100

200

300

400

500

600

700

800

900 1978

1979

1980

1981

1982

1983

1984

1985

1986

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

Year

mean

p90

p75

p50

p25

p10

Consumption started falling back earlier than income…….


Percentage Change in Food Expenditure: UK

Notes: Understanding Society Source: Blundell and Etheridge, 2014

-‐0.1

-‐0.08

-‐0.06

-‐0.04

-‐0.02

0

0.02

0.04

0.06

0.08

1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79 82 85 88 91 94 97 2012-‐2009

2012-‐2010

2010-‐2009


Percentage Change in Food Expenditure: 2010-‐2012 UK

-‐0.12

-‐0.1

-‐0.08

-‐0.06

-‐0.04

-‐0.02

0

0.02

0.04

1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79 82 85 88 91 94 97

Change in Food Expenditure by Cen4le and HH type: 2010 to 2012

All HHs

Pensioners

w/o children

with children

Notes: Understanding Society Source: Blundell and Etheridge, 2014

Prospects •  Younger workers and families are acHng as if they expect a long-‐run fall

in relaHve living standards, and quite far up the distribuHon.

–  evidence from consumpHon, savings and labour supply. •  The number of rouHne jobs near the middle of the earnings distribuHon

has declined steadily

–  more jobs are now professional or managerial –  ‘successful’ absorpHon of the higher numbers with degrees.

Employment shares of occupation groups


Source: Blundell, Green and Jin (2014) Notes: the discontinuities in occupation classification in 2001 and 2011 have been addressed in the following way. For the conversion of SOC 1990 to SOC 2000, we look at individuals who were surveyed in 2000Q4 and 2001Q2 and stayed with the same employer and infer the transition matrix from this group. For the conversion of SOC2010 to SOC 2000, we used the ONS two-way tabulation of the LFS 2007 Q1 sample by the major occupation groups under the two SOC systems.

Prospects •  Younger workers and families are acHng as if they expect a long-‐run fall

in relaHve living standards, quite far up the distribuHon. –  evidence from consumpHon, savings and labour supply.

•  The number of rouHne jobs near the middle of the earnings distribuHon has declined steadily –  more jobs are now professional or managerial, ‘successful’ absorpHon of the higher

numbers with degrees.

•  Suggests longer term earnings growth will mostly come from high-‐skilled occupaHons, enhancing the top of the income distribuHon.

•  Lirle evidence of earnings progression for lower skilled. •  Benefit changes have play a central role for low skilled

Source: Blundell, Dias, Meghir and Shaw (2013)

1.61.8

22.2

2.42.6

log w

age

20 30 40 50age

secondary further higher

Wage profiles by education and age – UK Women

Benefit changes are also important drivers SimulaHons up to 2015–16


-8%

-6%

-4%

-2%

0%

2%

4%

10 20 30 40 50 60 70 80 90

Inco

me

chan

ge

2007–08 to 2012–13

2012–13 to 2015–16

Percentile point

Note: Calculations using the Family Resources Survey 2007–08 and 2012–13 and TAXBEN, the IFS tax and benefit microsimulation model. Figure is an update of that in Brewer et. al. (2013), Fiscal Studies, Vol. 34, No. 2, pp. 179–201. See also Andy Hood’s presentation!

Benefit changes are also important drivers SimulaHons up to 2015–16


-8%

-6%

-4%

-2%

0%

2%

4%

10 20 30 40 50 60 70 80 90

Inco

me

chan

ge

2007–08 to 2012–13

2012–13 to 2015–16

2007–08 to 2015–16

Percentile point

Note: Calculations using the Family Resources Survey 2007–08 and 2012–13 and TAXBEN, the IFS tax and benefit microsimulation model. Figure is an update of that in Brewer et. al. (2013), Fiscal Studies, Vol. 34, No. 2, pp. 179–201. See also Andy Hood’s presentation!

Implications for Inequality •  Persistent changes

–  evidence from distribuHon of consumpHon, and labour supply responses, implies strong persistent/uninsurable shocks for certain key birth cohorts.

•  Longer term earnings growth will mostly come from high-‐skilled occupaHons, –  enhancing the top deciles of the income distribuHon.

•  Lirle progression and low producHvity for the low skilled part of the distribuHon –  incomes at the borom will only be maintained through family labour supply and the

benefit/tax-‐credit system (and younger adults staying at home).

•  Studying the the relaHonship between inequality in income, consumpHon, wages and earnings brings structure to the key features of inequality and living standards –  consumpHon falls much more than in past recessions, even for necessiHes, –  labour supply of women and older workers is maintained, even enhanced, –  for low skilled with few assets, family labour supply, family structure and welfare are

key insurance mechanisms.

•  Providing a picture of the dynamics of inequality and living standards over the austerity period and a sense of the likely evoluHon in coming years.

•  SHll need to do more on this with Understanding Society…..


That’s all!

Income Shocks, Consumption and Labour Supply: Understanding Inequality Dynamics


Thanks to Ben Etheridge, Andy Hood, Wenchao Jin, Robert Joyce, Cormac

O’Dea, Luigi Pistaferri, Ian Preston, and Itay Suporta




Extra Slides

Income Shocks, Consumption and Labour Supply: Understanding Inequality Dynamics


Thanks to Ben Etheridge, Andy Hood, Wenchao Jin, Robert Joyce, Cormac

O’Dea, Luigi Pistaferri, Ian Preston, and Itay Suporta



Employment and self-‐employment rate of older people

-‐10%

-‐8%

-‐6%

-‐4%

-‐2%

0%

2%

4%

2001Q1

2001Q3

2002Q1

2002Q3

2003Q1

2003Q3

2004Q1

2004Q3

2005Q1

2005Q3

2006Q1

2006Q3

2007Q1

2007Q3

2008Q1

2008Q3

2009Q1

2009Q3

2010Q1

2010Q3

2011Q1

2011Q3

2012Q1

2012Q3

2007Q4 = 0

employment rate 60-‐74 male self-‐employment rate 60-‐74 male

employment rate 60-‐74 female self-‐employment rate 60-‐74 female

(RPI deflated) Real wages by occupation group since 1993


Note: the low-skilled wage would end up around the 1993 level if we use CPI instead of RPI. Each log wage series is normalized to 0 in 1993.

NEET rate among young people

0%

10%

20%

30%

40%

50%

60%

18-‐21, GCSEs or above

18-‐21, no GCSEs or equilvants

22-‐24, degree or above

22-‐24, GCSEs, no degrees

22-‐24, no GCSEs

Median male real hourly wages by educaHon


Note: Average real hourly wages have fallen back – this picture stands even a^er adjusHng for selecHon.

Adjusted Median Wage by Year, Men


In contrast to previous recessions, real output per hour has at best been stagnant since 2008

90

95

100

105

110

115

120

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Quarter since the labelled one

Real output per hour

1979Q4 1990Q2 2008Q1 Linear trend (90Q2-‐08Q1)


What have we learned? •  Changes in the labour market are a key factor behind the changes in overall

living standards and inequality –  especially for younger families/singles

•  Real wages and employment tell much of the story –  real wages have fallen across the distribuHon, –  large increases in entry cohorts with BA have maintained their BA premium, even in the

recession but they have also suffered the overall decline in real wages,

–  the poor employment and earnings performance of young low educated.

•  Benefit changes are also important drivers –  key impact in the near future too …


Prospects - summary •  Younger workers and families are acHng as if they expect a long-‐run fall

in relaHve living standards, quite far up the distribuHon. –  evidence from consumpHon, savings and labour supply.

•  Real wages (and producHvity) have been slow to pick up –  expect the parern of lower real wages to conHnue, but with fairly buoyant

employment due to increased supply.

•  Most falls in real earnings have happened (but low/no real growth) –  fiscal contracHon implies large benefit cuts in near future.

•  The number of rouHne jobs near the middle of the earnings distribuHon has declined steadily –  more jobs are now professional or managerial, ‘successful’ absorpHon of the higher

numbers with degrees.

•  Suggests longer term earnings growth will mostly come from high-‐skilled occupaHons, enhancing the top of the income distribuHon.

•  Lirle evidence of earnings progression for lower skilled.

Prospects - II •  But sHll much to do in focussing on older workers in general, on return

to work for parents/mothers, and on entry into work.

•  There are sHll some potenHal big gains here, –  for example, as (higher skilled) women age in the workforce.

•  Tax/welfare reforms to enhance earnings (from Mirrlees): –  refocus incenHves towards transiHon to work, return to work for lower

skilled mothers and on enhancing incenHves among older workers.

•  ProducHvity is sHll the key, –  (financial) capital misallocaHon and potenHal investment returns.

•  Human capital and ‘on the job’ wage/producHvity complementarity –  note the relaHve importance of mismatch of entry skills in the recession.

•  ProducHvity and wages are closely related but note recent growth in the wedge between labour costs and hourly wages, –  the growing importance of pensions and NI in the UK, –  what will the trends in this wedge look like?

Employer contribuHons to pension funds – in constant prices terms

Source: Office for National Statistics Notes: Data for Q4 2012 is not yet published so has been estimated based on Q4 2011 to Q3 2012 data

The growth in durable and non-‐durable real expenditure


80

85

90

95

100

105

110

115

2008 Q1

2008 Q2

2008 Q3

2008 Q4

2009 Q1

2009 Q2

2009 Q3

2009 Q4

2010 Q1

2010 Q2

2010 Q3

2010 Q4

2011 Q1

2011 Q2

2011 Q3

2011 Q4

2012 Q1

2012 Q2

2012 Q3

2012 Q4

2013 Q1

2013 Q2

Non-‐durable Durables

Employment rate by highest qualificaHons achieved, 16-‐59 year olds


Note: level 2 is GCSE grade C or above.

50%

55%

60%

65%

70%

75%

80%

85%

90%

95%

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

degree or above

level 2 and 3

level 1 or below

ParHcipaHon rates over Hme

0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0% 70.0% 80.0% 90.0%

100.0%

23-64, male 23-64, female 65-84, male 65-84, female

•  In this notation, the transmission parameters φ and ψ subsume π and γ from the self-insurance model

•  This provides the key linkage between the evolution of distribution of consumption and the evolution of income

•  It describes how the consumption distribution updates to income shocks

ln 'it it it t it t it itC Z ϑ φζ ψ ε ξΔ ≈ Γ +Δ + + +

Partial insurance coefficient w.r.t. permanent shocks, 0≤φ ≤1

Excess sensitivity coefficient w.r.t. transitory shocks, 0≤ψ≤1

Need to generalise to account for additional ‘insurance’ mechanisms and excess sensitivity

Consumption dynamics (2)

% facing nominal wage freeze in the coming year by current wage quintile

Source: New Earnings Survey Panel Dataset 1975-‐2012. There are 20,000-‐30,000 observaHons underlying each data point.

0%

2%

4%

6%

8%

10%

12%

14%

16%

18%

Lowest-‐paid 20% 2nd quinHle 3rd quinHle

4th quinHle Highest-‐paid 20%

More nominal pay freezes in the absence of collecHve agreement, 2008-‐11

Change to the proporHon of 23-‐64-‐year-‐olds who are unemployed by recession

0.0%

1.0%

2.0%

3.0%

4.0%

5.0%

6.0%

0 1 2 3 4 5

male, from 1979

female, from 1979

male, from 1990

female, from 1990

male from 2008

female, from 2008

Source: LFS every year. No data point for year 1980, 1982. Quarter 2 is used for years since 1992.

Median male real hourly wages by educaHon (exl. Im.)


Can changes in the composiHon of the workforce explain the fall in real hourly wages?

•  Short answer: no –  unlike, to an extent, in other countries.

•  As in previous recessions, the composiHon of the workforce shi^ed towards more producHve workers, –  hence we would have expected the average real wage to increase

(other things being equal).

•  What has been different in this recession is that the returns to those characterisHcs have fallen substanHally.

•  This is true even amongst workers who keep their jobs, who have experienced nominal wage freezes/real wage falls, –  May have been facilitated by the reduced power of labour market

insHtuHons (e.g. unions) since previous recessions.

ProporHonate Fall in Each Component


65

70

75

80

85

90

95

100

105

110

Chan

ge since 2008 Q1 (2008 Q1 = 100)

Govt. purchases

Consumer durables

Nondurable consumpHon

Corporate investment

House Prices by Region


Components of GDP


-‐20

-‐15

-‐10

-‐5

0

5

10

15

Chan

ge since 2008 Q1 (£b) Net exports

Govt. purchases

Consumer durables

Nondurable consumpHon

Corporate investment

-‐1.0%

-‐0.5%

0.0%

0.5%

1.0%

1.5%

2.0%

2.5%

3.0%

3.5%

4.0%

10 20 30 40 50 60 70 80 90 Average an

nual re

al income chan

ge Parents and children

Working-age without children

Pensioners

Percentile point

...and on an a^er-‐housing-‐cost (AHC) basis

Notes and source: see Figure 5.2 of Living Standards, Poverty and Inequality in the UK: 2013

Median income changes by age (BHC, GB)


-4%

-3%

-2%

-1%

0%

1%

2%

3%

4%

0s 10s 20s 30s 40s 50s 60s 70s

Aver

age

annu

al g

row

th

2007–08 to 2012–13 2002–03 to 2007–08

Age

Average real hourly wages have also been stagnant

95

100

105

110

115

120

0 1 2 3 4 5 6

Real hou

rly wage (first y

ear o

f recession = 100)

Years since the year in which the recession began

Average real male hourly wage (using GDP deflator)

1979 1990 2008 Linear trend 1990-‐2008

Average real hourly wage by age group (RPI deflated)

0

2

4

6

8

10

12

14

16

18

1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

16-‐17

18-‐25

26-‐35

36-‐45

46-‐55

56-‐64

65+

Employment and labour market parHcipaHon -‐ employment rate of 23-‐64-‐year-‐olds by recession

-10.0%

-8.0%

-6.0%

-4.0%

-2.0%

0.0%

2.0%

0 1 2 3 4 5

Year since the start of recession

male, from 1979 female, from 1979 male, from 1990 female, from 1990 male from 2008 female, from 2008

Source: LFS every year. No data point for year 1980, 1982. Quarter 2 is used for years since 1992.

Employment status of 16-‐22-‐year-‐olds

0%

10%

20%

30%

40%

50%

60%

70%

80% % in work % unemployed

Source: LFS 2nd

quarter every year

Male employment rate by age group over Hme


3. Consumption

•  Expenditure falls have been deeper than in previous recessions. –  Note that the start of the fall is coincident with the fall in GDP (not income).

•  Unusually expenditure on consumer nondurables has fallen most –  Especially among the young families with children and to some extent among the

middle aged. Less for the old.

•  Homeowners have made the largest cuts. –  In past recessions there was lirle difference between owners and renters.

•  Saving raHos are lower than during the early 1980s and early 1990s –  but haven risen dramaHcally since 2008 (and pension contribuHons higher).

•  The data points to an expectaHon of a permanent/persistent fall in living standards, –  especially among the young and middle-‐aged.

•  The very recent growth is dominated by durable expenditure increases.


Income Shocks, Assets and Labour Supply: Understanding … · 2014. 9. 15. ·...

Documents

Transcript of Income Shocks, Assets and Labour Supply: Understanding … · 2014. 9. 15. ·...