Long-run Trends in Wealth Inequality and the Role of Inheritance Daniel Waldenström Uppsala University Winter School in Inequality and Social Welfare,

1

Long-run Trends in Wealth Inequality and the Role of Inheritance

Daniel WaldenströmUppsala University

Winter School in Inequality and Social Welfare, Canazei, 2012

2

Outline of talk

1. Introduction: Role of wealth in inequality trends

2. Wealth concentration over the path of development

– Cross-country evidence, 1774-2006

– The case of Sweden, 1873-2006

3. Inheritance − the role of "old wealth"

4. Concluding remarks

3

(Part of the) Income/Wealth relationship

Wealth(Level,

distribution)

Income(Level,

distribution)

Inheritance

Capital income(rW)

Savings(sY)

Other processes(opportunity, insurance etc)

4

Capital income share in total income, Sweden, top decile, 1912-2004

0%

10%

20%

30%

40%

50%

60%

70%

1912 1922 1932 1942 1952 1962 1972 1982 1992 2002

Ca

pit

al s

ha

re o

f to

tal i

nc

om

e (e

xc

l. c

ap

ita

l ga

ins

)

P90-99 P99-99.9 P99.9-99.99 P99.99-100

5

Wealth matters to income mobility: Intergenerational transmission in Sweden

Björklund, Roine and Waldenström (2011), “Intergenerational Top Income Mobility in Sweden – Capitalist Dynasties in the Land of Equal Opportunity?”,

6

Wealth matters to income mobility: Intergenerational transmission in Sweden

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

-0.2

0.0

0.2

0.4

0.6

0.8

1.0

P90-95 P95-99 P99-99.9 P99.9-100

Income

FinAss

Educ

IQ

Non-cog

Björklund, Roine and Waldenström (2011)

7

Link between inequality and development

• Large literature − no consensus– Equalization view (Stiglitz 1969; Becker & Tomes; Loury 1981)– Disequalization view (Ray; Mookherjee & Ray 2003)– History matters (Banerjee & Newman 1993; Galor and Zeira 1993)

• Kuznets (1953, 1955): Structural change– Shift from agericultural (trad.) to industrial (modern) sector

generates inverted U-like relationship– Also: concentration of capital and savings boost inequality

Inequality

Time

8

Is there a Kuznets Curve in income inequality?

0

5

10

15

20

25

30

1903 1913 1923 1933 1943 1953 1963 1973 1983 1993 2003

Sh

are

of

tota

l in

co

me

(%

)

USA

UK

Canada

France

Australia

Netherlands

Sweden (incl. CG)

Sweden (excl. CG)

Kuznetsinverted-U?

What about wealth inequality?

Long-run trends in wealth concentration:

Cross-country evidence

Ohlsson, H, J. Roine and D. Waldenström (2008), “Long-Run Changes in the Concentration of Wealth: An Overview of Recent Findings”, in Davies, J.B. (ed.), Personal Wealth from a Global Perspective, Oxford, Oxford University Press. Link

9

10

Starting point• Main question: Is there a long-term linkage between

development and inequality? (cf. Kuznets)

• Further questions:– Common vs. specific trends across countries

– Potential heterogeneity within the top wealth decile

• This study:– Reviews recent empirical findings of others– Presents new evidence on Nordic countries

Measurement and data

• Estate data– Incentives to minimze tax - but also to divide fairly– Homogenous source over time– Problems: Sample size, W of deceased (mortality multipl.)

• Wealth tax data– Large samples; Relatively homogenous over time– Problems: tax incentives, asset valuation, excludes items

• Survey data– Covers most asset items; Scattered points; Short history– Problems: sample size, response rates, top-coding

11

Measurement and data

• Wealth concept: Net worth– Real + financial assets less debts (excl. pensions, human

cap.)

• Wealth owners:– Wealth tax data: Households (tax units; >18yrs)

– Estate data: Adults (18 yrs + ); deceased

– Survey data: Households (everyone in dwelling)

– Historically, households difficult to define

12

13

Measurement and data

Main methodology:

• Compute shares of total household wealth (using Pareto interpolation) that goes to the top 1%, top10% etc of all potential wealth holders

• The shares of both wealth and population are defined in relation to reference totals– Reference total for wealth: All personal wealth in the

economy (not only taxed wealth)

– Reference total for the population: All potential tax units (not just those who file tax returns)

French wealth concentration, 1807-1994

0

10

20

30

40

50

60

1800 1840 1880 1920 1960 2000

Share

of

tota

l w

ealt

h (%

)

P99-100

P95-99

P99.9-100

P90-95

14

U.K. wealth concentration, 1774-2001

0

10

20

30

40

50

60

70

80

90

100

1740 1760 1780 1800 1820 1840 1860 1880 1900 1920 1940 1960 1980 2000

Share

of

tota

l w

ealt

h (%

)

P95-100

P99-100

P95-99

England and Wales (1740-1937) U.K. (1938-)

Lindert (2000) Atkinson et al IRS (2006)

15

10

15

20

25

30

35

40

45

1770 1790 1810 1830 1850 1870 1890 1910 1930 1950 1970 1990

Share

of

tota

l w

ealt

h (%

)

P99-100 (adults)

P99-100 (households)

P90-95 (households)

P95-99 (households)

U.S. wealth concentration, 1774-2001Shammas (1993) Lindert

(2000) Kopzcuk & Saez (2004)

Wolff (1987, etc)

16

Swiss wealth concentration, 1913-1997

5

10

15

20

25

30

35

40

45

50

1910 1920 1930 1940 1950 1960 1970 1980 1990 2000

Share

of

tota

l w

ealt

h (%

)

P99-100

P95-99

P99.9-100

P90-95

17

Danish wealth concentration, 1789-1996

0

10

20

30

40

50

60

1780 1800 1820 1840 1860 1880 1900 1920 1940 1960 1980 2000

Share

of

tota

l w

ealt

h (%

)

P99-100

P90-95

P95-99

P99.9-100

18

Norwegian wealth concentration, 1789-2002

0

5

10

15

20

25

30

35

40

45

50

1780 1800 1820 1840 1860 1880 1900 1920 1940 1960 1980 2000

Share

of

tota

l w

ealt

h (%

)

P99-100

P95-99

P99.9-100

P90-95

19

Swedish wealth concentration, 1800-2002

0

10

20

30

40

50

60

70

1800 1840 1880 1920 1960 2000

Share

of

tota

l w

ealt

h (%

)

P99-100 (wealth tax based) P90-99 (wealth tax based) P99-100 (SCB, 1975-2000)P90-99 (SCB, 1975-2000) P99-100 (estate based) P90-99 (estate based)P99-100 (Spånt, 1920-75) P90-99 (Spånt, 1920-1975)

20

Cross country top percentile (P99-100)

10

20

30

40

50

60

70

1740 1760 1780 1800 1820 1840 1860 1880 1900 1920 1940 1960 1980 2000

Share

of

tota

l w

ealt

h (%

)

UK*

Denmark

Sweden

France

USA

Switzerland

Norway

21

Cross-country "next four" (P95-99)

15

20

25

30

35

40

45

1740 1760 1780 1800 1820 1840 1860 1880 1900 1920 1940 1960 1980 2000

Share

of

tota

l w

ealt

h (%

)

UK*

Denmark

Sweden

France

USA

Switzerland

Norway

22

Heterogeneity between countries and within the top

Period: ≈1780-1914

Fractile: Top1% Next 4%

France Increase Flat

Switzerland - -

UK Increase Decrease

US Increase Flat?

Denmark Decrease Flat

Norway Decrease Increase

Sweden Flat Flat

23

1914-2000

Top1% Next 4%

Decrease Flat

Flat Flat

Decrease Flat

Decrease Flat?

Decrease Flat

Decrease Decrease

Decrease Decrease*

24

Summing up cross-country evidence

1. Industrialization: no uniform impact on wealth inequality– Inequality increased in the US, UK (top1%), France– Flat (or decreasing) inequality in Nordic countries– Role of country size? Timing of industrialization?

2. Other factors matter (20th experience)– Geopolitical shocks, Crises, Redistribtion

3. Little support for Kuznets-type "inverse-U"– Trends rather look like an inverse-J (or inverse-S)

Wealth concentration in Sweden, 1870-2006

Roine, J. and D. Waldenström (2009), “Wealth Concentration over the Path of Development: Sweden, 1873–2006”, Scandinavian Journal of Economics, 111(1), 151–187. Link

25

Questions

• Closer at changes in wealth concentration

• Evolution during episodes:– Industrial take-off (1870-1910)

– Rise of the welfare state (educational reforms, home ownership, redistribution) (1930-1970)

– Globalization and deregulation (1980- )

• Role of offshore wealth

26

187018711872187318741875187618771878187918801881188218831884188518861887188818891890189118921893189418951896189718981899190019011902190319041905190619071908190919101911191219131914191519161917191819191920192119221923192419251926192719281929193019311932193319341935193619371938193919401941194219431944194519461947194819491950195119521953195419551956195719581959196019611962196319641965196619671968196919701971197219731974197519761977197819791980198119821983198419851986198719881989199019911992199319941995199619971998199920002001200220032004200540

50

60

70

80

90

100

Share

of

tota

l house

hold

wealt

h (

%)

Top wealth decile in Sweden

Top 10%

27

18701871187218731874187518761877187818791880188118821883188418851886188718881889189018911892189318941895189618971898189919001901190219031904190519061907190819091910191119121913191419151916191719181919192019211922192319241925192619271928192919301931193219331934193519361937193819391940194119421943194419451946194719481949195019511952195319541955195619571958195919601961196219631964196519661967196819691970197119721973197419751976197719781979198019811982198319841985198619871988198919901991199219931994199519961997199819992000200120022003200420050

10

20

30

40

50

60

Shar

e of

tot

al h

ouse

hol

d w

ealt

h (

%)

Bottom 90%

Top 10-1%

Top 1%

Wealth distribution in Sweden, 1870-2005

Industrial take-offEducation, welfare state,

home ownership

28

29

0

5

10

15

20

25

30

35

40

1911 1916 1921 1926 1931 1936 1941 1946 1951

Wealt

h s

hare

s (

%)

P99.9-100 P90-99 P99-99.9

Interwar era: High-wage earners accumulate wealth

18701871187218731874187518761877187818791880188118821883188418851886188718881889189018911892189318941895189618971898189919001901190219031904190519061907190819091910191119121913191419151916191719181919192019211922192319241925192619271928192919301931193219331934193519361937193819391940194119421943194419451946194719481949195019511952195319541955195619571958195919601961196219631964196519661967196819691970197119721973197419751976197719781979198019811982198319841985198619871988198919901991199219931994199519961997199819992000200120022003200420050

10

20

30

40

50

60

Shar

e of

tot

al h

ouse

hol

d w

ealt

h (

%)

?Bottom 90%

Top 10-1%

Top 1%

Wealth distribution in Sweden, 1870-2005

Industrial take-offEducation, welfare state,

home ownershipGlobalization,deregulation...

30

31

What has happened since 1980?

• Sweden in the 1980s and 1990s:– High taxes on wealth, inheritance, property– Booming financial markets– Liberalized capital account (since 1989)

• Potential impact: Capital flight– Private wealth (and wealth owners) leaves Sweden– Wealth is moved to closely held companies

• If so, what is the effect on the wealth distribution?

32

Wealth concentration in Sweden since 1950 (official series)

1950

1951

1952

1953

1954

1955

1956

1957

1958

1959

1960

1961

1962

1963

1964

1965

1966

1967

1968

1969

1970

1971

1972

1973

1974

1975

1976

1977

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

15

20

25

30

35

40

45

50

Wealt

h s

ahre

(%

)

Bottom 90%

Top 1%

33

Wealth concentration in Sweden since 1950(adjusted for foreign wealth of Swedish hh's)

1950

1951

1952

1953

1954

1955

1956

1957

1958

1959

1960

1961

1962

1963

1964

1965

1966

1967

1968

1969

1970

1971

1972

1973

1974

1975

1976

1977

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

15

20

25

30

35

40

45

50

Wealt

h s

ahre

(%

)

Bottom 90%

Top 1%incl. foreign wealth

34

How estimate the wealth of the rich?

• We add wealth to the top percentile (≈50,000 hh's)

• Three additions:1. Household wealth abroad

• Accumulated mismatches in Balance of Payment• Unaccounted savings in the Financial Accounts

2. Wealth of super-rich Swedes in Sweden• Unlisted wealth; Source: Rich Lists in magazines 1983-

3. Wealth of super-rich Swedes abroad• Unlisted wealth; Source: Rich Lists in magazines 1983-

35

Balance of Payments mismatch, six countries

-30%

-20%

-10%

0%

10%

20%

30%

40%

50%

60%

70%

1975 1980 1985 1990 1995 2000 2005

Acc

umul

ated

BoP

mis

mat

ch a

s sh

are

of G

DP

Switzerland

USA

NorwaySweden

DenmarkFinland

36

Comparing top wealth percentile: Sweden vs USA

15

20

25

30

35

40

45

1945 1955 1965 1975 1985 1995 2005

Sha

re o

f to

tal w

ealth

(%

)

+ BoP mismatch

Official series

+ Rich Swedes abroad

+ Rich Swedes at home

USA

+ BoP error, dom. and for. super-rich

37

Summing up: Swedish wealth inequality trends

1. Industrial take-off had a small effect on wealth concentration – But possibly larger on the composition of the wealthy ...

2. 20th century: Equalization– Before 1950 the top 1% lost to the medium-rich

– 1950-1980 rise of "popular wealth"

– After 1980, wealth compression halts

3. Adding estimates of foreign wealth and large domestic unlisted fortunes reverses trend 1980-

The Importance of Old Wealth: Estimating Long-Run Flows of

InheritanceOhlsson, H., J. Roine and D. Waldenström (2012), “The Role of Inheritance in Sweden, 1860–2010”, mimeo.

Waldenström (2012), “Household Wealth in Sweden, 1810–2010”, mimeo.

38

39

How important is inheritance?

• Next question: How important are inheritances relative to other sources of income?

• Households get rich from two sources: – they save out of their income (new wealth) – they receive transfers from others (old wealth)

• Conventional story: Old wealth mattered in the past, today new wealth (growth-based) is all that matters.

• But is this really true...?

40

Annual inheritance flow as share of national income: France, 1820-2008

Source: Piketty, Thomas (2011), “On the Long-Run Evolution of Inheritance: France 1850–2050”, Quarterly Journal of Economics 126(3), 1071–1131.

41

Annual inheritance flow as share of national income: Sweden, 1860-2010

18601870188018901900191019201930194019501960197019801990200020100

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

Economic inheritance flow Fiscal inheritance flow

42

Basic approach

• Piketty (2011): the importance of inheritance depends on ratio between economic growth rate g and returns to capital r.– With r > g, old wealth accumulates faster than new wealth

is created

• Let– B = Aggregate inheritance flow– Y = national income, – W = aggergate private wealth– m = mortality rate– μ = ratio aveW of deceased/aveW of the living.

43

• Then, compute the inheritance flow over income as:

B/Y = μ m W/Y∙ ∙

• In a dynastic model, heirs save a fraction g/r of the return on their inerited wealth, making the wealth-income ratio W/Y stationary.

• Steady-state bequest flow B/Y = (W/Y)/H – H = generation length

• If W/Y = 400% and H = 30, then B/Y = 13%.

Basic approach

44

B/Y = μ m W/Y (France) ∙ ∙Mortality rate (m) in France

Wealth-Income ratio (W/Y): France

AveW of decendents/living (μ): France

Inheritance flow/Income (B/Y): France

Source: Piketty (2011)

45

Constructing W, Y, m, μ for Sweden

• Y: National income data exist (disposable less so)

• m: Decendents over population. Data exist.

• μ: AveW of decendents/living. Difficult. Estimated from estate and wealth tax sources.– Gift-correction: +15%

• W: Real and financial assets less debts. No solid annual data prior to 1970 (1950). Construct new series back to early 19th century...

46

Constructing W: Real assets

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

1810 1850 1890 1930 1970 2010Agricultural net assets (farmland, forestry, livestock) Real estate (buildings, housing) Consumer durables

47

Constructing W: Financial assets

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

1810 1850 1890 1930 1970 2010

Bank deposits and currency Shares and mutual funds Bonds

Other claims Private pension savings

48

Constructing W: Debt

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

1810 1850 1890 1930 1970 2010

Informal liabilities (Other debt) State Financial sector

49

Debt components up to 1970.

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

1810 1830 1850 1870 1890 1910 1930 1950

Commercial banks Savings banks

Mortgage associations, Credit companies Insurance companies

State Informal liabilities (Other debt)

50

Asset composition

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

1810 1850 1890 1930 1970 2010

Real assets Financial assets

51

Swedish household wealth over GDP, 1810-2010

0

0.5

1

1.5

2

2.5

3

1810 1850 1890 1930 1970 2010

Real assets / GDP Financial assets / GDP Debts / GDP

52

Wealth-Income ratio (W/Y): Sweden

1810182318361849186218751888190119141927194019531966197919922005200%

300%

400%

500%

600%

700%

53

B/Y = μ m W/Y (Sweden) ∙ ∙

Wealth-Income ratio (W/Y): Sweden

200%

300%

400%

500%

600%

700%

1810 1830 1850 1870 1890 1910 1930 1950 1970 1990 2010

AveW of decendents/living (μ): Sweden

0%

50%

100%

150%

200%

250%

300%

1860 1880 1900 1920 1940 1960 1980 2000

Mortality rate (m) in Sweden

0.0%

0.5%

1.0%

1.5%

2.0%

2.5%

1860 1880 1900 1920 1940 1960 1980 2000

Inheritance flow/Income (B/Y): Sweden

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

1860 1880 1900 1920 1940 1960 1980 2000

Economic inheritance flow Fiscal inheritance flow

54

So, does inheritance have a role?

1. Evidence from France and Sweden suggests that:– Inheritance was important up to WWI– After 1950, inheritance did not matter (small W/Y)– Today, inheritance flows are again become more significant

(W/Y is growing)

2. Piketty's main lesson: – "with r > g (say, r=4-5% vs g=1-2%), then wealth coming

from the past is being capitalized faster than growth, and inherited wealth dominates self-made wealth"

55

4. Concluding remarks

1. Before industrialization, wealth was highly unequal

2. Wealth equalized during 20th century– But not due to Kuznets-curve

– Instead: Political shifts (education, redistribution) and Exogenous shocks (wars, crises)

3. Current globalization and modest growth rates in Western world suggest increased future role of wealth and inheritance