Wealth Inequality in the United States since 1913 Emmanuel Saez (UC Berkeley) Gabriel Zucman (LSE) JRCPPF 4 th Annual Conference 19 February 2015
Is rising inequality in the United States only a labor income phenomenon? US Income inequality has increased sharply since the 1970s Mixed existing evidence on wealth inequality changes Is inequality increase solely driven by labor income? We capitalize income tax return data to estimate new annual series of US wealth concentration since 1913 Key result: Wealth inequality has surged, mostly because of the rise of the top 0.1% wealth share (=wealth above $20m)
Goal: distribute the total household wealth recorded in the Flow of Funds The composition of household wealth in the U.S., 1913-2013 50 40 % of national income 30 20 10 Equities Sole proprietorships & partnerships Pensions 1913 1918 1923 1928 1933 1938 1943 1948 1953 1958 1963 1968 1973 1978 1983 1988 1993 1998 2003 2008 2013 Currency, deposits and bonds Housing (net of mortgages)
To obtain wealth, we capitalize income tax returns How the capitalization technique works: Start from each capital income component reported on individual tax returns Compute aggregate capitalization factor for each asset class that maps total tax-returns income to Flow of Funds wealth Multiply each individual capital income component by capitalization factor of corresponding asset class Simple idea, but lot of care needed in reconciling tax with Flow of Funds data Key assumption of the capitalization method: within asset class, each group has the same average capitalization factor
Evidence suggests that within-asset class realized returns are flat Two potential issues: Maybe the rich have higher equity/bond returns (e.g., better at spotting good investment opportunities) level bias Maybe this differential has increased since the 1970s (e.g., due to financial globalization/innovation) trend bias We successfully test the capitalization method in three micro-datasets where we can observe both income and wealth All tests show that within-asset class returns are flat, so that capitalization method generates correct top wealth shares
We account for wealth that does not generate taxable income Owner-occupied housing Home values set proportional to property tax paid Home mortgages set proportional to mortgage interest paid We assume (based on SCF) that itemizers have 75% of home wealth and 8 of home mortgages Pensions Pension wealth set proportional to pension distributions and wages above 50th percentile Consistent with SCF and with direct information on IRA wealth (IRAs 3 of pension wealth) We each year cover 10 of the wealth reported in the Flow of Funds Distributional Flow of Funds
At the very top, US back to early 20th century wealth concentration levels 25% Top 0.1% wealth share in the United States, 1913-2012 2 % of total household wealth 15% 1 5% 1913 1918 1923 1928 1933 1938 1943 1948 1953 1958 1963 This figure depicts the share of total household wealth held by the 0.1% richest families, as estimated by capitalizing income tax returns. In 2012, the top 0.1% includes about 160,000 families with net wealth above $20.6 million. Source: Appendix Table B1. 1968 1973 1978 1983 1988 1993 1998 2003 2008 2013
Fortunes >$20mn grow much faster than average, fortunes of a few million do not 14% Top wealth shares: decomposing the top 1% 12% % of total household wealth 1 8% 6% 4% 2% Top 0.1%-0.01% Top 0.5%-0.1% Top 0.01% Top 1%-0.5% 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015
Mechanisms? 1. 1980s-1990s: upsurge of labor income at the top 8% 7% Share of income earned by top 0.1% wealth-holders % of total pre-tax income 6% 5% 4% 3% 2% 1% National income Labor income 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015 This figure shows the share of total pre-tax national income and pre-tax labor income earned by top 0.1% wealth-holders. Labor income includes employee compensation and the labor component of business income. Source: Appendix Tables B25 and B28.
2. Booming labor incomes saved at a high rate % of each group's total primary income 55% 5 45% 4 35% 3 25% 2 15% 1 5% -5% -1-15% 1917-19 Saving rates by wealth class (decennial averages) 1920-29 1930-39 1940-49 1950-59 1960-69 Top 10 to 1% 1970-79 1980-89 1990-99 Top 1% Bottom 9 2000-09 2010-12
3. Capital generates a sizable return snowball effect 2 Yield and total return on U.S. private wealth 15% Pure yield = capital income (including retained earnings) / wealth 1 5% 1913 1918 1923 1928 1933 1938 1943 1948 1953 1958 1963 1968 1973 1978 1983 1988 1993 1998 2003 2008 2013-5% Total return = pure yield + asset price effect -1 back
At the bottom: massive dissaving 15% Saving rate of the bottom 9 % of bottom 9 primary income 1 5% 1975 1980 1985 1990 1995 2000 2005 2010-5% -1
The average wealth of the bottom 9 is no higher today than in 1986 Real average wealth of bottom 9 and top 1% families Top 1% real average welath 14,000,000 12,000,000 10,000,000 8,000,000 6,000,000 4,000,000 2,000,000 Top 1% (left y-axis) Bottom 9 (right y-axis) 140,000 120,000 100,000 80,000 60,000 40,000 20,000 0 0 1946 1950 1954 1958 1962 1966 1970 1974 1978 1982 1986 1990 1994 1998 2002 2006 2010 Bottom 9 real average wealth Real values are obtained by using the GDP deflator, 2010 dollars. Source: Appendix Tables B3.
The rise and fall of middle-class wealth Composition of the bottom 9 wealth share 4 35% 3 % of total household wealth 25% 2 15% 1 Business assets Equities & fixed claims (net of non-mortgage debt) 5% 1917 1922 1927 1932 1937 1942 1947 1952 1957 1962 1967 1972 1977 1982 1987 1992 1997 2002 2007 2012 Pensions Housing (net of mortgages)
Conclusion: A first step toward DINA We constructed new, consistent series on the distribution of wealth fully consistent with Flow of Funds aggregates Next step: construct a microfile with individual-level income (pre-tax and post-tax) and wealth consistent with macro totals (in progress, with Piketty and Saez) = distributional national accounts (DINA), reconciling macro growth and inequality studies
DINAs make it possible to compute growth rates consistent with macro totals 70,000 Real average national income: Full adult population vs. bottom 9 Average income in constant 2012 dollars 60,000 50,000 40,000 30,000 20,000 10,000 2. 2. All adults 1.4% Bottom 9 adults 0.7% 0 1946 1950 1954 1958 1962 1966 1970 1974 1978 Real values are obtained by using the national income deflator and expressed in 2012 dollars. Source: Appendix Tables XX. 1982 1986 1990 1994 1998 2002 2006 2010 2014
Top 1% vs. bottom 9: from a factor of 20 to a factor of 40 1,400,000 Real average national income of bottom 9 and top 1% adults 70,000 Top 1% real average national income 1,200,000 1,000,000 800,000 600,000 400,000 200,000 0 1946 1.4% 2. 1950 1954 1958 1962 1966 1970 1974 1978 1982 1986 Top 1% (left axis) 3.5% 1990 0.7% 1994 1998 Bottom 9 (right axis) 2002 2006 2010 2014 60,000 50,000 40,000 30,000 20,000 10,000 Real values are obtained by using the national income deflator and expressed in 2012 dollars. Source: Appendix Tables XX. 0 Bottom 9 real average national income
Supplementary Slides
The concentration of taxable capital income has increased dramatically 45% 4 The top 0.1% taxable capital income share % of taxable capital income 35% 3 25% 2 15% 1 Including capital gains Excluding capital gains 5% 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015
Offshore tax evasion has probably increased since the 1970s 1 U.S. equities held by tax haven firms and individuals % of U.S. equity market capitalization 8% 6% 4% 2% 1940 1950 1960 1970 1980 1990 2000 2010 In 2012, 9% of the U.S. listed equity market capitalization was held by tax haven investors (hedge funds in the Caymans, banks in Switzerland, individuals in Monaco, etc.). Source: Zucman (2014) using US Treasury International Capital data. back
Check 1: In matched estates-income data, within asset class returns are flat 10. 9. Returns by asset and wealth class, 2007 (matched tabulated estates and income tax data) 8. 7. Dividends + capital gains 6. 5. 4. Dividends yield 3. 2. 1. 0. Interest yield up to $3.5m $3.5m-$5m $5m-$10m $10m-$20m $20m+ Total net wealth at death
The very rich did collect a lot of dividends in the 1970s 8. Dividend yield by wealth class in 1976 (matched micro estate and income tax data) 7. 6. 5. 4. 3. 2. 1. P90-95 P95-99 P99-99.5 P99.5-99.9 P99.9-99.99 P99.99-100 Fractiles of the distribution of net wealth at death
Capitalization method works for the SCF % of household wealth excluding pensions and owneroccupied housing 10 8 6 4 2 1988 Top 1 1990 Top household wealth shares: reported SCF wealth vs. capitalized SCF incomes Top 1% Top 0.1% 1992 1994 1996 Wealth 1998 Capitalized income 2000 The figure compares direct SCF wealth shares to wealth shares estimated by capitalizing SCF income. Wealth excludes pensions and owner-occupied net housing. Source: Appendix Table C1. 2002 2004 2006 2008 2010 2012
Capitalization works for foundations 8 7 Top 1% Top foundations wealth shares: reported wealth vs. capitalized income Wealth % of foundation net wealth 6 5 4 3 Top 0.1% Capitalized income 2 1985 1987 1989 1991 1993 1995 The figure compares top foundation wealth shares obtained by using balance sheet wealth data as reported to the IRS and obained by capitalizing IRS-reported income. Source: Appendix Tables C11 and C13. 1997 1999 2001 2003 2005 2007 2009
Wealth in 2012 is very concentrated Table 1: Thresholds and average wealth in top wealth groups, 2012 Wealth group Number of families Wealth threshold Average wealth Wealth share A. Top Wealth Groups Full Population 160,700,000 $343,000 10 Top 1 16,070,000 $660,000 $2,560,000 77.2% Top 1% 1,607,000 $3,960,000 $13,840,000 41.8% Top 0.1% 160,700 $20,600,000 $72,800,000 22. Top.01% 16,070 $111,000,000 $371,000,000 11.2% B. Intermediate Wealth Groups Bottom 9 144,600,000 $84,000 22.8% Top 10-1% 14,463,000 $660,000 $1,310,000 35.4% Top 1-0.1% 1,446,300 $3,960,000 $7,290,000 19.8% Top 0.1-0.01% 144,600 $20,600,000 $39,700,000 10.8% Top.01% 16,070 $111,000,000 $371,000,000 11.2%
Rates of returns on wealth around 7% No long-run price effects 14% Figure A8: Yield and total return on U.S. private wealth (decennial averages) 12% Pure yield 1 8% 6% 4% 2% Total return = pure yield + asset price effect 1913-19 1920-29 1930-39 1940-49 1950-59 1960-69 1970-79 1980-89 1990-99 2000-09 2010-13 back
Capital income missed by tax data 0.35 0.3 From reported to total capital income, 1920-2010 Retained earnings % of factor-price national income 0.25 0.2 0.15 0.1 Non-filers & unreported sole prop. profits Corporate income tax Imputed rents Income paid to pensions & insurance 0.05 Didivends, interest, rents & profits reported on tax returns 0 1920 1925 1930 1935 1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 back
Most trusts generate income taxable at the individual level 12% Wealth held in estates & trusts 1 % net household wealth 8% 6% 4% 2% Total estate & trust wealth Estate & trust wealth that does not generate distributable income 1952 1962 1972 1982 1992 2002 2012 back
Charitable giving follows top incomes Surge in top incomes is real Mean charitable giving of top 1% incomes / mean income back 9 8 7 6 5 4 3 2 1 1962 1966 Charitable Giving of Top 1% Incomes, 1962-2012 Mean charitable giving of top 1% divided by mean income [lea y- axis] Top 1% Income Share [right y- axis] 1970 1974 1978 1982 1986 Source: The figure depicts average charitable giving of top 1% inomes (normalized by average income per family) on the left y-axis. For comparison, the figure reports the top 1% income share (on the right y-axis). 1990 1994 1998 2002 2006 2010 25% 2 15% 1 5% Top 1% income share
Total returns of foundations grow with wealth but realized returns do not Figure C4: Return on foundation wealth, 1990-2010 average Returns including realized & unrealized gains 7% 6% 5% Realized return Unrealized capital gains 4% 3% 2% 1% 1m-10m 10m-100m 100m-500m 500m-5bn 5bn+ back
Wealth has always been concentrated 9 Top 1 wealth share in the United States, 1917-2012 85% % of total household wealth 8 75% 7 65% Capitalized income SCF 6 1917 1922 1927 1932 1937 1942 1947 1952 1957 1962 The figure depicts the share of total household wealth owned by the top 1, obained by capitalizing income tax returns versus in the Survey of Consumer Finances. The unit of analysis is the familly. Source: Appendix Tables B1 and C4. 1967 1972 1977 1982 1987 1992 1997 2002 2007 2012
Top 1% has gained more than top 1 55% Top 10-1% and 1% wealth shares, 1913-2012 5 % of total household wealth 45% 4 35% 3 25% 2 1913 1918 1923 1928 1933 1938 1943 1948 1953 1958 1963 1968 1973 1978 1983 1988 1993 1998 2003 2008 2013 Top 1 to 1% Top 1%
Top 1% surge is due to the top 0.1% 3 Top 1-0.1% and top 0.1% wealth shares, 1913-2012 25% % of total household wealth 2 15% 1 5% 1913 1918 1923 1928 1933 1938 1943 1948 1953 1958 1963 1968 1973 1978 1983 1988 1993 1998 2003 2008 2013 Top 1% to 0.1% Top 0.1%
Top 0.01% share: 4 in last 35 years 12% Composition of the top 0.01% wealth share, 1913-2012 1 % of total household wealth 8% 6% 4% 2% Equities 1913 1918 1923 1928 1933 1938 1943 1948 1953 1958 1963 1968 1973 1978 1983 1988 1993 1998 2003 2008 Fixed income claims Other
Wealth is getting older, but at the very top remains younger than in the 60s- 70s 6 Share of wealth held by elderly households (65+) 5 Top 0.1% % of each group's total wealth 4 3 2 Total population Bottom 9 1 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010
Our estimate for top 0.01% is consistent with Forbes rankings Top 0.00025% share (% of total household wealth) 3.5% 3. 2.5% 2. 1.5% 1. 0.5% 0. 1982 Forbes 400 (top.00025%) and top.01% Wealth Shares 1984 Top 0.00025%, Forbes magazine (left-hand scale) 1986 1988 1990 1992 1994 1996 1998 Top 0.01%, capitalized income (right-hand scale) The figure depicts the top.00025% wealth share as estimated from the Forbes 400 list on the left axis. For comparison, the figure reports our top 0.01% wealth share obtained by capitalizing income tax returns (on the right axis). Source: Appendix Table C3. 2000 2002 2004 2006 2008 2010 2012 14% 12% 1 8% 6% 4% 2% Top.01% share (% of total household wealth)
Estate tax returns fail to capture rising top wealth shares 25% Top 0.1% wealth share: comparison of estimates 2 Capitalized income % of total household wealth 15% 1 5% SCF adjusted SCF baseline Estate multiplier 1917 1922 1927 1932 1937 1942 1947 1952 1957 1962 The figure depicts the top 0.1% wealth share obained by capitalizing income, by using the Survey of Consumer Finances (SCF baseline and adjusted), and by using estate tax data (Kopczuk and Saez, 2004). Source: Appendix C4 and C4b. 1967 1972 1977 1982 1987 1992 1997 2002 2007 2012
SCF does not fully capture rising top capital income share Top 0.1% Capital Income Share in the SCF and Tax Data 45% Tax data 4 35% % taxable capital income 3 25% 2 15% 1 5% SCF 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 The figure compares the top 0.1% capital income shares estimated with the SCF data vs. the income tax data. Capital income includes realized capital gains, dividends, interest, net rents, and business profits. Source: Appendix Table C2.
Estate multiplier issue: mortality gradient by wealth within top 1 Relative Mortality by Age and Wealth Group, Men, 1999-2008 Mortality (relative to full population) 10 8 6 4 2 top 1 top 5% top 1% Kopczuk-Saez 30-49 50-64 65-79 80+ Age groups The figure depicts the relative mortality rate by age and wealth group for men in 1999-2008. E.g., male top 1% wealth holders aged 30-49 mortality rate is 4 of males aged 30-49 population wide. Kopczuk-Saez is based on the mortality of white college goers relative to population in the 1980s. The graph shows that mortality decreases with wealth (even within the top 1) and that the wealth mortality advantage decreases with age. Source: Appendix Table C7.
Estate multiplier issue: mortality gradient by wealth widens over time Mortality (relative to full population) 10 8 6 4 Evolu&on of Mortality Advantage, Men, Aged 65-79 top 1 top 5% top 1% Kopczuk-Saez 1979-1984 1984-1988 1989-1993 1994-1998 1999-2003 2004-2008 The figure depicts the relative mortality rate for men aged 65-79 by wealth group and period. E.g., male top 1% wealth holders aged 65-79 mortality rate is 9 of males aged 65-79 population wide in 1979-1984. Kopczuk-Saez is based on the mortality of white college goers relative to population in the 1980s. The graph shows that the wealth mortality advantage increases overtime and more so for the top 1% wealthiest. Source: Appendix Figure C7.
Bottom 9 wealth share decline due to (a) savings collapse, (b) income share fall Share of income and wealth of bottom 9 wealth holders 7 6 Income share 5 4 Simulated 1985-2012 wealth share (constant 3% saving rate and constant income share) 3 2 1 Simulated 1985-2012 wealth share (constant 3% saving rate) Observed wealth share 1917 1922 1927 1932 1937 1942 1947 1952 1957 1962 Since the 1980s the share of total household wealth owned by families in the bottom 9 of the wealth distribution has fallen proportionally more than the share of total pre-tax national income earned by these families. Source: Appendix Tables B1, B25 and B33c. 1967 1972 1977 1982 1987 1992 1997 2002 2007 2012
Table 2: Rates of growth, saving and return by wealth group Real growth rate of wealth per family Real growth rate of income per family Private saving rate (personal + retained earnings) Real rate of capital gains Total pre-tax rate of return g wf g yf s = S/Y q r + q 1917-1929 All 1.8% 0.5% 1 0.9% 9. Bottom 9-0.4% 0. 1% 0.2% 7.9% Top 1 2.3% 1.2% 23% 1. 9.2% Top 1% 3.6% 1.4% 28% 1.5% 10.5% 1929-1986 All 1.5% 2. 12% -0.6% 6.6% Bottom 9 3. 2.3% 6% -0.2% 6.2% Top 1 1. 1.4% 24% -0.9% 6.8% Top 1% 0.3% 0.5% 24% -1.1% 7.2% 1986-2012 All 1.9% 1.3% 9% 0.9% 7.5% Bottom 9 0.1% 0.7% 1.3% 7.5% Top 1 2.7% 2.3% 22% 0.7% 7.5% Top 1% 3.9% 3.4% 36% 0.9% 7.9%