Distributional National Accounts DINA Facundo Alvaredo Anthony B. Atkinson Thomas Piketty Emmanuel Saez Gabriel Zucman Meeting of Providers of OECD IDD Data OECD, Paris, February 18-19, 2016
Envision a realistic plan and timetable for harmonized, annual, global distributional national accounts It will probably take a long time before we are able to develop official, consensual DINAs. For many years decades to come, inequality statistics will still be produced by various groups (academics, statistics institutes ). It took a long time (1910-1950s) before scholars could hand over the computation of NI and GDP to official institutes. It took a long time (1950-2000s) before national accounts were able to standardize stocks accounts (first consistent balance sheets in Germany released in 2010); still many missing countries. In the WTID/WID series, despite efforts, the units of observation, the income concepts and the Pareto interpolation techniques were never made fully homogeneous over time and across countries. In the WTID/WID series, attention is restricted to the top decile income share, rather than the entire distribution.
Overall DINA agenda: Producing annual estimates of the distribution of income and wealth using concepts of income and wealth that are consistent with the national accounts. Producing synthetic micro-files on an annual basis (income, wealth, age, gender, savings) DINA and associated micro-files expected to be as consistent as possible across time, countries, income and wealth definitions, and cover the whole distribution (and not only the top). DINA as a new tool to connect inequality and macro Defining a clear common conceptual framework Incorporating results from economic theory Developing statistics techniques to fruitfully use information from all available micro sources Estimating income/wealth growth rates for tiny groups of the distrib.
W2ID Working Paper 2015-01 Distributional National Accounts (DINA) Guidelines: Concepts and Methods used in the W2ID * Facundo Alvaredo, Anthony B. Atkinson, Thomas Piketty, Emmanuel Saez, Gabriel Zucman This version: September 2 nd, 2015 (Preliminary and incomplete) Section 1. Introduction Section 2. Units of observation Section 3. Income concepts Section 4. Wealth concepts Section 5. Basic imputation methods US: Piketty, Saez, Zucman France: Garbinti, Goupille, Piketty UK: Alvaredo, Atkinson Section 6. Reconciling wealth inequality sources Section 7. Synthetic micro-files Section 8. Countries/years with limited data Section 9. Concluding comments References
Benchmark unit of observation: the adult 20+ individual In joint taxation countries we assume a 50-50% split Whenever possible: also equal-split inequality (full redistribution of resources between spouses) Intermediate splitting hypothesis possible Dependent children taken into account for the relevant cash and in-kind transfers Geographical scope: world/country/states Synthetic micro-files: can be aggregated from country to world and regional level with the appropriate population weights Dimensions: income, wealth, age, gender, saving
Four income concepts Personal factor income Personal factor income, which for simplicity we generally refer to as "factor income", and which could also be labeled "personal primary income", or "personal market income", is equal to the sum of all pretax personal income flows accruing directly or indirectly to the individual owners of the production factors, labor and capital, before taking into account the operation of the tax/transfer system, and before taking into account the operation of the pension system. Personal pre tax income (difficulty contribution-based social insurance and non-contribution based social assistance) Personal pretax income, which for simplicity we generally refer to as "pretax income", is equal to the sum of all pretax personal income flows accruing to the individual owners of the production factors, labor and capital, before taking into account the operation of the tax/transfer system, but after taking into account the operation of the pension system.
Personal disposable income Disposable income, which could also be labeled "personal disposable income", is equal to the sum of all income flows accruing to individuals, after taking into account the operation of the tax and transfer system. Personal fiscal income Definition close to that of the tax law and to what people have in mind when thinking about their personal income. Also, definition close to the income control in the WTID/WID.
Table 3. Personal factor income (DINA) Personal factor income = 1 661 101% = Net national income of Total economy (B5n, S1) 1 642 100% - Property income (D4) received by General government Sector (S13) - 22-1% + Property income (D4) paid by General government Sector (S13) 42 3% - Property income (D4) received by NPSIH (non-profit) Sector (S15) - 7 0% + Property income (D4) paid by NPSIH (non-profit) Sector (S15) 6 0% Personal factor income = 1 661 101% = Primary income of Household Sector (B5n, S14) 1 358 83% + Primary income of Corporations Sector (B5n, S11+S12) 112 7% + Taxes on production (net) (D2-D3) received by General govt. Sector (S13) 191 12% Personal factor income = 1 661 101% = Factor labor income 1 341 82% + Factor capital income 320 19% Personal factor labor income = 1 341 82% = Compensation of employees (D1) received by Household Sector (S1) 1 154 70% + Labor share (70%) of net mixed income (B3n + net D45, S14) 33 2% + Imputed Taxes on production (net) (in proportion to income) 154 9% Factor capital income = 320 19% = Capital share (30%) of net mixed income (B3n + net D45, S14) 14 1% + Net operating surplus (housing rents) of Household Sector (B2n, S14) 69 4% + Property income (D4 except D45) received by Household Sector (S14) 102 6% - Property income (D4 except D45) paid by Household Sector (S14) - 14-1% + Undistributed profits (Primary income of Corporations (B5n, S11+S12)) 112 7% + Imputed Taxes on production (net) (in proportion to income) 37 2%
Table 4. Personal pretax income (DINA) Net national income of Total economy (B5n, S1) 1 642 100% Personal pretax income (pension-based definition) = 1 639 100% = Factor Income 1 661 101% - Pension contributions (D6111+D6121+D6131+D6141, S14) - 309-19% - Investment income payable to pension entitlements (D442, S14) - 8 0% + Pension benefits (D6211+D6221, S14) 295 18% Personal pretax income (broad definition) = 1 652 101% = Factor income 1 661 101% - Net social contributions (employer and households) (D61, S14) - 333-20% - Investment income payable to pension entitlements (D442, S14) - 8 0% + Social insurance benefits (D621+D622, S14) 332 20% Personal pretax income (broad definition) = 1 652 101% = Pretax labor income 1 332 81% + Pretax capital income 320 19% Pretax labor income = 1 332 81% = Factor labor income 1 341 82% - Net social contributions (employer and households) (D61, S14) - 333-20% + Social insurance income (labor share) (in proportion to contributions) 324 20% Pretax capital income = = Factor capital income 320 19% - Investment income payable to pension entitlements (D442, S14) - 8 0% + Social insurance income (capital share) (in proportion to contributions) 8 0% Social insurance benefits (D621+D622, S14) 332 20% = Social security benefits in cash (D621, S14) 53 3% + Other social insurance benefits (D622, S14) 279 17%
Table 5. Disposable income (DINA) Nt Net national income of Total Ttl economy (B5n, S1) 1 642 100% Disposable income (cash income) = 1 300 79% = Pretax Income (simplified definition) 1 652 101% - Taxes on production (net) (D2-D3) received by General govt. Sector (S13) - 191-12% - Current taxes on income and wealth (D5) received by General govt. (S13) - 213-13% + Social assistance benefits in cash (D623, S14) 52 3% Disposable income (cash income + in-kind transfers) = 1 515 92% = Pretax Income (simplified definition) 1 652 101% - Taxes on production (net) (D2-D3) received by General govt. Sector (S13) - 191-12% - Current taxes on income and wealth (D5) received by General govt. (S13) - 213-13% + Social assistance benefits in cash (D623, S14) 52 3% + Social transfers in kind (D63, S14) 215 13% Disposable income (cash income + in-kind transfers + collective expenditure) = 1 684 103% = Pretax Income (simplified definition) 1 652 101% - Taxes on production (net) (D2-D3) received by General govt. Sector (S13) - 191-12% - Current taxes on income and wealth (D5) received by General govt. (S13) - 213-13% + Social assistance benefits in cash (D623, S14) 52 3% + Social transfers in kind (D63, S14) 215 13% + Collective consumption expenditure (P32, S13+S15) 169 10%
Basic imputations methods High quality administrative micro-data and combination with surveys Upscale incomes to the three income concepts Pragmatic but clear and explicit imputations (proportional upgrading, use of other available information, use legal information) Use of income and wealth surveys Synthetic micro-files Distribution according to the four income concepts Univariate (no information on joint distributions) Joint distribution o Simple functional form for the copula distribution G(yl,yk) such as the bivariate Pareto (Atkinson et al.)
Avrage income or capital by age (% average 18+) 180% 170% 160% 150% 140% 130% 120% 110% 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Figure 1. Age-income and age-wealth profiles (France DINA 2006) Total factor income Factor labor income Factor capital income Capital (private wealth) 20 30 40 50 60 70 80 90
Avrage income or capital by age (% average 21+) 180% 170% 160% 150% 140% 130% 120% 110% 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Figure 2. Age-income and age-wealth profiles (DINA 2006) Pretax Labor income Pretax Total income Pretax Capital income Capital 20 30 40 50 60 70 80 90
Table E2. Detailed income distribution table (template table based upon data from France 2010) Percentile p Number of adult individuals 20-yr-+ in percentile Percentiles of factor income Lower factor income threshold Average factor income Percentiles of pretax income Lower pretax income threshold Average pretax income Percentiles of disposable income Lower disposable income threshold Average disposable income Percentiles of net wealth Lower net wealth threshold Average net wealth Total pop. 50 366 272 33 551 24 284 182 572 0 503 725 0 0 0 0 0 0 1 503 643 0 0 0 0 0 0 2 503 616 0 0 0 0 0 0 3 503 762 0 0 0 0 0 0 4 503 634 0 0 0 0 0 0 5 503 697 0 0 0 0 0 0 6 503 669 0 0 0 0 0 0 7 503 768 0 5 0 2 0 0 43 503 633 12 487 12 966 17 415 17 613 60 402 63 076 44 503 605 13 466 13 959 17 811 17 999 65 319 67 765 45 503 844 14 455 14 976 18 180 18 359 70 018 72 013 46 504 079 15 509 16 045 18 535 18 699 73 746 75 603 47 503 130 16 593 17 196 18 869 19 044 77 333 78 750 48 503 686 17 782 18 375 19 216 19 385 80 342 81 808 49 503 669 18 980 19 596 19 555 19 721 83 156 84 555 50 503 676 20 203 20 807 19 891 20 060 86 153 87 489 51 504 280 21 426 22 053 20 226 20 398 88 785 89 952 52 503 125 22 676 23 291 20 565 20 735 91 197 92 613 53 503 607 23 905 24 484 20 900 21 070 93 838 95 061 99.992 596 3 193 306 3 334 714 1 522 239 1 606 414 34 800 000 36 300 000 99.993 411 3 500 309 3 657 988 1 650 506 1 724 779 38 200 000 40 100 000 99.994 504 3 830 937 4 056 416 1 802 877 1 895 493 42 000 000 44 500 000 99.995 504 4 316 785 4 617 444 1 998 960 2 136 060 47 600 000 50 900 000 99.996 503 4 959 455 5 467 883 2 287 483 2 527 626 54 500 000 60 100 000 99.997 502 5 974 735 6 460 247 2 814 833 3 258 277 66 400 000 72 200 000 99.998 503 7 197 656 8 668 643 3 888 071 4 808 121 79 300 000 96 600 000 99.999 505 11 000 000 20 700 000 5 855 208 10 900 000 123 000 000 233 000 000
Most capital income is missed by tax data 35% 30% From tax-reported to total capital income Retained earnings % of factor-price national income 25% 20% 15% 10% Non-filers & unreported sole prop. profits 2/3 missed by tax data Corporate income tax Imputed rents Income paid to pensions & insurance 5% 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
Agrowingfractionoflaborincomeis missed by tax data % of total NIPA compensation of employees 100% 95% 90% 85% 80% 75% 70% 65% 60% 55% From taxable to total employee compensation Employer payroll taxes Other 1/4 missed Pension by tax contributions data Reported taxable wages Health benefits 50% 1953 1958 1963 1968 1973 1978 1983 1988 1993 1998 2003 2008 2013
How we allocate taxes We follow standard tax incidence results: Labor taxes fall on labor; capital taxes on capital (and corporate tax on all capital assets: Harberger 1962) Reasonable if Y = F (K, L) haselasticityofsubstitution >> labor supply elasticity e L and capital supply elasticity e K Cross-country and time-series evolution of = rk/y and = K/Y broadly consistent with view that > 1ande L and e K relatively small in the long run e K But this is uncertain! will revisit if needed
How we construct micro-measures of capital income matching macro totals 1. Construct micro-measures of wealth Follow Saez and Zucman 2014: category by category, pragmatic approach Delivers micro-measures of family wealth matching macro totals Split capital 50/50 among spouses 2. Derive micro-measures of pre-tax capital income Compute aggregate pre-tax rates of return for each asset class that reconcile NIPA income flows with Flow of Funds wealth Apply these rates of return to individual assets
How we construct distributional estimates of national labor income 1. Start from reported wages; split income of spouses using W2 forms 2. Employer payroll taxes: apply schedule, e.g., in 2015: Medicare: 1.45% Social Security: 7.2% capped at $118,500 Unemployment insurance: vary by state 3. Pension and health insurance contributions: Available on W2 forms since 1999 for pensions, since 2012 for health
National income is more concentrated than tax income 55% Top 10% pre-tax income shares 50% % of total income 45% 40% 35% National income per adult (DINA) 30% 25% 1917 1922 1927 1932 1937 1942 1947 IRS family market income (Piketty-Saez) 1952 1957 1962 This figure displays the share of total pre-tax national income earned by top 10% adult income earners and the share of total IRS market income earned by top 10% family tax units. Source: Appendix Tables XX. 1967 1972 1977 1982 1987 1992 1997 2002 2007 2012
Less true for top 1% 25% 20% Share of national income earned by top 1% adult income earners National income per adult (DINA) % of total income 15% 10% 5% IRS family market income (Piketty-Saez) 0% 1913 1918 1923 1928 1933 1938 1943 1948 1953 1958 This figure displays the share of total pre-tax national income earned by top 1% adult income earners and the share of total IRS market income earned by top 1% tax units. Source: Appendix Tables XX. 1963 1968 1973 1978 1983 1988 1993 1998 2003 2008 2013
DINAs make it possible to compute growth rates consistent with macro totals 70,000 Real average national income: Full adult population vs. bottom 90% Average income in constant 2012 dollars 60,000 50,000 40,000 30,000 20,000 10,000 2.0% 2.0% All adults 1.4% Bottom 90% 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
The top 10% has grown three times faster than the bottom 90% since 1980 350,000 Real average national income of bottom 90% and top 10% adults 70,000 Top 10% real average national income 300,000 250,000 200,000 150,000 100,000 50,000 0 Top 10% (left axis) 1.9% 2.0% 2.3% 0.7% Bottom 90% (right axis) 60,000 50,000 40,000 30,000 20,000 10,000 0 1946 1950 1954 1958 1962 1966 1970 1974 1978 1982 1986 1990 1994 1998 2002 2006 2010 2014 Bottom 90% real average national income Real values are obtained by using the national income deflator and expressed in 2012 dollars. Source: Appendix Tables XX.