Dec 15 -- The U.S. Bureau of Economic Analysis (BEA) announced a series of changes in its prototype measures of distribution of personal income -- including adding data for 2017-18, introducing new data on the distribution of disposable personal income, and implementing several methodological improvements. BEA invites users to submit feedback on the updated prototype measures at DistributionofPersonalIncome@bea.gov.
BEA distribution of personal income webpage: https://www.bea.gov/data/special-topics/distribution-of-personal-income
"Measuring Inequality in the National Accounts" (December 2020) https://www.bea.gov/research/papers/2020/measuring-inequality-national-accounts
"Technical Document: An Updated Methodology for Distributing Personal Income" (December 2020) https://apps.bea.gov/data/special-topics/distribution-of-personal-income/technical-document-a-methodology-for-distributing-personal-income.pdf
1) With the December 2020 release, a distribution of disposable personal income (DPI) was introduced. The same set of measures produced for the distribution of personal income were also produced for disposable personal income. The methodology document describes how the DPI distribution is created. The Excel spreadsheets in the data and documentation section of the webpage provide side-by-side comparisons of the two distributions for each year.
2) The years 2017 and 2018 have been added for the distributional measures of personal income and disposable personal income.
Changes in Methodology:
1) The method for adjusting top incomes was changed to:
• Account for underreported IRS Statistics of Income (SOI) data.
• Include partnership information in business income.
• Decompose dividend income into individual dividends, dividends tied to S corporations (now distributed according to business income), and imputed dividends tied to pensions (now distributed along with imputed interest).
• Allocate households for SOI adjustment such that the proportion of tax units in the SOI with adjusted gross income of more than $500,000 matches that of the Current Population Survey (CPS) for each year.
2) BEA now imputes employer contribution to health insurance using Medical Expenditure Panel Survey (MEPS) tables for 2010 forward because the 2018 CPS revision dropped this imputed variable. These imputations vary by family size, plan type, firm size, and state.
3) Tax credits are now distributed more finely according to recipiency and calculated using the NBER TAXSIM program.