Comparison of BEA estimates of personal income and IRS estimates of adjusted gross income

Survey of Current Business, Nov, 2004 by Mark A. Ledbetter

New Estimates for 2002

Revised Estimates for 2001

THE Bureau of Economic Analysis (BEA) measure of personal income and the Internal Revenue Service (IRS) measure of adjusted gross income are both widely used measures of household income. Adjusted gross income (AGI) is the key, before-tax definition of income that is used by the IRS in the calculation of individual income tax liabilities. Personal income is the more comprehensive income measure; it represents current income received by persons from all sources, and it is often used in examining trends in national economic output, saving, investment, and consumer spending. (1)

A comparison of personal income and AGI for 2001 and 2002 is prepared because these two measures are often used jointly, although they are based on different concepts and serve different purposes. This comparison features the AGI gap, or the difference between the BEA-derived estimates of AGI and the IRS estimates of AGI. (For the definitions of selected terms, see the box "Key Terms.")

The AGI gap was $961.1 billion for 2002--the most recent year for which AGI data are available (table 1). The relative AGI gap--the gap as a percentage of BEA-derived AGI--was 13.7 percent. For 2001, the revised AGI gap was $834.4 billion (table 2); the revision reflects the incorporation of the results from the 2004 annual revision of the national income and product accounts (NIPAs) and of revised estimates of certain items that partly reconcile BEA-derived AGI and IRS AGI. IRS AGI was unrevised for 2001. The relative AGI gap for 2001 was revised up to 11.9 percent from 11.6 percent. (2)

The AGI Gap for 2002

The total AGI gap for personal income for 2002 is $961.1 billion (table 1). The misreporting adjustment accounts for $402.3 billion, so $558.8 billion is unexplained.

The AGI gap for wage and salary disbursements for 2002 is $308.2 billion. The misreporting adjustment accounts for $106.5 billion, so $201.7 billion is unexplained.

The AGI gap for farm proprietors' income for 2002 is $10.6 billion. For nonfarm proprietors' income, it is $395.4 billion. The misreporting adjustment accounts for $309 billion, so $86.4 billion is unexplained.

The AGI gap for rental income of persons for 2002 is $10.4 billion. The misreporting adjustment accounts for $1.3 billion, so $9.1 billion is unexplained.

The AGI gap for personal dividend income for 2002 is $74.4 billion.

The AGI gap for personal interest income for 2002 is $40.0 billion. The misreporting adjustment is -$14.4 billion.

The AGI gap for taxable pensions and annuities is $103.3 billion. For taxable unemployment compensation, the AGI gap is $10.7 billion. For taxable social security benefits, the AGI gap is $8.1 billion.

Revisions to the AGI Gap for 2001

The AGI gap was revised up $21.6 billion for 2001 (table 3). The upward revision was due to an upward revision of $11.1 billion to personal income and to net revisions of $10.6 billion to the reconciliation items that increased BEA-derived AGI by $21.6 billion. (3)

The $21.6 billion revision to the BEA-derived AGI is attributable to an upward revision of $8.0 billion to personal income that carried through to the BEA-derived AGI, to an upward revision of $12.6 billion to reconciliation items derived independently of NIPA revisions of personal income, and to an upward revision of $1.0 billion to AGI components. Total IRS AGI was not revised.

Of the $11.1 billion upward revision to personal income, $3.1 billion did not carry through to BEA-derived AGI because of offsetting revisions to the reconciliation items that are included in personal income. (4) All the revisions to the estimates of the reconciliation items that are estimated independently of personal income and of AGI carry through to the BEA-derived AGI (these estimates are derived from source data that are not used to prepare estimates of personal income or AGI). (5)

Revisions to AGI components (part of lines 9, 12, 14, and 15) were all carried through to the BEA-derived AGI. For example, capital gains and small business corporation income of AGI components are added, without adjustments, to personal income in the derivation of the BEA-derived AGI.

Appendix: Estimating the AGI Gap

Personal income and adjusted gross income (AGI) are based on different concepts, and each serves different purposes. Specifically, the estimates of personal income are available much sooner than the estimates of AGI, so they are used to extrapolate advance estimates of AGI. Because major changes in the tax laws affect the IRS estimates of AGI, the AGI estimates cannot be compared with BEA's historically consistent series of estimates of personal income. However, the estimates of personal income and the AGI gap can be adjusted to produce values that are comparable.

The AGI gap by type of income is estimated in three steps (tables 1 and 2). First, an estimate of BEA-derived AGI is prepared by adjusting the estimates of BEA personal income to conform to an IRS definition of AGI: Certain types of income are subtracted from personal income because they are not included in AGI (lines 3-9), and certain types of income are added to personal income to be consistent with AGI (lines 11-15).