The Tax Policy Center’s tables showing the distribution of President Obama’s new income tax proposals indicate that some middle-class households would pay more tax than under current law. The Administration says they wouldn't. The reason is that TPC and the White House disagree over what counts as income. The dispute centers on the President’s proposal to tax accrued, but unrealized, capital gains at death. We find that some middle-income households would pay this tax; the White House contends that all of the burden would fall on taxpayers with high incomes. OMB Director Shaun Donovan was quite pointed in his critique of our analysis on NPR’s Morning Edition today:

There’s some fundamental flaws [in the TPC analysis]. They actually assume that all of this income from capital gains isn't really income. It sort of defies logic to say that a family that has $500,000 of capital gains, that isn't income.

The Administration argues that unrealized capital gains held at the time of death become income in the year the law changes to make them taxable. It is certainly true that taxable income on a decedent’s tax return would increase under the President’s proposal, but our measure of income, called expanded cash income, is intended to track pre-tax economic status. That doesn't change just because a tax is levied on accrued gains. Treasury Deputy Assistant Secretary Adam Looney proposes a different income measure: adjusted gross income (AGI) plus unrealized capital gains taxed at death. Not surprisingly, using this income measure, nobody with income below $100,000 would be affected by the new capital gains proposals since they allow a $100,000 capital gains exclusion for singles ($200,000 for couples). Treasury’s distribution table is reproduced below. [[{"type":"media","view_mode":"default","fid":"128471","attributes":{"class":"media-image alignnone size-full wp-image-7765","typeof":"foaf:Image","style":"","width":"613","height":"421","alt":"Looney table"}}]] The problem is capital gains taxed at death represent a lifetime of accruals, not a single year’s income. The whole point of the Administration proposal is that these gains have entirely escaped taxation for years, so levying tax would close a giant loophole in the tax code. (I commented favorably on the merits of the proposal here. ) A Treasury Department working paper argued in 1999 (when I was Deputy Assistant Secretary) that the economically correct way to account for capital gains is to include assets’ annual increase or decrease in value in each year’s “economic income.” By that measure, only a fraction of capital gains taxed at death would be included in a decedent’s income, with the rest included in years prior to death. By extension, accrued tax liability should also be allocated as asset values increase or decrease. (Accountants use the concept of accrued tax liability to allocate deferred business taxes for company financial statements, so this isn't a radical economic concept.) If we classify households by counting gains allocated on an accrual basis, the President’s proposal would affect some households with moderate income. Data from the Federal Reserve Board’s Survey of Consumer Finances show that a small fraction of people with modest incomes have accrued unrealized capital gains large enough to be affected by the tax. (See Table 4 from our report on the distributional effects of the new proposals.) [[{"type":"media","view_mode":"default","fid":"128476","attributes":{"class":"media-image alignnone size-full wp-image-7766","typeof":"foaf:Image","style":"","width":"910","height":"405","alt":"SFC accrued gains"}}]] The bottom line is that while the President’s proposal is targeted at wealthy people, a small fraction of them have modest incomes measured on an annual basis. There are other reasons why our estimates and Treasury’s differ. But reclassifying exceptionally thrifty middle-class families to the top of the income distribution by counting a lifetime of unrealized gains in income when they die clearly overstates their well-being. Follow me on twitter.