Public Finance, 2nd Edition

most public sector economists had long believed that local property taxes

were at least mildly regressive. In the 1970s, a ``new view'' consensus emerged

that the property tax is almost certainly progressive.

3

Perhaps it is not surprising that empirical estimates of the incidence of

any tax should vary considerably, given the nature of the problem. Empirical

researchers must select what they think are the most important market

reactions to the tax from a staggering set of possibilities, and methods of

selection are bound to diVer. Unfortunately, empirical tax incidence analysis

appears not to be especially robust to assumptions made about sectors of the

economy not explicitly under examination. Another confounding factor,

mentioned earlier, is that researchers often employ diVerent theoretical meas-

ures of incidence as a basis for their empirical work, and it is the theoretical

diVerences that we wish to focus on here.

THEORETICAL MEASURES OF TAX INCIDENCE

Three distinct theoretical measures of incidence commonly appear in the

literature: incidence as impact, incidence as changes in certain relative prices,

and incidence as changes in welfare.

Impact Equals Incidence

Some research merely reports the pattern of tax payments by income

class and judges the equity of the tax on this basis alone, thereby equating the

impact and incidence of the tax. As noted above, most incidence studies of the

personal income tax employ this measure, on the assumption that income

taxes are essentially lump sum. For example, Joseph Pechman and Bernard

Okner allocate personal income tax burdens in this manner in their widely

cited Brookings studies, Who Bears the Tax Burden and Who Paid the Taxes,

1966±85?,

4

which appeared in 1974 and 1985.

Changes in Relative Prices

At the other end of the spectrum, a large group of incidence studies base

their measures of incidence on changes in certain market prices in response to

a tax. The change in the wage±rental ratio is the usual choice. Actual tax

payments inXuence this incidence measure because their impact and size

3

M. Feldstein, ``Incidence of a Capital Income Tax in a Growing Economy with Variable

Savings Rates,'' Review of Economic Studies, October 1974. Refer also to the American Economic

Review Papers and Proceedings, May 1974, articles by H. Aaron, ``A New View of Property Tax

Incidence,'' R. Musgrave, ``Is a Property Tax on Housing Regressive?,'' and comments. Also,

H. Aaron, Who Pays the Property Tax?, The Brookings Institution, Washington, D. C., 1975.

4

J. A. Pechman and B. A. Okner, Who Bears the Tax Burden?, The Brookings Institution,

Washington, D. C., 1974; J. A. Pechman, Who Paid the Taxes, 1966±85?, The Brookings Insti-

tution, Washington, D. C., 1985.

528 THEORETICAL MEASURES OF TAX INCIDENCE