Elasticity of the US Federal Personal Income Tax 1960 - 2015

The elasticity of a tax system is of interest as a proxy for progressivity, and also as a key parameter to estimate the response of tax revenues to changes in income. TAXSIM is an ideal way to calculate that elasticity, since tax liability can be calculated from a sample of actual tax returns, and then again after increasing all the incomes and deductions by 1%.

In these tables all dollar amounts are increased by an equal percentage, income and deduction amounts. See here for tables where deductions are unaffected.

Real Bracket Creep

Some time ago I was asked if I had any ideas about the causes of the inceased share of income tax revenues in GDP from 1995 to 2000, and my answer was that the I thought that the increase in real incomes probably explained a substantial part of it. Here is a simple analysis, which suggests that real income growth accounts for about a quarter of the increase through real bracket creep.

Here is the basic calculation for the change in share:

Assume a 16% increase in real GDP from 1995 to 2000. Then real GDP/taxpayer might incrase by 11%, allowing for 1%/year return growth. According to the elasticity formula, the tax liability in GDP will be expected to grow by 1.66 (a number from an earlier version of this table) times the growth in real per taxpayer income. This means the revenue grows by 18%, or 6% more than income growth. So the 8.12% share of liability in GDP for 1995 would be expected to increase to 1.06*8.12 or 8.6%. This accounts for a quarter of the distance from the 1995 share to the 10.2% estimate that I see mentioned in the CBO budget outlook for the share of individual income taxes in GDP for FY 2000.

Last update August 2016 by feenberg@nber.org