A new study published by AEI shows that American living standards have been rising steadily since 1947, despite assertions by some analysts that they have been declining. The new paper, Living Standards in the United States, also reports that inequality in the distribution of total expenditures is significantly smaller than differences in pre-tax family income.

The study, by Professor Daniel T. Slesnick of the University of Texas at Austin, argues that the income-based approach to measuring poverty and living standards, long used by the U.S. Bureau of the Census, leads to erroneous conclusions. The income approach understates what has been a continuous rise in living standards and exaggerates the differences in the well-being of American families, Slesnick finds.

Slesnick proposes that the government recalculate poverty using a consumption-based approach and switch from the consumer price index (CPI) for price-level adjustments to a price index calculated using prices reported in the National Income and Product Accounts. This method of measurement yields a poverty rate that is half the official rate.

"Economic theory unambiguously recommends that consumption, rather than income, be used to assess economic well-being," Slesnick says. Consumption rather than income provides a better measure of well-being over time, according to Slesnick, because it does not fluctuate as much as income does. Households compensate for short-term income disruptions that may be caused by economic fluctuations by borrowing money or by drawing from their savings. As a result, consumption provides a more accurate indicator of well-being over extended periods than does income.

Other problems arise when CPI-deflated family income is used as an indicator of well-being. These problems include failure to account for differences (for instance, in size and composition) across families; ignoring "unrelated individuals," who make up roughly one-third of all families and unrelated individuals; and downward biases in the growth rates of welfare indexes that result from overestimates of the CPI inflation rate.

Slesnick also concludes that moving to consumption-based social welfare statistics will help policymakers assess the case for new antipoverty programs and more accurately identify those who need the assistance.

The author holds the Rex G. Baker Jr. Professor in Political Economy title at the University of Texas at Austin and has published widely on welfare economics and applied econometrics.