We are unable to display your institutional affiliation without JavaScript turned on.
Browse Book and Journal Content on Project MUSE

Find using OpenURL

Comments and Discussion

From: Brookings Papers on Economic Activity
Fall 2012
pp. 184-200 | 10.1353/eca.2012.0023

In lieu of an abstract, here is a brief excerpt of the content:

Comment by Hilary W. Hoynes

This paper by Bruce Meyer and James Sullivan provides a comprehensive and informative analysis of poverty in the United States. Their analysis compares alternative measures of poverty over the more than five decades that poverty has been measured in this country. They compare official ("money income") poverty with alternative income-based measures as well as with consumption-based poverty measures. The focus in the paper is on the trend in poverty, motivated by whether one can conclude that the war on poverty has been won or lost. Meyer and Sullivan conclude that their preferred, consumption-based poverty measure shows significant improvement since the mid-1980s while income-based poverty shows little improvement. They argue that consumption-based poverty mea sures are preferred over income-based measures because the latter are biased by the underreporting of government transfers, which appears to have worsened over time. Additionally, they argue that consumption is preferable because it is closer to permanent income, and those classified as being consumption-poor appear more disadvantaged than those classified as income-poor.

As can be expected from these authors, the paper provides a detailed and serious analysis of a centrally important national statistic. The paper makes two central contributions. First, although consumption poverty is a well-identified alternative to income poverty, in practice consumption poverty measures have been used mainly in the developing-country setting. Meyer and Sullivan, here and in earlier work, have applied these principles to the United States. Additionally, many European countries include expenditure data—the basis for consumption poverty measurement—as part of their standard labor force surveys. In the United States the best source for expenditure data is the Current Expenditure (CE) Survey, which uses a much smaller sample than its labor survey counterpart the Current Population Survey. Notably, the public-use CE data do not allow for the identification of state of residence. Meyer and Sullivan's work is drawing more attention to data needs in this area.

Click for larger view
Figure 1. 

Real Expenditure per Capita on Cash and Near-Cash Safety Net Programs, 1980-2011

Second, throughout the more than 50 years of poverty measurement in the United States, tremendous attention has been given to the official poverty measure and its problems. The National Academy of Sciences (NAS) panel on poverty measurement recommended several alternatives (Citro and Michael 1995). Central in their recommendations were moving away from a money income measure and incorporating taxes and noncash benefits in the family resource measure. My figure 1, which updates figure 5 in my Brookings Paper with Marianne Bitler (Bitler and Hoynes 2010), illustrates the potential importance of this change. The figure plots real expenditure per capita from 1980 to 2012 for the three main cash or near-cash programs for low-income families: the earned income tax credit (EITC), Temporary Assistance for Needy Families (TANF, the successor to Aid to Families with Dependent Children, AIDC), and food stamps (now called the Supplemental Nutrition Assistance Program). During this period the combination of welfare reform, the expansion of the EITC, and the recent growth of the food stamp program have led to a dramatic change in the sources of government assistance for lower-income families. Because the official poverty measure counts only cash income, additional spending on tax-based assistance (through the EITC) or near-cash government assistance (through food stamps) has no effect on measured poverty.

In response to the NAS report, the Census Bureau implemented several "experimental" poverty measures, which are available for 1999 through 2010. More recently, the Census Bureau announced the Supplemental Poverty Measure (SPM) based on the NAS recommendations (Short 2011). Subject to funding, the bureau has committed to release the SPM each year. Many studies have followed and analyzed these changes (and many others not discussed here). Thus, much is known about the validity of different measures and how they perform over time. Meyer and Sullivan make a significant contribution by presenting additional poverty measures and (as no one had done previously) estimating a consistent series back to 1960.

The lens that the authors use to compare the alternative measures is the change in poverty over the entire 50-year period of...

You must be logged in through an institution that subscribes to this journal or book to access the full text.


Shibboleth authentication is only available to registered institutions.

Project MUSE

For subscribing associations only.