Using Assets to Improve Poverty Measurement: Comparing Official, Supplemental, Consumption, and Assets-Augmented Poverty Measures

Background and Purpose

Valid assessment of the effects of social programs on deprivation hinges upon accurate measurement of poverty. Poverty measures are also crucial for identifying the disadvantaged, determining program eligibility, and tracking changes over time in the prevalence and depth of poverty. There is, however, a longstanding debate on how to measure poverty and identify who is poor. Central to this debate is how to define the resources that are available to households to meet their basic needs. The U.S. government introduced the supplemental poverty measure (SPM) in 2010 to address many of the shortcomings of the official poverty measure (OPM). Although the SPM is widely considered to be superior to the OPM there are concerns that both income-based measures of poverty fail to capture the standard of living of households that have savings and other assets to draw upon. Two alternative definitions of resources—income augmented with the value of asset holdings and current consumption—are therefore appealing because they are conceptually closer to actual levels of material wellbeing than income.

We advance the debate on poverty measurement by examining whether incorporating assets in poverty measures improves identification of people who are poor by comparing the characteristics of people classified as poor using the OPM, the SPM, and three other poverty measures: asset-augmented OPM; asset-augmented SPM; and a consumption-based measure. We hypothesized that those classified as poor by the asset-augmented or consumption measures would have more disadvantages than the OPM and SPM poor.

Methods

We used data from a representative telephone survey of New York City residents aged 18 years or older. Respondents were recruited using random digit dialing with an additional random sample of clients drawn from a probability sample of social service agencies to include individuals of low-income unreachable via telephone.

The survey collected demographic characteristics of all household members and information on income, assets, debts, and consumption, including earned income, investment income, government transfers, bank balances, retirement account balances, home value, value of vehicles, mortgages, taxes, etc., which were used to calculate the OPM income, SPM income, and total consumption. We added annuitized net worth to the OPM income and the SPM income to obtain asset-augmented OPM and asset-augmented SPM resources, respectively.

We compared people categorized as poor under each of the five measures on six domains of disadvantage: general health, mental health, life satisfaction, neighborhood conditions, material hardship, and financial insecurity. 

Results

Augmenting OPM and SPM using net worth added to the OPM and SPM poverty rolls people with poorer general and mental health, more financial insecurity, more material hardships, lower life satisfaction, and poorer neighborhood conditions. The consumption-based measure had the worst performance of the five measures examined.

Conclusions and Implications

Including assets in the definition of household resources may improve the accuracy of poverty measurement. The results also suggest that assets allow households to minimize material deprivations or the effects of deprivations on wellbeing, and should be a component of efforts to understand causes and levels of poverty.