Working poverty is a widespread but under-analyzed and poorly-measured problem in the US

by Brian C. Thiede, Daniel T. Lichter, and Scott R. Sanders

Pundits and politicians in the U.S. frequently assume that poverty is the result of joblessness and idleness, and in contrast, upward social and economic mobility is possible for anyone given sufficient work effort. Such assumptions about work and poverty underpin many social policies. For example, minimum wages are largely designed to ensure that work is remunerated with a basic, presumably above-poverty standard of living. Stringent time limits and work requirements on many so-called safety net programs also presume that incentivizing work is an effective means of reducing poverty and public sector dependence.

In contrast to these assumptions, however, evidence suggests that a large share of the poor live in families which are attached to the formal workforce. Yet to date researchers have produced a rather limited set of empirical findings about the working poor. In our judgment, this knowledge gap has largely been due to inconsistent conceptualization and measurement of the working poor across studies. To address this gap, our research has examined the conceptual and normative assumptions behind different measurement choices for studying the working poor, and compared empirical estimates based on a range of measures.  Our findings demonstrate that working poverty is a widespread problem, but also show that estimates of the magnitude of the problem are sensitive to measurement choices.

Conceptualizing and measuring working poverty 

Any definition of the working poor requires a number of consequential assumptions. Which family members and members of society should be responsible for their economic welfare by working? How many hours per year does one need to work to be classified as a worker? Should income and poverty be defined at the individual or family level, and what poverty thresholds should be used? These are indeed technical questions, but their answers also reflect normative ideas about which populations are deserving of public support. As the 1996 welfare reform act demonstrated, ideas about the etiology of poverty have significant effects on policy.

These assumptions inform choices about how to define six necessary parameters for measuring working poverty:

Potential workers: the population eligible to be defined as a worker. Persons included in the universe of potential workers included: all individuals aged 18-64; heads of primary families; and all family members.

Weeks worked: the number of weeks worked in the previous year needed to be classified as a worker. We considered three alternative classifications of workers on the basis of weeks worked:  those working year-round (50+ weeks per year); at least half of the year (27+ weeks per year); and at least one week per year.

Hours worked: the number of hours per week usually worked in the previous year needed to be classified as a worker. We distinguished between persons who worked full-time (35+ hours per week); part-time or more (17+ hours per week); and any hours (1+ hours per week).

Income: we measure the working poor based on unadjusted annual income; and income adjusted to a full-time, full-year equivalent.

Poverty threshold: the level of income used to classify the poor and non-poor. We alternately distinguished these groups based on: the official poverty thresholds; 125 percent of the official poverty thresholds; 50 percent of the median U.S. family or individual income; and the Census Bureau’s supplemental poverty measure. 

Poverty unit: the group for which poverty status is determined. We consider income and poverty at both the individual and family levels. The former evaluates whether work provides sufficient income from earnings to lift a single person out of poverty. The latter measures when work lifts the family out of poverty.

Our research examines the conceptual and normative implications of choices regarding these parameters. We also conducted extensive analyses of the 2013 Current Population Survey that revealed how estimates of the size and racial composition of the working poor population were affected by measurement choices (across 126 distinct measures). We also drew upon our discussion and findings to identify a number of ‘best measures’ of working poverty.

Our analyses revealed a number of key results:

• The estimated size of the working poor population is highly sensitive to alternative assumptions. Headline rates of “working poverty” ranged from 2 percent to nearly 19 percent.
• Statistical analyses of alternative working poverty measures suggested that, empirically, the ‘best measure’ of working poverty defines workers as family reference persons’ working at least a half-time equivalent (27+ weeks per year and/or 17+ hours per week); defines income and poverty at the family level; and uses 125 percent of the official poverty threshold to capture the ‘near poor’, who are at high risk of falling below the official poverty line.
• Using this ‘best’ definition, 9.3-11.0 percent of working family heads were poor in 2013. This rate translated into 6.4-8 million working family heads in poverty, between 20 and 24 million people living in poor families with working heads.
• Average rates mask considerable racial and ethnic variability in poverty rates 6.0 percent-7.2 percent of non-Hispanic white workers are poor and 6.8 percent-7.5 percent of Asian workers are poor; but 16.2 percent-19.7 percent of non-Hispanic black workers are poor, and 19.9 percent-22.1 percent of Hispanic workers are poor. Racial and ethnic variability changes slightly in magnitude, but rarely in kind, across alterative measures. 

Social policies are based on a combination of ethnic assumptions and technical or utilitarian concerns. Our research has highlighted the linkages between these domains, and argued for renewed attention to the working poor. Improving research and policy regarding this vulnerable population will require close attention to the ways that approaches to measuring the working poor population (and thus assessing the extent of the problem) reflect or distort normative ideas about work, poverty, and the safety net. With high levels of mobilization both in support and opposition of increased minimum wages, both sides should think carefully about the measurement and ethnical choices that underlie their facts. 

Brian C. Thiede is a PhD Candidate in Development Sociology at Cornell University. Daniel T. Lichter is the Ferris Family professor in the Department of Policy Analysis and Management, Professor of Sociology, and Director of the Cornell Population Center. Scott R. Sanders is Assistant Professor of Sociology at Brigham Young University.

This article is based on the paper, ‘America’s Working Poor: Conceptualization, Measurement, and New Estimates’, in Work & Occupations.

Cross-posted at the USAPP – American Politics and Policy blog.