Printer Friendly

Work, poverty, and the working poor: a multifaceted problem.

Work, poverty, and the working poor: a multifaceted problem

In 1984, the poverty rate for all households in the United States was slightly less than the rate for 1967 and at about the same as that in 1971.(1) About one-fourth of all heads of household whom we classified as "expected to work' had low weekly earnings. However, about 60 percent of these households escaped poverty.

This article describes changes from 1967 to 1984 in the economic status of households headed by persons who are expected to work. It compares the situations of households that are "poor' with those headed by "low earners.' Excluded from the group expected to work are householders who are over age 65, the disabled, students, or women with a child under age 6.(2) Our results cast doubt on a common perception that most poor households are improverished because their heads, though capable of doing so, do not work.3

A household is classified as poor if its cash income, of all types and from all household members, falls below the official poverty line for a household of its size. We define low earners as household heads with weekly earnings below $204 per week in 1984 dollars. Such persons could not earn the poverty-line annual income for a family of four if they worked 52 weeks a year.4

Households headed by low earners are not necessarily poor. Whether or not the household is poor depends on the household's own poverty line and its annual cash income. Similarly, poor households do not necessarily have heads with low weekly earnings.5

As the following tabulation shows, there were 93.5 million households in the United States in 1984. Of these, 65.33 million had a head whom we classify as expected to work. Among this group of households, 17.03 million had low weekly earnings, but not all of them were poor:

By having a smaller family or other sources of household income, 10.94 million households were able to escape poverty. The heads of almost all poor households in which the head was expected to work had low weekly earnings (6.09 out of 6.66 million), while very few households in which the head did not have low weekly earnings were poor (0.57 out of 48.30 million). Thus, about 10 percent (6.66 out of 65.33 million) of all households in which the head was expected to work were poor in 1984.

Recent trends

Table 1 shows changes in the incidence of household poverty and the composition of all households and all poor households, classified by the characteristics of the household head, for selected years from 1967 to 1984. The poverty rate for all households declined from 17.1 percent in 1967 to 13.1 percent in 1979 and then rose to 15.2 percent in 1984, a rate that is quite close to that of 1971.

The proportion of all household heads expected to work declined slightly from 71.3 percent in 1967 to 69.9 percent in 1984. Among poor households, the proportion expected to work increased from 37.1 to 47.0 percent over this period, primarily because of the rapid decline in the incidence of poverty among the elderly. By implication, at most only half of the poverty population could be directly affected by increases in the demand for labor, inducements to supply more labor, or requirements to work.

The data in table 1 show that among all poor household heads expected to work in 1984, roughly equal proportions either did not work at all during the year, worked between 1 and 47 weeks, or worked all year. This is in contrast to 1967, when more than half of those expected to work (20.0 out of 37.1 percent) worked all year. An important cause of the decline in year-round work was the doubling of the unemployment rate between 1967 and 1984.

Incidence of low weekly earnings

Between 1967 and 1979, the incidence of household heads with low weekly earnings as a proportion of all household heads expected to work averaged 19.4 percent, but increased to 26.1 percent by 1984. In 1984, 21.1 percent of all such men and 44.9 percent of the women had low weekly earnings. Thus, a substantial proportion of household heads could not keep a family of four out of poverty even if they worked 52 weeks at their current weekly earnings.6 The following tabulation shows the incidence of low weekly earnings among male and female household heads who could be expected to work:

There were major differences in the incidence of, and trend in, low weekly earnings among male and female household heads. The incidence among men was much lower than that for women in every year studied. However, the incidence increased for men and decreased for women.

These trends did not change when we redefined the low earnings population as those whose weekly wage was insufficient to keep a family of three out of poverty--that is, the cutoff was lowered from $204 to $159 in 1984 dollars. The incidence of low weekly earnings for all household heads increased from 15.2 to 20.0 percent between 1967 and 1984 under this definition.

Escaping from poverty

Because the earnings of household heads can be supplemented by other sources of income and because the poverty threshold depends on family size, our measure of low earnings does not necessarily mean that the household is poor. Most poor household heads who are expected to work had low weekly earnings (91.4 percent), but most heads with low weekly earnings escaped poverty (64.2 percent).

Table 2 shows the methods by which households headed by those with low earnings escaped poverty. We classify low-earning household heads who escape poverty into one of the eight mutually exclusive categories shown. The classification is hierarchical: any household head who fits more than one category is classified only in the one closest to the top of the table.

Escape from poverty because of small family size indicates that even though the head's weekly earnings times 52 weeks was below the poverty line line for a family of four, actual annual earnings did exceed the poverty line for this household. In other words, this household had fewer than four members.

For each of the next three categories--earnings of other household members, other private income (which includes interest, dividends, rents, private pensions, and so forth), and public cash transfers--we use the following procedure. First, we compute the gap between the household's poverty line and the head's earnings. If the amount of income from the first of these sources exceeds this gap, the household was taken out of poverty by this source. If not, we compare the next source to the poverty gap. If no single source exceeds the gap, but some combination of sources does, the household is classified into the appropriate combination shown in rows 5 through 8. For example, consider a household of four persons, two parents and two children, with a poverty gap that is $2,000 after the head's earnings have been counted. If the wife earned more than $2,000 and the household received more than $2,000 in transfers, the household would be categorized as escaping from poverty because of the earnings of members other than the head (row 2). However, if the wife earned $1,999 instead, the household would be counted as escaping because of cash transfers (row 4). If both amounts were $1,999, then escape would be via the combination of other earnings and transfers (row 6).

In every year, at least 60 percent of households whose heads were low earners escaped poverty. The largest category of escape (row 1) indicates that the earnings of the head exceeded the household's poverty threshold, implying a household of fewer than four persons. The next most important source was the earnings of other household members. Other private income sources and cash transfers follow in roughly equal importance. However, the role played by cash transfers is small, owing in part to the hierarchical nature of our classification, but also to the unavailability of cash transfer programs for many of those expected to work and the relatively small average level of benefits for recipients.7

Characteristics of the poor, expected to work

Table 3 shows selected demographic and economic characteristics of poor households in which the head is expected to work. Although only about 10 percent of all households in which the head was expected to work were poor in 1984, they represented almost half of all poor households.

The top panel of table 3 further classifies poor households headed by persons expected to work by sex, race, Hispanic origin, and presence of children. In 1984, roughly half of these households (51.4 percent) consisted of single individuals or childless couples. Of the remaining 48.6 percent with children, 27.6 percent were white, 13.2 were nonwhite, and 7.8 percent were Hispanic. Thus, while a majority of poor households with children were white, nonwhites and Hispanics were overrepresented.

Between 1967 and 1984, households headed by women with children over age 6 increased from 13.4 to 17.7 percent of poor household heads who are expected to work. This trend toward the feminization of poverty was more pronounced among all poor households, as the percentage of such households headed by women with children under age 6 increased from 7.0 to 12.8 percent over the study period. (See table 1.)

The bottom panel of table 3 shows, for poor households in which there were children and a head expected to work, the proportion who received cash transfers, the weeks worked per year by the heads, the average amounts of household transfers, and the earnings of heads who worked, in constant 1984 dollars. Between 1967 and 1984, the share of those who received transfers increased from 17.1 to 38.2 percent for male-headed households, and from 48.9 to 61.5 percent for households headed by women. Most of the increase for men was attributed to the increased Social Security, disability, and unemployment insurance benefits; most, for women, to increased welfare receipts. The average transfer amount (in 1984 dollars) peaked at $3,336 for men and $5,425 for women in 1971. Between 1971 and 1984, the average benefit declined substantially for single mothers.

The fact that fewer than 40 percent of poor male household heads and only about 60 percent of all poor female household heads received transfers in 1984 indicates a substantial and growing gap in the safety net for many poor children.

There was a sharp decrease between 1967 and 1984 in the proportion of men who headed poor households and worked all year (from 71.8 to 45.6 percent) and a sharp increase in the proportion of those who did not work at all (from 7.8 to 17.0 percent). This undoubtedly reflects increased unemployment rates, but may also reflect increased participation in transfer programs.8 For women, the percentage who did not work at all diminished somewhat over the period, and the proportion working full year declined considerably (from 32.6 to 21.4 percent). Nonetheless, nearly half of the men and about 20 percent of the women (with children over age 6) who headed poor households worked all year in 1984.

For male household heads who worked, earnings in constant dollars declined substantially over the analysis period, reflecting both the decline in weeks worked and the increased incidence of low weekly earnings. Nonetheless, in every year studied, earnings were much more important to these households than were transfers. For female household heads who worked, earnings in constant dollars increased somewhat over the period. However, because more femalethan male-headed households received transfers and fewer had a working head, transfers were their most important income source.

Federal taxes of the working poor

While some able-bodied heads of poor households receive transfers, the majority of them work and pay taxes. Table 4 shows the amount of taxes that a hypothetical family of four at the poverty line (a now weekly earner by our definition) would have paid in Federal income tax and Social Security tax if he or she had worked all year and had had no other source of income.

In 1984, this family would have paid $366 in personal income taxes and $711 in Social Security taxes, or 10.1 percent of household income. Not only is this tax burden high in an absolute sense, but it is also high in comparison with the taxes imposed on similar poor households in earlier years. Although Social Security taxes steadily increased between 1965 and 1984, they were offset by reductions in Federal income taxes during the 1970's. The result was a decline in effective tax rates from 4.4 percent in 1965 to a low of 1.3 percent in 1975. This stands in sharp contrast to the steady increase in the effective tax rate on working poor households after 1975. Thus, if taxes were subtracted from the earnings data, the increased incidence of low weekly earnings over the 1967-84 period would be even greater than that shown in the text tabulation (p. 18). Moreover, a family with earnings only and a family with earnings and welfare benefits at a given income below the poverty level will be counted as equally poor in the census data, but the family with earnings only will have a lower spendable income owing to taxation.


Poverty for all households in 1984 was somewhat below the rate for 1967 and at about the same level as it was in 1971. Large changes have occurred, however, in the labor market characteristics of the poor during the period.

The majority of the heads of poor households are not expected to work because they are either over 65 years of age, disabled, students, or women with children under 6 years of age.

About a quarter of all household heads who are expected to work have low weekly earnings. About 60 percent of their households nevertheless escape poverty.

Among the remaining poor households with an able-bodied head, most have substantial labor market attachment. About half of all poor able-bodied mothers whose youngest child is over age 6 work at some point during the year, as compared with about 80 percent of men who head poor households with children.

Despite this work effort, poor households remain in poverty because of low annual earnings, which reflect both low weekly earnings and less than full-year work. And most of these households would remain poor even if their heads worked a full year at their current weekly earnings rate.

ACKNOWLEDGMENT: The authors acknowledge the contribution of Christine Ross and George Slotsve who provided valuable research assistance, and Elizabeth Evanson, editorial assistance.

1 A household consists of a family or an unrelated individual. This differs from the Census Bureau's definition of a household which "consists of all persons who occupy a housing unit.' Characteristics of Households and Persons Receiving Selected Noncash Benefits: 1984, Series P-60, No. 150, p. 109. For example, if an unrelated individual resides in the same housing unit as a family of four, we would have two households and the Census would have one. Our definition is consistent with the assumption that the family and the unrelated person do not pool their incomes; the Census definition is consistent with income-pooling.

2 While child care responsibilities may complicate labor market opportunities for single-parent households with a child over 6, we nevertheless classify such persons as expected to work because this is consistent with existing welfare policies.

3 Throughout this paper, we use the official measure of poverty as defined by the Census Bureau. This measure is based on cash income and does not account for the receipt of in-kind benefits, such as medicare, medicaid, and food stamps. Inclusion of benefits would lower the extent of poverty in any

year, but would not alter the trends in work effort and the incidence of low earnings discussed here.

Data for valuing in-kind benefits are available only for the years since 1979. All the data presented are based on computations by the authors from the computer tapes of the March 1968, 1972, 1980, and 1985 Current Population Surveys, conducted by the Bureau of the Census.

4 In 1984, the poverty level for a family of four was $10,609. We define any household head with weekly earnings below $204 as a low earner, regardless of his or her own household size. The poverty line for every family size is fixed in real terms and varies over time only because of changes in the Consumer Price Index. The same is true for our low-earnings threshold.

5 For example, a head of a household of four persons who earns $250 per week would not be counted as a low earner even if she or he worked only 10 weeks in the last year. If this were the household's only income for that year, the household would be poor. However, the householder would not be classified as a low earner because her or his household would escape poverty through full-year work. Also, consider the head of a two-person household who earns $150 per week for 50 weeks, or $7,500 per year. We classify this head as a low earner, but the household is not poor because the poverty line for a two-person household is $6,762.

6 Note that if a head did not work at all during the year, we consider him or her as a lower earner, along with those whose weekly earnings fell below our threshold.

7 When we recompute the low earnings cutoff on the basis of a poverty line for a family of three, the importance of family size obviously decreases. Nonetheless, a family size of less than three persons is still the largest single source of escape for households whose heads have low weekly earnings.

8 We have shown elsewhere that the increased transfers can account for, at most, small declines in work effort over this period. See Sheldon Danziger and Peter Gottschalk, "The Poverty of Losing Ground,' Challenge, May-June 1985, pp. 32-38.


Table: 1. Percent distribution of all households and poor households by selected characteristics of household head, selected years, 1967-84

Table: 2. Percent distribution of all low-earning households escaping poverty by source of escape, selected years, 1967-84

Table: 3. Selected characteristics of poor households headed by those expected to work, selected years, 1967-84

Table: 4. Federal direct tax bill for a family of four with poverty-line earnings, selected years, 1965-84
COPYRIGHT 1986 U.S. Bureau of Labor Statistics
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1986 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:employment of the poor
Author:Danziger, Sheldon; Gottschalk, Peter
Publication:Monthly Labor Review
Date:Sep 1, 1986
Previous Article:Computer manufacturing enters a new era of growth.
Next Article:The declining middle class: a further analysis.

Related Articles
The minimum wage: its relation to incomes and poverty.
Reasons for not working: poor and nonpoor householders.
Raising the minimum wage: effects on family poverty.
Working and poor in 1990.
Understanding the plight of the poor.
What does it mean to be poor in America?
Labor markets and public assistance programs.
Why don't we care about the poor anymore?
The working poor in 2001: about 6.8 million workers in 3.7 million families lived below the poverty level in 2001, an increase for the first time...

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters