Chapter Three


When we reflect on the poor in America, we are apt to think of people who are unemployed or on welfare. Families such as the Lamberts and the Bartelles seldom come to mind. In part, this is because both are hardworking families with at least one full-time worker and often with two. In part, it is because the government does not define either family as poor. The incomes of both the Lamberts and the Bartelles are about 50 percent above the government's official poverty line. And it is also because neither family receives any public assistance. The Lamberts and the Bartelles appear, in all these respects, to be beyond poverty and able to make their own way.

The American ethos holds that hard work will enable people to provide adequately for themselves. If we are to learn whether this ethos reflects reality in America, we must determine whether families such as the Lamberts and the Bartelles have indeed attained a minimally adequate standard of living. To what degree are these families economically self-sufficient? This would seem to be a simple question to answer, yet the idea of self-sufficiency means different things to different people. Most would agree that persons unable to afford basic necessities on their own incomes are not self-sufficient. The disagreements arise when we go one step further and ask what is meant by "basic necessities." Which necessities are basic, and how much income do families need to afford them?

This question of definition may seem technical, but it is crucial. The answers to it will determine how many hard working Americans are unable to obtain basic necessities and what kinds of experiences and backgrounds characterize such people. In turn, these answers will establish not only the size but also the nature and complexity of the problem of working poverty that the nation faces.

The government has attempted to measure poverty now for about thirty years with what has come to be called the official poverty line. Formulated at the birth of the War on Poverty, the measure starts with the idea that poverty signifies the inability of families to afford the basic necessities. The government chose an approach developed by Mollie Orshansky, then a little-noticed research analyst who was working for the Social Security Administration. Orshansky's own concern about the poor had deep personal roots, for she had come from a working-poor family. Although her father was a skilled tinsmith and repairman who nearly always had a job, she nevertheless recalls childhood days spent waiting in long lines to receive surplus food. aIf I write about the poor," she said, "I don't need a good imagination-I have a good memory.''1 Orshansky had worked at the Department of Agriculture early in her career. She turned to studies of food budgets she knew of there that estimated the least amount of money required to purchase a diet that would meet minimal nutritional requirements. The studies adjusted this minimum cost for food to take into consideration both family size and the ages of family members, since children and even adults of different ages have somewhat different nutritional needs. Armed with the knowledge that the cost of food in the 1960S made up one-third of the bud get for an average family, Orshansky set the poverty line for a household of given composition at three times the cost of the minimal food budget for that size household. The government adopted this formula as its measure of the official poverty line and thereafter updated it every year to take inflation into account. ln 1970, the official poverty line for a family of four (two adults and two children) was $3,968; in 1980, it was $8,414; and in 1990, it stood at $13,360.

The word "poverty" has a broad reach, of course. To some, it means destitution, that is, living at a level of sheer physical survival, as people do today in many parts of the Third World. But ordinarily it means something more.2 As originally conceived, the government's official poverty line signaled a standard higher than bare physical survival. Recall that in constructing the measure, Orshansky used the pro portion of the family budget (one-third) that the average American family spent on food rather than the proportion that poor families spent on food.3 The frame of reference originally was the budget and spending patterns not of the poorest families but of average Americans. That practice conveyed a definition of poverty that was tied to acceptable living standards.4 People were impoverished, even if they were not starving or destitute, when they had incomes beneath the amount required to purchase basic necessities at the most frugal standard of an ordinary family. This approximates Orshansky's definitions and stays well within the confines of conventional meanings of the term.6

From the beginning, however, the government did not emphasize this concept of poverty, and it soon got lost. Orshansky had feared precisely this result. She believed that the government's formal adoption of the measure would "tend to freeze the poverty line despite changes in buying habits and changes in acceptable living standards" that would take place in subsequent years.7 The problem is that in updating for inflation, the government in effect ignored the concept of poverty that underlay the original measure. Specifically, the way the government calculated poverty meant that the amount the average family spent on food continued to be about one-third of the family budget, almost as if it had been written in stone.8 Over the years, though, changes in the consumption patterns of the average American family and the relatively rapid rise in the costs of other necessities, such as housing and medical care, reduced the proportion of the budget that the typical family spent on food and increased the proportion it spent on other needs. By the early 1980S, for example, food had fallen to about one-fifth of the cost of the average family budget in America, although it remained at about one-third according to the poverty measure. With the passage of time, then, the official poverty line became ever more separated from its original mooring, which, through the budget of the average family, had tied it to minimally acceptable living standards. This, in turn, had decisive implications for official reports about the total number of Americans who were impoverished (reducing the number significantly) and the kinds of Americans living in poverty (far fewer working people were in poverty, for example), which powerfully affected the government's con clusions and the nation's thinking about who the poor were. This part of the story will begin to unfold in chapter 4.

The effect on the poverty line itself was dramatic. Had the official poverty line first come into being in 1980, and had it retained the approach to measuring poverty that was originally used (with the cost of nutrition in 1980 taking about one-fifth of the budget of the average American fam ily),9 the income necessary for a family of four to attain the poverty line would have been around $13,300. This level of Income is more than 50 percent higher than the $8,414 that the Census Bureau reported as the official poverty line for 1980. Calculated for 1990, when nutrition had fallen to about one-sixth of the average family budget, the income needed for a family of four ($22,300) would have been 67 percent greater than the Census Bureau's official poverty line, which was $13,360. According to this revised calculation, which employs the same concept of poverty used at the start, the incomes of both the Lamberts and the Bartelles would be beneath the official poverty line instead of well above it.10 What is more, having lost its connection with the original concept of poverty, and no alternative concept having been substituted, the government's official poverty line now had no solid tie to any concept of poverty.

The government should have had some inkling of the problem. One often hears that the federal government's left hand doesn't know what the right hand is doing. This applies, in some ways, to its measures of family budgets. For years, the government had two separate departments engaging in the development of these measures. The Census Bureau, in conjunction with the Office of Management and Budget, calculated the poverty line. Simultaneously, the Department of Labor compiled a series of other measures for family bud gets that it called a "higher family budget," an "intermediate family budget," and a "lower family budget." The Department of Labor designed these budgets to detail consumption of all necessities, not just food, and to describe what it cost a family of four to live. The department compiled its family budgets for the last time in 1981. For that year, it found that the minimum amount a family of four needed to live on the lower budget was $15,323.11 0ne might reasonably ask what could exist beneath the minimum amount required for a lower budget if not poverty, or some thing akin to it. Yet the amount was about 65 percent above the government's official poverty line. Something was wrong.l2

Unbeknownst to the government, the Department of Labor's lower budget actually lent support to the poverty line as it had originally been measured. Indeed, the two were similar in concept, although few observers were aware of it. Both attempted to measure, in different ways, the minimum amount of income an ordinary family required to buy basic necessities. The two arrived at roughly similar answers, as well. Recall that the minimum income needed to attain the Department of Labor's last lower budget, in 1981, was $15,323. This was within 10 percent of the $14,374 figure for the official poverty line that would have resulted in 1981(13) had the government adhered to the concept and method that had been the bases of the original measure.14

In an assessment of what it takes for families to afford basic necessities, no less important than the ideas and calculations of governmental officials are the views and daily experiences of the public. Numerous surveys of Americans concerning these matters have been carried out over the years. Surveys, of course, must be used with caution, because the answers respondents give may be shaped by how they interpret the question. The individual's own income and life-style might well influence what he or she considers to be basic necessities or a minimally decent level of living. People with higher incomes are likely to believe that it takes a relatively high income to afford necessities and to live minimally decently. The possibility of this kind of problem is suggested by William Bennett's comment when he declined the offer to become national chairman of the Republican party at a salary of $125,000 a year: "I didn't agree to take a vow of poverty," he is reported to have said.15 Consider, too, Robin Leach, star of television's "Lifestyles of the Rich and Famous," who was asked whether he considered himself to be a wealthy person. Responding that he did not, Leach said, "I have no servants. I do have a cleaning lady, like everyone else. I have no limo. In cities I usually hail a taxi like every one else."16 One person's luxuries may be another's necessities.

To avoid this problem, we might try a different strategy. We could ask respondents directly whether they have themselves experienced difficulty affording items generally considered to be basic necessities, such as food, clothing, and medical care. We can then connect the answers to the respondents' actual incomes and determine the level of income beneath which families report that they find it difficult to afford necessities. Unfortunately, this approach does not escape the problem. Although less obviously so, it still relies on the judgment of the respondents about what constitutes necessities. If respondents like fancy meals, they may eat out frequently. Respondents with these tastes who live on aver age incomes could easily answer that their income is insufficient for them to afford food. On the other hand, people with low incomes who eat frugally may believe that their incomes are sufficient to afford food without being aware that they may not be getting adequate nutrition.

Despite these shortcomings, survey results do tell us something important about the feelings of the public. Virtually every year going back to the 1950s, Gallup surveys have asked Americans what they consider the minimum amount of income a family of four needs to live in their communities. These polls reveal that the median response of the Americans surveyed never dropped beneath 140 percent of the official poverty line during the 1980s and reached a zenith of somewhat more than 160 percent of the official poverty line during the second half of the 1980s.l7 These responses approximate the income of the Department of Labor's lower budget and what the Census Bureau's official poverty line would have been if first constructed in the 1980s.

The same is true of the surveys that ask Americans about their ability to afford, food, clothing, and medical care throughout the year. In early 1984 and again in early 1987, the Gallup organization asked Americans, "Have there been times during the last year when you did not have enough money to buy food your family needed?" The surveys asked this same question about clothing and medical care. Comparing the respondents' answers with their yearly incomes and the size of their families reveals the percentage of the official poverty line at which a majority of respondents report that there were times during the year when they did not have enough money to afford one or more of these necessities. In both surveys, a majority of the families with incomes up through 150 percent of the official poverty line reported that they could not afford at least one of these basic items at some time during the year.19 Only when family incomes rose above 150 percent of the official poverty line did a majority of respondents say that they were able to afford all three of these necessities throughout the year (see table 2).

Table 2
YEAR, 1984 AND 1987(a)

Family Income
(in percentage of the
official poverty line)
Whole Population30.826.8
(b) It may be better to view the measures, taken very early in the year, as measures for 1983-84 and 1986-87. The table thus uses the average of the poverty lines for each of the respective two years in calculating respondents' incomes as a percentage of the poverty line.

This result parallels the experiences of the Lamberts and Bartelles, each of whom generally had an income hovering around 15O percent of the offficial poverty line.20 Both families faced times when they could not afford necessities. Even after receiving financial assistance from others, the Bartelles sometimes had no money for food-they ate potatoes or potatoes and eggs for several days running-and the Lamberts could not afford school clothes for their children or dental care for themselves or their children.

We pointed out earlier that respondents' own incomes and life-styles might affect their views on the minimum income necessary to live, and this turns out to be so. On the average, higher-income respondents say that somewhat higher minimum incomes are required. However, even respondents whose family income lies at the lowest level-beneath the official poverty line-state that the minimum income necessary for a family of four to live in their communities stands at nearly 140 percent of the official poverty line.2l

Not even residence in an urban or rural locale appears to change respondents' answers very much. In densely populated areas, something of a trade-off exists between housing and medical costs, which increase, and transportation costs, which often decrease because reliable mass transportation is more likely to become available. The evidence now available does not show, although it is often presumed otherwise, that urban-rural differences are strong or consistent enough to warrant making distinctions based on them in determining the income that the public feels is required for basic necessities. For example, the Gallup surveys of 1987 that asked Americans whether there were times during the year that they did not have the money to buy food, clothing, and medical care discovered little difference, on the average, in the responses of urban, suburban, and rural households. In all three locations, a majority of the households with incomes of up to 150 percent of the official poverty line reported that they ran short of money for one or more of these necessities during the year.22

The public's responses to questions about the minimum income necessary to live and the income required to afford food, clothing, and medical care throughout the year suggest that an income at or perhaps somewhat above 15O per cent of the present official poverty line is required to achieve self-sufficiency. So does the Department of Labor's lower family budget. Even the concept of poverty that launched the official poverty line itself, when applied to the 1980s, leads to this same conclusion. The results of these various measures for the 1980s are found in table 3.

Table 3
Official Poverty Line
Recalculated for the
1980s by Means of the
Orginal Measure(a)
Department of
Labor's Lower Budget(b)
Americans' Views on
the Minimum Income
Necessary to Live in
Their Communities(ce)
Income Level at Which
a Majority of
Americans Respond
That They Sometimes
Cannot Afford Food,
Clothing, or Medical
(a)See the discussion on pages 33-35, above, for a description of the concept and measure that originally underlay the official poverty line. The expenditure weightings for food used to calculate the figures in the table are 20.4 percent for 1982,18.8 percent for 1985, and 17.7 percent for 1987 and thereafter.
(b)Source: See note 11 of this chapter.
(c) Source See note 17 of this chapter.
(d)Source: See table 2.
(e)These measures span two years; the measure for 1981, for example, combines 1980-81, using the average of the poverty lines for the two years. We do this because the surveys are taken in the early months of the year for which the measure is reported.

Unhappily, the measures do not themselves provide a consistent alternative measure of self-sufficiency. For 1981-82, the measures register at 150 percent, 160 percent, 164 percent, and 166 percent of the official poverty line. In 1987-90, the available measures register at 150 percent, 160 per cent, and 167 to 172 percent of the official poverty line. An examination of these results and those reported in the table for the other years indicates that the four measures usually register above 150 percent and rarely surpass 170 percent of the official poverty line.

To determine a threshold of self-sufficiency, we need to step beyond these measures and create an actual family budget designed to enable people to get all of the basic necessities at the lowest realistic cost (the appendix to this chapter contains the definition of basic necessities that is used here) The family budget listed below-which we call the economy budget-describes the income an American family of two adults and two children would normally need in 1990 (see table 4). It shows graphically why an income substantially above the official poverty line is necessary for families to be able to afford the basic necessities even when purchased at the lowest realistic cost.

The economy budget is stringent. For example, the expenses for food are only those allowed in the federal government's thrifty food budget, the same one that the government calculates to be the lowest with which a family can satisfy minimal nutritional needs. Some evidence indicates that the thrifty food budget may not cover the actual cost of food in low-income areas,23 but we use it nevertheless. The housing portion of the economy budget is based on the level of rent for a two-bedroom unit defined by the federal government as a low-cost unit.24 Table 4 and the footnotes to it summarize and explain the major items in the family economy budget.

Table 4
Food(The lowest cost established by the federal government to provide minimally adequate nutrition for a family of four: $331 per month.)l$ 3,972
Housing(The rent level for a two-bedroom unit, including all utilities, defined by the federal government as a low-cost unit for its programs to assist low-income families, plus a telephone. The budget could comprise a monthly rent of $375, utility bills of $123, and a telephone bill of $18)(2) $ 6,192
Transportation(Fifteen percent beneath the average annual cost for one automobile over 10 years and 100,000 miles, including depreciation, financing, fuel, tires, repairs, insurance, and taxes. The average American family of four has 2.1 cars rather than one. Nearly 90 percent of all households have at least one car.)3$ 3,251
Medical(The cost of medical care on average, for an American family of four)4$ 1,568
Clothes($250 per family member over the year)$ 1,000
Personal(5)$ 400
Incidentals(6) $ 1,300
Tax(Federal income, social security, state income)7$ 2,975
Total: $20,658
1. In 1990, the cost of food for a family of four (two adults and two children) on the federal government's thrifty food plan is based on data provided by the Human Nutrition Information Service of the U.S. Department of Agriculture. For earlier figures, see Patricia Ruggles, Drawing the Line (Washington, D.C.: Urban Institute Press, 1990), p. 180.
2. The Department of Housing and Urban Development uses as its low-cost standard the 45th percentile of all two-bedroom rentals except new construction and public housing units. The figure for 1990 was $498 per month (for earlier figures, see Ruggles, Drawing the Line, p. 178), to which we added the cost of a telephone. Use of the 45th percentile of units seems appropriate both because some of the lowest-cost units are in rent-controlled apartments with long waiting lists (often they are rented by families with above-average incomes) and many other lowest-cost units are in dangerous neighborhoods that have high rates of violent crimes, areas that families may wish to avoid. See Christopher Jencks and Kathryn Edin, "The Real Welfare Problem," American Prospect (Spring 1990): 35. The housing cost that we use in the "economy budget," the 45th percentile, is an amount that may actually be spent for housing by many families with incomes at the official poverty line, equivalent to an income of $13,360 for a family of four in 1990, instead of the $20,658 of this economy budget (see note 32 below).
3. Bureau of the Census, Statistical Abstract of the United States, 1990, p. 609, table 1044. The figure is 85 percent of the reported cost for 1988, adjusted to account for the rise in consumer prices in transportation in the years 1988-90. The average four-person household has 2.1 cars (ibid., p.602, table 1028), and nearly 90 percent of all households have at least one car. See Robert Rector et al., "How Poor Are America's Poor," Backgrounder, Heritage Foundation, September 21, 1990, p. 8, table 4. For transportation price rises in the years 1988-90, see Council of Economic Advisers, Economic Indicators (Washington,D C: Government Printing Office, April 1991), p. 24.
4 Bureau of the Census Statistical Abstract of the United States, 1990, p. 99, table 151. The figure cited is the average medical care expense in 1987 of a four-person household, adjusted to account for inflation in medical care from 1987 to 1990. For medical care consumer price rises from 1987-90, see Council of Economic Advisers, Economic Indicators (April 1991), p. 24.
5.Personal includes soap, shampoo, toothpaste, toothbrushes, shaving cream and blades, feminine hygiene products, house cleansers, sponges, dish and wash towels, and other such items for a family of four for a year.
6 Incidentals include toys and presents ($200 per year to cover the four members of the family); repairs on household appliances, lamps, and furniture ($150 per year); purchase of used furniture, mattresses, bedding, towels, sheets ($300 per year); newspapers ($150 per year); school supplies ($100 per year for the two children); and such items as postage, dishes, utensils, glasses, napkins, stationery, and light bulbs.
7.Taxes include $1,075 for federal income tax, $1,510 for Social Security, and $390 for state income tax.

To purchase the necessities found in this economy budget, a family of four in 1990 needed an income of about $20,700, which amounts to approximately 155 percent of the 1990 official poverty line of $13,360 for a family of four. This standard broadly coincides with the four governmental and public survey measures of self-sufficiency we reviewed, falling toward the lower end of their range.

Other considerations, too, suggest that this standard of self-sufficiency is by no means excessive. Because the economy budget is based on the expenses necessary to meet basic needs at the lowest realistic cost, its stringency is revealed as much by what it excludes as by what it includes.

What is it like to live on the economy budget? Members of families existing on the economy budget never go out to eat, for it is not included in the food budget; they never go out to a movie, concert, or ball game or indeed to any public or private establishment that charges admission, for there is no entertainment budget; they have no cable television, for the same reason; they never purchase alcohol or cigarettes; never take a vacation or holiday that involves any motel or hotel or, again, any meals out; never hire a baby-sitter or have any other paid child care; never give an allowance or other spending money to the children; never purchase any lessons or home-learning tools for the children; never buy books or records for the adults or children, or any toys, except in the small amounts available for birthday or Christmas presents ($50 per person over the year); never pay for a haircut; never buy a magazine; have no money for the feeding or veterinary care of any pets; and, never spend any money for preschool for the children, or educational trips for them away from home, or any summer camp or other activity with a fee.

Nor is this all. The budget allots no money either for any absence from work because of illness or to pay for emergencies or other unanticipated major expenses; no money to help other persons, such as an ill or elderly parent, or to buy life insurance or to pay interest on consumer-credit borrowing except for car financing; and no money for savings for college for the children or to support a pension for retirement other than Social Security.

This is not to say that a family living on the economy budget will never purchase any of these items. It is to say that in order to do so, such a family has to follow one of three alternatives. It has to borrow money, figure out a way to pay less for the necessities included in the economy bud get, or find outside assistance. A family can borrow money, but there is no place in the budget for the repayment of debt (except for a car), so the repayment of loans requires the elimination of something else in the budget. The family might then turn to the second alternative: it can try to find a way to spend less for the necessities. Sometimes housing can be found at a lower price, although such housing may be in extremely poor condition.25 As often the housing options available actually cost more than has been provided for in the budget.26 Families may then decide to spend less on medical attention, a choice that many of them end up making.2 They may also try to spend less on food, sometimes at the risk of sending their children to school hungry.28 Or they can decide to go without automobile insurance. About two-thirds of the Americans who had no automobile insurance in 1987 lived in families with incomes beneath $I5,000.29 They may also decide to forgo a telephone. Nearly one in seven households on an income between 125 and 150 percent of the official poverty line lacks a telephone, a rate four times greater than that of households living on the median household income.30 Making these choices does free up money for purchases not contained in the economy budget. On the other hand, to forgo proper medical attention, to expose oneself to the dangers of driving without automobile insurance, and to have no telephone in a society dependent upon it is to live without necessities. One final alternative is to accept gifts or assistance from family, friends, church or synagogue, or some other source. This, the experiences of the Lamberts and Bartelles reveal, may be the only realistic choice.

Paul and Jane Lambert, heading a family of five, earned from about 150 percent to 160 percent of the poverty line, at the upper limit of the economy budget, for most of the period from 1984 to 1990. They did have some expenses not covered in the economy budget. Both Paul and Jane smoke. They also have two dogs. Although they live in a mobile home, their yearly housing costs are about $500 above the economy budget's expenses.3l They also have two automobiles instead of one. But their car expenses, at $3,000, are about equal to those allowed in the economy budget (both cars are more than fifteen years old, both have gone more than 300,000 miles, and one of them frequently is out of repair). The closest public transportation to the Lamberts is more than a mile away and not terribly reliable. No other expense that they have goes beyond what is allowed in the economy budget. They do not drink, eat out, or go out to movies. They have bought no new clothes for themselves in many years. They have not been on a vacation away from home with their children in the past decade. They feel they cannot even afford the gas to go for a Sunday drive, and never do so. They have neither life insurance nor a pension for retirement. Juggling bills from week to week, they often make ends meet by getting assistance from others. They regularly depend upon outside assistance to pay for their food and shelter and to get clothes for their children. Even then, they cannot pay fully for necessities such as car repairs and medical or dental care.

The Bartelles are little different. Their income, which averages nearly 150 percent of the official poverty line, also hovers near the upper end of the economy budget. Like the Lamberts, the Bartelles have few expenses that go beyond the economy budget (they do have two dogs, and they spend $150 each summer to send both their children to day camp for a week). They, too, have been unable to afford to eat out, go to a movie, buy new clothes for themselves or their children, go on a vacation, buy life insurance, or do many of the other things that most American families take for granted. To make ends meet, the Bartelles get help from others to buy clothes both for themselves and for their children. Even then, money is sometimes so tight that they eat meagerly for several days running in order to have enough left to pay overdue mortgage, electricity, or gas bills. They do not have any money to spare either to replace worn furniture that has become bitterly embarrassing to them or to prepare to fix their roof.

In fact, the vast majority of American families living on incomes at the upper end of the economy budget share these experiences. Consider the costs for shelter. Housing expenses are so high that most Americans who live at the top end of the economy budget-155 percent of the official poverty line-cannot pay their present housing costs without scrimping on food, clothing, health care, or other necessities.32 The Economic Policy Institute's detailed analysis of the American Housing Surveys reveals that about 90 percent of the families of four living at this income level in 1987 (equivalent then to about $18,000 for a family of four) did not, after spending only the minimal amounts alloted for food, medical, clothing, and other items on the economy