Quick viewing(Text Mode)

The Development and History of the Poverty Thresholds

The Development and History of the Poverty Thresholds

The thresholds are the primary version of the Federal poverty measure- rhe other version being the poverty guidelines. The thresholds are currently issued by the Bureau of the Census and are generally used for statistical purposes-for example, estimating the number of persons in poverty and tabulating them by type of residence, race, and other , economic, and demographic characteristics. The poverty guidelines,4 on the other hand, are issued by the Department of and and are used for administrative purposes-for instance, for determining whether a person or family is financially eligible for assistance or services under certain Federal programs.

The Development of the The Development and Poverty Thresholds of the Poverty Thresholds The poverty thresholds were developed by Gordon M. Fisher* in 1963-64 by Mollie Orshansky, an economist working for the Social Security In recent years there has been renewed interest in the United States in the Administration. As Orshansky later definition and measurement of poverty. In early 1992, the Committee on indicated, her original purpose was not to National Statistics of the National Academy of Sciences began a 30-month introduce a new general measure of study requested by Congress that includes an examination of statistical issues poverty,s but to develop a measure to involved in measuring and understanding poverty. Some 2 years earlier, in assess the relative risks of low economic January 1990, the Administration had approved an initiative on improving the status (or, more broadly, the differentials quality of . The current poverty measure was one of in opportunity) among different demographic groups of families with several dozen statistical series examined as part of that initiative. In April children.6 She actually developed two sets 1990, Urban Institute economist Patricia Ruggles published a book’ that of poverty thresholds+ne derived from urged the revision of the poverty line to reflect changes in consumption the Department’s patterns and changing concepts of what constitutes a minimally adequate plan and one derived from its . In July 1990, two private organizations concerned with the somewhat less stringent low-cost food poor and the elderly issued a report* reviewing current poverty measurement plan. She described an initial version of procedures and describing a Gallup poll in which a nationally representative these thresholds-for families with sample of Americans set an average dollar figure for the poverty line that was children only-in the July 1963 article higher than the current official poverty line. In view of these and other cited in footnote 6. She published an examples,3 it may be useful to reexamine the development and subsequent analysis using a refined version of the history of the current official poverty thresholds. thresholds-expanded to include thresholds for unrelated individuals and families without children-in a January 1965 article.’ *Office of the Secretary, Department of Health and Human Services. Because The Johnson Administration of the great interest in poverty and its measurement, the Bulletin asked Mr. Fisher announced its War on Poverty in January to write an article on the origin of the poverty thresholds. For related 1964, not long after the publication of information see, “Poverty Guidelines for 1992,” by Gordon M. Fisher, Social Orshansky’s initial poverty article. The Security Bulletin, Vol. 55, No. 1, Spring 1992, pp. 43-46. 1964 Report of the Council of Economic Advisers (CEA) contained a chapter on poverty in America.* The chapter set a poverty line of $3,000 (in 1962 dollars) for families of all sizes; for unrelated

Social Security Bulletin l Vol. 55, No. 4 l Winter 1992 3 individuals, the chapter implicitly set a (“economy level”) of Orshansky’s two of Agriculture began to issue an poverty line of $1,500 (a selection that sets of poverty thresholds as a working or “economy” food plan, costing only 75 was shortly made explicit). The $3,000 quasi-official definition of poverty in 80 percent as much as the basic low- figure was specified as being on the basis May 1965. As noted below, the cost plan, for “temporary or use when fUndsare low.“. .The food of before- annual money . thresholds were designated as the plan as such includes no additional There was a brief discussion of the Federal Government’s official statistical allowance for meals eaten out or other theoretical desirability of using estimates definition of poverty in August 1969. food eaten awayfrom home.16 of “total” -including nonmoney Orshansky did not develop the poverty elements such as the rental of thresholds as a standard budget-that is, To be more precise, what was used in owner-occupied dwellings and food a list of and services that a family developing the ihresholds was the dollar raised and consumed on farms-but it of a specified size and composition costs of the in the two food plans. was not possible to obtain such estimates. would need to live at a designated level Although the actual foods in both plans The CEA chapter pointed out that the of well-being, together with their provided a fully nutritious diet, families total of money plus nonmoney income estimated monthly or annual costs.” If spending for food at the dollar cost level that would correspond to the cash- generally accepted standards of of the economy food plan had about 1 income-only poverty line of $3,000 minimum need had been available for all chance in 2 of getting a fair or better would be somewhat higher than $3,000.9 or most of the major essential diet, but only 1 chance in 10 of getting a The CEA chapter referred to consumption items of living (for diet. ” Orshansky’s July 1963 article and its example, , medical care, The three steps Orshansky followed in $3,165 “economy-plan” poverty line for , and transportation), the moving from the cost of food for a family a nonfarm family of four. “Other studies standard budget approach could have to minimum costs for all family have used different baskets, many been used by costing out the standards requirements were (1) to define the of them costing more. On balance, they and adding up the costs. However, family size and composition prototypes provide support for using as a boundary, except for the area of food, no definitive for which food costs would be computed, a family whose annual money income and accepted standards of minimum (2) to decide on the amount of additional from all sources was $3,000. . . .“I0 This need for major consumption items income to allow for items other than passage has led some people to think that existed either then or today. food, and (3) to relate the cash needs of the CEA’s $3,000 poverty line was The “generally accepted” standards of farm families to those of comparable derived to a greater or lesser degree from adequacy for food that Orshansky used in nonfarm families. Orshansky’s $3,165 poverty line. developing the thresholds were the food Because of a special interest in the However, Robert Lampman (a member plans prepared by the Department of economic status of families with children of the CEA staff) had been working on Agriculture. At the time there were four and because income requirements are an analysis of poverty using the $3,000 of these food plans, at the following cost related to the number of persons in the figure as early as the spring of 1963l’-- levels: liberal, moderate, low-cost, and family, Orshansky estimated food costs several months before Orshansky’s economy. The first three plans had been separately for nonfarm families varying initial article was published. Instead, the introduced in 1933, and the economy in size from two members to seven or $3,000 figure was a consensus choice food plan was developed and introduced more. Families were further classified by based on consideration of such factors as in 196 1. Data underlying the latter plan sex of head and the number of members the minimum level, the income came from the Agriculture Department’s who were related children under age 18. levels at which families began to have to 1955 Food Consumption Among three-person families, for pay Federal income , and public Survey. I4 In developing her two sets of instance, there were separate assistance payment levels.‘* poverty thresholds, Orshansky used the subcategories with the following Orshansky was concerned by the CEA low-cost and economy food plans:15 compositions: three adults; two adults, report’s failure to adjust its poverty line The low-costplan, adapted to the food one child; and one adult, two children. for family size, which resulted in patterns of families in the lowest third Two-person families were further understating the number of children in of the income range, has for many classified by age of head as those under poverty relative to aged persons. This yearsbeen used by agencies age 65 and those aged 65 or older.‘* The prompted her to begin the that as a basis for food allotments for lower poverty thresholds that resulted for needy families and others who wished resulted in her January 1965 Social aged units of the smallest size were Security Bulletin article, extending the to keep food costsdown. Often, however, the actual food allowance for simply a mechanical consequence of two sets of poverty thresholds-at the families receiving public assistance having separate subcategories for two- “economy level” and at the “low-cost wasless than that in the low-cost person families with aged and nonaged level”-to the whole . This plan. spending as much as this heads, and of the fact that food plan costs article appeared just as the Office of food plan recommends by no calculated for the aged families were Economic Opportunity (OEO) was being means guarantees that diets will be lower than those for the nonaged established. The OEO adopted the lower adequate. .Recently the Department families. Orshansky did not claim that

4 Social Security Bulletin l Vol. 55, No. 4 l Winter 1992 necessary nonfood expenditures for the conducted at approximately lo-year where they equaled the cost of the aged are or should be lower than those intervals. The 1955 survey-the most economy food plan (or the low-cost food for the nonaged. As noted below, recent one then available-had found plan) for a family of that size, the family poverty thresholds for unrelated that for families of three or more would have reached the point at which individuals (one-person units) were not persons, the average expenditure for all its food expenditures were minimal but derived from food plan costs; instead, food used both inside and outside of the adequate, assuming that “the housewife they were calculated directly from the home during a week accounted for about will be a careful shopper, a skillful cook, thresholds for two-person families. one-third of their average money income and a good manager who will prepare all Calculating food plan costs for the after taxes. (Note that this finding the family’s meals at home.“24 nontarm family subcategories established relates to families at all income levels, Orshansky made the assumption that, at by Orshansky was a complicated process. not just those at lower income levels; one that point, the family’s nonfood The food plans included separate food of the most common errors made about expenditures would also be minimal but cost figures for 19 different age-sex the thresholds is to claim they are based adequate, and established that level of categories of persons.19 However, to a on a finding that “poor people spend a total expenditures as the poverty great extent data were not available to third of their income on food.“) threshold for a family of that size. Since show the distribution of persons by age Besides considering the Agriculture the family’s food expenditures would and sex within each family subcategory. Department’s 1955 Household Food still be one-third of its total expenditures, She used data distributions from the Consumption Survey, Orshansky also this meant that (for families of three or 1960 Decennial Census and made reviewed the 1960-6 1 Consumer more persons) the for a additional assumptions about Expenditure Survey of the Bureau of family of a particular size and characteristics of family members not Labor Statistics (BLS), which also composition was set at three times the shown in the census data distributions. provided an estimate of the proportion of cost of the economy food plan (or the For each subcategory of families with total after-tax income going for food. To low-cost food plan) for such a family. children, a combination of ages of use the BLS survey to derive a poverty The factor of three by which the food children was chosen that yielded a food measure would have resulted in a plan cost was multiplied became known cost that was higher than the food costs “multiplier” of just over four, rather than as the “multiplier.” of two-thirds of the (simulated) families three. However, the questions used by It is important to note that in that subcategory. “Because food BLS to get data on annual food outlays Orshansky’s “multiplier” methodology requirements for children increase had usually yielded lower average for deriving the thresholds was rapidly with advancing age and the food expenditures than the more detailed normative, not empirical-that is, it was plan cost is already critically low, this item-by-item checklist of foods based on a normative assumption protection was deemed necessary to consumed in a week used in the involving (1955) consumption patterns ensure adequate allowance for growing Agriculture Department survey.** of the population as a whole, and not on youngsters.“” Food plan costs for the 58 Orshansky finally decided to use the the empirical consumption behavior of nonfarm family subcategories were Agriculture Department survey, with its lower-income groups. calculated using January 1964 prices for one-to-three ratio of food expenditures to Different procedures were used to the economy and low-cost food plans. after-tax money income, in developing calculate poverty thresholds for one- To get from food plan costs to the poverty thresholds. person and two-person units in order to estimates of minimum necessary Orshansky started her food-costs-to- allow for the relatively larger fixed costs expenditures for all items, Orshansky total-expenditures procedure by that small family units face. Orshansky made use of the economic principle considering a hypothetical average noted that recent consumption curves known as Engel’s , which states that (middle-income) family, spending one- indicated that one- and two-person the proportion of income allocated to third of its income on food, which was family units, who as a group were less “necessaries,” and in particular to food, faced with a need to cut back on its homogeneous in composition, seemed to is an indicator of economic well-being.*’ expenditures.23 She made the be “out of line” with larger families with Orshansky made use of this law by assumption that the family would be able respect to spending pattems.25 For two- assuming that equivalent levels of well- to cut back its food expenditures and its person families, the 1955 Household being were reached by families (of three nonfood expenditures by the same Food Consumption Survey had found a or more persons) only when the proportion. This assumption was, of food/total-after-tax-money-income proportion of income they required to course, a simplifying assumption or first ratio of 0.27 rather than one-third; purchase an adequate diet was the same. approximation. Under this assumption, accordingly, a multiplier of l/0.27, or To determine the proportion of total one-third of the family’s expenditures 3.7, was used to derive poverty income that should be assumed to be would be for food no matter how far it thresholds for two-person families. spent for food, Orshansky used the had cut back on its total expenditures. To derive poverty thresholds for Agriculture Department’s Household When the hypothetical family cut unrelated individuals (one-person units), Food Consumption Survey, a survey back its food expenditures to the point no multiplier was used. Orshansky noted

Social Security Bulletin l Vol. 55, No. 4 l Winter 1992 5 that consumption data for unrelated corresponding nonfarm thresholds.*’ (As misunderstandings. Before 1969, the individuals were hard to interpret noted below, this figure was changed to thresholds were updated for price because of the heavy representation of 70 percent in 1965 when the Office of changes each year by the December-to- aged individuals not shown separately. Economic Opportunity adopted the December change in the per capita cost For the poverty thresholds derived from thresholds, to 85 percent in 1969, and to of the economy (or low-cost) food planz8 the low-cost food plan, thresholds for 100 percent (that is, the differential was The poverty thresholds were presented unrelated individuals were set at 72 eliminated) in 198 1.) as a measure of income inadequacy-“if percent of the corresponding thresholds It is important to note that it is not possible to state unequivocally for two-person families. For poverty at Orshansky’s farm/nonfarm distinction ‘how much is enough,’ it should be the “economy level”-that is, the was not the same as a rural/urban (or possible to assert with confidence how definition of poverty derived from the nonmetropolitan/metropolitan) much, on an average, is too little.“29 economy food plan that is still in use distinction. In April 1970, for instance, Orshansky accurately described her today-the thresholds for unrelated the Decennial Census found that of a poverty thresholds as a “relatively individuals were set at 80 percent of the total rural population of 53.9 million absolute” measure of poverty,30 corresponding thresholds for two-person persons, only 10.6 million (19.7 percent) inasmuch as they were developed from families. This procedure was based on lived on farms. The nonfarm poverty calculations that made use of the the premise that the lower the income, thresholds were applied to the rural consumption patterns (at a particular the more difficult it would be for one nonfarm population as well as to the point in time) of the U.S. population as a person to cut expenses such as housing urban population. It should also be whole. In the dichotomy between and utilities below the minimum noted that the reason for the relative and absolute definitions of expenses for a couple. Note that the farm/not&m distinction was not a poverty, one of the essential SO-percent factor was used to derive generalized “living-costs-are-cheaper- characteristics of a purely “absolute” separate thresholds (at the “economy in-farm-or-rural-areas” argument. definition is that it is derived without any level”) for male aged, male nonaged, With 62 detailed poverty thresholds reference to the consumption patterns or female aged, and female nonaged for nonfarm family units and 62 detailed income levels of the population as a unrelated individuals. However, because thresholds for farm family units, whole.31 However, although the poverty of the different weighting factors Orshansky had a total of 124 detailed thresholds were not a purely absolute involved for one-person and two-person thresholds at each of the two cost levels measure, they were also quite clearly not units, the weighted average poverty (economy and low-cost). Instead of a purely relative measure, such as the 50- threshold for a one-person unit is not doing a tabular presentation of 248 percent-of-median-income measure precisely equal to SO percent of the separate income cutoff figures, she proposed by Britain’s Peter Townsend in weighted average poverty threshold for a decided to present a smaller set of 1962 and (in the United States) by Victor two-person unit. weighted average thresholds (table 1). Fuchs in 1965.32 Having calculated poverty thresholds Note that calendar year 1963 has When Orshansky was developing the from each food plan for 58 categories of always been the base year for the poverty poverty thresholds, the Census Bureau’s nonfarm families and 4 categories of thresholds, both before and after the Current Population Survey (CPS) was nonfarm unrelated individuals, 1969 revision discussed below. Claims the only good source of nationally Orshansky had 62 detailed poverty that either 1959 or 1967 is the base representative income data. thresholds (from each food plan) for year of the series stem from Accordingly, Orshansky had to apply her nonfarm family units. According to the 1955 Household Food Consumption Survey, about 40 percent of the food Table 1.-Weighted average nonfamr poverty thresholds at the economy level and at items consumed by all farm families- the low-cost level for calendar year 1963 valued at prices paid by any families who Size of Thresholds al Thresholds at did buy them-came from their home family unit eco”omy level1 low-COSI level2 Ratio farm or garden rather than being I...... $1,539 . purchased for cash. In addition, Under age 65. . . 1,580 $1,885 1.19 Orshansky noted, farm families generally Aged 65 or older. 1,410 1,745 1.19 could count not only some of their food 2...... 1,988 Head under age 65. 2,050 2,715 1.32 but most of their housing as part of the Head aged 65 or older.. 1,850 2,460 1.33 farm operationz6 Because farm families 3...... 2,440 3,160 1.30 purchased for cash only about 60 percent 4...... 3,130 4,005 1.28 of the food they consumed, and because 5...... 3,685 4,675 1.27 of the issue of classifying farm housing 6...... 4,135 5,250 I .27 I or more. . 5,090 6,395 1.26 expenses as part of the farm business operation, Orshansky decided to set farm ’ Poverty thresholds under the current definition. poverty thresholds at 60 percent of the 2 Known until 1969 as the near-poverty or low-income thresholds.

6 Social Security Bulletin l Vol. 55, No. 4 l Winter 1992 poverty thresholds to the CPS income survey-was used to recalculate the percent higher than the existing data, even though the CPS used a before- (economy-level) thresholds. The revised thresholds. tax money income concept, while the economy food plan cost 8 percent more SSA started to implement its decision thresholds had been developed on the than the previous plan, so the thresholds to revise the poverty thresholds on the basis of the after-tax money income would have been raised in real terms by basis of the revised food plans, and by concept used in the Agriculture that percentage. The revised low-cost about mid-June had poverty population Department survey from which the food plan cost 4 percent more than the figures for calendar years 1966 and 1967 multiplier was derived. From the previous one and would have resulted in on both the old and the revised basis. beginning. Orshansky was aware of the a 4-percent rise in the near-poverty or However. during the review of a report inconsistency of applying after-tax low-income thresholds. One argument that was to use the revised numbers, thresholds to before-tax income data, but for using the revised food plans was that several agencies, notably OEO and the there was no other alternative. She they (rather than the previous food plans) Council of Economic Advisers, reasoned that the result would yield “a were being used in the family budgets repudiated the position taken by their conservative underestimate” of poveQ. that the Bureau of Labor Statistics was representatives on the advisory group, At that time, and for some years beginning to issue. and argued against publication of the thereafter, most families and individuals On April 26, 1968. SSA convened an new series figures. The issue became at the poverty level had little or no interagency meeting of technical staff intertwined with questions regarding the Federal incotne tax liability. from Federal agencies with an interest in effect of a new imputation procedure poverty. SSA personnel presented to the used by the Census Bureau to estimate The 1969 Revision group their proposal to use the revised 1967 income (conceptually unrelated, but food plans to recalculate the poverty and affecting all data tied to income) and As early as November 1965, Social low-income thresholds, and the group with questions relating to the preferred Security Administration (SSA) agreed to the proposal. method of adjusting the poverty index for policymakers and analysts began to Other subjects were also discussed at price change. A further problem was express concern about how the the meeting. One was an SSA proposal that OEO was using the poverty index to thresholds should be affected by the to adjust “the nonfood portion” of the determine eligibility in a number of its historical fact that poverty/subsistence poverty thresholds for by “the programs and a change would have lines have tended to rise in real terms as total CPI minus medical care,” while affected budgets and/or rules and the real incomes of the general presumably continuing to use the per regulations to a serious extent. population have risen.33 By late 1967, capita cost of the economy food plan to On July 16. 1968, the director of the SSA personnel were also concerned adjust the food portion of the thresholds. Office of Statistical Standards in the about a separate issue-that prices in This proposal is of interest because it Bureau of the Budget (BOB) sent a letter general (as measured by the Consumer suggests to the author the possibility that to the director of the Census Bureau Price index-CPI) had been rising more SSA assumed that (1) a family unit with directing that there be ‘no change in the rapidly than the (the per a poverty-level income should not have criteria for computing the poverty capita cost of the economy food plan) to meet medical costs out of its cash ‘thresholds’ for income year 1967.“ The that were then being used to adjust the income and (2) that such a family unit letter also said that BOB would appoint a poverty thresholds for inflation each should receive medical care, care task force “immediately” to start year. In other words, as tneasured by the under the Hill-Burton Uncompensated “[ilntensive work. .as quickly as CPI, the thresholds were actually Services Program, or care under the possible to develop concepts and tech- decreasing in real terms.‘4 In particular, relatively new programs of or nical information required to m-evaluate the per capita cost of the economy food . The interagency group the poverty thresholds for future use.“‘* plan-and thus the poverty thresholds (at favored a shift to CPI indexing of the The overruling of SSA’s decision to the economy- level)-did not change at thresholds but decided that it should not revise the poverty thresholds and the all from 1963 to 1964 and again from be implemented for processing data for appointment of a task force to reevaluate 1966 to 1967, even though overall the current year (1 967).36 them raised the question of which consumer prices did rise during those Another idea discussed at the meeting Federal agency should have primary periods.3S was to replace the poverty threshold responsibility for the thresholds. The Because of these concerns, SSA made multiplier of 3 (derived from the 1955 ultimate resolution of this question was a tentative decision early in 1968 to Household Food Consumption Survey) that while the Census Bureau was given adjust the poverty thresholds for the with a higher multiplier derived from the responsibility for publishing poverty higher general level of living by using 1965 survey.37 This idea was not part of statistics, no agency was given primary data from the 1965 rather than the 1955 the formal SSA proposal for revising responsibility for maintaining the Household Food Consumption Survey. the thresholds. Together with the definition of poverty- and doing research The recently revised version of the use of the revised food plan. this higher related to it.39 economy food plan-updated on the multiplier would have resulted in An interagency Poverty Level Review basis of consumption data from the 1965 poverty thresholds that were 25-30 Committee had been selected by late

Social Security Bulletin l Vol. 55, No. 3 l Winter 1992 7 September 1968. The Committee held l Farm poverty thresholds were for 1959-67 and the revised thresholds its first meeting on October 2, 1968. set at 85 percent rather than 70 for 1959-9 1.) The two changes above, There was also a separate technical percent of corresponding which comprise the 1969 revision in the group that supported the work of the nonfarm poverty thresholds. poverty definition, were explained in a Committee. Two assignments were Census Bureau publication issued on given to the technical group: (1) to Nonfarm poverty thresholds for the August 12, 1969. Because of this consider the matter of year-to-year base year 1963 were retained, and revision, poverty statistics from price adjustment and to recommend an the new annual-adjustment and documents dated before August 1969 index, or appropriate alternatives, and farm/nor&arm provisions were applied to should not be used, since they are not (2) to explore the subject of varying them to yield revised poverty thresholds comparable with current poverty poverty levels by geographic areas; for both earlier and later years; revised statistics. whether it was feasible to do this; if so, poverty population data for 1959 and On August 29, 1969, the Bureau of how many areas; and the appropriate subsequent years were tabulated using the Budget issued a memorandum that relationship between farm and nonfarm the revised thresholds. (Table 2 shows directed all Federal Executive Branch thresholds.40 the unrevised weighted average poverty agencies to use the revised-definition The Technical Staff presented papers thresholds for a nonfarm family of four poverty statistics and thresholds (as on the two assigned subjects to the Committee, which discussed the issues in Table 2.-Poverty thresholds for a nonfarm family of four (unrevised and question. On January 6, 1969, the revised defiiitions), 1959-91 Committee Chairman (who was from Unrevised (pre-1969) Revised (posl-1969) BOB) presented a proposal for 1969 thresholds-indexed thresholds-indexed (applicable to data for 1968) under YCiU by food plan by CPI which the poverty thresholds would be 1959 ...... $3,059 $2,973 indexed by applying the Consumer Price 1960 ...... 3,022 Index (rather than the per capita cost of 1961...... 3,054 1962 ...... 3,089 the economy food plan) to the poverty 1963 (base year)...... 3,128 3,128 thresholds at the economy level for the 1964 ...... 3,128 3,169 base year 1963. Under this proposal, 1965 ...... 3,200 3,223 which was accepted on March 7, no 1966 ...... I 3,335 3,317 change was to be made in either the 1967’...... 3,335 3,410 farm/nonfarm ratio or the food plans 1968 ...... 3,553 lY6Y ...... 3,743 used to calculate the base-year thresholds. It was also proposed and 1970 ...... 3,968 lY71...... 4,137 agreed that tabulations would be 1972 ...... 4,215 presented for two additional levels, one 1973 ...... 4,540 25 percent above and one 25 percent 1974 ...... 5,038 below the poverty level. Later, the 1975 ...... 5,500 Department of Agriculture representative 1976 ...... 5,815 1977 ...... 6,191 asserted that there was strong evidence to 1978 ...... 6,662 support a reduction in the farm/nonfarm 1979 ...... 7,412 differential. The Committee agreed to 1980 ...... 8,414 reduce that differential by raising the lY81...... 9,287 farm/nonfarm ratio from 70 percent to 1982 ...... 9,862 85 percent. 1983 ...... 10,178 Accordingly, the final form of the 1984 ...... 10,609 1985 ...... 10,989 changes in the poverty definition that the 1986 ...... 11,203 Committee agreed to make was: 1987 ...... 11,611 1Y88 ...... 12,092 l The annual change in the 1989 ...... 12,674 Consumer Price Index was made the basis for the annual 1990 ...... 13,359 1991...... 13,924 adjustment in the poverty thresholds. This replaced the ’ The last year for which thresholds indexed by the food plans were calculated. previous adjustment based on the Source: For unrevised thresholds, see U.S. Bureau of the Census, Current Population Reports, annual change in the per capita Series P-23. No. 28, p. 5: and Putnam, p. 278 inTechnical Paper I. For revised thresholds, see U. S. Bureau cost of foods in the economy of the Census, Current Population Reports, Series PXIO, No. 181. Appendix A, table A-3. For a table showing nonfarm poverty thresholds for families of different sizes since 1959, see table 3.El in theA?ulual &zfisfical food plan. Supplement, I!991 to the Social Security Bulletin.

8 Social Security Bulletin l Vol. 55, No. 4 l Winter 1992 issued by the Census Bureau) for terms “poverty” and “low-income.“4s The Subcommittee on Updating the statistical purposes. It was this action However, the Committee did not make Poverty Threshold, recognizing that that made the Orshansky thresholds (on any changes in the poverty definition. nutritional standards and consumption the revised-definition basis) the Federal In April 1973, the Offrce of patterns change over time, recommended Government’s official statistical poverty Management and Budget’s Statistical that the relationships contained in the thresholds. Policy Division requested the poverty series-in particular, the SSA’s low-income or near-poverty Interagency Committees on Income appropriate minimum standard of index was not included in the 1969 Distribution and on Poverty Statistics to nutrition and the multiplier derived from revision of the poverty definition. conduct a thorough review of Federal the average proportion of family income However, the 1969 revision did include a income and poverty statistics. spent on food-be updated every 10 provision that figures be published on the Subcommittees were formed to study the years, while retabulating poverty data for population below 125 percent of the following topics: updating the poverty the previous 10 years using the new poverty level. That income level was threshold, improving the measurement of thresholds to make comparisons over recognized as being “essentially the same cash income, and measuring noncash time possible. The Subcommittee as the SSA low-income index.“4’ By income. This was the last interagency reported that the logical time for the 197 1, the term “near-poverty” was being review of the poverty measure to make earliest updating of the threshold would applied to that income level, with the recommendations for significant changes be after the 1974 revision of the economy term “near-poor” being applied to the in the poverty measure. The three sub- food plan and that the food-to-total- population between 100 percent and 125 committees’ final reports were trans- income ratio from the Department of percent of the poverty level.42 mitted to OMB in September 1973.46 Agriculture’s 1965 Household Food The Subcommittee on Measurement Consumption Survey would be the most The 1970’s Studies of Cash Income made a number of appropriate of the available choices for specific recommendations for improving the derivation of a new multiplier.5i The the reporting of income on the Current In 197 1, the Office of Management Subcommittee considered this Population Survey. The subcommittee and Budget (OMB) established a Federal recommendation “a workable position also recommended “a separate income interagency Technical Committee on between having an out-of-date absolute Poverty Statistics. There were two items survey vehicle that would encompass measure and [having] an up-to-date many of the items not covered in the measure which is difficult to use for on the agenda for the Committee’s first CPS. . . .to collect better money (and making comparisons over time. . . .‘15* meeting: the possible replacement of the nonmoney) income data.“47 This Other recommendations included: term “poverty” by a term such as “low- recommendation was one of those that income;” and the definition of poverty ultimately led to the development of the areas to be used in future Census Survey of Income and Program The need for more frequent tabulations.43 Arguments in favor of Participation.48 studies of household food changing the term “poverty” to a more The Subcommittee on Non-Cash consumption in order to permit neutral term like “low-income” included Income discussed a proposed conceptual evaluation of the need for basic the claim that “low-income” was “a less framework for measuring noncash adjustment in the poverty value-laden. . .and. .more accurate income that was “worth further threshold at 5-year intervals description of the statistic.” In exploration;” agreed on some priority rather than the Subcommittee’s opposition to the change, it was argued areas for research; and reviewed recommended lo-year revision that the term “poverty” had been used for available data and data needs in the areas cycle. 8 years and was reaching the point of of food, health, and housing. The being a technical term and that changing Subcommittee agreed that noncash The need for consistency the label might lead to serious public income was received by families between the factor by which the relations problems. In July 197 1, Census throughout the income distribution and food budget is multiplied to Bureau reports began to use the term that an attempt should be made to value obtain the poverty threshold and “low-income” rather than “poverty” the income definition used for and “poor.” This terminology shift ended noncash income received for all the overall income distribution. in 1975. Although the term “low- recipients and not just those at the low end of the distribution. Members were “Thus, ifnoncash income is income” may still be used as a synonym also concerned that in many areas included in the income for “poverty” or “poor,” it is not the valuing noncash income at the cost to the distribution, it should be principal term for the Orshansky poverty provider would overstate the income included in both parts of the food concept.44 value that the recipient derives from it.49 to nonfood factor”-more In June 1972, the Technical The Subcommittee also supported the precisely, in both parts of the Committee completed a report on Cash Subcommittee’s recommendation food-to-total-income ratio from administrative and legislative uses of the for a new income survey vehicle.50 which the multiplier is derived.

Social Security Bulletin l Vol. 55, No. 4 l Winter 1992 9 l A long-term statistical research Index for Urban Wage Earners and discriminatory features of the poverty effort to provide a basis for the Clerical Workers (CPI-W). This definition could be eliminated within the development and evaluation of presented Census Bureau personnel with framework of the current measure. . . improvements in the the question of which version of the the Committee might wish to make measurement of poverty. index they should use to update the recommendations for other changes poverty thresholds annually for inflation. which would not be perceived as having l Deletion of the differentiation In 1979, Bureau personnel used the a significant impact on the number of between farm and nonfarm CPI-W to calculate poverty thresholds persons counted as poor, and which poverty thresholds. for 1978. However, it appears that at would not significantly modify the some point between November 1979 and structure of the current measure.” In However, no changes were made in October 1980, a decision was made in addition to the elimination of the male- the poverty definition as a result of the the Census Bureau to use the CPI-U headed/female-headed distinction, the 1973 review of poverty and income rather than the CPI-W to update the Committee also agreed on the statistics. poverty thresholds for inflation each elimination of the farm/nonfarm Section 823 of the year. No record of discussions or distinction and the “[elxtension of the Amendments of 1974 (P.L. 93-380) deliberations about this decision has been poverty matrix to families of up to 11 or required the Assistant Secretary of found. The decision may simply have more persons.“53 (In May 1980, the last Education in the Department of Health, mirrored the decision made at roughly recommendation was slightly modified Education, and Welfare (HEW) to the same time to adjust Census median for technical reasons to refer to families supervise a “thorough study of the family income data for inflation using of nine or more personss4) manner in which the relative measure of the CPI-U rather than the CPI-W. This After the 1981 change of presidential poverty for use in the financial assistance decision was the only change in the administrations, an earlier (June 1980) program authorized by Title I of the current official poverty definition that approval of the changes was reviewed Elementary and Secondary Education was not made by a Federal interagency and approved in November 1981 by the Act of 1965 may be more accurately and committee. new administration’s Working Group on currently developed.” At the end of Economic Statistics of the Cabinet 1974, an interagency Poverty Studies The 1981 Revision Council on Economic Affairs5’ The Task Force was established under the Census Bureau published an leadership of HEW to undertake an At a November 1979 meeting of the announcement of the changes in the intensive review of the current measure Interagency Committee on Income and poverty definition on page 62674 of the of poverty and of the implications of Distribution Statistics, it was December 28, 198 1, issue of the Federal various alternative measurement reported that the Justice Department’s Register. The three changes, as schemes. A final report, The Measure of Task Force on Sex Discrimination had described below, comprise the 1981 Poverty: A Report to Congress as found that Statistical Policy Directive revision in the poverty definition:56 Mandated by The Education No. 14 (the current version of the 1969 Amendments of1974, was submitted to BOB directive that Executive Branch The farm/nonfarm differential in Congress in April 1976. This report agencies use the Census Bureau poverty poverty thresholds was thoroughly explored all the issues thresholds for statistical purposes) eliminated by applying nonfarm involved in developing and revising discriminated against women; the reason thresholds to all families. (This poverty measures, gathering extensive for this finding was that the (detailed) was mathematically equivalent supporting information that was poverty thresholds for male-headed to setting farm thresholds at 100 presented in the report itself and in 17 families were higher than those for percent of nonfarm thresholds.) Technical Papers. (Technical Paper XI, female-headed families of the same size The detailed matrix of poverty “Update of the Orshansky Index,” was and composition. (Even though the thresholds had previously never published.) The report did not male-headed/female-headed distinction included separate sets of recommend specific changes in the had never been carried through to the thresholds for “Families with a current poverty measure. poverty guidelines used for program female householder, no husband eligibility purposes, “the use of separate present” and “All other A Shift in Indexing male and female thresholds may have a families” (the current Census significant impact on program evaluation Bureau terms for what used to be In January 1978, the Bureau of Labor uses.“) It was stated that “the called“female-headed” and Statistics introduced a second version of Committee should examine whether the “male-headed’ families). This its Consumer Price Index-the poverty measure appeared to distinction was eliminated, with Consumer Price Index for All Urban discriminate against women and. . . the two sets of thresholds being Consumers (CPI-U)-in addition to the determine whether agreement could be merged by averaging into a existing version, the Consumer Price reached on ways apparently single set.

10 Social Security Bulletin l Vol. 55, No. 4 l Winter 1992 l The detailed matrix of poverty if a national minimum welfare [public and Orshansky,“Statement. ,” in thresholds was extended to make assistance] benefit for low-income Redrawing the Poverty Line: Implications the largest family size category families with children were to be for Fighting and Poverty in “nine persons or more” rather established.) The panel will not set a new America-Hearing Before the Select Committee on Hunger. , U.S. Houseof than “seven or more persons.” poverty threshold, nor will it evaluate Representatives,Serial No. 10l-24, (This change was made possible any specific proposed program of public Washington,DC: U.S. GovernmentPrinting by the expanded sample size of assistance benefits. Instead, it will focus Office, October4, 1990 (printed in 199l), the CPS.) on concepts, information needs, and p. 5. measurement methods for such purposes. 60rshansky,“Children of the Poor,” Social These three changes are explicitly The panel will issue a report of its Security Bulletin, Vol. 26, No. 7, July 1963, described (p. 9) in the Census Bureau findings and recommendations by pp. 3-5; Orshansky,“Recounting the Poor- publication cited in footnote 55 as Summer 1994.58 A Five-Year Review,” Social Security “[s]ome of the changes explored in the After so many pages about the Bulletin, Vol. 29, No. 4, April 1966, p. 20. [ 1979 Fendler/Orshansky] paper” (the definition and measurement of poverty, ‘Orshansky,“Counting the Poor: Another one cited in footnote 32). perhaps the most appropriate way to Look at the Poverty Profile,” Social Security close is with a quotation from Mollie Bulletin, Vol. 28, No. 1, January1965, Subsequent Developments Orshansky: “Unlike some other pp. 3-29. calculations, those relating to poverty 8“The Problemof Poverty in America,” in During the 1980’s, there were have no intrinsic value of their own. Economic Report of the President extensive discussions and debates about They exist only in order to help us make Transmitted to the Congress January 1964 poverty measurement-particularly them disappear from the scene. .With Together With the Annual Report of the about proposals to count government imagination, faith and hope, we might Council of Economic Advisers, Chapter 2, noncash benefits as income for succeed in wiping out the scourge of Washington,DC: U.S. GovernmentPrinting without making poverty even if we don’t agree on how to Office, 1964,pp. 55-84 corresponding changes in the poverty measure it.“59 9“The Problem of Poverty. ,” thresholds. However, there was no pp. 58-59. The last sentenceis further official committee review of the poverty Notes explicatedin the following commentabout poverty measurement:“It shouldbe noted thresholds, and no changes were made in that the [possible]shift from moneyincome the official poverty definition. ’ Patricia Ruggles, Drawing the Line: to either consumptionexpenditures or In January 1990, the Administration Alternative Poverty Measures and Their personalincome [defined in the previous approved an initiative on improving the Implications for , Washington, sentenceas “includ[ing] suchnonmoney quality of’ Federal economic statistics. DC: UrbanInstitute Press,1990. incomeas home-produced food and imputed One of several dozen statistical series *William O’Hare, TayniaMann, Kathryn rent”] is tantamountto a loweringof the that was examined as part of this Porter,and RobertGreenstein, Real Life poverty line just assurely as is adoptinga initiative was the Federal poverty Poverty in America: Where the American lowermoney-income line”-Robert J. measure.57 In June-July 1990, an Public Would Set the Poverty Line (A Center Lampman,“Population Change and Poverty interagency working subgroup on on Budgetand Policy Prioritiesand Families Reduction,1947-75,” in Poverty amid Affluence, Leo Fishman(editor), New Haven: poverty, composed of employees from a USA FoundationReport), July 1990. Yale University Press,1966, p. 19. number of Executive Branch agencies, %ee,for instance,Spencer Rich, “Drawing prepared a draft agenda for possible the Line on Poverty: CensusBureau ‘O“TheProblem of Poverty. .,” pp. 57-58. research on poverty measurement and MeasurementSparks Criticism From Many I1Arthur M. Schlesinger,Jr., A Thousand Quarters,” Washington Post, October30, submitted it to the Council of Economic Days: John F. Kennedy in the White House, 1989, p,,Al 3; JulieKosterlitz, “Measuring Advisers. No further actions have been Boston: HoughtonMiMin Company,1965, Misery,” National Journal, Vol. 22, p, 1011;James L. Sundquist, and taken by Federal Executive Branch No. 3 1, August4, 1990, pp. 1892-1896;and Policy: The Eisenhower, Kennedy, and agencies regarding poverty in the contexl JasonDeParle, “In RisingDebate on Johnson Years, Washington,DC: Brookings of this initiative. Poverty, The Question:Who Is Poor?,” New Institution, 1968, pp. 135-l36; and Sar A. In June 1992, the Committee on York Times, September3, 1990,pp. 1 and 10. Levitan, The Great ’s Poor Law: A National Statistics of the National 4Fora brief descriptionof the poverty New Approach to Poverty, Baltimore:Johns Acadetny of Sciences convened a panel guidelines,see Gordon M. Fisher,“Poverty HopkinsPress, 1969, pp. 13-14. of academic experts to conduct a 30- Guidelinesfor 1992,” Social Security ‘*Personalcommunication with Robert month, two-part congressionally Bulletin, Vol. 55, No. 1, Spring1992, Lampman,September 10, 1987. pp. 4346. requested study. One part is an “Judith Innesde Neufville, Social examination of statistical issues involved ‘Personalcommunication with Mollie Indicators and Public Policy: Interactive in measuring and understanding poverty. Orshansky,June 14, 1988;Orshansky et al., Processes of Design and Application, New (The other part is a consideration of “MeasuringPoverty: A Debate,” Public York: ElsevierScientific Publishing statistical issues that would be involved Wefire, Vol. 36, No. 2, Spring 1978,p. 54; Company, 1975, p. 86; Orshansky,“Family

11 SocialSecurity Bulletin l Vol. 55, No. 4 l Winter 1992 Budgets andFee Schedulesof Voluntary confusedwith Friedrich EngeLs,the without committingourselves to the view that Agencies,”Social Secun’ty Bulletin, Vol. 22, collaboratorof Karl Marx.) A moreprecise the limit itself is sufficiently high”-Rev. No. 4, April 1959, p. 10. statementof the law (in Englishtranslation) JohnA. Ryan, “What Wageis a Living “‘Federal Register, December1, 1975, is as follows: “The pooreris a family, the Wage?’ Catholic World, Vol. 75, No. 445, p, 55650;U.S. Departmentof Health, greateris the proportionof the total outgo April 1902, p. 4. Education,and Welfare, TheMeasure of [family expenditures]which mustbe usedfor “‘Orshansky,‘Some Thoughtson CETA Poverty: A Report to Congress as Mandated food. .The proportionof the outgoused for Requirementsfor Price Statisticsand Poverty by The Education Amendments of 1974, food, other thingsbeing equal, is the best Measures”(unpublished), a paperpresented Washington,DC: U.S. GovernmentPrinting measureof the materialstandard of living of at the Bureauof Labor Statistics’North Office, April 1976,pp. 8 and 39. (This a population” (CarleC. Zimmerman,“Ernst AmericanConference on Labor Statistics, reportis cited in later footnotesas The Engel’sLaw of Expendituresfor Food,” Hyannis,Massachusetts, June 25, 1974,p. 9; Measure of Poverty.) Quarterly Journal of , Vol. 47, Orshanskyet al., “MeasuringPoverty: A No. 1, November1932, p. 80). 15TheAgriculture Departmentreplaced the Debate,” p. 54; andOrshansky, economyfood plan with the thrifty food plan **Forfurther discussionof this point, see “Commentary:The Poverty Measure,” in 1975at the samecost level. The thrifty Alan Haber,“Poverty Budgets: How Much Is Social Security Bulletin, Vol. 51, No. 10, plan is usedto determinethe amountof the Enough?” Poverty and Human Resources October 1988, p. 23. maximumfood stampallotment. However,it Abstracts, Vol. 1, No. 3, May-June 1966, ““An absolutestandard [of poverty] means shouldbe notedthat the thrifty food plan was p. 6, and The Measure of Poverty, p. 75. onedefined by referenceto the actualneeds not usedat any time in either the Yhe understandingof Orshansky’s of the poor andnot by referenceto the developmentor the updatingof the present proceduredescribed in this and the following expenditureof thosewho are not poor”-Sir official poverty thresholds. paragraphwas not madeexplicit in her July Keith Joseph[Secretary of State for ‘@‘Countingthe Poor: Another Look. ,” 1963and January 1965 articles. See, Educationin the Cabinetof PrimeMinister p, 6. The bibliographicsource for the however, Orshansky,“Who Are thePoor?” MargaretThatcher] and J. Sumption, “temporaryor emergencyuse” quotation is p. 94 in Sixth Biennial Workshop on Public Equality, London:John Murray, 1979, Betty Peterkin,“Family FoodPlans, Revised Welfare Research and Statistics: pp. 27-28, quotedin JoannaMack and 1964,” Review, October Proceedings-July I S-22,1966, Orshansky StewartLansley, Poor Britain, London: 1964, p. 12. The wordsin questionwere et al., “MeasuringPoverty: A Debate,” GeorgeAllen & Unwin Ltd., 1985,p. 16. usedabout the plan not becauseit wasnot p. 46; Orshansky,“Who’s Who Amongthe ‘IPeter Townsend,“The Meaning of nutritious(the foodsin the plan would Poor: A DemographicView of Poverty,” Poverty,” British Journal of , providea nutritiousdiet), but becausethe Social Security Bulletin, Vol. 28, No. 7, July Vol. 13, No. 3, September1962, pp. 221 selectionof foodsin the plan could become 1965, p. 9; and Orshansky,“How poverty is and 223. (Specifically, Townsendproposed monotonousover an extendedperiod of time measured,”Monthly Labor Review, Vol. 92, that a poverty line be setat half or two-thirds (personalcommunications with Betty No. 2, February1969, p. 39. of averagehousehold or family income.) Peterkin,February 20, 1990,and April 22, ““Children of the Poor,” p. 8. Victor R. Fuchs,“Toward a Theory of 1992). The sourcewhich containedthe food 2s“Countingthe Poor: Another Look. ,” Poverty,” in The Concept of Poverty, Task plansin the form in which Orshanskyused p. 9. Forceon and themhad statedthat “The economyfood Opportunity, Washington,DC: Chamberof *PIffarm housingis countedas part of the plan, .is essentiallyfor emergencyuse” Commerceof the United States, 1965, p. 74. farm operation-that is, if farm housing (EloiseCofer, Evelyn Grossman,and Faith For commentsby Orshanskyon the expensesare classifiedas part of farm Clark, Family Food Plans and Food Costs 50-percent-of-median-incomestandard, see (business)operating expenses-the result is For nutritionists and other leaders who Orshanskyet al., “MeasuringPoverty: A to reducethe amountof net farm self- develop or use food plans (Home Debate,”p. 54; cf. CarolFendler and employmentincome, and thusthe amountof EconomicsResearch Report No. 20), Orshansky,“Improving the Poverty total moneyincome, as definedby the Census Washington,DC: Consumerand Food Defition,“American Statistical Bureau. EconomicsResearch Division, Agricultural Association: 1979 Proceedings of the Social ResearchService, U.S. Departmentof *“‘Counting the Poor: Another Look. , ” Statistics Section, p. 643. pp. 9-10. Agriculture,November 1962, p. 25). “See, for instance,Orshansky, “Recounting i7BettyPeterkin and Faith Clark, “Money 280rshansky,“Recounting the Poor. ,” the Poor. ,” pp. 21 and22; and Ida C. Valueand Adequacy of Diets Comparedwith p. 22; The Measure of Poverty, p. 7. Merriam, “Conceptsand Measuresof theUSDA FoodPlans,” Family Economics 29“Countingthe Poor: Another Look. ,” Welfare,” in American Statistical Review, September1969, pp. 6-8 p. 3; seealso Orshansky, “Living in Association: Proceedings of the Social ‘*“Countingthe Poor: Another Look. ,” :A ModerateStandard for an Statistics Section, 1967, p. 182. For articles p. 6. Elderly Couple,”Social Security on this historicalphenomenon, see Robert W. Kilpatrick, “The IncomeElasticity of the ‘91bid., p. 6. Bulletin, Vol. 3 1, No. 10, October 1968,p. 4. For an exampleof intelligent, socially Poverty Line,” Review of Economics and loIbid., p. 7. concernedminds running (independently) “in Statistics, Vol. 55, No. 3, August 1973, *‘This law wasdeveloped by ErnstEngel the sametrack,” comparethe following pp. 327-332; andDavid Hamilton, (1X2 l-96), a Germanstatistician who was statementfrom an obscurearticle published “Drawingthe Poverty Line at a Cultural directorsuccessively of the statisticalbureaus yearsbefore Orshansky was born: “We can SubsistenceLevel,” Southwestern Social of Saxonyand Prussia.(He shouldnot be definea limit belowwhich it is wrong to go, Science Quarterly, Vol. 42, No. 4, March

12 SocialSecurity Bulletin l Vol. 55, No. 4 l Winter 1992 1962, pp. 337-345. Even without the benefit Standards, [Bureau of the Budget] to A. Ross Poverty Statistics,” Statistical Reporter, of quantitative studies, many U.S. analysts Eckler, Director, Bureau of the Census. Office of Management and Budget, October and scholars writing about poverty and Copies of the letter were sent to CEA, OEO, 197 1, p. 67; July 1, 197 1, memorandum consumption patterns between 19 17 and and SSA “in order that the major agencies from Bette Mahoney (for Margaret Martin, about 1970 were quite aware-indeed, concerned in this problem will be informed.” Chairman of the Committee), PRE/R [the accepted as almost a truism-that 390rshansky, “Statement. , ” in U.S. Research Division of the Office of Plarming, poverty/subsistence lines generally rose in House of Representatives, Census and Research, and Evaluation of the Office of real terms as the real incomes of the general Designation of Poverty and Income: Joint Economic Opportunity], to Members of the population rose. Hearing Before the Subcommittee on Census Interagency Poverty Review Committee- %rshansky, “Memorandum for Dr. and Population of the Committee on Post Subject: Meeting on Tuesday, July 13th. Daniel P. Moynihan-Subject: History of O&e and Civil Service and the ““The following exception should be noted: the Poverty Line,” July 1, 1970, reprinted in Subcommittee on Oversight of the Committee In the context of the Department of Housing Mollie Orshansky [compiler],“Documentation on Ways and Means. . , Post Office and and Urban Development’s housing assistance of Background Information and Rationale for Civil Service Committee Serial No. 98-28, programs and Current Poverty Matrix,” Technical Paper I Ways and Means Committee Serial No. Block Grant program, as well as housing of The Measure of Poverty, Washington, DC: 98-87, Washington, DC: U.S. Government assistance programs of the Agriculture lJ.S. Department of Health, Education, and Printing Office, May 15, 1984, p. 12; and Department’s Home Administration, Welfare, 1976, p. 235 (this paper is cited in Orshansky, “Statement. .” (January 30, the terms “low-income,” “very-low-income,” later footnotes as Technical Paper I); Israel 1986), in U.S. House of Representatives, and “lower-income” are defined in temis of Putnam, “Poverty Thresholds: Their History Census Bureau Measurement of Noncash perentages of a metropolitan or and Future Development” [November 19701, Benefits: Hearings Before the Subcommittee nomnetropolitan area’s median family in Technical Paper I, p. 276; and The on Census and Population of the Committee income, adjusted for family size; in these Measure ofPoverfy, pp. 6 and 9. on Post Ofice and Civil Service. . , Serial contexts, these terms do not refer to the 3sOrshansky, “Who’s Who Among the No. 99-5 1, Washington, DC: U.S. Orshansky poverty concept. Poor. ,” p. 4; Orshansky, “Progressing Government Printing Office, 1986, p. 129. 4SJune 2 1, 1972, memorandum from Bette Against Poverty,” Research and Statistics 400ctober 3, 1968, memorandum for the Mahoney, Consultant, Statistical Policy Note No. 24 (1968 series), U.S. tiles, from Ida C. Merriam-Re: First Division, [Office of Management and Department of Health, Education, and meeting of the Budget Bureau Poverty Level Budget] to Members of the Technical Welfare, Social Security Administration, Review Committee, October 2, 1968; Committee on Poverty Statistics-Subject: Office of Research and Statistics, October 9, 1968, memorandum from Working Paper on Administrative and December 10, 1968, p. 2, footnote 1; and Lawrence N. Bloomberg, Secretary, to Legislative Use of the Terms “Poverty” and The Measure of Poverty, p. 12, Figure 1. Members, Poverty Level Review “Low Income;” The Measure of Poverty, p. 7. Orshansky has indicated (personal Committee-Subject: Minutes, Meeting of 46September 4, 1973, memorandum from communication, May 29, 1990) that the fact October 2, 1968. Robert W. Raynsford, Statistical Policy that the cost of the economy food plan did 41Cf. Orshansky, “Statement. .” Division, [Office of Management and not increase for one year [presumably the [November 15, 19731, in U.S. House of Budget] to Paul F. Krueger-Subject: 1966-to-1967 non-increase] was a motivating Representatives, Student Financial Consolidated Report of Subcommittee factor behind SSA’s 1968 efforts to revise the Assistance (Theory and Practice of Need Chairmen: Review of Poverty Statistics; and poverty thresholds. Analysis): Hearings Before the Special Bette S. Mahoney, “Review Of Poverty 36May 3, 1968, memorandum from Ida C. Subcommittee on Education of the Committee And Income Distribution Statistics,” Merriam, Assistant Commissioner, Office of on Education and Labor. . .Part 1: Theory Statistical Reporter, No. 74-7, Statistical Research and Statistics [SSA]-Subject: The and Practice of Need Analysis. . , Policy Division, U.S. Office of Management SSA poverty and low-income cut-off points Washington, DC: U.S. Government Printing and Budget, January 1974, p. 117. for 1967 incomes; and May 7, 1968, Oflice, 1974, p. 66. 47June 28, 1973, memorandum from Murray memorandum from Robert L. Stein, [Bureau @Bureau of the Census, Current Population S. We&man [Census Bureau], Chairman, of the Census] to Professional Staff, Poverty Reports, Series P-60, No. 77, Poverty Subcommittee on Measurement of Cash Statistics Section [Bureau of the Census]- Increases by 1.2 Million in 1970 (Advance Income, to Julius Shiskin, Chief Statistician, Subject: Possible changes in poverty data from March 1971 Current Population Office of Management and Budget-Subject: definitions. Only the Merriam memo Survey), Washington, DC: U.S. Government Final Report of the Subcommittee on provides the specifics of SSA’s proposal Printing Office, May 7, 197 1, Measurement of Cash Income; see also about the indexing of the thresholds. p. 8, table 9, “Persons Below the Near- pp. 118-120 of the January 1974 Statistical 37For later discussions of revised, higher Poverty Level. ;” and Bureau of the Reporter article cited in footnote 46. poverty thresholds based on a multiplier Census, Current Population Reports, Series 48Martynas A. Yeas, Income Survey derived from the 1965 survey, see The P-60, No. 81, Characteristics ofthe Low- Development Program Staff, “An Measure of Poverty, pp. 75-77; Orshansky Income Population, 1970, Washington, DC: Introduction to the Income Survey et al., “Measuring Poverty: A Debate,” U.S. Government Printing Office, November Development Program,” unpublished report pp. 47 and 48; and Fendler and Orshansky 1971, p, 14 (section titled “Persons between (revised October 1979) p. 2; and personal (cited in footnote 32), pp. 642-643 and 645. 100 and 125 percent of the low-income communication with Bette Mahoney, 38July 16, 1968, letter from Raymond T. level”). ca. June 10, 1987. Bowman, Assistant Director for Statistical 43Bette Mahoney, “Technical Committee on 49Cf. Janice Peskin, “In-Kind Income and

13 Social Security Bulletin l Vol. 55, No. 4 l Winter 1992 the Measurement of Poverty,” Technical Clearinghouse Review, Vol. 14,No. 8, Paper VII of The Measure of Poverty, November1980, pp. 736-738. Washington,DC: US. Departmentof 54May22, 1980,memorandum from Milo B. Health,Education, and Welfare, October Sunderhauf,[Department of Commerce, 1976, pp. 2-3. O&e of FederalStatistical Policy and ‘OThesource for all itemsin this paragraph Standards],Chair, to Members,Interagency was: September24, 1973,memorandum Committeeon Incomeand Wealth from Bette Mahoney,Chairman of the DistributionStatistics-Subject: Technical Subcommittee,Statistical Policy Division, Revisionin the StatisticalDefinition of [Oftice of Managementand Budget], to Paul Poverty. Krueger-Subject: Final Reportof the 55Bureauof the Census,Current Population Subcommitteeon Non-CashIncome. See Reports,Series P-60, No. 133, alsopp. 120-121of the January 1974 Charactetistics of the Population Below the Statistical Reporter article cited in Poverty Level: 1980,Washington, DC: U.S. footnote46. GovernmentPrinting Office, July 1982, s’Cf. the followingcomment by Orshansky p. 3. Cf. alsoW[illiam] P. O’Hare, duringthe April 30, 1973,joint committee “MeasuringPoverty,” Clearinghouse Review, meetingbefore the division into three Vol. 15, No. 8, December1981, subcommittees:“. .not only shouldthe food pp. 648-649. budget[food plan] be allowedto changeover S6Foradditional details, see Bureau of the time, soshould the incomefood ratio”-May Census,Current Population Reports, Series 14, 1973,memorandum from Bette Mahoney, P-60,No. 133, Characteristics of the Secretaryto the Committees,Statistical Population Below the Poverty Level: 1980, Policy Division, [Office of Managementand Washington,DC: U.S. GovernmentPrinting Budget],to Membersof the Interagency Of&e, July 1982, pp. 2-5, 9, and 186. The Committee[sic] on IncomeDistribution and three changesreduced the numberof cells in Poverty Statistics-Subject: Minutes of the the detailedmatrix of poverty thresholds Meeting of April 30, 1973. from 124to 48. ‘*Thesource for all itemsin this paragraph 57Councilof EconomicAdvisers, (includingthe four recommendationsat the “Improving the Quality of Economic end)was: August2, 1973,memorandum Statistics,”press release4 pages, from Milo B. Sunderhauf,Statistical Policy January 25, 1990; Appendix B, “Improving Division, OMB, to RobertRaynsford, the Quality of EconomicStatistics,” OMB-Subject: Recommendationsof the in Economic Report of the President Subcommitteeon Updating the Poverty Transmitted to the Congress February 1990 Threshold.Some but not all of the Together With the Annual Report of the recommendationsof this Subcommitteewere Council of Economic Advisers, Washington, alsodescribed on p. 118of the January1974 DC: U.S. GovernmentPrinting Office, Statistical Reporter article cited in footnote pp. 281-285;and the DeParleNew York 46. Times story cited in footnote3. ‘%e sourcefor all precedingitems in this 5gNationalResearch Council, Commission paragraphwas: February26, 1980, on Behavioraland Social SciencesEducation, memorandumfrom Milo B. Sunderhauf, Committeeon NationalStatistics, “Panel on [Departmentof Commerce,Office of Poverty and Family Assistance:Concepts, FederalStatistical Policy and Standards], InformationNeeds, and Measurement Chair,to Members,Interagency Committee Methods,” May 1992. on Incomeand Wealth Distribution 590rshansky,“Demography and Ecology of Statistics-Subject: Recommendationsfor Poverty,” in Proceedings of a Conference Changesin Poverty Definition. (An on Research on Poverty (submittedto The attachmentsummarized the November7, Center for the Study of SocialProblems, 1979,meeting of the Committee.) For the National Institute of ,under changesrecommended by the Committee,see provisions of a grant from NIMH), also“Recent Proposed Changes to the Washington,DC: Bureau of SocialScience Ofticial Definition of Poverty,” Research,Inc., June 1968,p. 28.

14 SocialSecurity Bulletin l Vol. 55,No. 4 l Winter 1992

Top View