Family Equivalence Scales in the Poverty Thresholds

Michael Oakleaf and Kristen Voskuil

Adults eat more than small children do, and larger families spend more on some necessities than smaller families do. It therefore makes sense that poverty thresholds based on a percentage of median consumption should somehow account for these differences. An equivalency scale allows the poverty status of families of various sizes and compositions to be compared on an equal basis. The current poverty thresholds used by the Census Bureau and the redesigned thresholds proposed by the Panel on Poverty and Family Assistance of the National Academy of Sciences (the Panel) take different approaches in adjusting for the needs of families of different sizes and composition.

Current Poverty Thresholds

The current poverty thresholds, developed by Social Security staff economist Mollie Orshansky in 1965, contain an implied equivalency scale. The thresholds themselves, however, were not constructed using an equivalency scale. They are based instead on the U.S. Department of Agriculture's (USDA) Economy Food Plan, which estimated the costs of minimally adequate food budgets for families of various sizes and compositions. Orshansky then multiplied these individual budgets times three, based on the 1955 Household Food Consumption Survey, which found that the average family of three or more people spent in total three times what they spent on food. The same survey showed that the average family of fewer than three people spent an average of 27 percent of their total expenditures on food. For these families, Orshansky multiplied the food budgets by a factor of 3.70, rather than 3.0, to account for the smaller portion that these families spent on food relative to their total spending.

Critics point to some problems with the way the current thresholds treat families of differing sizes and compositions. First, the "food times three" formula is problematic. Spending on food cannot be expected to vary at the same rate among different families as spending on other goods, yet that is the assumption implicit in basing the thresholds entirely on food budgets. For example, the thresholds are lower for elderly people because they were assumed to spend less on food, but the "food times three" formula also implies that seniors must spend less on other necessities by the same proportion.

There are also anomalies in the differences between how much the thresholds increase as family size increases. For example, Ruggles (1990) shows that an increase from one person to two increases the threshold by 28 percent, from two to three by 23 percent, and from three to four by 28 percent. If larger families are able to take advantage of economies of scale, then as family size grows the thresholds should grow at a decreasing rate.

The Panel's Proposal

The Panel proposes an explicit family size equivalency scale. The proposed reference family consists of two adults and two children because, while no longer the most common family type in the United States, more people are in this family type than any other. The Panel uses a formula to make adjustments for families of different sizes and compositions: (A + PK)F, where A is the number of adults in the family, K is the number of children under 18, P is the percent value that each child contributes to a family's consumption relative to the adults, and F is a factor to account for economies of scale. The Panel further recommends a value of 0.70 for P (an assumption that children contribute to consumption at 70 percent the rate of adults) and a range of 0.65 to 0.75 for F. The scale factor for each family type is divided by the scale factor for the reference family. The threshold for the reference family is multiplied by the resulting ratio to determine the threshold for each separate family type. Table 1 displays the ratio of the scale factor of selected family types to that of the reference family. As Table 1 shows, using a scale factor of 0.65 results in somewhat larger thresholds for families of one to three people, and somewhat smaller thresholds for families of five or more, than a scale factor of 0.75.

Comparison of Scale Factors
Family Type Current Official Scale 0.65 Scale
Economy Factor
0.75 Scale
Economy Factor
One person 0.513 0.451 0.399
Married couple 0.660 0.708 0.672
Plus one child 0.794 0.861 0.841
Plus two 1.000 1.000 1.000
Plus three 1.177 1.130 1.151
Plus four 1.318 1.251 1.295
Plus five 1.476 1.367 1.434
Source: Adapted from Citro and Michael, Measuring Poverty: A New Approach, p. 181, Table 3-4.

The Panel suggests that its method for determining the equivalency scale is superior to the current thresholds implicit equivalency scale. The Panels recommended formula explicitly acknowledges that children consume less than adults do. The formula also adjusts for economies of scale in a more consistent manner than the current thresholds, such that the thresholds for larger families increase at a steadily decreasing rate.

The Panel acknowledges, however, that their method contains shortcomings, even going so far as to call it "crude." The Panel acknowledges that one may argue with the values suggested for P and F, and that assigning a single value for P makes the problematic assumption that all children up to the age of seventeen contribute an equal amount to the consumption of the family. The Panel also admits that the factor for economies of scale is not in any way tied to the commodities that produce the effect (such as housing). Finally, the Panel recommends further research on omitted variables such as relative prices, location, and variations in scale value that derive from the family's standard of living.

Cohabitation and Families

Another change to the poverty measure that the Panel recommends is the inclusion of cohabitants in the definition of "family." Currently, the unit of analysis for the measurement of poverty is the family (rather than either the individual or household), and is defined as people who are living together and related by blood, marriage, or adoption.

The rationale for using the family as the unit of analysis is that all adult members of the family have control over the resources, and all members of the family benefit from those resources. This pooling of resources allows for the economies of scale that are encompassed in the thresholds. Under the current methodology, cohabiting couples are treated for poverty measurement purposes as two single people. The Panels recommendation for treating cohabitants as a two-person family is based on the belief that cohabitants pool resources and consume commodities in much the same way that married couples do. This change would affect both the threshold and resource sides of the poverty measure, as the cohabitants would be counted in the family size, and the income contributed would be counted in the family's resources.

The Census Bureau estimated that there were 3.3 million unmarried-couple households in 1992, a significant increase over the levels that existed when the current poverty measure was developed (Citro and Michael 1990). Bauman (1997) estimates that including cohabitants in the definition of family would decrease the poverty rate by 6 percent. There may be some political resistance to this inclusion from groups who feel that public policy should promote traditional families with married parents.


An equivalency scale is a means to allow for comparisons of the poverty status of families of different sizes and compositions. The current thresholds were based on food budgets developed by the USDA. The equivalency scale is implied, since the individual budgets were based on the food needs of families of different sizes and composition.

The Panels proposal contains an explicit scale. They suggest using a reference family of two adults and two children, and a formula, (A + PK)F , to derive the thresholds for families that contain different numbers of adults and children from the reference family.

The Panel also proposes changing the definition of "family" to include cohabitants. They believe that cohabiting adults share resources in ways similar to married couples.


Bauman, Kurt. 1997. Shifting Family Definitions: The Effect of Cohabitation and Other Nonfamily Household Relationships on Measures of Poverty. Institute for Research on Poverty [Online]. Available: [1999 March 21].

Citro, Constance F. and Robert T. Michael, eds. 1995. Measuring Poverty: A New Approach. Washington, D.C.: National Academy Press.

Ruggles, Patricia. 1990. Drawing the Line: Alternative Poverty Measures and their Implications for Public Policy. Washington, D.C.: The Urban Institute.