This report argues that American poverty should be measured relatively to our society’s prevailing standards. The article shows why relative measures of poverty are superior to their alternatives, and especially those called “absolute” measures. It makes this case based on extensive international research, theoretical arguments, and concrete examples. By advocating for a relative measure, I encourage the United States to follow the approach used by the European Union, many other international organizations like the Luxembourg Income Study (LIS), and by the overwhelming majority of international poverty researchers.1

The report has the following sections. First, I provide a clear and simple definition of poverty as a shortage of resources compared to needs. Second, I contend that resources and needs must be defined within the social context of a time and place. Third, I explain that relative measures define poverty according to the prevailing standards of a time and place. Fourth, I show how relative measures better fit leading theoretical definitions of poverty. Fifth, I review evidence showing that relative measures of poverty are more strongly associated with health, well-being, and life chances.

Before proceeding, please let me define some key terms. A relative measure is calculated compared to the current distribution of resources and needs. Typically, a relative measure is calculated relative to the median income. The most popular and conventional relative measure is to say a person is poor if their income is below 50 percent of the median.2 Using that measure and the LIS, a family of three would be poor if their income was below about $34,000 in the United States in 2018 (after incorporating all taxes, tax credits, and welfare transfers).3 An absolute measure is calculated based on a timeless or uniform set of needs. For instance, one could calculate an absolute measure based on the cost of a basket of necessities like food and shelter. According to the U.S. official poverty measure (OPM)—which is considered an absolute measure—a family of three would be poor if their pretax cash income was below about $20,780 in the United States in 2018.4

In the process of making the case for relative measures of poverty, it should be clear this piece strongly advocates against the OPM. The OPM is irredeemably flawed, unreliable, and deeply problematic. It was devised in the 1960s based on an estimate of an emergency food budget in the 1950s, and has not been updated to reflect subsequent dramatic changes in a family’s needs. Most poverty researchers agree the OPM sets the poverty line far too low. Shawn Fremstad’s recent piece in this series of TCF reports on poverty, “The Defining Down of Economic Deprivation,” thoroughly and unequivocally critiques the OPM.5 Therefore, I will not reiterate the many reasons we should abandon the OPM and why the OPM should not be taken seriously in social science or public policy. Rather, this piece’s critique of the OPM is largely implicit, as it is clear that relative measures are theoretically and conceptually superior to any absolute measure like the OPM.

I. Poverty Is a Shortage of Resources Compared to Needs

The simplest and clearest definition of poverty is a shortage of resources compared to needs.6 For instance, one might lack enough of the resource of money to meet the need of paying rent. Another way of saying this is to say one’s resources fall below a threshold, the line under which you cannot meet your needs (sometimes referred to as a “standard of needs”). To escape poverty means one’s resources have risen above that threshold. For example, many critique the OPM because even at (or above) the threshold, most families cannot meet their needs.7

Poverty is a shortage of resources, and a shortage of resources usually triggers low well-being. Yet, some people have low well-being without being poor and some are poor without low well-being.

There is a difference between well-being and poverty, as economist and philosopher Amartya Sen persuasively explains. Poverty is a shortage of resources, and a shortage of resources usually triggers low well-being. Yet, some people have low well-being without being poor and some are poor without low well-being.8 Well-being may include health, life satisfaction, food security, and access to decent health care, for example, or a neighborhood with minimal pollution. As Sen explains, “Poverty is not a matter of low well-being, but of the inability to pursue well-being precisely because of the lack of economic means.” 9 Purchasing well-being is a need, and money is a resource that enables this purchase. Hence, poverty usually undermines well-being, and this is why we care about poverty. But a lack of resources to meet needs is not well-being per se.

II. Resources and Needs Must Be Defined within the Social Context of a Time and Place

If poverty is a shortage of resources compared to needs, we first must clarify the definition of resources. Most agree resources should be defined as comprehensively as possible. A common definition of resources is current income adjusted for taxes and including transfers and tax credits.10 This measure of income is often called “post-fisc” or “disposable” income. The measure embraces the reality that people live and are only able to consume in a “net” post-tax income world. Welfare transfers and tax credits clearly have real value to people. Income is also a positional good: it defines where one stands in society. We use income to compete with other people, who are also aiming to meet their needs. Income confers status in communities and society, and status is always relative to that of others. Thus, the value of income itself is inherently relative (as opposed to absolute). Indeed, any resource’s value is always relative.

The definition of needs also requires clarification. While resources are inherently relative, some might argue that needs are absolute across time and place. Often, there is a perception that there is a list of essential necessities that we can both identify and agree on. But the social science literature has not been able to reach a consensus on an objectively defined set of needs.11 Many assume that the U.S. official poverty measure is based on a scientific and objective standard of needs. This is simply not true.12 Almost always, poverty measures are not linked to anything like physiological or caloric requirements, or any objective budget of basic necessities. The main reason is because doing this is much harder than one might anticipate. As Patricia Ruggles shows, consumption patterns changed so dramatically from the 1950s to the 1990s that agreeing on the basic needs of American families is quite difficult.13

I am not saying that we cannot agree that a desperate level of deprivation does exist. There is surely a threshold of extreme deprivation under which we would all agree families are definitely poor. People are certainly poor if they are homeless or starving, or cannot access clean water. The problem is that it is practically impossible to define a valid and reliable standard of needs above such desperate levels of deprivation. Such a minimal standard sets the line so low that only a very small share of the population would be poor in the United States or any other wealthy nation.14 The moment one tries to define needs above such minimal subsistence levels, it quickly becomes clear that any definition of absolute needs will be questionable.

Think for a moment about how difficult it would be to define all necessities across any geographically large place over any reasonable period of time. Needs might include housing, food, water, and protection from extreme cold. But what quality of housing is considered sufficient to meet one’s needs? Exactly how much food is necessary across a population with varying heights, weights, health conditions, and ages? And necessary for what—just survival, or growth and thriving? Aldi Hagenaars points out that even nutritionists cannot agree about levels of calories needed for all the various ages, sexes, occupations, and living conditions.15 Does the need for water include bathing needs, and if so, how often per week? And, if we agree people should be able to avoid extreme cold, should they also be able to avoid extreme heat? If people are old or have certain health conditions, air conditioning might be an essential need. But if it is, how old and how ill must they be for air conditioning to be actually essential?

It doesn’t take much imagination to think up several other needs, such as transportation and communication. Are cars essential? What if mass transportation is available and cheap, do cars then become non-essential? How about a phone, and if a phone is essential, should that be a smart phone with internet access? And, what about health care or child care? If those are needs, how much is necessary, and at what quality? How old must the children be before child care stops being a need, and should the age cutoff vary depending on the safety in the neighborhood? What if it is culturally normative that grandparents live with adult children and grandchildren: should one then argue that child care outside the family is not a need?

The debate becomes endless and almost existential. If one is sad that a neighbor has a nicer car, that probably is not sufficient deprivation to make a fine car a need. But at what exact point is one’s housing sufficiently deprived to trigger enough suffering such that it rises to the level of an essential need? Precisely how many increments of health would have to be worse in order for housing to be insufficiently meeting one’s needs? In short, we cannot clearly differentiate between various views of what needs are truly essential. Worse, it is never possible to quantify all these needs and come up with a threshold that fits all the heterogeneities and diversities of modern societies.

We cannot clearly differentiate between various views of what needs are truly essential. Worse, it is never possible to quantify all these needs and come up with a threshold that fits all the heterogeneities and diversities of modern societies.

Working through these sorts of decisions makes clear that the definition of needs is overwhelmed by uncertainty. For exactly this reason, the vast majority of international poverty researchers have become skeptical of absolute measures of poverty. As Hagenaars contends, “the resulting estimates are not as absolute and objective as they are claimed to be.”16 Similarly, Peter Townsend concludes, “Any rigorous conceptualization of the social determination of need dissolves the idea of ‘absolute’ need.”17 As Lee Rainwater and Timothy Smeeding explain, “The more experience countries have with absolute poverty definitions, the more obvious becomes the absurdity of the rationale for them.”18 Therefore, lists of basic needs that should be applied across all cultural and historical contexts have failed to stand up to scrutiny.

Recall that the definition of poverty is a shortage of resources compared to needs. Therefore, despite all these uncertainties, we still need to move forward and come up with a usable definition of needs. The consensus solution is that we must define needs within each geographic, historical, and cultural context. Obviously, this is true in terms of the cost of living.19 As an extreme example in the United States, housing costs vary dramatically between rural areas in the Midwest and cities on the coasts. To meet the need of housing, ideally the threshold would be higher in Santa Monica, California than Sioux Falls, South Dakota. As a society experiences economic development or an upswing in the business cycle, the costs of needs tend to rise.

The very meaning of what counts as a need is always defined within a place and time. Rather than an absolute list of physiological and caloric necessities that transcends time and place, needs become “needs” as a society or community evolves.

Even more generally, the very meaning of what counts as a need is always defined within a place and time. Rather than an absolute list of physiological and caloric necessities that transcends time and place, needs become “needs” as a society or community evolves. Given such a realistic acknowledgement that one’s cultural and historical context shapes what we view as a need, it becomes clear why poverty should be defined relatively.

III. Relative Measures Define Poverty according to the Prevailing Standards of a Time and Place

A relative measure defines poverty as a shortage of resources relative to needs defined by the prevailing standards of a time and place.20 The prevailing standards of the time and geographical place thus give us a useful definition of needs.21 Hence, relative measures can assess whether people have fallen far below the living conditions and customary standards of the majority of the contemporary society or community in which they reside.

This definition of poverty is not arbitrary. Well-designed relative measures acknowledge there is always going to be some uncertainty about where to precisely draw the threshold (e.g. some advocate for 40 percent of the median rather than 50 percent). But a key advantage of relative measures is they make that uncertainty transparent, and this in turn makes the threshold more concrete. Indeed, the average person naturally thinks of their social context when they define poverty. When asked in a survey how to define poverty, most people come up with a poverty threshold that is roughly equal to the relative threshold of 50 percent of the median income.22 Further, the average person’s definition of “necessities” tends to rise as living standards rise.23

In the classic The Other America, Michael Harrington compellingly illustrates how the meaning of poverty must be contextualized relative to the living standards of the mainstream of contemporary society:

Shall we say to them [the American poor] that they are better off than the Indian poor, the Italian poor, the Russian poor?. . .In the nineteenth century, conservatives in England used to argue against reform on the grounds that the British worker of the time had a longer life expectancy than a medieval nobleman. This is to say that a definition of poverty is, to a considerable extent, a historically conditioned matter. Indeed, if one wanted to play with figures, it would be possible to prove that there are no poor people in the United States, or at least only a few whose plight is as desperate as that of the masses in Hong Kong. There is starvation in American society, but it is not a pervasive social problem as it is in some of the newly independent nations. There are still Americans who literally die in the streets, but their numbers are comparatively small. . .Those who suffer levels of life below those that are possible, even though they live better than the medieval knights or Asian peasants, are poor. . .The American poor are not poor in Hong Kong or in the sixteenth century; they are poor here and now, in the United States. They are dispossessed in terms of what the rest of the nation enjoys. . .To have one bowl of rice in a society where all other people have half a bowl may well be a sign of achievement and intelligence; it may spur a person to act and to fulfill his human potential. To have five bowls of rice in a society where the majority have a decent, balanced diet is a tragedy.24

Similarly, the equally influential scholar, Peter Townsend, eloquently elaborates:

Poverty is a dynamic, not a static concept. Man is not a Robinson Crusoe living on a desert island. He is a social animal entangled in a web of relationships at work and in family and community which exert complex and changing pressures to which he must respond, as much in his consumption of goods and services as in any other aspect of this behavior. . .Our general theory, then, should be that individuals and families whose resources over time fall seriously short of the resources commanded by the average individual or family in the community in which they live. . .are in poverty.25

Hence, poverty is a matter of falling behind the prevailing standards of one’s contemporary society.

While relative measures define poverty contextually within one’s time and place, some nevertheless advocate for absolute measures defining poverty independent of any time or place. These absolute measures break sharply with the relative approach. Above, I mentioned that such absolute measures have never been based on some objective list of physiological necessities. Such absolute measures tend to result in obviously problematic thresholds.

The most prominent recent example of this is Bruce Meyer and James Sullivan’s consumption-based absolute measure of poverty.26 In collaboration with the Trump administration’s Council of Economic Advisers and the American Enterprise Institute, they define poverty based on the living standards of the poor in 1980. Applying that standard to recent years, Meyer and Sullivan claim that less than 3 percent of Americans were poor in 2018. They claim they find such low poverty rates because their methods: (a) more comprehensively incorporate taxes and transfers, (b) advantageously use a consumption-based absolute measure, and (c) better adjust for inflation. To be clear, the international poverty literature has already agreed for decades to comprehensively incorporate taxes and transfers. Also, it is perfectly reasonable to consider consumption-based measures or alternative inflation adjustments. One could even go beyond Meyer and Sullivan and actually measure material deprivations like hunger instead of merely on consumption spending.27

However, all of this distracts from what really drives their low, 3-percent poverty rate. The reason they find so little poverty is because they set the threshold absurdly low to begin with. Meyer and Sullivan define a family of four in 2018 in the United States as NOT poor if they consume more than $12,935.28 They arrive at this threshold by starting with the fact that the OPM poverty rate in 1980 was 13 percent. They then move over to consumption data, and find the dollar amount that corresponds to the bottom 13 percent in consumption spending in 1980, which was $5,982. They then adjust that dollar amount of $5,982 for inflation forward to 2018. This is how they come up with the threshold of $12,935 in 2018. They then say if a family in 2018 spends more than $12,935 in 2018, they are not poor. Recall that poverty scholars agree the OPM threshold is far too low. Yet, at $25,100, the OPM threshold for a family of four in 2018 was almost twice as high as Meyer and Sullivan’s threshold. Thus, there really is no justification for Meyer and Sullivan’s threshold.29

Meyer and Sullivan thus define resources and needs and set the threshold in a way that is radically detached from the United States in 2018. Rather than the prevailing standards of today, they simply assert that the definition of needs was set thirty-eight years earlier. They have no basis for why the threshold should be based on what the bottom thirteenth percentile consumed in 1980. The fact that the threshold lacks face validity and is so absurdly low demonstrates precisely why it is preferable to define resources and needs within a time and place.

IV. Relative Measures Better Fit the Leading Theoretical Definitions of Poverty

Any concrete social science measure should be based on a theoretically rigorous definition of the concept to which it pertains, and measures of poverty should be no exception. One reason that relative measures are preferred is that they better fit leading theoretical definitions of poverty. This can be demonstrated by reviewing how several classical theorists define poverty. Then, I will discuss two of the leading contemporary theoretical definitions of poverty: social exclusion and capability deprivation.

Most recent debates about ethics and political philosophy have been influenced by political scientist John Rawls and especially his highly regarded Theory of Justice.30 In his well-known “difference principle,” Rawls proposes that societies should be judged by how the society treats the “least advantaged” or “least fortunate group in society.”31 Rawls was concerned that the least advantaged should be “drawn into the public world and see themselves as full members of it.”32 Rawls clarifies that the difference principle is a matter of relative, not absolute, deprivation, writing that it implies a “social minimum [that] depends on the content of the public political culture.”33 Quite consistent with relative poverty measures, Rawls claims that society could define the “least fortunate group” as those with less than half of the median income and wealth. Rawls says this could form a meaningful poverty standard.34 This social minimum “is not given by the basic needs of human nature taken psychologically (or biologically) apart from any particular social world.”35 Throughout his writings, Rawls is concerned with equal citizenship and how it is threatened by inferiority and deference, inequalities, and social status.36

Preceding Rawls, one can find justifications of relative measures in classical economists. In The Constitution of Liberty, the neo-classical conservative Friedrich Hayek concedes, “Poverty has, in consequence, become a relative, rather than an absolute concept. . .Most of what we strive for are things we want because others already have them.”37 Much earlier than Hayek, in The Wealth of Nations, Adam Smith defines poverty relatively:

Whatever the custom of the country renders it indecent for creditable people, even of the lowest order, to be without. A linen shirt, for example, is, strictly speaking, not a necessary of life. The Greeks and Romans lived, I suppose, very comfortably, though they had no linen. But in present times, through the greater part of Europe, a creditable day-labourer would be ashamed to appear in public without a linen shirt, the want of which would be supposed to denote that disgraceful degree of poverty, which, it is presumed, no body can well fall into without extreme bad conduct. Custom, in the same manner, has rendered leather shoes a necessary of life in England. The poorest creditable person of either sex would be ashamed to appear in public without them. . .In France, they are necessaries neither to men nor to women. . .Under necessaries therefore, I comprehend, not only those things which nature, but those things which the established rules of decency have rendered necessary to the lowest rank of people.38

In a careful study of Smith’s writings, Geoffrey Gilbert explains that Smith defined poverty relatively and that basic necessities matter mostly for status and esteem in one’s community.39 Gilbert highlights that Smith characterizes poverty in terms of psychic pain, social isolation, and social inferiority: “Thus poverty, as addressed by Smith in 1759, did not subject the individual to hunger, malnutrition, disease, lack of clothing or shelter; rather, it shamed him through a pained awareness of his inferior position in the social scale.”40

Turning to contemporary theoretical definitions, “social exclusion” has been a useful concept for defining poverty. In a masterful review, Hilary Silver explains that social exclusion is essentially the opposite of solidarity.41 Social exclusion means marginalization and isolation from one’s community. Others characterize social exclusion as “the multi-dimensional character of disadvantage and exclusion in modern market economies;”42 multiple deprivation or “cumulative misery;”43 those “who suffer from an accumulation of disadvantage which cannot be reached by macro-policies;”44 and, those difficult to reach with social policy.45 Anthony Atkinson exemplifies social exclusion as lacking a telephone in the home: “A person unable to afford a telephone finds it difficult to participate in a society where the majority have telephones.”46 Today we might think of access to the Internet. In sum, social exclusion means incomplete or disadvantaged access to the status, benefits, and experiences of typical citizens in society.47

Social exclusion unites many of the implicit definitions of poverty conveyed by poverty scholars. Social exclusion echoes Harrington’s concern that “the poor are losing their links with the greater world.”48 Atkinson explains that Rawls’s least fortunate were disadvantaged in ways very compatible with social exclusion: disconnected from mainstream society and unable to participate as normal citizens.49 In addition, social exclusion is consistent with William Julius Wilson’s concept of “social dislocation,” which he describes as limited differential opportunities for economic resources, political privileges, organizational influence, and cultural experiences.50 Similarly, in his classic The Affluent Society, John Kenneth Galbraith defines poverty as when people’s income “falls radically behind that of the community.” Galbraith emphasizes that poverty involves more than simply having enough to physically survive, and is better understood as lacking what the “community regards as the minimum necessary for decency.”51

The concept of social exclusion has been pivotal to European measurement and debates about poverty. In 1984, when the European Commission constructed measures of poverty, the Council of Ministers explicitly linked their measures to social exclusion by defining poverty as “persons whose resources are so limited to exclude them from the minimum acceptable way of life in the Member State in which they live.”52 Ever since, and because of this definition, the European Union has monitored member countries for the share of each country’s population “at risk of poverty,” which they define as below 60 percent of the median income.53

Beyond social exclusion, and building on the pioneering work of Sen and Martha Nussbaum, many define poverty with the concept of “capability deprivation.”54 Nussbaum describes capabilities as, “what people are actually able to do and to be.”55 Rod Hick and Tania Burchardt clarify that capability refers to the ability or capacity to function effectively in society.56 Sen’s definition of poverty focuses on the poor’s lack of substantive freedom of choice to achieve valuable “functionings” and the capability to acquire well-being.57 In turn, a functioning member of society must have basic freedoms to participate fully and equally in the mainstream of society.58 The concept of capability also links to Rawls’s argument that basic liberties should be prioritized in a society.59 A society that deprives people of basic liberties can be understood as depriving those people of capability.

Recall that Sen drew a distinction between resources and well-being. He explained that poverty is a shortage of resources to purchase well-being. Because capabilities are the freedom or capacity to pursue one’s well-being, capabilities conceptually sit between resources and well-being. This builds in an intermediate step into the definition of poverty. Some people might have limited capabilities, for example, because of a health condition. Therefore, such people would need greater resources to “convert” their resources into well-being.60 Thus, resources are instruments that need to be converted to enable the capabilities that secure one’s well-being. Despite this added complexity, this approach has productively forced some governments to come up with lists of capabilities that would be (a) compromised by insufficient resources, and (b) are essential for well-being.61

Finally, Atkinson integrates both concepts of social exclusion and capability deprivation by explaining that poverty involves, “people being prevented from participation in the normal activities of the society in which they live or being incapable of functioning.”62 Returning to poverty as a shortage of resources compared to needs, it is helpful to think of capabilities and social inclusion as important needs. This also recognizes how one is poor relative to one’s context and community. Lacking sufficient economic resources compared to others in one’s society results in one suffering from social exclusion and capability deprivation.

V: Relative Poverty Is More Consequential to Health, Well-Being, and Life Chances

If relative measures are superior, they should better distinguish between who is actually deprived and who is not. As I explained earlier, we care about poverty because it produces harm and undermines well-being. Hence, the better poverty measure is the one that is more strongly associated with vital outcomes like health and other aspects of well-being. While there is still a need for further research, the preponderance of extant research strongly shows that a relative measure better predicts and differentiates health, well-being, and life chances.

In the mid-1990s, the National Research Council (NRC), which is the research arm of the National Academy of Sciences, conducted a scientific review of the OPM. As part of that review, they proposed a relative measure of poverty that was a proportion of prevailing levels of consumption. With a few modifications, the proposed measure is similar to what became the Supplemental Poverty Measure (SPM).63 A few studies have analyzed how well the NRC’s proposed relative measure (like the SPM) predicts outcomes compared to the OPM. For instance, Carolyn Hill and Robert Michael demonstrate that the NRC’s measure is more strongly associated with a variety of children’s outcomes: school grades and suspensions, achievement test scores, expectations of completing college, and avoiding pregnancy.64

In a related literature, health scholars assess if relative or absolute income is more strongly associated with health outcomes.65 One conclusion of this literature is that absolute deprivation only matters up to a point. Once countries are developed, relative rather than absolute deprivation becomes most salient to well-being. For instance, Malavika Subramanyam and colleagues use the U.S. Current Population Surveys and find that beyond absolute income, individual relative deprivation in income and position in the income hierarchy are associated with poor health.66 A few demonstrate that subjective perceptions of relative deprivation undermine health.67 Several show that inequality or relative poverty matter to health and well-being aside from a country’s level of economic development or level of poverty.68 Scholars have also shown that even though having a higher personal income is beneficial, health and well-being are still undermined by residing in a community with a higher mean income.69 Thus, one’s relative standing within one’s local context matters to one’s health beyond one’s own absolute income.

In a recent working paper, this author and colleagues provide a novel, rigorous test of the comparative effects of relative and absolute income for health and well-being.70 In particular, we uniquely exploit variation over time within individuals to assess how well absolute and relative income respectively are associated with the following: self-rated health, self-rated poor health, psychological distress, high psychological distress, and life satisfaction. Using the Panel Study of Income Dynamics, which is the longest running nationally representative dataset following people over time, enables us to examine individuals over a fairly long period while adjusting for age and time period differences. Our results unequivocally show that relative income is more important than absolute income to health and well-being. We find that relative income is significantly associated with all five outcomes, while absolute income is only significantly associated with self-rated health and poor health. Furthermore, relative income has substantially larger effects than absolute income for all outcomes.

In sharp contrast with the research demonstrating how relative measures better predict well-being, there has been remarkably little evidence that absolute measures are superior. Instead of actually supplying empirical evidence that absolute measures better differentiate between poor and non-poor, proponents of absolute measures tend to merely assert, without evidence, that their measures are better. For instance, Meyer and Sullivan simply claim that consumption is “well-being,” and then declare there is little poverty because low-income families consume more than in the past. Again, it needs to be stressed that one can certainly use consumption as the measure of resources and still use a relative measure of poverty. Indeed, the SPM is a relative measure because the threshold is based mostly on the bottom 33 percent of consumption expenditures. Again, one major problem is that Meyer and Sullivan’s particular consumption-based measures are based on absurdly low absolute thresholds.

As mentioned above, Meyer and Sullivan find less than 3 percent of Americans are poor because they define a family of four as not poor if they consume at or above $12,935 in 2018. However, they never provide any evidence that this particularly low absolute threshold demarcates who has sufficient resources to purchase well-being.71 In other words, Meyer and Sullivan never empirically compare their measure against a relative measure, and show their measure is more associated with health, well-being, and life chances. Indeed, they never even engage the literature described above that shows that relative measures are more strongly associated with those outcomes.

It should also be noted that if one is really concerned with measuring absolute deprivation, there is a far more productive approach in Europe that examines material deprivation.72 This approach actually measures concrete indicators of being unable to “consume” what one needs—such as hunger or insufficient living conditions—and is likely to be far more useful than focusing on consumption spending.73 Ultimately, advocates of absolute measures have never provided evidence that absolute measures are superior to relative measures.

Conclusion

This report contends that relative measures of poverty are superior to alternatives. Relative measures are anchored in the here and now of one’s time and place rather than some distant past such as the 1960s. Extensive literature on poverty measurement, the overwhelming majority of international poverty researchers, and most international organizations support relative measures over the alternatives. This report reviews and catalogues the many and varied justifications for relative measures. In the process, it critiques absolute measures and endorses abandoning the OPM. The report aims to provide a clear set of arguments, hopefully without too much academic jargon, for why one should prefer a relative measure of poverty.

This essay makes its case across five sections that each cover a major theme in building the argument for relative measures. First, I provide a clear and simple definition of poverty as a shortage of resources compared to needs. Second, I contend that resources and needs must be defined within the social context of a time and place. Third, I explain that relative measures define poverty according to the prevailing standards of a time and place. Fourth, I show how relative measures better fit leading theoretical definitions of poverty. Fifth, I review evidence showing that relative measures of poverty are more strongly associated with health, well-being and life chances.

Public policy, as well as the social sciences, would benefit from further objective tests of whether absolute or relative measures better predict health, well-being, and life chances. Advocates of absolute measures often simply assert their measures are superior without presenting empirical evidence that vindicates absolute measures versus relative measures. This is puzzling, considering that existing objective evidence strongly favors relative measures. Hence, beyond offering a theoretical justification and arguments for relative measures, hopefully this piece can encourage subsequent research that objectively adjudicates between relative and absolute measures. Unless such objective tests show convincingly otherwise, the most important conclusion is that existing social science evidence and theoretical arguments strongly favors relative measures over alternatives. Poverty in America should be measured relatively.

Notes

  1. David Brady and Linda M. Burton, The Oxford Handbook of the Social Science of Poverty (New York: Oxford University Press, 2016); Luxembourg Income Study (LIS) Database, http://www.lisdatacenter.org (United States 2018; key figures consulted March 8, 2021).
  2. David Brady and Linda M. Burton, The Oxford Handbook of the Social Science of Poverty (New York: Oxford University Press, 2016).
  3. Luxembourg Income Study (LIS) Database, http://www.lisdatacenter.org (United States 2018; key figures consulted March 8, 2021). For comparison, a single mother making $15 per hour for forty hours in a week and fifty weeks in a year would make $30,000.
  4. Sort of in between relative and absolute measures are what is called “anchored” measures. An anchored measure defines relative poverty in a given year and then updates for subsequent years with inflation alone. See Timothy Smeeding, “Poverty Measurement,” The Oxford Handbook of the Social Science of Poverty, edited by D. Brady and L.M. Burton (Oxford: Oxford University Press, 2016).
  5. Shawn Fremstad, “The Defining Down of Economic Deprivation: Why We Need to Reset the Poverty Line,” Bernard L. Schwartz Rediscovering Government Initiative, Bernard L. Schwartz Rediscovering Government Initiative, 2020, https://tcf.org/content/report/defining-economic-deprivation-need-reset-poverty-line/.
  6. Timothy Smeeding, “Poverty Measurement,” The Oxford Handbook of the Social Science of Poverty, edited by D. Brady and L.M. Burton (Oxford: Oxford University Press, 2016).
  7. Shawn Fremstad, “The Defining Down of Economic Deprivation: Why We Need to Reset the Poverty Line,” Bernard L. Schwartz Rediscovering Government Initiative, Bernard L. Schwartz Rediscovering Government Initiative, 2020, https://tcf.org/content/report/defining-economic-deprivation-need-reset-poverty-line/.
  8. Amartya Sen, Inequality Reexamined (Cambridge, MA: Harvard University Press, 1992); Amartya Sen, Development as Freedom (New York: Anchor Books, 1999).
  9. Amartya Sen, Inequality Reexamined (Cambridge, MA: Harvard University Press, 1992), 110.
  10. Current income has many valuable attributes as a measure of resources. Among other reasons, current income (including taxes and transfers and adjusting for household size) proxies long-term economic resources far better than do alternative measures of current resources; see David Brady, Marco Giesselmann, Ulrich Kohler, and Anke Radenacker, “How to Measure and Proxy Permanent Income: Evidence from Germany and the U.S.,” Journal of Economic Inequality, volume 16, 2018, 321–345, https://doi.org/10.1007/s10888-017-9363-9. Most poverty researchers prefer income, while some prefer assets or assets plus income. The OPM focuses only on cash income, whereas international researchers agree that income should also include “near cash” transfers like SNAP and tax credits like the EITC. Ideally, one would find a way to monetize and assess the value of any resource a household has at its disposal. For instance, imagine a family has a grandmother living with them and the grandmother provides free child care. That free labor is a resource, and that family would have more resources than a family with the same income and without the free labor. Hence, accurately counting the family’s resources would mean we were able to monetize and calculate the value of all resources. As this example illustrates, however, any measure of resources can only approximate the true resources a family has at its disposal. Because any measure of resources is imperfect, we strive for a measure of resources that is as comprehensive as possible.
  11. Poverty researchers have not found a consensus on an objective list, but it should be noted that typical people can get closer to coming up with a workable list. For instance, in the United Kingdom, some have advocated for a minimum income standard (e.g. https://www.jrf.org.uk/report/minimum-income-standard-uk-2020). Notably, such minimum income standards would put the poverty threshold far higher than that set by the OPM in the United States.
  12. Despite popular impressions, the standard of needs underlying the OPM does not actually have a scientific basis. See David Brady, Rich Democracies, Poor People (New York: Oxford University Press, 2009); Shawn Fremstad, “The Defining Down of Economic Deprivation: Why We Need to Reset the Poverty Line,” Bernard L. Schwartz Rediscovering Government Initiative, Bernard L. Schwartz Rediscovering Government Initiative, 2020, https://tcf.org/content/report/defining-economic-deprivation-need-reset-poverty-line/; Michael B. Katz, The Undeserving Poor (New York: Pantheon, 1989); and Alice O’Connor, Poverty Knowledge (Princeton, NJ: Princeton University Press, 2001). There was never a strong justification for multiplying food times three. Using data from the mid-1950s, there was evidence that food amounted to roughly one-third of expenses for typical households on average. The evidence was not clear that this applied to low-income households. Further, the Johnson administration ended up using the “economy food plan,” which was about 25 percent below the “low-cost food budget” used by Orshansky; seeMichael B. Katz, The Undeserving Poor (New York: Pantheon, 1989). The economy food plan was meant for emergencies and on a temporary basis. Also, the food budgets were not subsequently revised. A few years later, the government began updating the OPM thresholds using the consumer price index rather than calibrating the thresholds according to changing food budgets. This had the consequence of severing any tie to the food budget as a standard of needs. Indeed, Katz quotes Orshansky as writing: “This meant, of course, that the food–income relationship which was the basis for the original poverty measure no longer was the current rationale;” see Michael B. Katz, The Undeserving Poor (New York: Pantheon, 1989), 116. Moreover, as is well known, food is certainly much less than one-third of household expenses today. As a result, the OPM effectively ignores the costs of important household needs like child care and health care, which were less essential or much cheaper when the OPM was created.
  13. Patricia Ruggles, Drawing the Line: Alternative Poverty Measures and Their Implications for Public Policy (Washington, D.C.: The Urban Institute Press, 1990).
  14. David Brady and Zachary Parolin, “The Levels and Trends in Deep and Extreme Poverty in the U.S., 1993-2016,” Demography, volume 57, 2020, 2,337–2,360, https://doi.org/10.1007/s13524-020-00925-0..
  15. Aldi J.M. Hagenaars, “The Definition and Measurement of Poverty,” Economic Inequality and Poverty: International Perspectives (Armonk, NY: M.E. Sharpe, 1991), 141.
  16. Aldi J.M. Hagenaars, “The Definition and Measurement of Poverty,” Economic Inequality and Poverty: International Perspectives (Armonk, NY: M.E. Sharpe, 1991), 146.
  17. Peter Townsend, “Research on Poverty,” Wealth, Income and Inequality, edited by A. B. Atkinson (New York: Oxford University Press, 1980), 299–306, 300.
  18. Lee Rainwater and Timothy M. Smeeding, Poor Kids in a Rich Country (New York: Russell Sage Foundation, 2004), 9.
  19. As Smeeding argues, “In the broadest sense, all measures of poverty or economic need are relative, because context is important to the definition of even ‘absolute’ needs.” See Timothy Smeeding, “Poverty Measurement,” The Oxford Handbook of the Social Science of Poverty, edited by D. Brady and L.M. Burton (Oxford: Oxford University Press, 2016), 27.
  20. This simple definition can be easily extended to deep and extreme poverty as well. As Parolin and the author have put it elsewhere, “To say that individuals are in ‘deep’ or ‘extreme’ poverty simply means that the gap between resources and needs is deeper or more extreme.” David Brady and Zachary Parolin, “The Levels and Trends in Deep and Extreme Poverty in the U.S., 1993-2016,” Demography, volume 57, 2020, 2,344, https://doi.org/10.1007/s13524-020-00925-0.
  21. Peter Townsend, “Research on Poverty,” Wealth, Income and Inequality, edited by A. B. Atkinson (New York: Oxford University Press, 1980), 299–306.
  22. John Hills, Inequality and the State (New York: Oxford University Press, 2004); Lee Rainwater and Timothy M. Smeeding, Poor Kids in a Rich Country (New York: Russell Sage Foundation, 2004).
  23. John Hills, Inequality and the State (New York: Oxford University Press, 2004).
  24. Michael Harrington, The Constitution of Liberty (Chicago: University of Chicago Press, 1960), 18, 187–188.
  25. Peter Townsend, “The Meaning of Poverty,” British Journal of Sociology, volume 13, 1962, 219 and 225).
  26. Bruce D. Meyer and James X. Sullivan, “Annual Report on U.S. Consumption Poverty: 2018,” October 18, 2019, https://leo.nd.edu/assets/339909/2018_consumption_poverty_report_1_.pdf.
  27. Basak Kus, Brian Nolan, and Christopher T. Whelan, “Material Deprivation and Consumption,” The Oxford Handbook of the Social Science of Poverty, edited by D. Brady and L.M. Burton (New York: Oxford University Press, 2016), 577–601.
  28. They also use the more conservative inflation adjustment CPI-URS. As with other absolute measures like the OPM, there is zero objective and scientific basis for saying the bottom 13 percent of the consumption expenditure distribution is any sort of meaningful threshold. Indeed, they provide no real evidence at all for this arbitrarily chosen and rather absurdly low threshold. See Bruce D. Meyer and James X. Sullivan, “Winning the War: Poverty from the Great Society to the Great Recession,” Brookings Papers on Economic Activity, Fall, 2012, 133–200, https://www.brookings.edu/wp-content/uploads/2012/09/2012b_meyer.pdf.
  29. It is noteworthy that one has to dig into appendices to find their thresholds, as they do not make explicit and clear how low their thresholds are. It seems reasonable to conclude they do not want to make their thresholds transparent to the reader because it would be quite difficult to defend them. Indeed, one has to read other articles of theirs (e.g. Bruce D. Meyer and James X. Sullivan, “Winning the War: Poverty from the Great Society to the Great Recession,” Brookings Papers on Economic Activity, Fall, 2012, 133–200, https://www.brookings.edu/wp-content/uploads/2012/09/2012b_meyer.pdf) to fully understand their measure, because their presentation is opaque, unclear, and obfuscating in their 2018 report. Their 2018 report does offer another threshold based on the poor’s living standards in 2015. But even their alternative higher thresholds presume a family of four in 2018 is not poor with $20,000. All these thresholds are far below the OPM thresholds, which are already well-known to be too low. See Shawn Fremstad, “The Defining Down of Economic Deprivation: Why We Need to Reset the Poverty Line,” Bernard L. Schwartz Rediscovering Government Initiative, Bernard L. Schwartz Rediscovering Government Initiative, 2020, https://tcf.org/content/report/defining-economic-deprivation-need-reset-poverty-line/.
  30. Particularly important to bringing Rawls into poverty measurement are the economists Anthony Atkinson and Amartya Sen. Despite Rawls’s influence on poverty and inequality debates, he made very few overt references to poverty; see Anthony B. Atkinson, “Chapter One: Social Exclusion, Poverty and Unemployment,” Centre for Analysis of Social Exclusion, case paper no. 4, January, 1998; Amartya Sen, Inequality Reexamined (Cambridge, MA: Harvard University Press, 1992); Amartya Sen, Development as Freedom (New York: Anchor Books, 1999); and John Rawls, A Theory of Justice, revised edition (Cambridge, MA: Harvard University Press, 1971, 1999). Atkinson notes that the word “poverty” does not even appear in Rawls’s extensive index (John Rawls, A Theory of Justice, revised edition (Cambridge, MA: Harvard University Press, 1971, 1999)); Anthony B. Atkinson, “On the Measurement of Poverty,” Econometrica, volume 55, 1987, 749–764. He discusses poverty a bit more later (e.g. John Rawls, Justice as Fairness: A Restatement (Cambridge, MA: Harvard University Press, 2001)), but mainly one needs to rely on his discussion of economic inequality.
  31. In his related “veil of ignorance,” Rawls argues that people should judge society as if they could be in that least fortunate group. According to Rawls, if that veil of ignorance is in place, we would choose societies that are most fair to that least fortunate group.
  32. John Rawls, Justice as Fairness: A Restatement (Cambridge, MA: Harvard University Press, 2001, 129–130).
  33. John Rawls, Justice as Fairness: A Restatement (Cambridge, MA: Harvard University Press, 2001), 132, emphasis added.
  34. John Rawls, A Theory of Justice, revised edition (Cambridge, MA: Harvard University Press, 1971, 1999), 84.
  35. John Rawls, Justice as Fairness: A Restatement (Cambridge, MA: Harvard University Press, 2001, 132, emphasis added.
  36. One could use Rawls’s maximin criterion to argue that Rawls endorsed absolute measures of poverty. But, Rawls (e.g. John Rawls, Justice as Fairness: A Restatement (Cambridge, MA: Harvard University Press, 2001), 68) wrote much about how justice depends on democratic equality and that this was compromised where there were large differences between classes. For example, he wrote, “Significant political and economic inequalities are often associated with inequalities of social status that encourage those of lower status to be viewed both by themselves and by others as inferior. . .It is close to being wrong or unjust in itself that in a status system, not everyone can have the highest rank. Status is a positional good, as is sometimes said. High status assumes other positions beneath it; so if we seek a higher status for ourselves, we in effect support a scheme that entails others’ having a lower status.” See John Rawls, Justice as Fairness: A Restatement (Cambridge, MA: Harvard University Press, 2001), 131.
  37. Emphasis added. Friedrich Hayek, The Constitution of Liberty (Chicago: University of Chicago Press, 1960), 44-45.
  38. Adam Smith, An Inquiry into the Nature and Causes of the Wealth of Nations (New York: The Modern Library, Random House, 1937; originally 1776), 821–822).
  39. Emphasis added, Geoffrey Gilbert, “Adam Smith on the Nature and Causes of Poverty,” Review of Social Economy, volume 55, 1997, 274.
  40. Geoffrey Gilbert, “Adam Smith on the Nature and Causes of Poverty,” Review of Social Economy, volume 55, 1997, 275.
  41. Hilary Silver, “Social Exclusion and Social Solidarity: Three Paradigms,” International Labour Review, volume 133, 531–578.
  42. Bea Cantillion, “The Challenge of Poverty and Exclusion,” Social Policy Studies, no. 21: “Family, Market, and Community: Equity and Efficiency in Social Policy,” 1997, 115–116 and 130.
  43. C. Schuyt and A. Tan, “De Maatschappelijke Betekenis Van Armoede. Deel II,” Op Zoek Naar Armoede En Bestaansonzekerheid Langs Twee Sporen, Nationale Raad Voor Maatschappelijk Welzijn, Rijswijk, 14.
  44. Ralf Dahrendorf, The Modern Social Conflict – An Essay on the Politics of Liberty (Berkeley, CA: The University of California Press, 1990), 151.
  45. G. Engbersen, “Moderne Armoede: Feit En Fictie,” Sociologische Gids, volume 37, 1991, 7–23.
  46. Anthony B. Atkinson, “Chapter One: Social Exclusion, Poverty and Unemployment,” Centre for Analysis of Social Exclusion, case paper no. 4, January, 1998), 20.
  47. Charles Gore, “Chapter 1 Introduction: Markets, Citizenship and Social Exclusion,” Social Exclusion: Rhetoric Reality Responses, edited by Gerry Rodgers, Charles Gore, and Jose B. Figueiredo (Geneva: International Labour Organization, 1995), 1–40.
  48. Michael Harrington, The Constitution of Liberty (Chicago: University of Chicago Press, 1960), 11.
  49. Anthony B. Atkinson, “On the Measurement of Poverty,” Econometrica, volume 55, 1987, 749–764.
  50. William Julius Wilson, “Studying Inner-City Social Dislocations: The Challenge of Public Agenda Research,” American Sociological Review, volume 56, 1991, 1–14, https://doi.org/10.2307/2095669.
  51. John Kenneth Galbraith, The Affluent Society (New York: Mariner Books, 1958), 235.
  52. Anthony B. Atkinson, Poverty in Europe (Malden, MA: Blackwell, 1998), 2.
  53. Hilary Silver, “Social Exclusion and Social Solidarity: Three Paradigms,” International Labour Review, volume 133, 531–578; Lee Rainwater and Timothy M. Smeeding, Poor Kids in a Rich Country (New York: Russell Sage Foundation, 2004).
  54. Amartya Sen, Inequality Reexamined (Cambridge, MA: Harvard University Press, 1992); Amartya Sen, Development as Freedom (New York: Anchor Books, 1999)); Martha C. Nussbaum, “Poverty and Human Functioning: Capabilities as Fundamental Entitlements,” Poverty and Inequality, edited by D.B. Grusky and R. Kanbur (Redwood City, CA: Stanford University Press, 2006), 47–75.
  55. Martha C. Nussbaum, “Poverty and Human Functioning: Capabilities as Fundamental Entitlements,” Poverty and Inequality, edited by D.B. Grusky and R. Kanbur (Redwood City, CA: Stanford University Press, 2006), 49).
  56. Rod Hick and Tania Burchardt, “Capability Deprivation,” The Oxford Handbook of the Social Science of Poverty, edited by D. Brady and L.M. Burton (Oxford University Press, 2016), 75).
  57. Although Sen argued that there was value in absolute poverty measures when studying developing countries, he was clear that poverty should be measured relatively in developed countries like the U.S. Indeed, Sen stressed that the relatively poor in rich countries are capability deprived, even though they are well off in comparison to most of the people in the world. Sen explains, “Relative deprivation in terms of incomes can yield absolute deprivation in terms of capabilities” (emphasis in original). See Amartya Sen, Development as Freedom (New York: Anchor Books, 1999), 89.
  58. Brian Barry, “Social Exclusion, Social Isolation and the Distribution of Income,” case paper no. 12, Centre for Analysis of Social Exclusion, London School of Economics, 1998.
  59. John Rawls, A Theory of Justice, revised edition (Cambridge, MA: Harvard University Press, 1971, 1999); see also Anthony B. Atkinson, “On the Measurement of Poverty,” Econometrica, volume 55, 1987, 749–764.
  60. Rod Hick and Tania Burchardt, “Capability Deprivation,” The Oxford Handbook of the Social Science of Poverty, edited by D. Brady and L.M. Burton (Oxford University Press, 2016), 75–92.
  61. Rod Hick and Tania Burchardt, “Capability Deprivation,” The Oxford Handbook of the Social Science of Poverty, edited by D. Brady and L.M. Burton (Oxford University Press, 2016), 85–86).
  62. Anthony B. Atkinson, Poverty in Europe (Malden, MA: Blackwell, 1998), 27.
  63. Shawn Fremstad, “The Defining Down of Economic Deprivation: Why We Need to Reset the Poverty Line,” Bernard L. Schwartz Rediscovering Government Initiative, Bernard L. Schwartz Rediscovering Government Initiative, 2020, https://tcf.org/content/report/defining-economic-deprivation-need-reset-poverty-line/.
  64. Carolyn J. Hill and Robert T. Michael, “Measuring Poverty in the NLSY97,” Journal of Human Resources, volume 36, 2001, 727–761, https://doi.org/10.2307/3069640.
  65. Michael Marmot, The Status Syndrome (New York: Owl Books, 2004).
  66. Malavika Subramanyam, Ichiro Kawachi, Lisa Berkman, and S.V. Subramanian, “Relative deprivation in income and self-rated health in the United States,” Social Science & Medicine, volume 69, 327–334, https://doi.org/10.1016/j.socscimed.2009.06.008.
  67. Sandeep Mishra and R. Nicholas Carleton, “Subjective Relative Deprivation Is Associated With Poorer Physical and Mental Health,” Social Science & Medicine, volume 147, 144–149.
  68. Johan Fritzell, Johan Rehnberg, Jennie Bacchus Hertzman, and Jenni Blomgren, “Absolute or Relative? A Comparative Analysis of the Relationship Between Poverty and Mortality,” International Journal of Public Health, volume 60, 2014, 101–110, https://doi.org/10.1007/s00038-014-0614-2; Marianne M. Hillemeier, John Lynch, Sam Harper, Trivellore Raghunathan, and George A. Kaplan, “Relative or Absolute Standards for Child Poverty: A State-Level Analysis of Infant and Child Mortality,” American Journal of Public Health, volume 93, 2003, 652–657, doi:10.2105/ajph.93.4.652; Jon M. Jachimowicz, Barnabas Szaszi, Marcel Lukas, David Smerdon, Jaideep Prabhu, and Elke U. Weber, “Higher Economic Inequality Intensifies the Financial Hardship of People Living in Poverty By Fraying the Community Buffer,” Nature Human Behavior, volume 4, 2020, 702–712.
  69. Alpaslan Akay and Peter Martinsson, “Does Relative Income Matter for the Very Poor? Evidence from Rural Ethiopia.” Economics Letters, volume 110, 2011, 213–215; Mohammad Niaz Asadullah and Nazmul Chaudhury. 2012. “Subjective Well-Being and Relative Poverty in Rural Bangladesh.” Journal of Economic Psychology volume 33, 2012, 940–950; Mary C. Daly, Daniel J. Wilson, and Norman J. Johnson, “Relative Status and Well-Being: Evidence from U.S. Suicide Deaths,” Review of Economics and Statistics, volume 95, 2013, 1480–1500.
  70. David Brady, Michaela Curran, and Richard Carpiano, “Relative Income Is More Important Than Absolute Income for Self-Reported Health and Well-Being,” working paper, 2021.
  71. The closest they get is to show cross-sectional differences between income and consumption poverty for some measures of deprivation and well-being. They mainly concentrate on showing how those “misclassified” with one or the other measure would differ in terms of well-being. For instance, they try to show that those defined as poor with an income-based measure but not poor with the consumption-based measure might have modestly better well-being than those defined as poor with the consumption measure. However, these comparisons fail to provide an adequate test of absolute versus relative measures of poverty. These comparisons also appear quite cherry-picked (e.g. it is not at all clear why certain well-being measures are shown and many alternatives in the datasets are not shown).
  72. Basak Kus, Brian Nolan, and Christopher T. Whelan, “Material Deprivation and Consumption,” The Oxford Handbook of the Social Science of Poverty, edited by D. Brady and L.M. Burton (New York: Oxford University Press, 2016), 577–601.
  73. Meyer and Sullivan occasionally declare and assert that consumption better reflects permanent income or long term economic resources than income. This reflects a widely rehearsed claim by many that refers back to Milton Friedman’s permanent income hypothesis. However, there is remarkably little—indeed, perhaps zero—evidence that consumption actually better proxies permanent income than income. In one of the few actual empirical studies using long-term panel data from both Germany and the United States, the author and colleagues find that current income predicts permanent income far better than a wide variety of alternative proxies; see David Brady, Marco Giesselmann, Ulrich Kohler, and Anke Radenacker, “How to Measure and Proxy Permanent Income: Evidence from Germany and the U.S.,” Journal of Economic Inequality, volume 16, 2018, 321–345, https://doi.org/10.1007/s10888-017-9363-9. Meyer and Sullivan have never provided evidence that consumption actually proxies permanent income better than current income.

Tags: ,