According to legend, Michael Harrington’s infamous 1962 book on poverty, The Other America, was launched on a nation busily celebrating its universal affluence and confronted it with all the poverty hidden in its midst. It was always a little hard to believe this tale. Poor people are never very far away in the physical sense, even if they were written out of public discourse. But no one important thought it worth the trip. Nobody wants to make that trip today, either. Defining poverty remains a slippery issue, rife with politics and based on what, at best, could be described as shaky scientific foundations.
Four decades ago, when Harrington’s book came out, The Poor didn’t exist in any official sense. It wasn’t until the Johnson administration, which launched a war on poverty as a domestic echo of its war on Vietnam, that the government unleashed its formidable statistical machinery on the problem. Just like McNamara’s whiz kids deployed state-of-the-art information science on the Viet Cong, the domestic warriors needed some good definitions and numbers.
Though the U.S. didn’t get an official poverty line until 1965, the concept has a long history. The broad project of studying household budgets of the poor and working class goes back at least to the 1870s, a project motivated in large part by fear of unrest from below. It was hoped that the collection of statistics might lead to an amelioration in the condition of the poor — and calm potential unrest — but of course there’s no simple relation between information-gathering and policy. Now we know a great deal about the composition and incomes of the officially poor, but all that knowledge doesn’t inspire the powerful to do much but moralize.
While upper class interest in the welfare of the lower classes was motivated largely by self-preservation, much of the actual research was done by advocates for the poor, ranging from paternalistic liberals to revolutionary socialists. A disproportionate number of them were women — something that continues to be the case — and much of the work has been done not by economists, most of whom are interested in things far more elevated than human welfare, but by sociologists, social workers, union researchers, and government statisticians.
Initially, “poverty” was defined not as material deprivation, but dependence on charity. But around the turn of the last century, the definition shifted towards the lack of income. Around the same time there was a shift in prevailing theories about the causes of poverty, from individual moral failings of the poor to structural economic and social reasons.
Curiously, poverty theory and practice over the last 20 years has revived 19th century thinking, as conservative theorists like Charles Murray and George Gilder revived moral explanations. But liberals did their part, emphasizing not the tendency of the U.S. economy to produce lots of crummy jobs, but alleged skills deficiencies among the indigent. Also, with the end of welfare — thanks in large part to a Democratic president, Bill Clinton, and not just right-wing troglodytes like Murray — during the late 1990s, interest shifted to getting the poor off the welfare rolls. Most poor people remain quite poor after they’re kicked off welfare, but no one in a position of intellectual or political authority really cares.
Many pre-World War II efforts to devise a poverty line were based on assembling a basic market basket of goods and services necessary to keep body, if not soul, together, with minimal food budgets playing a prominent role. They varied considerably in generosity, ranging from the bare mimimum necessary to sustain life to something approaching a reasonable degree of comfort. Often researchers strategically opted for the minimal, so as not to frighten the better sorts with their generosity. During the years immediately after World War II, however, most poverty lines were just dollar figures plucked more or less out of the air.
The most influential of these efforts turned out to be those of Mollie Orshansky, a researcher with the Social Security Administration, who published an article in the July 1963 of the Social Security Bulletin describing a set of proposed poverty thresholds for families with children. Orshansky didn’t intend to devise a formal poverty line; she was merely trying to assess the relative opportunities for families of varying sizes.
Her basic technique was simple. The Department of Agriculture had long published several food budgets at varying levels of extravagance. The two Orshansky looked at were called the “economy” and “low-cost,” the economy version being extremely minimal, intended by its designers as a temporary emergency standard for families on hard times; it allowed just under 25 cents per meal in 1964 dollars, which would amount to about $1.38 today. The low-cost plan was only slightly more generous, allowing less than a third more, or $1.78 per meal in today’s dollars.
Clearly, it would require extreme discipline and ingenuity to survive on that kind of budget. (Today’s low-end food budgets are almost identical; the Department’s “thrifty” food budget, which is what Food Stamp recipients are expected to live on, allows $1.41 a meal, less than the cost of a frappuccino.) Since the Ag Department also had research — dating from 1955 — showing that the average family spent a third of its income on food, a poverty line might be conceived of as three times the minimal food budget.
Orshansky followed up her 1963 paper with another published in 1965. Preprints of that paper were circulating around Washington in late 1964, just as the Johnson administration was looking for figures to set a poverty line. Of course, they used the lower of her estimates — the one based on the painfully stringent “economy” plan — and multiplied that level times three. They ignored more recent surveys showing that the share of income the average household devoted to food had declined from a third to a quarter, suggesting that the multiplier should be four, not three, which would translate into a significantly higher poverty line. The original low Orshansky numbers were adopted in 1965 (though projected backwards starting in 1959), and, with minor modifications, have simply been adjusted for inflation ever since. Conceptually, today’s poverty line is designed to represent the same purchasing power as the day it was adopted, despite broadly rising incomes. The poverty line for a family of three was 43% of median income in 1959; in 2001, it was 27% of the median.
In the jargon of the trade, the U.S. poverty line is an “absolute” one, based on a purchasing power fixed at a point in time and adjusted only for inflation over the years. The alternative is a “relative” measure, one defined as a percentage of average incomes that grows along with incomes.
A relative measure has several advantages. First, price indexes are notoriously devilish to work with: the recent controversy over the alleged overstatement of inflation by the consumer price index brought some of these difficulties into plain view. How do you adjust for improvements in quality? If the price of an item goes up, but its quality improves, did it really get more expensive? If the price of apples goes up, and that of pears goes down, what does that do to the price of fruit? The problems multiply over the long term. There were no DVDs or PDAs in 1965 -- what does their development mean to the changed cost of living 35 years later? A relative measure avoids all these problems.
And second, relative measures comport much more closely with the way people perceive themselves. People evaluate themselves and others as poor or nonpoor by comparing themselves to their neighbors, not to some number invented by statisticians 40 years ago and adjusted by another set of statisticians every year. Suuport for this point of view comes from no less revered an authority than Adam Smith. The father of laissez-faire economics said in 1776 that poverty was characterized by the want of “necessaries,” which he in turn defined as “not only the commodities which are indispensably necessary for the support of life, but whatever the custom of the country renders it indecent for creditable people, even of the lowest order, to be without.” Ironically, Smith’s self-identified intellectual descendants, who generally prefer wearing images of him on their neckties to reading his words, almost always reject this in favor of absolute definitions. Some conservatives even argue that we have no poor people in the U.S. today, if you use the 19th century urban proletariat as your standard.
It’s common among academic researchers to draw the poverty line at half the national median income. This habit is confirmed by results from Gallup polls, which periodically ask people what their subjective impressions of a poverty line are; the results come pretty close to the half-the-median standard (which is about where Orshansky’s original line fell at the time). If we computed poverty this way, the rates would be one-and-a-half to two times as high as they are by official measures — a gap that would widen over time, because a relative poverty line would rise with average incomes, while the absolute one stays stuck in 1959.
With all these caveats, what do those official measures look like? In 2001, there were 33 million officially poor people in the U.S., almost 12% of the population. That was up a bit from the previous year, the last year of the great boom. That boom did lower the poverty rate, which was over 15% in 1993, at the end of the first Bush recession. So, the boom helped remove about six or seven million people from the ranks of the certified destitute. As nice as that sounds, it only brings us back to levels reached in the late 1970s, before the early 1980s recession drove it up. The 1980s expansion brought it down a bit, to around 13% by the end of the decade, but the 1990s recession drove it back up over 15% again. The 1990s boom brought it down, but now we’re heading northwards again.
There were a lot of numbers in that preceding paragraph, dense enough to numb the numerophobic. So let me put it this way: the poverty rate, even by the rather nondemanding official definition, was about the same in 2001 as it was in 1978, even though the size of the U.S. economy doubled over the same period. And this was a time when the U.S. was hailed as the economic model for the world to envy. Worse, now we know that its modest achievements in bringing down poverty in the late 1990s were the fortunate byproducts of an unsustainable speculative bubble.
Of course, unlike in Harrington’s day, we have plenty of information about poverty. We know, for example, that women are more likely to be poor than men; that families with children are much more likely to be poor; that blacks are three times as likely to be poor as whites; and that female-headed families are three times as likely to be poor as other kinds.
And we have a pretty good idea how much it would take to bring the destitute up to the poverty line: about $56 billion, or 0.5% of GDP. That’s roughly what the Pentagon spends in two months. Or just 3% of the income of the richest fifth of households (those, that is, with incomes above $84,000). Or, if America suddenly came down with a burst of generosity, we could double the incomes of the poorest fifth of households — a closer approximation of who’d be poor under a more reasonable standard than the official measure — by appropriating a mere 7% of the upper fifth’s income, which would bring them down from an average of $146,000 to $136,000. That would reduce their share of the national pie to 1991’s levels, hardly a period when the upper orders were suffering. But clearly it’s much more important that the affluent be able to buy Lincoln Navigators ($53,480 MSRP).
