Poverty Rates: How a Flawed Measure Drives Policy

Poverty Rates: How a Flawed Measure Drives Policy

Justin Sullivan/Getty Images

On November 7, the Census Bureau published new estimates of the poverty rate based on experimental definitions. Calculation of the official poverty rate, which was released in September 13, remains unchanged. Conservatives charge that the Census Bureau is trying to inflate the poverty rate and ignores indicators showing improvement in the economic condition of the poor.

It’s important to remember that the official poverty rate came about largely by happenstance and was not the result of a carefully thought through analysis. An economist named Mollie Orshansky at the Social Security Administration made the first estimate of the poverty rate in 1963. Ms. Orshansky, who died in 2007, had some data from the Department of Agriculture on food budgets for families of different sizes and incomes. She saw that food constituted about a third of spending by poor families and thus assumed that three times the budget for food would approximate the poverty rate.

This back of the envelope calculation was seized upon by the White House under Lyndon Johnson, which turned Orshansky’s figure into the official measure of poverty. Since that time, the original poverty measure of $3,000 for a family of four has simply been increased by the rate of inflation. In 2010, the official poverty threshold for a family of four (two adults and two children) was $22,113.  

Obviously, this method of measuring poverty is extremely problematic, regardless of one’s political philosophy. Liberals typically point to these problems.

  • The poverty rate is not adjusted for differences in the geographical cost of living. Everyone knows that it costs more to live in big cities than in rural areas and it’s cheaper to live in some regions of the country than others. But the Census Bureau assumes that the poverty threshold is the same everywhere.
  • Patterns of consumption have changed over time. The real cost of food has fallen while the cost of housing, energy and health care has risen. Moreover, many families today have expenses that didn’t exist back when the poverty rate was established. For example, few mothers worked outside the home 50 years ago; today most do and thus have out of pocket childcare expenses that are not included in the poverty calculation.
  • The poverty rate doesn’t take into account changes in the nature of the population and the structure of families. There are many more single parent families today as well as many more single-person households, many of them elderly persons with different expenses than young people.

Conservatives typically point to a different set of problems with the way poverty is measured:

  • The poverty rate is based on before-tax income and excludes non-cash benefits. Thus, things like housing subsidies and food stamps are not counted even though they clearly add to the resources available to the poor.
  • The poverty rate doesn’t take into account the fact that many of the elderly may own their homes free and clear and may have substantial savings to draw down for consumption. Looking only at their cash income makes many more of them appear poor than is actually the case.
  • Looking at the actual consumption of poor families shows many with possessions considered luxuries in the recent past. These include air conditioning, microwave ovens, DVD players, cable television, HD televisions and others, according to a recent Heritage Foundation report.

Depending on one’s point of view, reasonable adjustments to the poverty calculation may either raise the rate or lower it. Conservatives fear that accounting for liberal reforms would raise the poverty rate and increase pressure to increase federal spending on programs for the poor. Liberals believe the conservative proposals would artificially reduce the poverty rate and understate the true incidence of poverty, thus justifying cuts in programs for the poor. This is why the poverty calculation has been on automatic pilot for so many decades. The easiest thing to do is just keeping doing the same thing year after year and put off the political headache of doing something that one side or the other is certain to object to.

Nevertheless, there is a strong case for rationalizing the poverty rate and both sides make reasonable points. The Census Bureau’s supplementary poverty measure is an effort to balance different concerns and begin the process of perhaps calculating the official poverty rate in a different way. The process will, no doubt, take years before a politically doable consensus develops. Among the proposals suggested by the Census report are these:

  • Change the basic unity of measurement to include all related individuals living at the same address and unrelated children that may be under a family’s care such as foster children.
  • Adjust for regional costs of living, especially housing costs.
  • Update the inflation rate used to adjust poverty thresholds.
  • Revise the basket of goods and services consumed by the poor to take account of changes in relative prices and shares of consumption.

By and large, the Census Bureau’s supplementary measure of poverty would raise the rate slightly, from 15.2 percent of the population to 16 percent. More importantly, our picture of the population living in poverty changes significantly. Among the findings:

  • There are fewer poor people under age 18 and more over age 65.
  • There are fewer blacks in poverty and more of Asian descent.
  • There are more poor people living in urban areas and fewer in rural areas.
  • There are more poor people in the Northeast and West and fewer in the Midwest and South.

Such results clearly have political implications that go beyond a simplistic left-right divide. Insofar as the federal government will probably be spending less on poverty reduction in coming years, as deficit reduction measures are implemented, it will be more important to target increasingly limited government resources more efficiently. Reprogramming federal programs to aid those most in need while scaling back benefits to those that may be relatively well off will limit the impact of aid cuts that probably cannot be avoided. This is the best reason to have a more accurate measure of poverty.