Most systems for measuring poverty use a so-called poverty rate or poverty line (that’s the case in the U.S. for instance, or in the UN’s Millennium Development Goals). That’s a level of income (or consumption etc.) which is considered to be the minimum that’s necessary for a decent human life and for the satisfaction of basic needs. These systems are also called “headcount” measures of poverty: they simply count how many people fall below the fixed point that determines the difference between poverty and non-poverty.
You can see the problem coming: according to these systems, you’re either poor or you ain’t. They just tell us how many people are poor, not how poor they actually are. This is a big problem in developed countries that use such poverty measurement systems. The poverty rates in those countries are rather high in dollar terms. For example, the thresholds in the U.S. are, as of 2008:
- One adult: $11,200 annual income, not including the EITC or non-cash benefits (Food Stamps, Medicaid, housing assistance, employer health-insurance contributions, etc.), and including taxes
- Two adults: $14,400
- One adult, two kids: $17,300
- Two adults, two kids: $21,800.
By “rather high” I don’t mean to say that the people under those poverty lines aren’t really poor and that the U.S. measurement system is too generous (if anything, it’s the contrary). What I want to say is that in developed countries, people need a substantial income in order to escape poverty. If you want a job, you’ll probably need a car, a phone, internet connection, child care etc. If you want a place to live, you’ll need to spend a huge amount of money on a house, and so on. Poverty lines in developed countries are therefore not so low that being poor means being on the brink of starvation. They are set at such a level that being poor means being unable to afford a job, quality housing, healthcare and education.
Given the fact that poverty rates are rather high, there’s a lot of space below them. Hence, you have different kinds of poor people: there are those who have a job and an income, albeit a rather low one, but who struggle to survive because of their expenses; and there are those who just live on the street. You have people who are poor for some years and people who are poor their entire lives.
All these people are equally poor in the measurement system we’re discussing here. This system doesn’t provide data on the distance from the poverty line, or, in other words, on the depth of poverty. In the worst case, people who are already poor according to the system could become much poorer, without any change in the headcount of poverty. If the 13% or so of Americans who are currently under the poverty line all became homeless beggars, you wouldn’t see a change in U.S. poverty statistics.
In order to solve this problem, people have come up with the concept of the poverty gap (incidence * depth of poverty): the mean distance separating the population from the poverty line (with the non-poor being given a distance of zero), expressed as a percentage of the poverty line (see also here). Unfortunately, this hasn’t become a very popular number. It’s probably too complicated. A clear and simple poverty line is much more appealing yet deficient.
More post on the problems of poverty measurement are here.