Poverty Measurement

According to CQ:

.... Barack Obama .... has endorsed the idea of updating the federal measure of poverty, a proposal that is slowly gaining some traction after years of being confined to quiet talk among poverty experts.

New York mayor Michael Bloomberg called for a new poverty measure this week, and Democratic Rep. Jim McDermott of Washington held a hearing on his own proposal yesterday in the House Ways and Means Subcommittee on Income Security and Family Support, which he chairs.

But Obama’s support for the idea has the potential to advance the idea significantly, if he wins the election and pushes for it aggressively.

“Senator Obama knows that the federal poverty guidelines, which were developed decades ago, simply do not take into account the rising costs of child care, health care, transportation, and housing that make it difficult for many families to make ends meet in our globalizing economy,” campaign spokesman Nick Shapiro told me yesterday.

“Senator Obama believes that we should modernize the federal poverty guidelines to more accurately reflect the costs of living and the economic pressures on American families. Without an accurate measure of poverty and economic insecurity in America, we will not be able to fully tackle the effects of these problems on our children and families.”

This week, Mayor Michael Bloomberg proposed changing the formula for measuring poverty. Now Congressional leaders are holding a hearing on adopting a similar change at the national level.

This is all much ado about almost nothing.

At the national level, the proposal would move the bar ever so slightly, so that being poor equals household income of less than $21,818 for two adults with two children, up from the current $20,444.

To be sure, the new measure would be a better reflection of the rate of material deprivation in our nation than the current one.

Today’s measure still uses a formula based on 1950s household expenditures – before housing and transportation costs went up and two-income households became the requirement, causing average child care expenses to soar. In contrast, the new measure would allow for regional reflection of differences in the cost of living and would count some federally-funded employment benefits as income for the first time.

Yet, it’s important to note that we aren’t really accomplishing what we desire with this goal – even with a more accurate formula! This is especially true since this high-level reconsideration of the formula is occurring at a time when there are numerous calls for a national goal to reduce poverty by 50 percent over the next ten years.

What’s wrong with expending a whole lot of energy on this discussion?

First, it’s only a proposal to measure income and not the other resources that communities need for a strong economy and full participation in our democracy and civil society. The proposal isn’t about quality education or clean air or reasonable housing costs or access to health care or reducing prejudice…. and so on.

Second, a if we want a measure of income, a relative measure would be a much more useful test of how well our nation is doing at making sure all residents can contribute to a strong society. As higher income earners do better, low-wage workers must see increases in income relative to the higher earners – otherwise poverty increases. As one leading paper said of a new relative measure:

Certainly, the…relative poverty measure is hard to budge…Yet, when all the research shows that it is how one's income compares to the average that drives one's health, happiness and opportunity, the target must be the right one.

Third, while we do need a better standard for measuring progress as a nation on income deprivation, we’re not likely to succeed in achieving the goal of better policy outcomes if we insist on maintaining a subsistence standard. Indeed, if the goal is based on any measure of “poverty” as it is currently understood in this country – material deprivation blamed on immoral or ill-considered personal choices – we should not expect much policy progress on efforts to strengthen our economy.

At The Mobility Agenda, we’re engaged in a conversation about developing a goal that is more consistent with widely supported policy proposals – which tend to go way beyond income deprivation and which include paid time off at work, worker voices at the table for establishing workplace policy, fair wages, and access to affordable health insurance.

When we put the poverty headline over these policy options, policymakers face real resistance created by the widely-held public beliefs about causes of poverty. We cannot change these beliefs by adopting a goal to end or reduce poverty – regardless of which formula we use to define the term.

Of course, we should adopt a more current measure of income deprivation. Mayor Bloomberg and members of the House should be applauded for their effort to make the more up-to-date but unofficial Census measure more official.

Unfortunately, we're not doing so well on the policy front as it is, and changing the formula will not have much of an impact on this reality. Progress on policy requires a different goal and new measures for testing our progress toward that goal.

Cross-posted at http://www.dmiblog.com/.

I want to second John's concern that the new measure proposed by Mayor Bloomberg, while a "useful step forward" from the current poverty measure, is also likely "to narrow and technocratic to have much resonance outside of policy circles."

My own personal favorite is a measure that sets poverty based on a percentage of median income, say 50 or 60 percent, the lineage of which goes straight back to Adam Smith. The basic idea is that poverty can only be defined in connection to the economic mainstream. Notably, this type of measure produces poverty thresholds that are much more consistent with where the public believes income poverty starts, according to public opinion surveys.

The median income approach to defining poverty is commonly called a relative poverty measure, but it's actually more objective in social terms than either the current US poverty measure or measures like that proposed by Mayor Bloomberg. As social scientists Lee Rainwater and Tim Smeeding note in their book, Poor Kids in a Rich Country, "Anchoring the poverty line in terms of the median is a way of focusing on mainstream incomes and talking them as a point of departure in measuring poverty."

NYC Mayor Mike Bloomberg unveils a new poverty measure for New York City:

Under the federal formula, which is $20,444 for a family of four, some 19 percent of New Yorkers are considered poor, city officials said. Under the new formula, which takes into account New York's high cost of living, the poverty line is at $26,138 and 23 percent of New Yorkers are below it.

The new measure indicates a larger proportion of the city's poor is elderly, and that more working families are under the poverty line than in the federal measure. It has slightly lower poverty rates for children living in single-parent homes and people living in extreme poverty, indicating government programs are helping, officials said.

The new measure appears to be a variant of the one proposed by the National Academy of Sciences in the 1990s. Although an improvement on the outdated and basically arbitrary federal poverty measure, Bloomberg still sets the poverty threshold too low. EPI's basic family budget for a family of four in NYC is about $58,000, and half of median income (a common poverty measure in wealthy nations) in New York State for a family of four is about $37,000.

One political objection to revising our outdated and essentially arbitrary poverty line is that it would result in an increase, substantial under some proposed revisions, of the number of Americans who are classified as living in poverty. I don't doubt that many members of Congress would object to revisions on this ground, but it's worth noting that more than 2/3rds of the public already sets the poverty line more than 40 percent higher than the current measure.

This chart, from a talk I did earlier today at the National Low Income Housing Coalition's annual conference, graphs some recent public opinion data on this score. In October 2007, NPR, Kaiser, and Harvard conducted a survey that asked if a four-person family living at various income level was poor. Another question asked if families at various incomes were middle class. The blue bars are the percentages of those surveyed who agreed that a particular income level made a family poor; the green bars are the percentages who agreed that an income level made a family middle class.

Some 68 percent of those polled said that an income of $30,000—a figure 42 percent higher than the current poverty line of $21,200—made a family of four poor. One conclusion to be drawn from this chart is that an inequality-based poverty measure—one that sets the poverty line at around 50 percent of median income—is more in sync with what the public thinks poverty is than either the current measure or even the revised measure proposed by a National Academy of Sciences panel in the 1990s.

As for the middle class, it looks from this data that a bare majority would place the minimum income threshold for entering it at somewhere between $40,000 and $50,000. There's also a grey area around $40,000, which most Americans don't fit into either poverty or the middle class.

A tipster sends us the latest report on the status of the UK goal to halve the number of children living in poverty within a decade:

... less than three years before the first major deadline in 2010, it appears ministers are falling well behind in meeting their own bold targets.

According to the latest report from the Joseph Rowntree Foundation, although the number of children living in poverty has fallen by 600,000 since the government made its pledge, it remains 500,000 short of the target it should have reached in 2004/5.

The report's authors say the most serious setback was an increase of 200,000 children living in poverty in 2005/06 - taking the total to 3.8 million, or one in three children, if housing costs are taken into account.

....

Although half of all children in poverty are in working families, according to the report, the redistribution of income through tax credits has not kept pace with the growing inequality in incomes across Britain.

"That is a sign that tax credits alone are not quite sufficient," says Mr Kenway. "They are asked to carry more burden than any single policy."

However, while he admits there is no "silver bullet", he would like to see the government tackle low pay, especially in the public sector, the tax burden of low-income working-age households - such as council tax - and the education of young adults aged between 19 and 20 who don't have a minimum level of education.

The Rowntree Foundation report, Monitoring Poverty and Social Exclusion 2007, also concludes that "overall earnings inequalities are widening" and "disability rather than lone parenthood is the factor most likely to lead to worklessness."

In response to misleading terminology (often used here in the US) that describes the standard poverty measures used in the UK and the rest of the EU as "relative" measures, and essentially arbitrary measures like current US measure as "absolute" ones, the authors of the report make a point that deserves emphasis:

In our view, one of the worst things that has happened to the discussion of poverty in recent years is the way in which the poverty measured in this report is now often referred to as 'relative' poverty.

This is wrong in principle and betrays a misunderstanding of what poverty is: in short, poverty is measured relative to average income because poverty itself is inherently relative, that is, when someone is so short of resources that they are unable to attain the minimum norms for the society in which they live. So what is being measured is not some lesser thing called 'relative poverty' but poverty itself.

It is perfectly possible – and one of our indicators does this – to measure progress relative to a fixed poverty line, that is, one that does not change from year to year. In the statistics for the latest year, for the first time in many years, the fixed threshold measure has important information to impart. But it is not in any sense an 'absolute' poverty measure and it plays only a supporting role.