The Problem with Poverty (Measures)

Flickr - Dominic Alves

Remember this outdated tool? Flickr – Dominic Alves

This post is the first of a series of articles about the recently released 2015 Children By the Numbers Data Bulletin.  In this series, we tell the stories behind the numbers and statistics featured in the data bulletin.

On occasion, I have been known to give talks about statistics in Westchester County.  Whenever I encounter a group of people who are not from the nonprofit/social service sector, I like to ask this question:

“How much income do you need to support a family of four in Westchester County?”

I usually get guesses around the ballpark of $80,000/$90,000.  On occasion, people will guess an income of over $100,000, which isn’t surprising considering that the 2013 median income of Westchester families with children under 18 is estimated to be $108,174 according to the American Community Survey One-Year Estimates (Table B19125 to be exact).

Then I inform them about the US poverty level, which basically determines a family’s poverty status.  If you make an income at or below the poverty level, you are considered poor according to the US government.  Above the poverty level = not poor.

Then I ask,

“What do you think the poverty level is for a family of four in Westchester County?”

And before they can answer, I tell them. For 2015, the official poverty level for a family of four is:

$24,250*

Imagine trying to raise a family of four on $24K a year in Westchester.

And yet, the poverty rate is, for the most part, based on levels such as the one stated above.

*NOTE:  This number is different than what was used in the 2015 bulletin, which cites the 2013 poverty level for a family of four in order to more closely align with the rest of the economic data displayed in the bulletin.  For th majority of the economic data, the most recent figures we can obtain are for the year 2013.

The not-so-secret secret

Poverty Measure: Map of Median Incomes by Census Tract

We often use median family incomes to better demonstrate areas of high need in Westchester

I’m going to let you in on a not-so-secret secret – the poverty rate is an inadequate measure of need, ESPECIALLY in Westchester County.

Here’s another not-so-secret secret – experts have known this for years.

The poverty rate basically represents the percentage of people in a given population that live at or below “the poverty threshold”.  The poverty threshold is the minimum income needed for adequate survival.   In the 1960s, Mollie Orshansky of the Social Security Administration developed a methodology to develop a US poverty threshold. According to historian Michael B. Katz in his book “The Undeserving Poor,” she intended to use this methodology as a research tool.  She never intended for it to be used as an instrument of policy.

But an instrument of policy it is today.  A simpler derivative of the poverty threshold, the poverty guideline, is used to determine eligibility for poverty programs such as TANF (Temporary Assistance for Needy Families) and school-based Free/Reduced Lunch.  (NOTE:  When people talk about “poverty levels”, they’re usually referring to the poverty guidelines.  A discussion of poverty guidelines vs.poverty thresholds can be found here.)

However, the drawbacks of the poverty threshold, as developed by Orshansky, are many.

The drawbacks

1)      The poverty threshold fails to address regional costs of living.  The poverty thresholds are the same for a family of four in Mississippi as they are for a family of four in Westchester.  As a result, the poverty thresholds may be overstated in areas where the cost of living is low, and may be understated in areas where the cost of living is high (take Westchester, for example…).

2)      The poverty threshold itself is based on an outdated family living in the Mad Men era. The threshold presumes that food is the largest expenditure in the family budget, and it currently assesses whether a family’s pretax cash income is less than 3 times the cost of a minimally adequate diet.  At first glance, this sounds like a pretty reasonable basis for a poverty threshold, but at the time Ms. Orshansky created the poverty threshold in the early 1960s, food expenditures took up one-third of a family’s total budget.  In this day and age, however, housing costs take precedence over food.

The poverty threshold also does not take into account the cost of child care and other expenses that are relevant today. Orshansky used the 1955 USDA household survey when creating the methodology for the poverty threshold, and families that participated in the USDA survey generally had one wage earner and one stay at home parent.  As we know, that is not the case today.  Additional expenses, such as internet access (which did not exist in 1955), are not included as well.

3)      Public assistance income is not included in the poverty threshold.  As Jesse Willis of the Oregon Center for Public Policy once stated, “If the value of food stamps, publicly provided health insurance benefits, and cash welfare payments were counted as income in the poverty calculation, many people would no longer be considered poor.”

The US Census Bureau issued an alternative poverty measure, called the Supplemental Poverty Measure (SPM), to address the shortcomings of the official poverty measure.  The US Census Bureau released its first SPM report in 2011, but even now it still doesn’t get much attention in the media. Organizations such as the Annie E. Casey Foundation have taken note of the SPM.  Just last week, Annie E. Casey released KIDS COUNT Data Snapshots based on the SPM.

There are also other attempts to find a better measure that can more effectively determine need, such as the Self-Sufficiency Standard by Dr. Diana Pearce, the Housing Wage Calculator from the National Low Income Housing Coalition, and the Family Budget Calculator from the Economic Policy Institute.  All are worth taking a look at.

So as the war on poverty continues, the attempt to measure it continues to be hampered by an outdated methodology that is only now being slowly replaced by other alternatives.  This is just the beginning of the conversation.