By One Count, Government Programs Cut Poverty By 42.3 Million People in 2012

Yesterday, the Census released its supplemental poverty measurement for 2012. The supplemental measurement differs considerably from the official poverty measurement. Basically, the supplemental measurement takes the income from the official measurement, adds non-cash benefits and tax credits, and then subtracts taxes, out-of-pocket medical expenses, child care expenses, work expenses, and child support. This new income is then put against a new poverty line. This new poverty line is sensitive to geographic differences in living expenses and tracks the cost of food, clothing, shelter, and utilities. When you make all these modifications, a new poverty figure pops out. In 2012, the supplemental poverty rate was 16 percent, 1 percentage point higher than the official poverty rate.

I assume the usual suspects will exhaustively cover the supplemental poverty data. So I want to do something a little different here. In yesterday’s release, the Census also updated its March supplement microdata file to include all of the non-cash benefits, taxes, and tax credits. With that file and using the official poverty line, I want to detail just how much things people call “government programs” actually cut poverty. I put the term in quotes here because all income is derived from government programs. But whatever.

The first thing to do is establish a baseline. As we know, there are officially 46.5 million people below poverty. This figure includes a laundry list of income sources, some from government programs and some not. To derive a figure for non-government income, I added the following income sources: wages and salaries, self-employment, farm, private retirement income, private survivor’s income, private disability income, private education assistance, interest, dividends, rental income, child support, alimony, private financial assistance, and other income (catch all). This is not purely non-governmental. For instance, I can tell that the Alaska Permanent Fund dividend is being counted in “other income” for the lucky Alaskans. But it’s pretty close. Counting only this income, 74 million people fell below the poverty line in 2012.

The next thing to do is to deduct taxes paid. For that, I take the income amount for each family and deduct from it FICA payroll taxes (or federal retirement payroll deductions for federal employees), federal income taxes before credits, and state income taxes before credits. So counting only the income mentioned above minus the taxes mentioned here, 79 million people fell below the poverty line in 2012. This will be my baseline figure.

Now let’s bring in the government programs–cash benefits, non-cash benefits, and tax credits–and see what happens. For my purposes here, I include only the following: Earned Income Tax Credit, Child Tax Credit, unemployment compensation, worker’s compensation, Social Security, Supplemental Security Insurance, public assistance or welfare (e.g. TANF), veterans’ benefits, food stamps, school lunches, and housing subsidies. I am not dealing with health care stuff here for simplicity purposes. When I add in these government programs to the baseline income described in the above paragraph, the number of impoverished people falls to 35.8 million, a decline of 43.2 million people from the baseline.

So yes, government programs are pulling massive numbers of people out of poverty each year. And this is even considering that survey participants systematically understate their benefit incomes.

You may be wondering how many people each program listed above pulls out of poverty. This is a conceptually complicated question because the programs work in concert. So, for instance, if two programs combine to put someone’s income over the line, which one of them is pulling the person out of poverty? Or, if a family receives income from two programs, both of which are sufficient alone to pull them above the line, which program do you attribute the poverty reduction to?

So you can’t chop up all of those brought out of poverty and assign them to a specific program. But you can ask yourself this question for each program: if you ended just that program, how many people would suddenly find themselves below the poverty line? Given all of the assumptions made thus far and using 35.8 million impoverished as the baseline, this is the answer:

As you can see, Social Security is a total beast. Perhaps we should use that as a model to cut poverty further.