The left continues to lie about poverty in America. Statistical trickery works just fine on low info voters.

The Number Of Americans Living On $2 A Day Or Less Is Zero

There’s an interesting new book out talking about how many Americans are absolutely poor. It’s being taken seriously too: the New York Times runs a review of it today. Sadly, the central premise of the book, $2.00 a Day: Living on Almost Nothing in America is, depending upon how you want to look at it, either not true or a piece of little more than statistical trickery. That hasn’t stopped the usual sources getting all hyper about it, of course it wouldn’t. But it’s worth examining what the authors are actually saying about poverty in America. And much more importantly, examining why they’re not saying what all too many people seem to think they are saying. For the truth is that there just isn’t, to any reasonable approximation, anyone at all living on less than $2 a day in the United States. Yet the general reaction to this book is that the authors’ estimates of 1.5 million families living in this sort of poverty are true. I’m sorry, but it simply isn’t true. Firstly because of the details of the measures that they’re using and secondly because they’re not actually looking at the right measure at all to decide upon living standards.

Here, for example, is something from that NYT review: 

Shaefer analyzed the census data, which is based on annual interviews with tens of thousands of American households, to determine the growth of the virtually cashless poor since welfare reform. His results were shocking: Since the passage of TANF in 1996, the number of families living in $2-a-day poverty had more than doubled, reaching 1.5 million households in early 2011.

That simply is not true.

Similarly, this at CBS:

The number of U.S. residents who are struggling to survive on just $2 a day has more than doubled since 1996, placing 1.5 million households and 3 million children in this desperate economic situation. That’s according to “$2.00 a Day: Living on Almost Nothing in America,” a book from publisher Houghton Mifflin Harcourt that will be released on Sept. 1.

The measure of poverty isn’t arbitrary — it’s the threshold the World Bank uses to measure global poverty in the developed world. While it may be the norm to see families in developing countries such as Bangladesh and Ethiopia struggle to survive on such meager income, the growing ranks of America’s ultrapoor may be shocking, given that the U.S. is considered one of the most developed capitalist countries in the world.

That first paragraph is not true and that second is hugely misleading at best.

Here’s what is true. There’s a number of families that have incomes of less than $2 per day per member of family for some period of the year. Agreed, we probably don’t want this to happen, we quite probably ought to do something about it. But that is not the same statement as people are trying to live in America on less than $2 a day. Because cash income is not the same as total resources available to live upon. For example, in that larger number of 1.5 million families the estimation method is to look only at cash income. Thus most of the things that are done to help poor people (SNAP, Section 8 vouchers, Medicaid and so on) are not in fact included. This is thus to fall foul of Worstall’s Fallacy. To measure the situation before whatever it is we do to change it and to use that preliminary measurement to insist that more must be done. Instead of what we should be doing which is to look at what is the situation after whatever it is that we do to alleviate the problem and then consider whether more or less should be done.

So here’s what the authors have really done to get to this lovely shocking number. Firstly, they’ve taken the World Bank number for poverty. That at least is correct: the $2 a day number. And that’s the number that hundreds of millions of people really do live under around the world. And that pays for everything: housing, health care, clothing, food, everything. At current (actually, 2005, but still) US prices. This is real, deep, biting, poverty. Stand in Walmart with $2 per member of your family and feed, clothe, shelter, heat, educate them for that.

The first little trick when looking at the US is to change the definition. The World Bank number is what people get to consume. The authors here have decided instead to use income. These are not the same thing. You can consume vegetables grown in the back yard without having an income for example.

The second is to use an odd definition of income. One that doesn’t include all income in fact. They look only at cash income. But the US spends $500 billion a year on things like Medicaid and other health care services for poor people. That’s obviously got to be part of poor peoples’ incomes as well then. And then there’s Section 8 housing vouchers which pay for habitation for many poor people. This is also excluded. In their most restrictive version of income they don’t even include Food Stamps.

So what they’re really measuring is the cash incomes of the poorest people without accounting for pretty much all the things that are done to increase the total incomes (such total income being equal to consumption possibilities of course) of those poor people. The only part of welfare that they do include is cash welfare under TANF.

And that’s where their large rise in this form of poverty comes from. They are, as above, measuring only the cash income of poor people, including cash from welfare. But this system was deliberately changed in the 1990s to reduce the amount of cash welfare people got and increase the amount of in kind welfare (ie, Section 8, EITC, SNAP and so on) that people got. So, what they’re actually recording, in their increase, is what Congress voted for. Stop giving people cash and give them more in other ways instead. I don’t think the original change was a good idea myself but it’s still something of a statistical trick they’re pulling off here, not a real change in what people can consume.

Finally, they’re deliberately blurring how this sort of poverty persists or does not. When we look at those World Bank numbers for other countries then we’re looking at people who live that way all the time. In this research done about the US we’re not looking at that at all. Rather, at people who have, for some however short period of time, had very little cash income. Just as an example (and I emphasise, this is only an example, just to illustrate the logic here) imagine a family with one person in work earning $100,000 a year. Then they get fired. It takes them 2 months to find another job paying $100,000 a year (say, a decent software engineer in a company that goes bust). They do not file for any welfare while unemployed (perhaps because they have savings, that sort of thing). Under the method used to calculate those 1.5 million people living on less than $2 a day our thoroughly upper middle class software engineer qualifies. Because there was a period of one or two months where cash household income was under $2 a day per family member.

Do I think the US welfare system is perfect? Nope, most certainly not. It’s also true that the US is more unequal than other advanced countries. But the real incidence of $2 a day poverty in the US is zero. The numbers cooked up for this book look only at cash income, not any other form of aid that people might get. They are also measuring transient populations flowing through a rocky patch, not some vast underclass. And finally they are measuring income, not the thing that we really want to measure, consumption possibilities.

And I’m afraid that it’s not just me, free market radical and running lackey dog of the plutocrats that I am saying this. Here’s the Brookings Institute on the same book:

Based on an assessment of consumption in the fourth quarter of 2011, we obtain a much lower $2 a day poverty
rate of only 0.07 percent. To verify this result, we rerun our calculation using a more selective definition of
consumption with the same survey data. This yields a similar poverty rate of 0.09 percent.

That is, to any reasonable level of accuracy the level of $2 a day poverty in the US is zero.

And again, when Brookings uses the World Bank’s method of measuring poverty:

When we replicate this process using data from the U.S. CEX, our estimate of $2 a day poverty falls all the way
to zero.21 In other words, if we measured poverty in the U.S. as if it was a developing country, we would conclude
that no-one falls under the $2 threshold.

Which is why the World Bank lists the US as having no $2 a day poverty.

What should be done about poverty in the US is what many of us have been saying for some time now. Just bring in a universal basic income for all adults and be done with it. Instead of, say, inventing one’s own poverty measure, ignoring just about everything that is currently done to alleviate poverty and then shouting that there’s 1.5 million American families as poor as the poorest in the world. There ain’t. 

My latest book is “The No Breakfast Fallacy, why the Club of Rome was wrong about us running out of resources.” Amazon and Amazon.co.uk. $6.99 and relevant prices in other currencies.
                                                     ...............



How do the authors square with the fact that a family of four will receive the equivalent of $60,000 a year in benefits. It's why many wish to come here. If it were $2/day who would come?