Zero Hedge recently went on a tear about “soaring poverty in America”, compiling a list of 29 items which “proved” the case. Tim Worstall indulges in a bit of fisking over the mythical claims:

2. New numbers have just been released, and they show that the number of public school students in this country that are homeless is at an all-time record high. It is hard to believe, but right now 1.2 million students that attend public schools in America are homeless.

No, I’m afraid not. In fact, this isn’t possible in the slightest. The numbers for homelessness are here.

In January 2012, 633,782 people were homeless on a single night in the United States. Most (62 percent) were homeless as individuals and 38 percent were homeless as persons in families.

At any one time we have 630,000 homeless people. We cannot therefore have twice that number of children alone being homeless at any one time. This is not one of the mathematical possibilities that this universe offers us.

What is actually being said in the original number is that over the course of a a year 1.2 million children will suffer one or more incidents of homelessness. We can still think that this is appalling, that it shouldn’t happen and that we ought to be doing more about it. But it is not true that “right now 1.2 million students that attend public schools are homeless”.

3. When I was growing up, it seemed like almost everyone was from a middle class home. But now that has all changed. One recent study discovered that nearly half of all public students in the United States come from low income homes.

Well, if you grow up in a middle class area then of course those you grow up with are likely to be middle class. But that nearly half of students coming from low income homes is indeed true. But it’s extremely uninteresting that it is. Here’s the definition of low income they are using:

To be crystal clear, the researchers were not analyzing poverty rates per se. Rather, they tracked at the percentage of children in each state who received free or reduced school lunches, which are only available to students whose families earn below 185 percent of the poverty line. For a family of four, that amounted to about $41,000 in 2011.

As it happens, median household income is around two times that federal poverty line. So, 185% of the federal poverty line is going to be pretty close to the median household income. Median household income being the amount that 50% of households get more than and 50% get less than. So, yes, we would pretty much expect that 50% of children are coming from low income families. Because it’s only in Lake Woebegon that all the children are above average. This is a statistical artifact of the measure they are using as low income, nothing else.

And, of course, the traditional mis-measurement of poverty in the US compared to pretty much every other western country:

6. According to the U.S. Census Bureau, approximately one out of every six Americans is now living in poverty. The number of Americans living in poverty is now at a level not seen since the 1960s.

No, this is not true, as I’ve pointed out in these pages many a time. Many, many, many, many times in fact.

The 15% in poverty measure is the number of people who would be in poverty if we didn’t help them by giving them money, food, housing and health care. It is not the number living in poverty. It is the number in poverty before we help them out of poverty. And as to the comparison with the 1960s, in the 1960s it really was the number in poverty after we helped, now it’s the number before we help.

As to why this is when we calculate the number below the poverty line we look only at cash incomes. This includes any actual money that people might be given to alleviate their poverty. In 1960s America pretty much the only poverty alleviation that was done was to give money to poor people (“welfare”). Today we give very little money directly. But we spend a vastly greater amount in giving people things. Medicaid, Section 8 housing vouchers, SNAP (or food stamps). We also give aid to the low paid through the tax system, with the EITC. However, we do not count all of those things, nor the EITC through the tax system, in our calculation of the number below the poverty line. So, since the 1960s the US measure of poverty has changed. From being one of people who are indeed living in poverty after whatever help they get to one of the number of people who would be in poverty before any help that they get.

When we correct for this the US poverty rate is significantly lower now than it was in the 60s. One estimate is that it is about half the 60s number.