You'll be aware that the shocking rise in inequality in the US is being used as the justification for all sorts of nonsense: from something must be done all the way to this is something therefore let's do this. The problem is with the initial statement: that inequality has risen. For it isn't actually obvious that it has. The reason being one of those little points that I bang on about so often: the results of your measurements depend entirely upon what it is that you're measuring.

It is true that inequality as measured by market incomes has increased in the US. It is also true that consumption inequality [3] hasn't changed very much at all. How can we reconcile these two views? Well, that's after we answer the more important question, which inequality should we care about? How much people get in their paycheques or how much people get to eat? Personally I go for the latter but agree that it is possible to be both righteous here and differ. The reason is, in large part:

Studies of income alone often exclude things like Social Security, Medicare, and food stamps.

Well, sorta, but yes and no. Social Security, being a cash payment, is included in market incomes. But the EITC (our tax credits) Section 8 housing vouchers (our housing benefit), SNAP (food stamps) and Medicaid (health care for the poor) are not included in those market incomes. And the US spends some $400 billion a year I think on those programmes (Medicaid being the largest, the EITC near $100 billion perhaps, SNAP is over $50 billion, these are serious sums) on poverty alleviation through these schemes. When we're measuring consumption inequality we are including the effects of these transfers. When we're measuring market income inequality we're not: because, as in kind and through the tax system transfers these things are deliberately and specifically excluded from the incomes of the poor.

There's one more thing. Since the mid 1970s the US poverty alleviation system has been deliberately moving from cash transfers to these in kind and through the tax system transfers. This is also about the time that the recorded increase in market inequality happened.

Which leads to an interesting conclusion. We see that consumption inequality hasn't changed very much at all. Yet market income inequality has, substantially. At least some part of this difference is simply the way that the US measures incomes: by deliberately and specifically not including all of the money they swap around to reduce income inequality.

Something that is, I think you'll agree, a rather strange thing to do but that is what they do do. A goodly part of the US rise in market income inequality is simply down to the way that they measure it, not to any actual increasing consumption inequality.