Is it undeniable? Last I heard there was a lot of noise from Picketty and Saez, which was based on fundamentally invalid and unsound scholarship. To put it in

Is it undeniable? Last I heard there was a lot of noise from Picketty and Saez, which was based on fundamentally invalid and unsound scholarship. To put it in computing terms, it's on a par with benchmarking your software for performance improvements, but upgrading your laptop partway through and calling that a massive speedup. A one-time methodology change in the methodology of the census bureau's Current Population Survey (in 1992) is responsible for roughly a third of Picketty's claimed increase in the Gini coefficient. It's undeniable that this has been in the news and on the lips of many partisans. And there was some short-term increase in the aftermath of the latest recession, I suppose. That's not really controversial, though its durability has not yet been proven. (Also, I'm not sure the "especially in the US" label is necessarily deserved.) Really, the "undeniable" label is a tendentious one: it argues, more than it describes. Its purpose is to turn a little solid evidence for a small change into what seems like an ocean of evidence for an overwhelming change. I call it deceitful.