REUTERS/China Daily
He has previously blogged on the evidence that inflation is systematically overestimated by statistical authorities.
A range of different studies that Yates mentions suggest inflation is repeatedly exaggerated upwards, especially in Europe. Mark Wynne at the Dallas Fed estimated in 2005 that there could be "an upward bias of between 1.0 and 1.5 percentage points per annum on average".
That might not sound like a lot, but with inflation currently at just 0.4% in the eurozone, even a small downward adjustment would put the currency union in deflation, with prices actually falling. A 1.5 percentage point shift would put the true rate at -1.1%, and mean the continent dropped into negative territory way back in summer 2013.
This all sounds a bit terrifying: the four big central banks in advanced economies all target 2% inflation, so investors interpret a big difference between say, 3% and 1.5% inflation. But a lot of economists think that difference could just be down to statistical measurement problems.
There are some huge problems when an economy suffers from deflation. Demand falls off and growth comes to a halt as consumers think, why buy something today when it will be cheaper tomorrow? But whether the true rate in the eurozone is just extremely low, or actually falling, might not matter too much. Dean Baker at the Center for Economic and Policy Research has made a similar point before: deflation isn't a chimera. Low and falling inflation has the same effect as deflation, just in a slightly more muted way:
Going from a positive rate of inflation to deflation is a move in the wrong direction, but it is not qualitatively different from a drop in the rate of inflation to a still positive rate. The problem is simply too low a rate of inflation, there is nothing magical about crossing zero. (As a practical matter, there is enough measurement error in price indices that a low reported positive rate is in fact consistent with a true negative rate.)