Overseas QE and Treasury yields: Brad Setser piles on

A reader kindly sent me a link to a recent piece by Brad Setser, bearing on the question of whether overseas QE might explain why Treasury yields in the US are still low, 2 ½ years after QE here ended.

My own view, as you probably know, is that Treasury yields are not influenced much by efforts at demand/supply manipulation, but reflect deeper fundamentals.  To me, that makes QE a weak tool.

That QE is a weak tool is now an almost-uniform consensus, but not for the reasons I have argued. I believe QE has always been a weak tool, and boiled down mostly to signaling that could have easily been done without actual bond purchases — and eventually was.

Many of its defenders claim it is was a potent force, but one that is now spent, because yields are so low. Once the term premium, whatever that is, goes negative, there is not much left to do, they say. And there are still some hangers on, some with pedigree, who claim QE is still a potent tool. But I think it is fair to say they are a small minority.

The interesting thing, to me, about the Setser piece is that it deepens the puzzle, at least from the perspective of a US QE enthusiast.  Setser points out reserve-manager sales of G4 government debt securities are equivalent to a negative QE, particularly in dollars.  Taking account of this, total official purchases of Treasuries (Fed QE plus overseas reserve manager operations) have actually swung negative, while official purchases of all G4 securities have slowed a lot. And yet Treasury yields have fallen. That is pretty hard on the QE story – so thanks emailer!

screen-shot-2016-09-18-at-11-50-19-amIn fairness it is not decisive, though. If you think that the market for duration risk is global and that it is the stock and not flow that matters, then maybe a QE enthusiast could lean on this still.  They would then have extreme trouble with the taper tantrum, but never mind.

Getting a bit more granular, the main swing factor in reserve-manager positioning in G4 government bonds has been China. They went from rapidly accumulating to selling. This convinced many that Treasury yields would have to spike, as you may recall.

But predictably, it is not that simple. People overlook, egregiously, the REASON China has been selling Treasury debt. They have been doing so to offset the impact of private capital flows in precisely the opposite direction.

Ex-ante, these private capital flows have predictably been the more important in the capital market repricing. And they have reflected the deeper fundamental, which is that perceived financial risks in China have been going up.

That fundamental is not a reason to expect Treasury yields to go up, regardless of what the bean counters observe in official flows. I often wonder why that is not obvious to people. My somewhat snarky answer is that people prefer to add up shiny items over thinking. China in crisis. Treasury yields up! Well, obviously. 😉

Having said that, the Setser piece is close to the aggregation I was looking for.  I wish I had it in levels, 10-year equivalents and expressed as a share of global or rich-country GDP. Oh, also measured relative to expectations. That would be nice too.