Event study event study, yet again

Anybody who follows this blog, either one of you, will know I don’t have much use for event studies applied to macro, especially when the sample size is low, like 1.

When Larry Summers inferred from the short-run correlation between Brexit odds and FTSE that the latter would collapse on a Brexit vote, I scoffed.

When Justin Wolfers used a similar approach, in greater detail, to claim Trump would collapse the S&P, I also scoffed, although I figured he at least had the direction right! Oops.

My distaste for event studies is perhaps not best practice. They have been used to support the idea that QE durably depressed Treasury yields, which did not fit my priors — or any common sense at all.  I figured, if these things are why people believe in QE nonsense, then I need to look into why they are wrong. Full disclosure.  (It is a bit similar to how guys convincingly claiming that the NK model predicts Neo-Fisherian effects has helped discredit the Taylor Principle no New Keynesian model.)

Um, guys, those event studies are unreliable because the rates signal cannot be separated from the portfolio balance effect and because — in any event — the announcement effect is a crap measure of the enduring effect. Bit ballsy to assume the EMH in extremis AND assume that supply manipulation would matter.  Mike Woodford once pointed that contradiction out, but too politely for people to notice much.

So, today, the Fed apparently hinted at an earlier move to asset sales and Treasury yields went up, I mean down, sharply.  What are the odds that the QE apologist will stick this in their packet of evidence to trot out when needed? Low to zero.

Going down the chain of causation here, we have a few subjects. First there it the tool, I mean policy instrument, and then there is the high-prestige tool, I mean Fed official proposing its use, and then there are the sheep who fall in line to bleat its importance.

The rest, apparently, follows from that.  Let’s just say that monetary policy analysis has not covered itself in glory over this issue. Talk about Paul Homer’s follow-the-leader effect! Egregious.

Screen Shot 2017-04-05 at 6.05.00 PM

So what explains this short-run reaction in yields today.  Damned if I know. It is not unusual, though, for  QE news to influence risk-on / risk-off, I assume through behavioral channels.

If equities take a hit in the short run, because the Fed proposes to cast a smaller spell of magic on the economy, then fixed income follows with the opposite sign. And vice versa on occasions when the Fed is increasing the magic.

It is precisely the opposite of what QE is supposed to do, but by now we are quite used to that.  That applications of QE often have precisely the opposite effect of what is indicated on the label will daunt nobody. QE advocates are made of stronger stuff than mere respect for evidence!  They will ignore even their beloved event study approach if it does not fit. Ok, well at least that gets them right on event studies once. Baby steps.

What they will say is that the tool is SO powerful that it moves expectations of the economy directly, which makes bond yields go the opposite way of what is supposedly QE’s direct mechanical effects, which PROVES QE works.

Accordingly, today, bond yields going down on a smaller balance sheet shows that the portfolio balance effect is so widely recognized as being potent that expectational effects prevent it even being observed!!!! Science.

I shit you not. You can read it in Bernanke.  I have used this quote before, for two reasons. First, it is SO great. And second, it seems like people generally don’t recognize how great it is, which supports my claim of the Paul Romer effect. To me, it is pretty amusing, emphasis added:

A new era of monetary policy activism had arrived, and our announcement had powerful effects.  Between the day before the meeting and the end of the year, the Dow would rise more than 3,000 points – more than 40 percent – to 10,428.  Longer-term interest rates fell on our announcement, with the yield on ten-year Treasury securities dropping from about 3 percent to about 2.5 percent in one day, a very large move.  Over the summer, longer-term yields would reverse and rise to above 4 percent. We would see that increase as a sign of success.  Higher yields suggested that investors were expecting both more growth and higher inflation, consistent with our goal of economic revival. 

Must be nice to be so authoritative that how yours senses work actually determines reality.

But I should not be too hard on Bernanke.  FWIW, I think he did a pretty good job as Chairman, regardless of the fibs he told along the way. And he belatedly provided an accurate — and far out of consensus — interpretation of the taper tantrum.