Turtles All the Way Down: Are Fed Promises “Actions”? How About Promises to Make Promises? Game Theory Edition

The frequently brilliant Ashwin Parameswaran makes The Case Against Monetary Stimulus Via Asset Purchases, and Nick Rowe responds. Ashwin replies, analyzing “monetary policy as a threat strategy.”

Nick’s view (in Ashwin’s words):

…a credible threat will cause market expectations to adjust and negate the need for any actual intervention in markets by the central bank.

Ashwin counters this view [my slight tweaks]:

…expectations[-setting] only work[s] when there is a clear and credible set of actions that serve as the bazooka(s) to enforce these expectations.

I think the key hinge here is the difference between the declarative, conditional, and subjunctive moods: the assumption that the actions would serve as a bazooka, if….

Nick seems to be assuming that the Fed’s promise is always credible — regardless of whether they have ever (or reliably) fired the bazooka in the past (followed up on their promises with OMO/QE actions/balance sheet changes) — because in theory the Fed could fire the bazooka, and (again, in theory) the bazooka would have the desired effect.

The promise would be credible if: 1. the Fed has shown willingness to engage in those actions in the past in similar situations — by actually taking those actions — and 2. Those actions had the desired effect.

Nick’s view might make sense if the Fed were playing a one-round game. (They could do it, so market players have to assume that if necessary they can and will do it, and it will work.)

But it’s a repeated/ongoing game. I sed at Nick’s place:

It’s the threat that I will shoot you that causes you to turn over your wallet. Not the actual shooting. (The shooting can have the desired effect, but it’s unnecessary if the threat is credible.)

But you and I both know that I’m going to be stealing your wallet every day for the rest of our lives. If I don’t shoot you today when you refuse to hand it over, the threat won’t be credible tomorrow.

So yeah: the threat generally does the job. But the shooting can also do the job, and periodic actual shooting is necessary for the threat to do the job.

Nick replies:

I threaten to go to the North of the herd of cattle, to drive them South. But if my threat is credible, and they head South, I actually follow the cattle south.

To which I replied:

But if it’s a repeated game, if you never head north the threat isn’t credible.

You have to head north sometimes.

(This is all ignoring the question of whether actual balance-sheet changes even would have the desired effect, which market players seem decidedly uncertain about.)

Max, replying to me in the comments at Ashwin’s, seems to perfectly encapsulate Nick’s (some-variety-of-)monetarist position:

Steve, raising the inflation target is an “action”.

Try this:

The Fed promises to raise inflation or NGDP to X (target or relative-to-trend level), doing whatever is necessary in the future to achieve that target/level. They do nothing else — make no significant changes to their balance sheet.

What they’re saying:

“If we have not achieved that target/level at our meeting six months or two years from now, we promise that we will again promise to do whatever is necessary in the future to achieve that target/level.”

Max: when the Fed promises to make future promises, is that an “action”?

It’s turtles all the way down.

I hereby propose a new acronym to add to the lexicon of Fed and monetarist watchers: TATWD.

Nick takes his thinking to its logical conclusion (as he is quite admirably wont to do). Working with his north/south cattle metaphor, he says:

if the conditional threat to buy assets is credible, the central bank would actually sell assets.

So the Fed:

“We’re so confident in our ability to boost NGDP that we’re going to announce our intention to do so, and when we (inevitably) see the lead cow tending south, we’ll start selling in anticipation of our threat working perfectly. The rest of the herd, convinced by our displayed confidence, will turn and follow the leader.”

Does this seem likely?


“We’ll make our threat, and watch with satisfaction as the whole herd (inevitably) turns south. Then we’ll follow them by selling. We’ll never have to buy.”

As Market Monetarists will be the first to tell you (and as all will I think agree), the Fed is extremely, even excessively reluctant to turn north (more QE). The cattle — market players — know that.

If the Market Monetarists are right, Nick’s (and their) logic has a broken link: the cattle won’t believe the threat to go north because the cattle know that the Fed really, really doesn’t want to go north. (My explanation of why they don’t want to.)

Or, Nick’s logic is just obvious: if the economy gets better (the herd turns south) — for whatever reason(s) — the Fed will tighten by selling assets. No duh.

Cross-posted at Angry Bear.






3 responses to “Turtles All the Way Down: Are Fed Promises “Actions”? How About Promises to Make Promises? Game Theory Edition”

  1. […] sheet, sometimes — rather than just promising to make future promises — to avoid the Turtles All The Way Down […]