Does QE do NOTHING?

[NCR85]

Let me get some really basic facts straight about QE.

If the Fed credibly promises to keep rates at 0% for the next 5 years and there is an absolute guarantee that this promise will be upheld, 5 year rates converge on zero*. This is basic logic, because it means that people can arbitrage the short term rate to the long term rate for those 5 years without having a risk of the short term rate rising (after all we're talking about an absolute guarantee here).

The qualification here is of course that there are no absolute guarantees. So instead the Fed can buy down the long term rate so as to force it down, the pretense being that this is somehow a more "final" move that can not just be reneged upon.

But where on earth does this idea come from, that it is more final than forward guidance, i.e. promising to control the short term rate a certain way at future dates? Any treasuries bought using QE can be sold at any moment. So this action is one that is just as easy to reverse as the breaking of a promise.

Does any forward guidance (policy through promises) basically swamp the effect of QE in this way? Does QE do absolutely nothing? Is it a placebo?

* I may be slightly wrong here; is it zero or JUST the 5 year term premium? In any case the idea is that short term rates would go down as far as they can go.

[NCR85]

Re: Does QE do NOTHING?

OK, one area where I think QE has some different effects from forward guidance is in that if the market realizes that the Fed will be forced to renege on the promise to keep rates low before the Fed itself realizes this, it will pre-empt the Fed's long term treasury sales and pocket a large part of the proceeds that the Fed would otherwise have gotten. So this is money that the Fed would have had and now doesn't have because market participants beat them to the sale. In a sense, the Fed puts its own funds on the line. It is like a promise with a bet behind it. "I'm going to keep rates low AND I'm betting that I will do so". If QE is done too liberally and reversed to late, it can damage the Fed's balance sheet by quite a bit. And that can limit the extent to which the Fed can halt inflation, since reducing inflation requires the sale of assets.

Thanks, but I'm beginning to think this thread was a false alarm?

[Chris]

Re: Does QE do NOTHING?

Thanks, but I'm beginning to think this thread was a false alarm?

Maybe; but I found the thought experiment interesting.

[NCR85]

Re: Does QE do NOTHING?

Cool. I have a few more ideas about QE of a "thought experiment" nature that I might as well dump here now that the topic is raised.

The idea behind QE is normally that it funcions like an asset swap: the Fed buys treasury paper from the secondary* market with the intention of unloading them again at a later point. The secondary dealer receives money in the form of a credited bank account and the Fed gets treasury paper on it's balance sheet that it can at a later point use to rein in any inflation given rise to by the newly generated money. As such, the money injection is rendered temporary. However, it is temporary only to the extent the Fed doesn't lose money on the transactions. If the asset that is bought (treasury paper) loses value in the meantime, there is at least some inflation that is not neutralized unless additional sales occur. Likewise a profit can be made on the transaction. But the basic issue is: the Fed has to protect it's balance sheet to be able to keep inflation under control.

I'm interested in a situation where a permanent disinflationary influence causes the Fed to need to suppress interest rates, either on the long or the short end, indefinitely. I'm mainly thinking about an unproductive debt overhang as a result of past asset bubbles of a kind that Steve Keen regularly talks about. While in the normal situation, the Fed first loads up on treasury paper and then unloads again, in this situation the treasury paper is put on the Fed's balance sheet and never returned to the market.

What happens to the treasury paper in this instance is operationally equivalent to ripping up the paper: any interest paid to the Fed on the treasury paper is sooner or later returned to the treasury. This is what the law necessitates be done with Federal reserve profits above a certain level. When treasury paper is perpetually kepts on the Fed's balance sheet and the interest payments from the treasury to the Fed are continually returned back to the treasury, the paper is 100% irrelevant and may as well already be gone.

The interesting question is of course, to what extent does the current situation in Japan and the USA resemble this description? Is Japanese and/or US debt basically already ripped up for all intents and purposes? Is it such an outrageous idea when you look at Japanese bond yields that are at levels you totally wouldn't expect them to be at given a 230% of GDP public debt level?

* I think it's the secondary market, not primary?

[NCR85]

Re: Does QE do NOTHING?

Another thing about this: when treasury paper is "bought" without the intention of ever selling it, QE goes from being an asset swap to working plainly like a gift. And this changes the political picture around the use of QE COMPLETELY. While a loan is a transaction without a winner and a loser, a gift is a transaction with a clearly defined winner and loser. Since the loser in this gift transaction is a public institution, it is a politically highly legitimate question as to who the gifts are and are not given to.

And this, I think, is where Steve Keen's idea of a debt jubilee comes in: on the course it's going on right now, the Fed IS engaging in gift-giving. Now let's make sure the gift giving is distributed in a way that has some measure of fairness to it. Obviously this can not be done in a "perfect" way but it should be easy to come up with a distribution scheme that is better than "hand it all over to wall street".

[NCR85]

Re: Does QE do NOTHING?

OK, I have to take a small step back to question whether this is right at all because I am thinking very much along MMT ish lines and I trust itulip's critcism of MMT even though I don't always find it easy to really wrap my head around the argument of how the MMT theory breaks down. I think it has something to do with the way interest rate policies between domestic and foreign economies interact, and there is probably also some spillover of inflation into the commodity markets via the Hotelling price effect. So the idea that treasuries can be kept under wraps forever is probably not very realistic.

[aaron]

Re: Does QE do NOTHING?

There are trillions of loans out there that must be paid back with "real" dollars. The dollar demand is huge to pay back these loans.
Does collecting U.S. treasuries in electronic vaults lead to national growth and security?

It seems to me that owners of treasuries have done extremely well. Asia has had very good growth for many decades. The more treasuries they collect, the better. It seems logical that buying treasuries would be good for our country as well, no?
Treasuries are the gift that keeps on giving. You get paid every month to hold pieces of paper.

[NCR85]

Re: Does QE do NOTHING?

---

John Cochrane and MMTers arguing largely the same thing: http://www.cnbc.com/id/101306602