Saturday, June 09, 2012

A Non Blast from the Past

update: welcome Thomists. Prof Thoma has done it again, linking to a post which I thought was by me for me and put here in obscurity. I do maybe have something to say about QE2 effects, but it is here.

In my personal totally non-expert view QE 2 and Twist did not amount to much, because they included the same key error -- purchases of Treasury Securities not Agency issued mortgage backed securities. My sense is that they had small effects on interest rates, because treasuries are quite safe, so private demand for treasuries is highly interest elastic. In contrast, I think QE 1 which consisted of purchases of MBS was highly effective. MBS were feared so demand is not highly elastic. Their price responded to Fed purchases*.

So why did the Fed shift from MBS to Treasuries ? I think one reason must have been small c conservatism. The Fed normally buys treasuries, so the FOMC wanted to return to a portfolio of treasuries. To learn more I read Bernanke's August 27 Jackson Hole speech which included the first public hint of QE2.

Bernanke used the word "reinvest" because the choice to invest in treasuries was explained before the mention of QE2 (as the acronym is currently used) when Bernanke was discussing reinvesting proceeds from maturing MBS. This is a separate issue involving an estimated $ 400 Billion in 2011 decided prior to the decision to issue another $600 Billion in Fed liabilities to buy 7 year Treasury Notes**.

We decided to reinvest in Treasury securities rather than agency securities because the Federal Reserve already owns a very large share of available agency securities, suggesting that reinvestment in Treasury securities might be more effective in reducing longer-term interest rates and improving financial conditions with less chance of adverse effects on market functioning. Also, as I already noted, reinvestment in Treasury securities is more consistent with the Committee's longer-term objective of a portfolio made up principally of Treasury securities. We do not rule out changing the reinvestment strategy if circumstances warrant, however.

Well "might" makes right, but the argument in the first quoted sentence makes no sense. The effect of further purchases should depend on the stock of MBS held by private investors not on the stock held by the Fed. It depends on the variance of the value of the stock of MBS held by private investors (own variance) and the covariance of that value with other assets (including human wealth) held by those investors. The idea that risk associated with mortgages, mortgage delinquencies and house prices held by private investors on August 27 2010 was minor, because the Fed had already assumed so much of that risk is nonsense. Also the Fed's holdings of Treasuries on August 27 2010 was comparable to its holdings of MBS (very roughly $800 Billion vs $ Trillion IIRC).

It is clear that the decision to buy treasuries was made, because that is what the Fed normally does. I think the decision has had gigantic costs.

Later Bernanke suggests that there will be at least some QE2 (or QE2.2 if reinvesting were QE2.1)

Policy Options for Further Easing

Notwithstanding the fact that the policy rate is near its zero lower bound, the Federal Reserve retains a number of tools and strategies for providing additional stimulus. I will focus here on three that have been part of recent staff analyses and discussion at FOMC meetings: (1) conducting additional purchases of longer-term securities, [skip]
I believe that additional purchases of longer-term securities, should the FOMC choose to undertake them, would be effective in further easing financial conditions.

Not a promise, but a very broad hint by Fed Chairman standards. Here, Bernanke is careful not to use the expectations channel. I tend to defend Bernanke, but even I am shocked by the casualness with which he dismisses the possibility that higher expected inflation might be a good thing.

Another concern associated with additional securities purchases is that substantial further expansions of the balance sheet could reduce public confidence in the Fed's ability to execute a smooth exit from its accommodative policies at the appropriate time. Even if unjustified, such a reduction in confidence might lead to an undesired increase in inflation expectations. (Of course, if inflation expectations were too low, or even negative, an increase in inflation expectations could become a benefit.) To mitigate this concern, the Federal Reserve has expended considerable effort in developing a suite of tools to ensure that the exit from highly accommodative policies can be smoothly accomplished when appropriate,

"Of course" an increase in inflation expectations might be a benefit and of course the Fed expended considerable effort making sure that no such increase occurred.

Unlike many, I don't think this is the key issue. I have extremely no expertise so who cares what I think, but I think that the people who parse every word that comes out of Bernanke's mouth are traders trying to guess what other traders will think tomorrow. I am extremely unconvinced that managers of firms deciding on fixed capital investment pay as much attention, and I am quite sure that very few consumers deciding how much to spend pay any attention at all.

I think the key choice was buying treasuries not MBS not promising to exit when the economy recovers and make it clear that we can. However, no rationale for either choice was presented.

*I might add down here where no one is reading, that risk matters more because of covariances than own variances. By bearing mortage default risk, the Fed made private entities more willing (more nearly willing ?) to loan to home buyers. In contrast, the risk in 7 year Treasury Notes (QE 2) or 30 year Treasury Bonds (Twist) is the risk of high shorter term nominal interest rates in the future, perhaps due to high inflation. This makes such notes and bonds a hedge against another recession. By bearing that risk, the Fed removed that hedge. In any case, the risk in 7 year and 30 year treasuries is clearly perceived to be small as is shown by the small effects on yields of huge changes in the outstanding stock.

**QE2 could refer to both policies and therefore amount to an estimate $ Trillion over all. The first estimated $400 billion (QE2.1 ?) was announced in FOMC notes issued August 10 2010. The program normally called QE2 (as demonstrated by the fact that people normally say QE2 involved only $600 Billion and not around $ Trillion) was not mentioned in the August 10 minutes. The exact amount and the exact security to purchase were decided later with the final announcement IIRC November 3 2010.

1 comment:

Mike Kimel said...


Hi. I tend to agree with you on most things, but I think you're wrong from a big picture perspective. In my view, QE 2 and Twist didn't amount to much because they didn't address a relevant issue - slack demand. Finding ways to compensate financial players for making bad bets and hoping it trickles down to the pocketbooks of the average Joe works as well as any other trickle down program - not at all.