May to August 2006:
Some version of:
Recent indicators suggest that economic growth is moderating from its quite strong pace earlier this year, partly reflecting a gradual cooling of the housing market and the lagged effects of increases in interest rates and energy prices.September to December 2006:
Some version of:
Economic growth has slowed over the course of the year, partly reflecting a cooling of the housing market. .January 2007
Recent indicators have suggested somewhat firmer economic growth, and some tentative signs of stabilization have appeared in the housing market.March to June 2007:
Some version of:
...the adjustment in the housing sector is ongoing.August to December 2007:
Some version of:
...the housing correction is ongoingJanuary to April 2008:
... deepening of the housing contraction...June to September 2008:
...the ongoing housing contraction...
After that, housing was generally mentioned with regard to household wealth, housing starts, and the Fed's MBS purchases. Clearly housing was important, as it should have been.
Anyway, it bothers me a bit, because every Tom, Dick, and Harry, including those with much better credentials than me, has a criticism of the Fed, and I find most of those criticisms to be horrendously wrong. Why do I think I'm any different?
Yet, despite the fact that most everyone is wrong, the Fed is important. Actually, because most everyone is wrong, the Fed becomes important. What 'r ya gonna do?
I have recently been critical of rate hikes because of declining inflation and flatlined lending (although very recent bank real estate lending has turned up). I fear the Fed Funds target is already pushing above a neutral level, and if that is the case, the Fed will certainly be too slow to reverse course.
And, I think this issue with the balance sheet will make it worse.
I really don't understand much of the academic treatment of Fed policy. I certainly could be wrong. I am not formally trained in these matters, so there is much I am certain to miss. Please correct me in the comments if you feel that I am wrong and salvageable.
Much of the treatment of Fed policy seems to treat long term rates as if they are simply a product of the sum of the intervening forward rates (that much is somewhat defensible) and that those intervening rates are purely a product of future Fed policy stances in a ceteris paribus world. It seems to me that more than even a few quarters forward, rates will be increasingly the product of factors out of the Fed's control or factors in the Fed's control that push rates in the opposite direction of short term Fed machinations.
Low long term rates were, famously, a "conundrum" for Fed officials in 2005-2006. The fact that they see this as a mystery seems like part of the problem to me. Policy rates during the time were basically neutral. There was no reason for forward rates to rise because expectations were muted. Far forward rates were actually declining during the time. That's not because as the Fed was rapidly raising rates from 1% to 5.25% everyone was becoming convinced of their commitment to loose policy. It's because there was a sharp shift in sentiment, a massive buildup of savings, and a lack of risk-taking investment. Part of this is visible in sharply dropping housing starts and homeownership rates at the time. Yet, the Fed generally seems to have viewed those low rates as stimulative.
The other thing I think is strange about reviews of monetary policy is that the liquidity effect seems to be treated as the overwhelming effect on interest rates. This is especially weird because these are extremely liquid markets. So, the idea generally seems to be that when the Fed loosens, it buys treasuries, which injects buying pressure into fixed income markets, pushing up prices. When the scale of these effects is measured, it seems to usually be done as a sort of event study. This is common in financial markets. But, that is because financial markets are generally perceived to be extremely liquid, with little room for a liquidity effect. Prices shift immediately because future buying pressure is presumed as part of the current price. So, the effect of policy changes is measured by measuring the liquidity effect at the point of the shift. But, if the liquidity effect can be anticipated and priced by markets, then how can there be any liquidity effect at all?
So, my mind boggles when I see so many analyses of Fed policy that claim the QEs were effective because at the point of the policy shift, forward interest rates declined. This is a backwards interpretation of the signal forward rates are giving, based on a signal that shouldn't exist.
Now, I must admit that in the very short term there does appear to frequently be a shift in long term rates in the same direction as a short term policy shift that doesn't really make sense if, say, loose policy signaled by lower short term rates should cause long term sentiment, inflation expectations, and therefore rates, to rise. I can't say forward rates confirm my complaints on a second-by-second basis.*
Source |
One of the great failures of my trading life was that I was set up with a position that was highly sensitive to rising rates in the summer of 2013, and I managed to screw up the execution so badly that I ended up with nothing to show for it.
Anyway, what is odd to me is that conventional analysis of the QEs reverses this. So, unwinding the balance sheet is expected to raise long term interest rates. Here is Gavyn Davies in the Financial Times:
...the Fed will shed only around one third to one half of the assets it accumulated during the expansion phase, implying that the balance sheet will drop by $1.2-1.8 billion over several years. The total effect of this might be to increase 10 year bond yields by about 40-60 basis points...The article later says:
Janet Yellen has suggested that the expectation of balance sheet normalisation has already increased the bond yield in 2017 by 15 basis points, which she says is equivalent to two 25 basis point increases in the fed funds rate. The market seems to think that the balance sheet run down will have an even larger effect on short rates than Yellen implies, which is perhaps why it is so reluctant to price in the full rise in rates implied by the FOMC’s “dot plot” for 2018-19.So the liquidity effect, which must assume some sort of short term inefficiency in bond prices, has been anticipated. Rates have already risen by 15 bp because spot prices reflect the inability of future spot prices to anticipate temporary shifts in Fed buying. (Please tell me how I have this wrong! I really would love to know that I am wrong.)
So, just because of institutional inertia, and the fact that rent inflation is wrongly attributed to monetary policy, I would expect the Fed to hold the Fed Funds Rate too high for too long, just like they did in 2007. But, here, we have an added problem. The expectation is that long term rates will rise as they unwind the balance sheet that was built up by the QEs. But, I don't think they will. As the program is implemented, long term rates will fall. Maybe not by much. But, I think they will fall a little bit. And, instead of seeing that as a problem, FOMC members will say, "Well, we were worried, because we thought unwinding the balance sheet would raise rates, but rates are falling. This will stimulate asset markets, which will allow us to continue raising rates."
Of course, these events would lead to an inverted yield curve, which is a bad sign, and, according to research at the New York Fed that I (and many other finance folks) do find useful, is associated with recessions (and, thus, declining interest rates).
Will the Fed raise rates as the yield curve starts to flatten? Surely they wouldn't, I want to think. But, there is 2006 and 2007. And, the consensus FOMC view seems to be that, if anything, rates should have risen higher and sooner.
This all is moving in slow motion, and there is some chance that some sort of economic momentum will outrun Fed tightening. So, this may come to pass in 2018, later, or never. But, it seems to me that a defensive posture is increasingly called for, and that, at some point, probably after another rate hike, a speculative position on sharply falling rates will be lucrative. I suppose, if I'm not crazy here, there is a variation on the old cliché at work. "The Fed can remain irrational longer than you can remain solvent." On the other hand, I generally prefer the corollary to that old cliché, which I heard someone say a while back. "The market can remain solvent longer than you can remain irrational." This probably describes the bulk of the speculative losses that have been excused by the original version of the cliche, and might also apply to mine. You should probably hope that's true.
* On the other hand, there are some pretty epic, and sort of funny, counter-examples in favor of the POV I am describing here. I remember hearing a press conference with Ben Bernanke in June 2013, when, in response to the "taper tantrum" the Fed decided to promise to stretch out QE3 a little longer. He said, "These large and growing holdings will continue to put downward pressure on longer-term interest rates." Interest rates shot up literally as the words were coming out of his mouth. From the day before to the day after that FOMC meeting, the yield curve moved up about 40 basis points. Expectations destroyed the liquidity effect. To add a wrinkle to that, the move was effectively all in real rates - not in inflation expectations. The "taper tantrum" a month earlier had also been all in real rates. That might seem strange. But, when the primary effect of tightened monetary policy has been to collapse real investment, should we expect a reasonable loosening to be inflationary? I would like to attribute the May 2013 rise to improved real economic expectations, rather than to the proposed slowing of QE3, and the June rise to monetary accommodation. I realize that is a "just so" story. (Note, however, that the rise in May was gradual while the rise in June was a sharp reaction to the FOMC meeting.) But, if you judge that, then you need a story that explains coherently how rates moved sharply higher as the Chair of the Fed explained that today's policy shift is intended to move rates lower.
Excellent blogging.
ReplyDeleteI have an even weirder view on QE than you, and it seems to me obviously true to me, too.
I say QE was stimulative. Not much due to interest rates.
It was because people who sold Treasuries or MBS to the Fed could consume a part of the proceeds, or invest in property.
The orthodox view is that sellers into QE only bought other financial instruments.
Another way to look at it: Before QE, if Kevin Erdmann sold a $100k Treasury bond, someone else would have to give up the right to immediate consumption to buy the bond. So I buy Kevin's bond, and give up present consumption.
Okay, but with QE, Kevin effectively sells his bond to the Fed (who printed or digitized money). No one has to give up consumption.
Kevin may invest or consume his $100k in proceeds, or buy real estate.
I contend the $4 trillion in QE was stimulative, and the final stage of QE---when the Fed said it would be open-ended, depending on results---were effective.
Add on: If QE is truly inert, then by all means the Fed should conduct a few trillion more in QE. This unburdens taxpayers.
Add on, add on: Really, I wonder if QE plus federal deficits is all that different from money-financed fiscal programs, aka helicopter drops.
Keep in mind, QE didn't really inject much permanent cash into the economy. The Fed loaned cash to the government by buying treasuries, but then just borrowed the same cash back from the banks on similar terms. I don't think it can be said that QE funded real estate, although I do think it temporarily lifted inflation during implementation.
DeleteJust for fun: Unit labor costs in the last four quarters are….down.
ReplyDeleteYes, down.
https://www.bls.gov/news.release/prod2.nr0.htm
"Unit labor costs decreased 0.2 percent over the last four quarters."
Labor shortages! No workers! $96,000 a year for construction workers!
All consistent with declining unit labor costs.
Great find! This is useful for my Phillips Curve thesis.
Deletehttps://www.wsj.com/articles/with-merger-deal-blackstone-starwood-bet-on-being-americas-biggest-home-landlord-1502361000
ReplyDeleteinteresting
On QE:
ReplyDeleteWell, this gets insanely complicated, but before QE the Fed first created a entity that lent money to the 21 primary dealers. (Selgin says this was illegal).
My version is the Fed prints (digitizes) money, gives it to the primary dealers, who buy bonds from their clients, giving them cash.
What the clients did with the cash is unknown, except that the banked, invested it, consumed it or bought real estate, or converted it into actual paper cash.
The primary dealers collect cash from the selling bondholders and deposit into commercial banks, and that swells deposits.
But remember! The client of the primary dealers have $4 trillion in cash, and also there is $4 trillion in extra reserves.
Before QE, you had bondholders who owned $4 trillion in bonds, which can be thought of as "inert money." If I own a bond, I am inert.
Once I sell the bond, I become ert. hah-ha.