Why Does Expansion Feel So Much Like Recession?
The technicians will be happy with today’s price action in equity markets. The movement of the S&P clearly above the summertime highs will clearly signal to some that the “inverted head and shoulders” on the daily chart – which projects to new equity market highs – has been completed and we are about to take a massive run to the upside.
I am less convinced, although the technical damage done to the bearish case by the 1.5% rally is undeniable and I would no longer lean short on a tactical basis. But volumes remained weak (912mm shares, still under 1bln even with a breakout), bonds rallied, and the VIX refused to move to new lows. By the way, the scale of the “inverted head and shoulders” is out of proportion with the length of the move into the formation, so technically it isn’t a reversal pattern at all. However, because the pattern is recognizable by all investors as well as dogs and some species of fish, there will be much ado about this move…especially this near to quarter-end.
The impetus for the rally? It probably was not the sharp widening of credit spreads for Ireland and Portugal. Greek Prime Minister Papandreou said that for Greece to default on its bonds would be a “tragedy” and he promised it would not happen. Didn’t we already move on from that? I hate it when someone promises something that we already took for granted. It makes it seem like we were prematurely taking it for granted, because otherwise he wouldn’t need to promise, right? (Of course, this is tongue-in-cheek. It would be a grievous error to assume that Greece will not default, or that all of the dealers who own piles of Greek bonds and are telling us it’s time to buy them are necessarily giving arm’s-length advice.)
A better candidate for goosing the rally, although it was already well under way at the time, was the announcement by NBER that the recession is over…actually, that it was over in June of last year. That is a curious decision in cycle dating. The Unemployment Rate rose another half-point after last June, and is only back down to 9.6% because of the decennial Census. The U-6 measure, which adds discouraged and marginally attached workers, is still 0.2% higher than in June of 2009 (see Chart).
Initial Claims has come down from the 550k-600k level seen in mid-June, but as I have written previously that merely indicates that the Lehman bankruptcy did not actually end the world. It was almost impossible for Claims to continue at that level; the spike was related to the crisis but the recession is distinct from the crisis. And Initial Claims remain at levels comparable to those seen at the worst points of the prior two recessions (see Chart below).
Capacity Utilization clearly bottomed in June 2009 (see Chart below), thanks to huge government spending, but fewer Americans toil in manufacturing than used to so surely this is less important than it once was.
I guess the real question here is, what’s the hurry? With total output still not up to the level it was in 2008, declaring an end to the primary recession mainly creates the possibility (which will become a probability unless the Bush-era tax rates are extended) that the single recession will shortly be declared to be two recessions. I submit that the only reason to do this now, the day before the FOMC meets and less than two months before the elections, is political. It isn’t like most of us were sitting around waiting for the NBER’s pronouncement, but I would expect the political advertisements of the incumbents are already being re-written to incorporate the declaration that “Senator so-and-so helped end the recession…”
It will certainly make it a little bit harder for the FOMC, when it meets tomorrow, to make any overt move towards QE2. I didn’t really expect any overt move this month, since it makes so much more sense at the next meeting (the day after the election), but it will be extra-difficult to be dovish when the NBER has just said that we are in an expansion that is already 15 months old.
Housing Starts tomorrow (Consensus: 550k from 546k) will help remind them, and other observers, that the economy isn’t exactly booming. The NAHB Housing Market Index, out today, remained mired at 13 despite expectations for a small pickup. That is lower than it was in June 2009, so this expansion has led somehow to worse housing market conditions (even with tax incentives!). The all-time low in January 2009 of 8 is not likely to be exceeded any time soon, but neither is the high of 2005 (at 72). Nor is the 2006 high of 57, or the 2007 high of 39. I think you get my point. If this isn’t a recession, then it is a pretty sorry excuse for an expansion.
In inflation-related news, the Barclays Index Advisory Council is having its annual meeting tomorrow. According to the Wall Street Journal, among the topics is whether to include TIPS in the Barclays (neé Lehman) Aggregate Bond Index. This is apparently partly what has been behind the recent bump higher in inflation-linked bonds, although it doesn’t explain the bump in European inflation swaps!
It makes very little sense to include TIPS in the Agg. It makes exactly as much sense to include inflation swaps, which have only inflation duration, as it does TIPS, which have only real duration, in an index of nominal bonds. Let me explain.
Irving Fisher’s equation says that
(1+nominal rates)=(1+real rates)(1+expected inflation)(1+risk premium).
For brevity, we usually combine the risk premium with expected inflation and say:
which reminds us that nominal rates are a function of real rates and expected inflation. Now, recall that duration describes the change in a bond’s price caused by a change in its yield to maturity. We can go further (and Waring and Siegel did, in their landmark 2004 article “TIPS, the Dual Duration, and the Pension Plan”), and observe that TIPS have only real duration – meaning that changes in their value today depends only on changes in their real yield to maturity – and inflation swaps have only inflation duration – their value depends only on changes in expected inflation. Nominal bonds are an equally unique case in which the inflation duration is approximately equal to the real duration. That is, a 1 basis point change in real yields will cause roughly a 1bp change in nominal rates; a 1 basis point change in expected inflation will cause roughly a 1bp change in nominal rates as well. Nominal bonds have both inflation duration and real duration, in (more or less) equal helpings.
By adding TIPS to the Aggregate Index, it will no longer be possible to calculate the Aggregate’s nominal duration. It will be a meaningless number. TIPS have an indeterminate nominal duration! By that, I mean that if you tell me that nominal interest rates changed by 1bp, I can not tell you anything about what happened to the price of TIPS. It depends whether that 1bp change in nominal rates came from a change in real rates (in which case TIPS will move), a change in expected inflation (in which case TIPS will not move), or a combination of these two in which case TIPS might move more, or less, or in the opposite direction.
For example: suppose that inflation expectations rise 2bps and real rates fall 1bp. This means:
- The nominal interest rate, which is approximately their sum, will rise by about 1bp.
- TIPS will appear in this case to have a negative nominal duration, because their prices will rise since real yields fell.
By adding TIPS, the Agg will become a confusing mishmash of unlike instruments. So I really hope the committee decides not to do it.
Two groups, however, would find such a change in the index very good news indeed. The first is the U.S. Treasury, which would find a sudden increase in demand for TIPS from bond fund managers who are trying to track the Barclays Agg. The second group is the community (fairly small!) of inflation experts, who will have a much larger audience for our services. Perhaps this is a good time to point out that I do consulting…