It was another low-volume day on the stock exchange, but at least there was some interesting news and some intriguing undercurrents. And, of course, CPI day tomorrow!
On the data front, strong Housing Starts (596k vs. 539k expected) kicked off the day. The market reacted with characteristic giddiness to the 14.6% rise month-on-month but as the oft-shown chart below shows, the correct response is "who cares."
Wow, 14% improvement. Wheeeeee!
On the flip side, Industrial Production was surprisingly weak although off an upwardly-revised prior print. Overall production was -0.1%, but that represented a rise of 0.3% in manufacturing output and a strange drag from mining and utilities (strange because, if I remember correctly, January was pretty dang cold). This, and the PPI release generally (core PPI was an unpleasantly-high surprise, but with the usual caveat that "it's PPI"), ought to be jointly tossed into the bin of irrelevancy.
There were some geopolitical events. Oil jumped 0.9% when the Israeli Foreign Minister suggested that two Iranian combat ships are planning to sail to Syria through the Suez Canal imminently. This hasn't happened in quite a while, and since it represents two of Israel's implacable enemies greeting each other with a brotherly embrace they are unsurprisingly concerned. Even with the Egyptian pinch point past, the region continues to look like a frying pan on which some drops of water have been sprinkled, with sizzling and jumping all over the place. (I suppose that's why it's called a 'hot spot.') There were demonstrations in Libya, too. Was "Iranian warships provoke a military confrontation with Israel" on anyone's list of possible buzzkills for the market? This is why value, and a margin of safety, matter so much - after you have planned for every foreseeable eventuality, you need to prepare for the unforeseeable ones (although, in the grand scheme of things, I suppose that Iran trying to tick off Israel isn't exactly a bolt from the clear blue sky).
The Fed provided some entertainment today in their release of the minutes from the January meeting. According to the Fed, yields rose during the intermeeting period "...in response to data releases that generally pointed to some firming of the economic recovery," but inflation breakevens "...moved up, likely pushed higher by rising prices for oil and other commodities and by the firming of the economic outlook." As far as anyone can tell from the minutes, the Fed thinks QE2 is hardly being noticed (which raises the question, "then why did they do it?").
Now, in reading this next section keep in mind the great confidence that has been evinced by the Fed Chairman (and other Fed speakers) publicly when the topic of tightening comes up. We are told that it's not a problem because the Fed is keeping an eye on inflation and can pull the plug at will.
"Regarding risks to the inflation outlook, some participants noted that increases in energy and other commodity prices as well as in the prices of imported goods from EMEs posed upside risks. Others, however, noted that the pass-through from increases in commodity prices to broad measures of consumer price inflation in the United States had generally been fairly small. Some participants expressed concern that in a situation in which businesses had been unable to raise prices in response to higher costs for some time, firms might increase them substantially once they found themselves with sufficient pricing power. In any case, the factors affecting the ability of businesses to pass through higher prices to consumers were viewed as complex and hard to monitor in real time. Most participants saw the large degree of resource slack in the economy as likely to remain a force restraining inflation, and while the risk of further disinflation had declined, a number of participants cited concerns that inflation was below its mandate-consistent level and was expected to remain so for some time. Finally, some participants noted that if the very large size of the Federal Reserve's balance sheet led the public to doubt the Committee's ability to withdraw monetary accommodation when doing so becomes appropriate, the result could be upward pressure on inflation expectations and so on actual inflation. To mitigate such risks, it was noted that the Committee should continue its planning for the eventual exit from the current exceptionally accommodative stance of policy."
This is a treasure trove of interesting clauses. "Hard to monitor in real time" is not exactly comforting given the sums of money involved and the stakes at risk if the FOMC is wrong about inflation. And the last sentence reads, to me, more like "we all decided that it is important to keep talking about how well-prepared we are to unwind all this stuff."
Consonant with this intention, Dallas Fed President Fisher said today that the most logical way for the Fed to begin pulling back the stimulus, when appropriate - and Fisher is likely to be one of the earlier advocates of that, by his own admission - is "to undo what you did last." That is, the Fed should consider selling off their Treasury portfolio as a first step in tightening. This is eminently sensible, except for one objection - if they do that, the bond market may collapse. Now, perhaps not, but as I have pointed out before the difference between the Fed buying $100bln per month and the Fed selling $100bln per month is the difference between 0 net issuance of Treasuries and $2.4 trillion/year.
It is still a reasonable thought experiment, though. If the policy to date isn't reversible, what does that say about the real point of the policy? If rates will end up much higher, it implies the market is, in fact, close to saturated with Treasury securities and that in turn implies that the Fed did, in fact, act as an enabler to a profligate Administration and Congress. The Fed needs to unwind these purchases if only to prove that they are indeed independent.
.
Tomorrow is a big day in inflation. We start with CPI (Consensus: +0.3%, +0.1% ex-food-and-energy) and end with the auction of $9bln 30yr TIPS. Oh, there's also Initial Claims (Consensus: 400k from 383k) and the Philly Fed index (Consensus: 21.0 vs 19.3).
The consensus estimates for both Philly Fed and Claims imply not just growth, but accelerating growth. The cycle high for Philly Fed was 21.2 in May of last year when the Census was hiring workers with both hands. But more to the point, values much about 20 are not all that common. In fact, since 1985 a 21.0 would be at the 84th percentile of all values of the Philly Fed Business Outlook (see Chart). So 21.2 isn't "a little improvement over last month." This is a rate-of-change survey, and getting much above 20 (at least on a sustained basis) suggests a booming economy. I don't see it.
Philly Fed much above 20 is not an easy bet.
The 400k estimate for Claims, too, represents expectations that most of the 27k downward surprise last week was real and that the economy is kicking to a higher gear. This may well be so, but again I don't see it. The average for the last 4, 8, and 12 weeks is between 415k and 419k on Initial Claims. That looks closer to the central tendency to me. I suspect last week's low level of Claims was snow-induced. We will see. Another low print would force me to reconsider that hypothesis, because (ex-blizzard) we are in a part of the year that is usually easier to seasonally-adjust.
Now for CPI, clearly the star of tomorrow's show. For a review of last month's number and a general outlook for the year, see my comment from Jan 16th and a subgroups-summary here. The consensus figures would push the annualized headline number from 1.5% to 1.6% and core from 0.8% to 0.9%. The consensus forecast for headline actually is closer to 0.25%, just barely rounding up to 0.3% although comfortably bumping the y/y number higher. It is that extra edge with respect to rounding that makes the risk of an 0.2% or an 0.4% about even. A true 0.3% month/month change should cause the year/year rate of headline to go to 1.7% actually.
Here's a fun exercise. If we take the year-on-year rates of change for each of the eight major subgroups of CPI (Food & Beverages, Housing, Apparel, Transportation, Medical Care, Recreation, Education and Communication, and Other) for Nov '09 to Nov '10, look at how those rates of change evolved last month (Dec '09 to Dec '10), and extend all 8 of those trends one more month, the headline y/y tomorrow would be 1.9%. The core figure would be 0.9%, but just barely. For example, two months ago the y/y change in Housing CPI was +0.01%. Last month, the y/y change had risen to 0.287%. If that improvement continues to 0.564% this month, it will go a long way to offsetting any drags from the declining/decelerating groups. But this also shows you what sort of momentum the declining groups have had.
The risk this month in the headline is clearly to the upside. The risk in core is muddy because of the conflicting crosscurrents, but it has been more than a year since we saw a bona fide 0.2%. It is hard to disagree strongly with the consensus this month.
After CPI is out of the way (be sure to read tomorrow's commentary where I will break down the subgroups a bit), the Treasury will be auctioning $9bln of 30y TIPS at something around 2.25% real yield. Recall that last month, a 10y TIPS auction tailed 6bps and then sank in the aftermarket to consume that entire "Dutch treat." That was sloppy, and people are concerned about the Treasury's ability to sell $9bln 30y TIPS (so is Treasury, which is why they're selling just 9bln and not 10bln).
But here's what you need to know about a 2.25%, 30-year inflation-linked bond yield: on a global basis, that's a pretty fat long-term yield at the moment. The French 2040eis yield 1.70%. The Canadian 2041s yield 1.48% (if you can find them) and the 2044s about the same. In the UK, inflation-linked bonds maturing in 2040 flash a sporty 0.85%; the 2042s are 0.80%; the 2047s are 0.74%; the 2050s are 0.71%, and the 2055s are 0.68%. And that's about it for long-dated issuance (unless you feel game to invest in Italy for 30 years, in which case you can earn 3.15% plus European inflation...whatever that means in 30 years). So if you want long-dated inflation-linked bonds, you can invest sub-1% in the liquid UK market, or 1.4% in the illiquid Canadian market, or roll the dice on Italy. Or, you can buy 2.25% or so from the U.S. Treasury.
There is ample global demand for very long-dated inflation-linked returns, but the supply is scarce. Buying the US instead of the UK gives you a 1.4% advantage. Since most of what causes inflation in both of these countries are factors in common and the rest can be hedged,[1] it should be a popular investment for non-US investors. Despite the rotten performance at the 10y TIPS auction last month, I think that this auction will see good end-user demand and dealers ought to be comfortable owning some.
[1] Well, it isn't easy to do but Enduring Investments has a method.
