Home > Causes of Inflation, CPI, TIPS > Price-A-Palooza


The monthly price-a-palooza from the Bureau of Labor Statistics, coinciding with the auction of $15bln in new 10-year TIPS, also shared the day with Housing Starts and Initial Claims. Dispensing with those two details first: Housing Starts was 657k, which was a disappointing shortfall after the strong NAHB number on Wednesday. Initial Claims looked strong, at 352k, but this comes with two caveats that are really the same caveat. The first is that the prior week’s claims were over 400k; the second is that year-end seasonal adjustment to these figures makes them nearly useless. It is probably most-useful (which is to say, only barely useful at all) to think of the last two weeks together and figure that the current pace of Claims is “about” 375k, right in the middle. This would coincidentally be consistent with the level of claims prior to the year-end volatility (see Chart, Source Bloomberg).

Also out was CPI, of course. Headline CPI was unchanged on a seasonally-adjusted basis, but Core CPI came in as-expected at +0.1% keeping the year-on-year rate at 2.2%. “Stabilization!” screamed the pundits, but it isn’t yet so. Core CPI was actually up 0.145% before rounding, which means we were a mere 0.005% from what would have been considered a surprise for the cheerleaders on CNBC. The year-on-year figure, too, rose nearly a tenth, from 2.153% for the year ended in November to 2.230% for the year ended in December. Rounding giveth and rounding taketh away! Yes, a rise in the year/year pace of only 0.08% represents a modest slowing, but that would still add a full percent to core CPI over the next year were it to persist.

It probably will not persist, because housing is going to begin to act like ballast on the number over the next few weeks, but core ex-housing is already at 2.46% and I see few reasons to expect it to not continue its rise.

In any event, we should remember that the 1.6% rise in core inflation over the last 14 months is the fastest such rise since 1984. A little respect, please, for the inflation process. I know it doesn’t always act like an instant-gratification video game, but looking at the chart of 14-month changes, below, can you tell me that this advance is so shaky that stabilization is automatic from here? I didn’t think so.

Rolling 14-mo absolute change in core CPI


A chart of the current y/y changes and the previous y/y changes is shown below.

Weights y/y change prev y/y change
 All items




  Food and beverages
















  Medical care








  Educ & Communication




  Other goods and services




As you can see, Food & Beverages, Medical Care, Recreation, and Education & Communication, which collectively represent 34.1% of the basket, are still accelerating. Transportation (mostly because of energy), Apparel, and Other (24.4%) are decelerating. Housing looks like it is a wash, but it isn’t really, yet:

Weights y/y change prev y/y change








   Fuels and utilities




   Household furnishings & operations




As you can see, most of the apparent slowdown in Housing is also in the energy sector, while Shelter is still rising. If we put 32.4% as “accelerating” and 5.1% as “decelerating”, then we still have 2/3rds of the basket accelerating. Again, that won’t be the case for long, but it is early to call the end of the inflation rise. Note also that the Median CPI put out by the Cleveland Fed actually ticked up to 2.3%, so it is above core CPI (although for all intents and purposes, tied).

There is a reason that many models are calling for a flattening out of core “soon.” The most-honest reason is that some models establish an important role for “anchored” inflation expectations. I am familiar with the literature on inflation anchoring, and I find it completely unpersuasive. I also do not believe in poltergeists. Both theories seem to explain certain phenomena, but neither has compelling empirical data to back them up. While it does seem that poll respondents “anchor” their responses (and it seems they anchor them to the most-recently-released figure that all media trumpet), there is not any evidence that consumers and producers actually change their behavior at all because of that “anchoring.” However, if it’s in your model, it’s one reason you might think that core inflation above 2% ought to begin reining itself in.

The sneakier reason that some economists are calling for a flattening out of core inflation is that we all can see the conditions in the rental markets, and that follows the recent renewed downturn in housing prices which is due to the inventory overhang. So it’s easy to say “Core won’t reach 3% soon.” It would be remarkable if it did. Indeed, it’s remarkable it’s already this high given that it has the unwind of an epic bubble to deal with. The current 2.23% core rate is above what our models expected to see realized for 2011, because shelter inflation rose more than we expected. What’s more interesting is to forecast what is going to happen to core ex-shelter, which is already above the Fed’s target and rising.

Our models take note of the fact that 52-week M2 money growth is now at 10.71%, just slightly high of a one week peak above that level in January 2009. Before that, and a dramatic one-week spike in September 2001 (wonder what that could have been?), you have to go back all the way to 1982 to find faster year/year money growth. Unlike last year, too, it’s not all going into the vault – corporate credit growth over the last year is now up over 3% and still rising. So in my opinion, is probably too early to send the hawks back to the eyrie.

Remarkably, my measure of the spread of perceived inflation over actual measured inflation – I think of it as sort of an ‘angst’ index – reached an all-time low (going back 12 years) this month. The index is driven by, among other things, the volatility of price changes and the dispersion of price changes. In other words, inflation has been rising in a comparatively stealthy, orderly way, which tends to diminish our sensitivity to it. Not unlike a frog being cooked in water that is being brought slowly to boil, come to think of it.

And yet, with everyone telling us not to worry about inflation, with 10-year real yields negative, with dealer risk-appetites still low, and with headline inflation tumbling back down to only 3%, the Treasury today sold $15bln 10-year TIPS 3bps better through the screens at the bidding deadline. Dealer demand was strong, as was the overall bid:cover. Someone wants inflation protection here!

On Friday, Existing Home Sales (Consensus: 4.65mm vs 4.42mm last) is the only data. I think we are also supposed to hear about the private-sector cram-down from Greece. The word was that there was supposed to be an agreement “by the end of the week,” but come to think of it I guess maybe they didn’t say which week! In any event, conditions look good for a return of the 10-year yield above 2% (closed today at 1.97%). While calamity can strike at any time, a fair amount of calamity is already priced into Treasuries. Moreover, it’s only calamity of a deflationary kind, not a calamity of an inflationary kind…and it isn’t at all clear that that is the most likely kind of calamity here.

  1. onebir
    January 20, 2012 at 2:47 am

    How is the housing element done in the US?

    If housing loan rates get factored in, that and the exchange rate both seem to pose major upside risks to core CPI if (/when?) investors decide they’ve had enough T bonds.

    (Russia & China seem to have decided this already.)

    • January 20, 2012 at 10:16 am

      onebir, I wrote a piece on how shelter inflation is measured here: http://seekingalpha.com/article/251393-under-the-hood-of-the-cpi-shelter-inflation

      The short answer is that loan rates are not factored in, for about the last 30 years or so. And I believe that’s true for all major economies now. The loan is an investment expense. Now, it might seem like “well, it’s a cost for me to live there,” but the fact is that you’re financing both your investment in the house AND your consumption of the house (living there). It turns out that it makes more sense to simply measure directly your cost of living in a house, which means to measure your effective rent. Because you would choose to ACTUALLY rent rather than live in a house, if it was much cheaper to do so (and vice-versa), deviations in the cost of home-ownership are likely due to the value of the home as an investment. The bubble in the 2000s was not due to the fact that it was becoming more expensive to live in a house, but because it was becoming more expensive to INVEST in a house (and obviously people wanted the investment value and believed it had value…otherwise, they would have rented!).

      It’s an argument that sounds convoluted to the uninitiated, but it makes sense. I think I explain it a little better in that link above.

      • onebir
        January 22, 2012 at 2:59 am

        Sure – that makes sense. It’s just that the UK RPI included a mortgage interest component (part of the difference between RPIX and RPIY). But since the BoE switched to targetting CPI, which I think is based on EU Harmonised Index of Consumer Prices methodology, RPI isn’t so important & I think your comment about all major economies is de facto correct.

        Incidentally, the chart of housing CPI vs housing inventories illustrates the other point I was trying to make. In a long recession, the element of housing inventories in very poor condition is likely to grow. That probably puts less downward pressure on rents (& house prices) than would otherwise be the case. So it might be worth looking out for signs that the relationship is changing.

      • January 22, 2012 at 10:08 am

        I am not sure how to leave a reply to a reply to a reply! This is in reply to your comment, onebir, #3 here.

        Not only did the BoE choose to target CPI, all of the new bonds issued by the UK are linked to CPI rather than RPI.

        I think it just makes more sense to try and separate the two main considerations of homebuyers – a place to live and an investment. The rental-equivalence method might not be the purest way (although I think it’s pretty good and can’t think of a better one off the top of my head…it performed well in the bubble!), but I think it’s a success story for the BLS.

        I appreciate the comments! Thanks for bringing in the UK DMO to this discussion!

  2. Frank R.
    January 20, 2012 at 9:55 am

    Is there anything in this week’s Fed overall SA revisions for 2011 to M2 that may be worthy of note?

  1. No trackbacks yet.

Leave a Reply

%d bloggers like this: