One of the gems of net-borne financial journalism is the FT's Short View currently assembled by John Authers. A couple of days ago he did a piece called Gold glitters once more in which this chart appears:

Now, inflation vs deflation is one of those debates occupying many minds; and easily one of the best discussions I have read is to be found in two blogs at Cassandra Does Tokyo (including the comments). By contrast this post is semantic, not strategic.

One funny thing with data relationships during crises is that many of them break down: chaos and linearity are generally incompatible bedfellows. The TIPS break even spread, I would suggest, is such a linear relationship that has become somewhat confused the deeper the financial chaos becomes.

The nominal 10 year US bond market is highly liquid. The inflation protected equivalent less so. Investors pay a premium for that feature of the nominals meaning their yield, other things being equal, is relatively smaller than those of the TIPS. Even in "regular" times the derived real inflation measure is therefore "artificially" low. In times of panic the effect is exaggerated. Hence, I think, that deflationary point on the chart.

Anoraks will tell you that there is also an inflation premium built into the nominals mitigating this effect. That's true (the Cleveland Fed has a full explanation here and a further background paper from the KC Fed is here). But during crises the magnitudes are not, probably, comparable.

That suggests that the unadjusted-for-these-effects Short View chart underestimates real inflation expectations and is more a demonstration of the extent of the flight-to-liquidity panic. The recent upturn possibly reflects this overshoot as well as the (consequent) growing enthusiasm for TIPS, led most notably by Bill Gross, as many persuade themselves that the deflation threat is overstated.

Course, those are opinions and all nuance is lost without discussion of timing and lags: just because deflation may be an exaggeration at this stage does not mean it will remain so.

Nonetheless, it used to be that such views as Mr Gross' could be tested (somewhat) using an adjusted TIPS measure conveniently provided by the Cleveland Fed. Guess what the Cleveland Fed has to say about it today:

"We have discontinued the liquidity-adjusted TIPS expected inflation estimates for the time being. The adjustment was designed for more normal liquidity premiums. We believe that the extreme rush to liquidity is affecting the accuracy of the estimates." (link)

Still, so what? Between adjusted and unadjusted it's all a question of magnitude not direction - isn't it? Well, it wasn't back in May the last time, unfortunately, I had a look at adjusted vs unadjusted US inflation expectations (chart below):

At the time mainstream commentators overlooked this divergence since not many were aware of the adjusted series. Which does not mean by any stretch, given what subsequently happened, that the adjusted model was in whack. But somewhere there is still a place for the point that what seems obvious isn't (maybe) always. In a chaotic crisis this is even more true.

Bookmark and Share


  1. Anonymous // 2/24/2009 03:07:00 AM

    A chart while impressive just hints at casual relations ships to limited data point econometric data is NOT a method. For what it's worth, my experience has taught me the following lessons. Be a creator of a reactionary not predicitionary system and tested it extensively. To be the
    developer, quant and creator is the only way to get this "stick to it iveness" strategy. Get yourself a good porfolio level backtesting engine and a high quality historical dataset like from these guys and begin your journey.

Related Posts with Thumbnails