I ran across a fine posting tonight at Illusion of Prosperity which raised a topic that has been puzzling me for some weeks. Why are TIPS yields so low, can they go any lower, and are the “inflation expectations” implied in the TIPS-Treasuries spreads meaningful? The implied inflation expectations have been trending down through 2010, into the 1-1.5% range and well outside the 2-3% zone that the market lived in from 2004-2008, and the Fed is clearly worried about this.
Mark pointed out that those worried about deflation will pick Treasuries since in that case TIPS have zero yield. Those worried about serious inflation will invest in hard assets. So why are so many people buying TIPS?
I had two comments regarding the TIPS:
First, not everyone is absolutely convinced that deflation is imminent, nor is everyone convinced that inflation is imminent. But many are convinced that we’re going to have one or the other, but are just not sure which, as that appears to be a political choice. For those people, TIPS until recently made a decent “safe either way” option, particularly for tax-protected accounts. [Note: at current yields I no longer think TIPS are a safe choice against inflation… we are currently letting our TIPS ladder run off and finding other uses for the money until TIPS yields recover. ]
Second, I think it’s a serious flaw in Fed policymaking to assume that the TIPS and Treasuries markets are efficient enough to produce a meaningful “inflation expectations” measure through the relative yields. One of the great lessons of the 2008 crash is that modern portfolio theory and the efficient markets hypothesis are both deeply flawed. I suspect it would be very interesting to plot the “TIPS/Treasuries Inflation Expectations” chart against actual 5, 7- and 10-year CPI growth, and see how accurate the metric has(n’t) been! For instance, in 2003 the implied 5/7/10 year inflation expectation was about 1.5%/year, and the actual 5/7 year inflation was 2.5-3.0%. Similarly, in 2008 the implied inflation swung from 2.5% to -1% within a few months, and of course most of that range will prove in a few years to have been grossly wrong.
For my part, I find the TIPS yield alone to be very provocative. Bonds historically are expected to return 2-3% above inflation. So the current 10-, 20- and 30-year TIPS yields of 0.7-1.5% could be taken, in and of themselves, to imply a negative inflation rate. Alternatively, the 0% yield on the 5-year (and the other anomalously low yields) could be seen as implying that other asset classes look so overpriced (or corrupt and unappealing?) as to make even a 5 year zero real return rate look attractive!
Or maybe it just means that Treasury needs to issue more TIPS relative to Treasuries, to bring supply back in line with demand! In that mindset, one can take the “implied inflation expectations” with a grain of salt, since Treasury controls the relative supply and can therefore radically influence the yield spread! The U.S. recently tripled the annual issuance of Treasuries… and the Fed is in QE mode… so what else would one expect but low real TIPS yields? The real mystery may not be why are TIPS yields so low, but why are Treasury yields so low? Higher Treasury yields are vital to increasing the “inflation expectations” implied in the Treasuries-TIPS spread… yet the Fed is actively buying Treasuries as it rolls over its (illegal) MBS holdings, and presumably (if it goes into the QE mode promised in this week’s FOMC statement) will buy even more!
A tangled web, indeed…