Some notes on market measures of inflation

I learnt some interesting facts about inflation-linked investment products. To begin with Sober Look had an intriguing chart showing an inversion in the TIPS yield curve.

Michael Ashton at Epiphany had an interesting explanation for this. Basically, short-dated TIPS begin to trade like gasoline futures. Like zero-coupon inflation swaps, TIPS are indexed to headline inflation, not core inflation. The most volatile component of headline inflation are gas prices, although in general large changes in gasoline prices will mean-revert to core inflation. We've had a large fall in gas prices, so near TIPS have fallen in value. More distant TIPS price in an expectation of gasoline reverting to core, and therefore are less sensitive to the fall in gas prices.

Michael explains here why inflation swaps are a better indication of true inflation than TIPS.

I learn here that the 5Y 5Y forward inflation curve is the market price for an inflation swap that starts in 5years and ends in 10 years.

In this post at David Glasner's blog, I point out that the fall in TIPS prices might not necessarily be a negative indicator. Starting with Michael's point about the sensitivity of near TIPS to gasoline prices, how much of the fall in gasoline prices is linked to fears of deflation, and how much to the massive rise in crude oil inventories coursing through North America? Domestic crude oil production is skyrocketing due to the technical combination of horizontal drilling and multiple stage fracturing. If this technological leap has contributed to the fall in energy prices, and the fall in energy prices has contributed to the fall in TIPS prices, then there is some component of ingenuity, and not wholly pessimism, at the core of the fall in TIPS prices.

Inflation swaps and TIPS

Econbrowser has a post on inflation expectations as measured via TIPS spreads. This links back to an earlier comment I made on Glasner's blog, in which inflation swaps are posited as an alternative to TIPS spreads. Here is the comment from Econbrowser:

"One might worry about characteristics in the TIPS market distorting the estimates of the real yields."

 and

 "The real yield curve starts at five years, so one can’t be sure what the real yield curve suggests for the horizon less than five years. "

 Why not use zero coupon inflation swap prices? They (supposedly) don't suffer from some of the same distortions as TIPS (liquidity premia), and you can get shorter terms.

 For instance, here is the 2 year inflation swap.

 http://www.bloomberg.com/apps/quote?ticker=USSWIT2:IND

Just subtract the 2 year swap rate from the nominal 2 year rate and you have the 2 year real rate. 

See the Cleveland Fed's explanation of their methodology for measuring inflation expectations, which includes the use of  inflation swaps

Liquidity options, liquidity premium, natural interest rate, TIPS, and inflation swaps

Commented on David Glasner's Once Again The Stock Market Shows its Love for Inflation:

The only problem here is the one that we talked about in a previous post (Unpleasant Fisherian Arithmetic) concerning the liquidity premium that assets carry. The reason for the rising TIPS spread could be (though not necessarily must be) that the liquidity premium on treasuries is shrinking relative to that of TIPS, and therefore TIPS are rising in price relative to Treasuries. This makes it hard to pass judgment on the hypothetical rate of return on capital.

 Anyways, you have already commented on this problem in your paper: “One possible cause of distortion in the yield on TIPS bonds and in the TIPS spread during the autumn 2008 financial crisis is that the yield on conventional Treasuries was depressed because of a liquidity premium. Even though the ex ante real interest rate was likely negative, because TIPS bonds were perceived as much less liquid than conventional Treasuries, TIPS bonds could not be sold unless they were discounted, so that their yields rose well above the (unobservable) ex ante real rate on holding real assets.”

 We were talking in the comments of the “Unpleasant Fisherian Arithmetic” post about how one could measure liquidity. I thought a bit about this. If there were a market for financial products, say options, that managed to price the pure value of an underlying asset’s liquidity premium, then you would be able to measure the value that the market placed on assets’ relative liquidity premiums and, from there, get a better idea for what portion of an assets total return is provided by an own-rate and what is provided by a liquidity premium. These sort of financial assets don’t exist, but if they did they would probably be sort of like credit-default swaps… more like liquidity-guarantee swaps.

Note that is follows from a previous comment I had concerning the impossibility of computing the natural rate of interest because liquidity interferes. See Unpleasant Fisherian Arithmetic

David responded:

Have you looked at how the Cleveland Fed tries to extract the inflation expectation from the TIPS spread?

I responded:

 I looked at it this morning. It seems to me that the Cleveland Fed is using alternative measures to extract inflation expecations given the problems posed by illiquidity in TIPS markets. They are including the inflation swaps market. In an inflation swap, one party pays a fixed interest rate, the other pays the inflation rate.

 Apparently swap markets didn’t suffer as much as TIPS markets did from liquidity problems in 2008, and for that reason the Cleveland Fed is using swap rates as one of their main indicators of inflation expectations.

 That being said, swaps are still traded contracts and bear some sort of liquidity premium, so it is still empirically impossible to back out a real rate without knowing what that premium is.

 For your records, here is the quote page for the 2 year USD dollar inflation swap spread: http://www.bloomberg.com/apps/quote?ticker=USSWIT5:IND

The Cleveland Fed discuss their methodology here. http://www.clevelandfed.org/research/workpaper/2011/wp1107.pdf