20 Sep 2012

TIPS Breakeven Bask

Federal Reserve, Inflation 7 Comments

Hey kids, I just want to make sure I’m interpreting these Bloomberg charts right. As usual, please be clear when you chime in, if you actually know what you’re talking about, or if you are just guessing. Guessing is welcome, but just be clear if that’s what you’re doing, please.

So I think this chart shows that the average expected price inflation over the next 5 years was about 1.9% in the beginning of the month, then surged to about 2.4% right after QE3, and then fell back to about 2.17% as of right now.

This chart, in contrast, shows that the expected rate of price inflation in 2017 (i.e. not the average, just the one-year rate five years from now) was about 2.4% in early September, then surged to about 2.85% after QE3, and now is about 2.80%.

Am I reading these charts right? In particular I’m worried that the number reported for these securities, actually translates into the percentage price inflation rate. Is that the right way to read them?

(Also, I’m aware of all the pitfalls in these methods of calculating “the market’s expectation of price inflation.” But I just want to make sure I’m reading the charts right.)

7 Responses to “TIPS Breakeven Bask”

  1. Chris Ridder says:

    Dr. Murphy,

    My opinion is that your understanding is correct. It has been 7 years since I worked with creating forward rates so take this as a best guess. Look at the description of the first chart.

    “Rate Profile Information for USGGBE05

    Yields are yield to maturity and pre-tax. Indices have increased in precision as of 5/20/2008 to 4 decimal places. The rates are United States breakeven inflation rates. They are calculated by subtracting the real yield of the inflation linked maturity curve from the yield of the closest nominal Treasury maturity. The result is the implied inflation rate for the term of the stated maturity. Please see {YCGT0169 Index DES} 2 for “Members” of the US Treasury Inflation Linked curve. The pricing source used for both inflation and nominal is “BGN”. You can locate the rates under {ILBE}”

    This seems to be what you describe.

    The second chart seems to be a 5 year forward rate; simple enough, but is the forward rate from a 1 year bond, 5 year, 10 year?

    5 year forward, 1 year out
    5 year forward, 5 year out
    5 year forward, 10 year out
    ???

    I think you would have to look at the paper listed under the description.

    My guess here is that this would be the inflation rate expected for 1 year, 5 years into the future.

    I hope this helps.

  2. Nick says:

    One caveat. It shows expected inflation if there is a free market.

    Now if pension funds have to invest in treasuries, and have to match inflation, they end up matching liabilities using gilts. That’s an artificial market. If there is a shortage of inflation linked products compared to demand, it drives down the ‘expected inflation’

  3. Jeff Lonsdale says:

    Fed5Year is the Fed’s version of the 5 year breakeven 5 years forward. So it is 2017 through 2022. It looks a lot like USGG5Y5Y but varies slightly due to how the Fed smooths out the treasury yield curve compared to bloomberg’s methodology.

  4. Matthew Corso says:

    Here are some BE plots against (& seems to predict) deltaCPI – http://bit.ly/Pgpsvm

    Made sure to bit.ly it as Dr. Murphy seems to be allergic to the EPJ-related anonymous that reside in the very Austrian fight club of the remnant.

  5. Ghengis Khak says:

    ‘Bask’ isn’t a stellar portmanteau for the purpose in which you use it since it already has a completely unrelated meaning. When I see titles like XYZ Bleg, my brain immediately tells me that you basking in victory on topic XYZ which isn’t true.

    How about ‘bleg’?

    • Ghengis Khak says:

      Sigh…

      XYZ Bleg ==> XYZ Bask.

  6. John says:

    I always wonder what Scott Sumner is talking about when he says that inflation is below expectations based on this five year TIPS spread. However, interestingly, the Cleveland Fed, using TIPS spreads and some other measure, are reporting 10 years inflation expectations only around 1.37%. I have no idea how there could be such a large difference.

Leave a Reply