Streetwise Professor

October 25, 2010

At a TIPping point?

Filed under: Economics,Financial crisis — The Professor @ 8:48 pm

Tyler Cowen notes that yields on inflation protected securities, TIPS, have turned negative.  He links to this NYT story for more details.

Tyler argues that this negative TIPS yield indicates that expected real interest rates are negative.  But as I note in a comment to his post, that’s not right.  TIPS have an embedded option resulting from an asymmetric exposure to positive and negative inflation rates.  If inflation is positive, the Treasury gives you the contractual coupon on the TIPS on a principal that is increased by the inflation rate.  If inflation is negative, Treasury gives you the contractual coupon, but doesn’t adjust the principal.  So you get a real return equal to the coupon when inflation is positive, and a real return equal to the coupon minus the rate of inflation if the price level declines; that’s bigger than the coupon.  In contrast, with a nominal Treasury, your real return is the coupon minus the rate of inflation regardless of whether that inflation rate is positive or negative.

Investors bid down the coupon on the TIPS to reflect the value of this option.  The time value of this option is particularly big when (a) expected inflation is approximately zero (as this makes the option at the money), and (b) there is a lot of uncertainty about the future inflation rate.

That’s a pretty fair description of current market conditions.  Inflation is hovering around zero now, but there are real risks of either a big spike in inflation or a deflation.  There is intense debate in the investing and policymaking communities about whether inflation or deflation is now the bigger risk.  You look at the ballooning monetary base, and you recognize that if velocity picks up even a bit and the Fed doesn’t react quickly, and prices could shoot up.  The plummeting dollar and rising commodity prices are also inflationary signs.  But there is also real concern about deflation, a liquidity trap, etc.

So to me, that means that there is huge uncertainty about the future course of the price level, and that translates into a big value of the option embedded in TIPS.  That in turn, in a situation of an expected inflation rate of zero and relatively low real returns given the sluggishness of the economy, can produce negative TIPS yields.  As a result, I don’t see the negative yields as an indication of negative real returns.  Instead, they are an indication of the huge uncertainty about the future course of prices.

Here’s what I posted as a comment on Tyler’s site:

You need to take into account the terms of the TIPS when drawing conclusions about expected real rates of return. Taking these terms into account, a negative TIPS yield does *not* necessarily mean a negative real rate of return.

The crucial thing to remember is that the principal amount of TIPS is increased when inflation is positive, but the principal amount of TIPS is *not* decreased when there is deflation. As an example, if I own TIPS and there is a 10 percent deflation, I earn a 10 percent real rate of return on my TIPS holdings because the principal amount on the TIPS stays constant, but the real value of that principal rises by 10 percent.

Thus, negative yield could be signaling a high probability of deflation. This means that you cannot conclude that a negative TIPS yield implies a negative real rate of return. If the expected change in the price level is -2 percent, and the TIPS yield is -.5 percent, the expected real rate of return is 1.5 percent.

But an expected deflation should affect nominal yields too. If the real rate of return is 1.5 percent, given an expected change in the price level of -2 percent, nominal bonds should be yielding -.5 percent too.

Right now, TIPS in the 2 year maturity range are yielding about -.85 percent. Nominal 2 year yields are about +.35 percent. TIPS in the 5 year maturity range are yielding about -.3 percent, whereas 5 year nominal yields are about 1 percent.

The features of TIPS can result in negative TIPS yields and positive nominal yields when there is *uncertainty* about inflation. TIPS have an option-like nature. TIPS investors are protected against inflation when the price level goes up, but can benefit from deflation. In contrast to this asymmetry, nominal bond investors are hurt by inflation and helped by deflation in a symmetric way. That is, a TIPS is like a nominal bond plus a call option on inflation struck at zero.

There’s another way to look at this. In real terms, the payoff to the TIPS is its yield minus min[price level change,0]. The payoff (in real terms) to the nominal bond is its yield minus (price level change). The “min” expression in the TIPS payoff and its absence in the nominal bond payoff means, again, that the TIPS embeds an option.

This embedded option reduces the TIPS yield relative to the nominal yield. If the expected inflation rate is zero, then the nominal yield equals the real return. But the TIPS yield must be below the real rate of return to compensate for the value of the embedded option. The more uncertainty about price level changes, the more valuable is this option, and the more depressed is the TIPS yield. The difference between the nominal and TIPS yield is basically the time value (or the “extrinsic” value) of the price level change option. This time value is greater, the more volatile the inflation rate.

So right now, where a zero expected inflation rate is not an unreasonable assumption, the yield on nominal Treasuries is a more reliable estimate of the real rate of return. I would interpret the present situation as being: (a) low but positive real rates; (b) approximately zero expected inflation; but (c) substantial uncertainty about the future change in the price level. This substantial uncertainty is understandable, given the unprecedented nature of the current economic situation, and the equally unprecedented nature of monetary policy.

Print Friendly

13 Comments »

  1. I am going to move from commodities into rates and foreign exchange pretty soon. Very instructive! Have shared it on facebook.

    Comment by Surya — October 25, 2010 @ 10:36 pm

  2. Thanks Surya. Feel free to friend me on FB if you’re interested (though I’d understand if you want to maintain your anonymity).

    The ProfessorComment by The Professor — October 25, 2010 @ 10:57 pm

  3. “there is huge uncertainty about the future course of the price level, and that translates into a big value of the option embedded in TIPS. That in turn, in a situation of an expected inflation rate of zero and relatively low real returns given the sluggishness of the economy, can produce negative TIPS yields.”
    So Paul Krugman seems to be wrong writing
    “Financial markets seem convinced that quantitative easing will be highly effective at solving at least one problem: inflation running well below the Fed’s 2-percent-or-so target”
    http://krugman.blogs.nytimes.com/2010/10/26/do-investors-expect-too-much-from-bernanke/
    The markets just feel the rise in the uncertainty b4 FOMC.

    Comment by a.russian — October 26, 2010 @ 4:01 am

  4. […] This post was mentioned on Twitter by Ashwin Parameswaran, burton hollifield. burton hollifield said: At a TIPping point? http://bit.ly/dyk3xZ. Explaining? Negative TIPS yields […]

    Pingback by Tweets that mention Streetwise Professor » At a TIPping point? -- Topsy.com — October 26, 2010 @ 5:14 am

  5. Similar approach is used by Jens Christensen, a senior economist at the Federal Reserve Bank of San Francisco:
    “In the TIPS contract, deflation protection comes from the fact that the investor is paid at maturity the greater of the original principal or the inflation-adjusted principal”
    http://www.frbsf.org/publications/economics/letter/2010/el2010-32.html

    Comment by a.russian — October 26, 2010 @ 7:21 am

  6. Thanks Prof. This is the first thing I’ve read that’s made any sense of it. Good blog.

    Comment by DonutBoy — October 26, 2010 @ 8:24 am

  7. Thanks for the analysis. Helpful in understanding.

    Comment by Michael — October 26, 2010 @ 9:49 am

  8. Thanks Michael and DonutBoy. If you want to read something funny, check out the WSJ’s attempt to explain the negative yields. Complete gibberish.

    The ProfessorComment by The Professor — October 26, 2010 @ 12:01 pm

  9. By comparing on-the-run TIPS yields (which, as you point out, have an “embedded option”) against inflation swaps (which don’t), it should be possible to put a number on the price of that embedded option. The implied volatility of that option would then quantify the market’s uncertainty about future inflation.

    The main caveat to this idea, I think, is that TIPS have a higher liquidity premium than inflation swaps. Maybe just compare against off-the-run TIPS instead of inflation swaps?

    Might make an interesting topic for a paper.

    Comment by Nemo — October 26, 2010 @ 7:57 pm

  10. Nemo–I had that thought exactly, but don’t have ready access to inflation swap quotes. On vs. Off the run comparisons (holding maturity constant) would be a decent way to isolate the liquidity effect.

    The ProfessorComment by The Professor — October 26, 2010 @ 8:11 pm

  11. Nemo–inflation swaptions would also provide useful information. Indeed, you’d really need the vol information in a swaption to value the option embedded in the TIPS.

    The ProfessorComment by The Professor — October 26, 2010 @ 8:51 pm

  12. Bloomberg provides current rates for inflation swaps for free; e.g. 5-year. (I have links for other durations on the sidebar of my site.)

    But I know of no source for inflation swaption quotes, and I agree that would be a better way to obtain this number.

    This seems like the sort of indicator the Fed would want to monitor, since they pay so much lip service to “inflation expectations” being “well-anchored”.

    Comment by Nemo — October 27, 2010 @ 12:10 pm

  13. Nemo–Thanks. I have my RA working on this right now. I’ve seen quotes on Bloomberg (screen shots, actually, of Bloomberg screens from RBS) of inflation caps and floors. She’s looking for those too. I’ll let you and other readers know what she finds. If I get decent data, I’ll write a new post.

    The ProfessorComment by The Professor — October 27, 2010 @ 2:36 pm

RSS feed for comments on this post. TrackBack URI

Leave a comment

Powered by WordPress