This is the first of a couple of blogs that look at why inflation breakevens may not be a measure of ‘true’ inflation.

One of the most useful relationships in finance is the Fisher equation.  In its original form it states:

Nominal yields (n) = real yields (r) + expected inflation (f) + risk premium (p – for unexpected inflation)

Over the years the risk premium and expected inflation components were often shunted together and expressed as a single factor and loosely termed ‘breakeven inflation’.  This is often widely interpreted as a measure of inflation expectations.  However, this definition misses a number of subtleties that these notes will seek to address.

Note that the breakeven is an implied value.  It is possible to observe nominal yields (e.g. from non-inflation-linked bonds) and real yields (from inflation-linked bonds).  So the difference between the two is the breakeven.

Suppose that in the UK market the 0.125% Aug 31 linker is currently yielding -2.825% and its comparator is the 4.75% Dec’ 2030 nominal bond which is yielding 1.024%.  Subtracting the real yield from the nominal yield returns a value of 3.85%, which should be referred to as the breakeven spread.  Think of it as being analogous to the credit spread between the yield on a corporate bond and a Treasury.  Indeed this approach is useful for a trader who wishes to express a view on the evolution of breakeven spreads.

In its original formulation the Fisher equation would state the relationship as:

(1+n) = (1+r)(1+f)(1+p)

So ignoring p for the moment and using our observed values we get:

(1.01024) = (0.97175)(1+f)

Rearranging for f returns a value of 3.96%.  This is referred to as the Fisher breakeven and is used to determine if an investor should buy a linker or buy a nominal.  If an investor believes that realised inflation will on average be greater than 3.96%, they should buy the linker.  If they think it will be less than this value, they should buy the nominal.  I have written about this previously here with a simple worked example.

First of all, in an era where there is an increase in demand for inflation protection, it is logical that price will increase accordingly.  The majority of inflation-linked bonds are traded in real terms (real prices and real yields).  So given the inverse relationship between price and yield, an increase in real linker prices will push down real yields.  All other things being equal this will increase the breakeven spread.  But is this a circular argument? Is the change in demand and supply for linkers caused by an increase in inflationary expectations?  But maybe there are circumstances where investors become less willing to hold relatively illiquid assets (subject of another post) perhaps in a period of market stress.  Real prices would fall, real yields increase, and the breakeven spread would decrease.

The population of inflation-linked bonds have different maturities and therefore trade with different degrees of market risk.  Given that linkers have lower real coupons and lower real yields than their nominal comparators, then mathematically, they will trade with a higher degree of market risk for a change in real yields.  There is a debate over whether real yields are more or less volatile than their nominal equivalents but that is another topic for another time!  However, it is possible that investors will demand a higher yield to hold longer-dated instrument.  This will cause real yields to increase and so the breakeven spread will decrease.  This is independent of inflation expectations.