• Log in
• Register free
• Subscribe
• Help
• Tour & Trial IC Advantage

• Share Tips
• Trading
• Companies
• Markets Data
• Columnists
• Funds Data
• Covered Warrants
• Research Tools
• Investment guides
• Q and A
• Blog
• Video

Subscribe today

Celebrate our 150th anniversary with fantastic savings on a joint magazine and IC Advantage subscription.

Subscribe today and save £150

Risks to inflation expectations

Created:

26 February 2010

Updated:

9 March 2010

Written by:

Chris Dillow

Many market watchers are puzzled by the breakeven inflation rate, the gap between conventional and index-linked gilt yields. For some, a breakeven rate of 3.4 percentage points for 14-year yields is too low, given that the Bank of England has printed money and is insouciant about inflation being above-target. For others, it's too high given that the recession should eventually force inflation below its target.

To understand what's going on, we should forget about how you or I think about inflation, and ask instead: how does the gilt market form its inflation expectations?

And the answer is: quite simply. In the 17 years to August 2008 (just before the collapse of Lehman Brothers), annual changes in the breakeven inflation rate were closely correlated with a handful of economic variables. In early 2009, breakeven inflation fell by more than these variables predicted. This was because gilts priced in a risk of deflation, which they have since priced out. In the last 12 months, breakeven inflation has risen by half a percentage point - bang in line with what we'd expect, given the 1991-2008 relationship between breakeven inflation and observable macroeconomic conditions. This implies that it's quite possible that the recession hasn't permanently changed the way the gilt market's inflation expectations are formed.

This gives us three reasons why breakeven inflation is low now:

1. Bank rate is low. Traditionally, breakeven inflation has risen and fallen as Bank rate has risen and fallen. The gilt market's attitudes to future inflation are similar to the Bank of England's, so circumstances in which the Bank worries about inflation enough to raise Bank rate are circumstances in which the market's inflation expectations rise.

2. US Treasury yields are low. These are a measure of global inflation expectations. When these are low, so too are the gilt market's expectations for UK inflation.

3. The economy's weak. At the worst point of the recession - the 12 months to February 2009 - manufacturing output fell by 14.2 per cent. 1991-2008 relationships say this should have cut breakeven inflation by 0.8 percentage points. With output recovering only slowly, this has not been reversed.

Two things, though, have partly offset these factors. One is the weakness of the pound, which raises import prices and hence inflation fears. The other, weaker influence is the price of gold; a high gold price is a sign of high worldwide inflation fears.

All this also gives us reasons to suspect that breakeven inflation might rise over the next 12 months. One reason for this is simply that the economy should recover. It's also possible that, at least later this year, Bank rate will rise. In the past, each percentage point rise in this has added 0.17 percentage points to breakeven inflation.

A third reason is quite simply that breakeven inflation is low right now. It's 1.3 percentage points below RPIX inflation. In the past, a breakeven inflation rate below RPIX inflation has led to breakeven inflation rising in the following 12 months.

A further risk to breakeven inflation comes from US yields. If these rise - say, in response to rises in the fed funds rate or to the economic recovery - there'll probably be a knock-on effect onto breakeven inflation.

What could offset all this? One possibility would be if sterling were to fall significantly. A less likely possibility - because it has a relatively weak effect upon UK inflation expectations - would be if the gold price falls sharply.

Personally, though, I suspect the risks point more to breakeven inflation rising than falling - particularly later this year, when the recovery should become better established. This implies that conventional gilts could well underperform index-linked ones.

The model

This table shows the model used in my chart. The dependent variable is the annual change in the breakeven inflation rate, which is defined as the generic 10-year gilt yield minus the yield on all index-linked gilts with a maturity of over five years.

The co-efficients imply that a 10 per cent fall in sterling would add 0.19 percentage points (on average) to breakeven inflation, that a 10 per cent rise in gold would add 0.09 percentage points, and so on. I don't think we should be hung up about the precise size of the co-efficients (although they are all statistically well defined). What matters more is their rough magnitude.

You might be puzzled by the intercept. It implies that breakeven inflation falls by 0.63 percentage points a year, other things equal. The answer to this puzzle lies with the notes and coin (narrow money) term. On average, these grow by around 6 per cent a year, adding 0.66 percentage points to breakeven inflation. Taking both terms together, they net out.