Inflation compensation and risk sentiment

Matt Roberts-Sklar.

Inflation breakevens and inflation swap rates have fallen a lot in recent years. Big falls have often occurred amid deteriorating risk sentiment. This isn’t a new phenomenon. Looking across markets and time periods, I show that measures of financial market inflation compensation tend to fall when risk sentiment worsens. What’s more, this effect is asymmetric – inflation compensation doesn’t rise by as much when risk sentiment improves.

Financial market inflation compensation has fallen a lot since 2014. Chart 1 shows how 10-year inflation compensation – implied by inflation breakevens and inflation swaps – has changed for the UK, US and euro area. These moves have been much discussed, including their correlation with oil price moves.

Chart 1: Cumulative changes in 10-year inflation compensation (January 2014 to July 2016)

Source: Gürkaynak, Sack and Wright (2008), Bloomberg, and Bank of England calculations.

Some of this decline may well reflect lower inflation expectations. But inflation breakevens and inflation swaps are not the same as inflation expectations. As Fed Chair Yellen remarked at the June FOMC press conference: “I always try to be careful to call it inflation compensation rather than inflation expectations, because they’re not inflation expectations. Inflation expectations influence those market measures, but there’s also an inflation risk premium. And there are actually good reasons to think that the inflation risk premium could have declined significantly and may be depressing those measures”.

In this post, I’m not going to try and split out moves in inflation compensation into expectations and various risk premia. Instead, I’m going to highlight an empirical fact frequently pointed out by market participants:  when risk sentiment worsens, measures of inflation compensation tend to fall.

This can be seen from Charts 2a and 2b, which show (for the US and UK respectively) the sensitivity of changes in 10-year nominal yields, real yields and inflation breakevens to a change in the VIX (a proxy for risk sentiment) since 1999. The effect is clearest for the US: a one percentage point change in the VIX leads nominal yields to fall by twice as much as real yields, pulling down on breakevens.

Chart 2a: Sensitivity of changes in US 10-year yields to changes in the VIX in either direction

Chart 2b: Sensitivity of changes in UK 10-year yields to changes in the VIX in either direction

+

Regression of daily change in yield on a constant and daily change in VIX, January 1999 to July 2016. Coefficient on change in VIX with two Newey-West standard errors shown. TIPS (Treasury Inflation-Protected Securities) and ILGs (Index-linked Gilts) are real bonds in the US and UK respectively.

Source: Gürkaynak, Sack and Wright (2008), Bloomberg, Bank of England and author calculations

You get a similar result from running these same regressions just for the most recent fall in inflation compensation (Chart 3). And the result holds for inflation swaps as well as breakevens, though to a slightly lesser extent.

Chart 3: Sensitivity of change in 10-year inflation compensation to change in the VIX in either direction

Regression of daily change in yield on a constant and daily change in VIX, January 1999/January 2014 to July 2016. Coefficient on change in VIX with two Newey-West standard errors shown.

Source: Gürkaynak, Sack and Wright (2008), Bloomberg, Bank of England and author calculations

So what’s behind this effect? In part it could be purely mechanical, arising from differences between nominal and real bonds, rather than reflecting genuine changes in inflation expectations or perceptions of inflation risks per se. Nominal bonds – mainly in US but also in other advanced economies such as the UK – are seen as a safe asset, so investors are willing to accept a lower yield on these bonds, for reasons unrelated to their expected cashflows. When risk sentiment worsens, investors often move into these safe assets – so-called ‘flight to safety’ flows. Real bonds are less liquid – the market is smaller and often dominated by domestic institutional investors – so don’t tend to benefit as much from these flight to safety effects. This means that when risk sentiment deteriorates, nominal yields tend to fall by more than real yields and so inflation breakevens, which are the difference between nominal and real yields, mechanically fall.

If that were the whole story, then you might not expect to see a similar effect in inflation swaps. But, notwithstanding the fact that some price discovery in inflation swaps may occur in breakevens, we do see inflation swap rates fall when the VIX rises. An alternative – and complementary – explanation is that shifts in near-term risk sentiment affect investors’ perceptions of distribution of future inflation outcomes, or the price investors attach to inflation risks (Gürkaynak, Levin & Swanson, 2010).

Risk sentiment tends to ebb and flow in financial markets. So does that mean there’s no lasting impact on inflation compensation? No, there appears to be an asymmetric impact: breakevens tend to fall by more when the VIX rises than they rise when the VIX falls (Chart 4).

Chart 4: Sensitivity of changes in 10-year inflation breakevens to increases and decreases in the VIX

Regression of daily change in yield on a constant and the absolute value of the daily change in VIX (splitting out increases and decreases in the VIX as separate variables), January 1999 to July 2016. Coefficient on change in VIX with two Newey-West standard errors shown.

Source: Gürkaynak, Sack and Wright (2008), Bloomberg, Bank of England and author calculations

For the US, a 1p.p. increase in the VIX pushes down on 10-year breakevens by 0.8bps, but a 1p.p. decrease in the VIX only pushes up 10-year breakevens by 0.4bps. What’s more, UK 10-year breakevens don’t respond to falls in the VIX, only to increases. In the short run at least, this implies some stickiness from the impact of risk sentiment on inflation compensation. So the falls in measures of inflation compensation seen in recent years may be slow to reverse as and when risk sentiment picks up.

Matt Roberts-Sklar works in the Bank’s Macro Financial Analysis Division.

If you want to get in touch, please email us at bankunderground@bankofengland.co.uk. You are also welcome to leave a comment below. Comments are moderated and will not appear until they have been approved.

Bank Underground is a blog for Bank of England staff to share views that challenge – or support – prevailing policy orthodoxies. The views expressed here are those of the authors, and are not necessarily those of the Bank of England, or its policy committees.