4 min read
Financial markets tensions and breakeven inflation
The behaviour of long bond yields and breakeven inflation calculated from TIPS yields suggests that market participants are worried about the outlook for inflation. Stefan Gerlach examines how much weight should be attached to breakeven inflation as a measure of expected inflation and whether it provides clean signals.
Market participants are worried about the outlook for inflation. The main driver of these concerns appears to be the massive fiscal stimulus programme launched by the Biden administration, which is expected to lead to a sharp boost to activity. This has led to upward revisions to GDP by the OECD and FOMC member’s median projection for PCE inflation in 2021 rose from 1.8% to 2.4%. As a result, market-based measures of inflation expectations have risen sharply.
But how much weight should be attached to breakeven inflation as a measure of expected inflation? Does it provide clean signals of changes in inflation expectations or is the information it contains hidden by other factors?
To address these questions, Figure 2 shows monthly data on headline CPI inflation (which inflation-protected bonds are indexed to), and breakeven inflation calculated as the difference between 10-year nominal bonds and TIPS. Both series show large declines around the time of the Global Financial Crisis in 2008-9 and during Covid in 2020, which is compatible with the idea that deep downturns lead to a fall in inflation. Interestingly, the average annual inflation rate during the 18-year period shown in the graph is 2.05%, almost exactly identical to average breakeven inflation, which is 2.03%. This is striking since inflation measures price changes over the past 12 months, while breakeven inflation pertains to the next 10 years (which is why it is much more stable).
The idea that the spread between yields on nominal and inflation-indexed bonds measures expected inflation assumes that differences between them are due solely to inflation. In practice, that is not the case: the market for indexed bonds is smaller and much less liquid.
This suggests that nominal bonds are more attractive in episodes of financial stress when market participants put a premium on liquidity. Thus, when markets come under stress, the yields on indexed bonds rise to compensate for their relative illiquidity. Breakeven inflation therefore spuriously appears to fall when financial market tensions rise. Might this effect be so large as to impair the information content of breakeven inflation?
To explore this question, the figure below shows the VIX (inverted scale) together with 10-year breakeven inflation. Importantly, the surge in the VIX in the spring of 2020 as the covid pandemic struck seems associated with a collapse of breakeven inflation. And as the VIX declined, breakeven inflation rose as financial market tensions subsided. The figure thus suggests that part of the recent increase in breakeven inflation reflects a normalisation in financial markets since last spring rather than an increase in expected inflation.
How important might changes in market tensions have been for breakeven inflation during the covid pandemic? Figure 4 provides a close-up of breakeven, and VIX-adjusted breakeven, inflation in 2020-21. In interpreting the latter series, it is important to note that it is an estimate and therefore subject to some uncertainty.
While breakeven inflation rose by 1.65% between March 2020 and March 2021, VIX-adjusted breakeven inflation rose by 0.93%. Adjusting for the state of financial markets reduces the increase in expected inflation almost by half. That said, the estimate of the breakeven inflation rate is about 0.2% higher at the end of the time period considered.
The above analysis leads to several conclusions. First, breakeven inflation falls in periods of financial market stress and rises when such episodes end. While expected inflation has surely increased since the spring of 2020, it has done so by far less than breakeven inflation suggests. Second, while adjusting breakeven inflation for changes in the VIX has a large effect during peiods of market stress, it is clear that expected inflation has been generally lower after 2014 as the Fed has struggled to keep inflation close to its 2% objective.
Download the full edition of our Infocus publication here.