The Fed is committed to maintaining very accommodative monetary conditions and unchanged interest rates until the economy has returned to full employment and inflation has stabilised above its 2.0% target. But the Fed has so far been vague on what determines the pace of its asset purchases. It is clear that these will have to decline long before it raises its key rates. But when and on what basis? Tapering has its risks but also its benefits. The challenge for the Fed will be to steer a gradual steepening as history shows that steepening episodes can be quite abrupt and generate volatility on markets.

The recent upward movement in US longterm interest rates is mainly linked to the rise in inflation expectations and not to the rise in real rates. In a scenario where US GDP growth is expected to accelerate (5.2 to 5.7% in our central scenario in 2021, followed by +2.6 to 3.2% in 2022), real rates could adjust, as well. In this context, the Federal Reserve’s strategy will become a key parameter to watch, with implications on portfolio construction.
Since its strategic review, the Fed’s reaction function has changed. By targeting 2% inflation on average over a cycle, the Fed aims to anchor short-term interest rates at their current level. The FOMC has also committed not to tighten monetary conditions until the economy has returned to full employment (defined on the basis of multiple criteria, such as minority, woman or age group employment), and inflation prints are above target. Since June 2020, the Fed has bought $120bn of eligible assets each month ($80bn in Treasuries and $40bn in MBS). Its balance sheet is therefore growing at a rate of nearly $1500bn per year (7.5% of GDP). This balance sheet expansion is necessary to maintain very accommodative monetary and credit conditions via low long-term bond yields in the recovery phase.

As the US economy emerges from the crisis, and if inflationary pressures materialise sooner than expected on the back of fiscal stimulus and pent-up demand, the Fed may want to revisit its policy. After all, the Fed never pre-committed to keeping the pace of its asset purchases unchanged. And it is clear that the asset purchase program must be reduced well before the Fed increases its policy rates. As a result, the expected level of asset purchases will ultimately define the path for long-term Treasury yields this year. Hence, the fear of a “taper tantrum 2.0” in reference to the one that occurred in April 2013 (the 10-year yield had then risen by 120bp in four months on the back of Ben Bernanke’s announcement).
It is clear that it is not in the Fed’s interest to destabilise the US bond market and the US financial system with it. The main argument is that there is still too much leverage in the economy for that (federal government, corporations, and households). A soaring debt burden would not be “sustainable” and could jeopardise the recovery. However, it is equally clear that the Fed might consider it is time to reduce the degree of monetary accommodation if nominal GDP growth accelerates and the pandemic ends. The Fed never committed itself to keeping longterm bond yields stable. Several arguments may play in favour of “early” tapering: (i) avoid a decline in real rates in a recovery phase; (ii) restore a role for markets in anchoring inflation expectations; (iii) smooth out the rise in bond yields; and (iv) restore some policy space, should the situation deteriorate again. Some FOMC members have begun to discuss this option for 2021. The Fed needs to clarify its strategy sooner than later, since this ‘taper tantrum’ thematic is likely to gain ground as the GDP returns to its pre-crisis level.

One of the many consequences of a shift of investor’s expectations regarding a Fed’s tapering, would be further steepness in the Treasury yield curve. Therefore, the challenge for the Fed will be to steer a gradual steepening of the Treasury curve, as history shows that steepening episodes can be quite abrupt and generate market volatility.
There have been four great episodes of steepening of the US Treasury yield curve over the past 40 years. Market participants are using several measures – from threemonth bills to 30-year bonds. We think that the spread between UST 10y and 2y is most relevant for long-term comparisons, and we consider a steepening phase is starting when the curve is fully flat or inverted. Although irrelevant for the analysis, the 1980s witnessed a gradual transition from a high-inflation to a low-inflation regime, leading to high volatility in yield-curve steepness (see chart).
During the last three episodes (89-92, 00- 03, 06-10), the YC (10y-2y) steepened with astonishing regularity. Indeed, the average length of steepening was 3.3 years with only a few months of difference for an average rise of 310bp in a tight range, too. Excluding periods where the yield curve was inverted does not change the outcome (275bp over around three years).
The ongoing UST steepening phase started in September 2019. It is already a year and a half old with a 100bp move so far. Should history repeat itself, we can expect the UST YC (10y-2y) to steepen by another 200bp over the next two to 2.5 years. Assuming that the Fed keeps rates at zero over the period (Amundi’s scenario), this would mean bringing the UST 10-year bond yield close to 3%. Obviously, it is not because of the yield curve that long-end yields have to move, the causality is the other way round. However, it is interesting to use the yield curve as a reference point to assess the minimum level of Treasury yield over this new cycle.