The steady-state slope of the yield curve is tied to the level of the steady-state term premium. Indeed, when the economy is settled at full employment and price stability and the funds rate is at its neutral level, the slope of the yield curve is entirely pinned down by the term premium. We normally think of the term premium as being positive, the compensation for investors taking duration risk. The slope of the yield curve is therefore generally upward sloping. But we have had what appears to be a regime change in the last couple of decades. The fear of inflation has diminished and fears of stagnation and deflation have increased. As a result, we suspect the steady-state term premium has declined, perhaps all the way to zero. In this case, we would be headed to a flat yield curve as a new normal. We wrote this commentary with guidance from Jonathan Wright, a senior adviser at LH Meyer.
Jonathan Wright wrote a commentary for us on how the term premium had evolved in the post-Great Recession period. Jonathan argued that, in effect, there has been a regime shift with respect to the source of the term premium—less risk of inflation, more risk of secular stagnation and deflation—and that shift has lowered the term premium, maybe all the way to zero.
The Evolution of the Term Premium, 1960 to Today
The term premium is defined as the difference between an n-period zero-coupon Treasury bond yield and the expected short-term rate over the maturity of the longer-term bond. In Figure 1, we plot a measure of the term premium based on the Kim-Wright model.1 The term premium in Figure 1 is the difference between a zero-coupon ten-year Treasury note and the expected Treasury bill rate over the ten-year duration of the bond. The series extends from 1990 (the beginning of the sample used in the Kim-Wright paper) through today, and it is updated regularly on the Federal Reserve Board website. To develop a measure of the term premium that extends back to the 1960s, we extrapolated the Kim-Wright measure back to 1964.
Historically, the term premium has been positive, reflecting the compensation investors demanded taking duration risk, which, in turn, was principally related to inflation risk. Note, however, that the term premium has a real risk as well as an inflation risk component. First, the Kim-Wright proxy rose, we suspect principally as a result of rising inflation risk from the late 1960s, peaking near 5% a couple of times in the early-to-mid 1980s. It declined thereafter as Volcker implemented a monetary policy that lowered inflation, and Greenspan finished the job bringing inflation down to around 2%.
1 See Finance and Economics Discussion Series 2005-33, “An Arbitrage-Free Three-Factor Term Structure Model and the Recent Behavior of Long-Term Yields and Distant-Horizon Forward Rates,” Don H. Kim and Jonathan H. Wright.
The Kim-Wright Measure of the Term Premium
Sources: MPA, Federal Reserve.
Jonathan sees the decline in the steady-state term premium over recent years to closer to zero as largely reflecting the change in the source of the term premium, what investors sought compensation for in taking duration risk. As we noted, earlier, inflation risk was the principal source of duration risk. After the recent recession, inflation risk was no longer what investors most wanted to protect themselves from. Investors now wanted to protect themselves from long periods of a near-zero funds rate, secular stagnation, and, hence, deflation risk. So, investors might even pay a premium to hold longer-term securities.
Asset Purchases, the Term Premium, and the Yield Curve
The term premium has also been affected by the Fed’s asset purchase programs, beginning in late 2008 and ending in October 2014, and the anticipated and actual normalization of the balance sheet that is ongoing. In Figure 2, we show a measure based on Board staff studies of the peak cumulative effect of asset purchases on the term premium—100 to 120 basis points—and how that cumulative effect has declined and is expected to continue to decline as balance sheet normalization continues.2It is clear in Figure 1, however, that the path of the term premium since the end of 2014 has been uneven, apparently affected by forces other than balance sheet normalization, including cyclical factors and, importantly, by changes in the term premium attributed to foreign influences.
2 Engen, Eric M., Thomas Laubach, and David Reifschneider (2015). “The Macroeconomic Effects of the Federal Reserve’s Unconventional Monetary Policies,” Finance and Economics Discussion Series 2015-005. Washington: Board of Governors of the Federal Reserve System, http://dx.doi.org/10.17016/FEDS.2015.005. Bonis, Brian, Jane Ihrig, and Min Wei (2017). “The Effect of the Federal Reserve’s Securities Holdings on Longer-term Interest Rates,” FEDS Notes. Washington: Board of Governors of the Federal Reserve System, April 20, 2017, ttps://doi.org/10.17016/2380-7172.1977.
The Term Premium, Asset Purchases, and Balance Sheet Normalization
Sources: MPA, Federal Reserve.
We will hopefully get a clearer reading of the steady-state term premium when (and if) the fund’s rate reaches and is expected to stabilize at its neutral rate and the balance sheet is more fully normalized, provided that foreign influences diminish.3
The Term Premium and Recession Risk
Historically, a flattening of the slope of the yield curve has been seen as weakening the growth outlook, and an inverted yield curve has been seen as a harbinger of an impending recession. With a sizeable positive term premium, an inverted yield curve is only possible if monetary policy is significantly tighter than neutral, and that naturally increases recessionary risk. But the slope of the yield curve depends on both the stance of monetary policy relative to neutral and the level of the term premium. As Jonathan has emphasized in his
research, a decline in the slope of the yield curve as a result of a drop in the term premium is, in itself, stimulative, not a sign of increasing recession risk. It follows that the flatter the steady-state yield curve is, the less an inverted yield curve signals recession risk.
Toward a Flat Steady State Yield Curve
The decline in the actual term premium over much of the period shown in Figure 1 was likely driven by a decline in the steady-state term premium as investors required less compensation for inflation risk. That compensation may now be zero, or even negative, as we noted above. When the fund’s rate eventually stabilizes at its long-run equilibrium level—if that ever happens—and absent a shift in foreign influence on the U.S. term premium, the term premium may be close to zero. A zero steady-state term premium, in turn, means a flat steady-state yield curve. Figure 3 is illustrative—and only illustrative—of a situation in which the yield curve becomes flat after about 2020: We assume that the funds rate reaches its long-run equilibrium level and is expected to remain there and that the steady-state term premium is zero. But in this scenario, a flat yield curve is just about two years off. And if the term premium is still below its steady-state and somewhat negative, the flat yield curve could be coming sooner. Given the historical relationship between the slope of
3 Note, however, that the normalized balance sheet is likely to be much larger (in ten-year equivalents relative to GDP) than in the pre-crisis period, leaving the steady-state term premium a bit lower than otherwise.
the yield curve and recessions, it is hard not to view this with some trepidation. However, a flat or inverted yield curve caused by a near-neutral funds rate coupled with a zero or negative term premium should not be taken as a clear indicator of an impending downturn.
Toward a Flat Yield Curve