The Term Premium Turnaround

As is often in economics, simple ideas inspire complex explanations. Justifying the need for a golf trip to one’s partner is one example. The ‘Term Premium’ another. A modest idea at heart, it relies on intricate and evolving models to produce an estimate. And although recent geopolitical events have overshadowed its recent rise, it is noteworthy that the Term Premium recently reached a decade high after years of negativity. Why does this matter? A return to normalcy has implications for traditional asset allocation and might prompt the need for alternative sources of diversification.

The Term Premium: What is it?

The definition of the Term Premium is straightforward. It is the additional return long-term bondholders receive beyond rolling at short-term rates. The concept of a Term Premium clashes with proponents of the Expectations hypothesis¹ – If markets are rational and price short term rate paths correctly, why should long-term bond investors earn more? But as is apparent to casual market observers, the path of short-term rates is never certain², and neither are longer-dated bond determinants such as inflation. It feels logical that long-term investments come with added risk, similar to an illiquidity premium, and added return.

Although the idea is simple, accurately estimating this premium is challenging. Not only are short-term rate expectations not observable and need a model themselves, but interest rate drivers are complex, global, and often undefined. This led econometrists to pursue various approaches to derive Term Premium estimates. Current models span purely statistical³, to macroeconomic-heavy⁴, to international-cross-border dynamic⁵. Despite differing methodologies, these models agree that following the Global Financial Crisis (GFC), the Term Premium saw nearly a decade of negativity. During this period of negativity, investors essentially paid for the privilege of bearing interest rate risk. The culprit behind this phenomenon? Global central bank intervention. Depressing long-term interest rates came at the expense of adequately compensating long-term bondholders’ risk. Duration-heavy portfolios in 2022 can corroborate this view.

So why does this recent increase matter?

Well first, duration isn’t dead. Rather than paying to be out on the curve, investors are now being paid again. This aligns with historical norms, where the Term Premium was positive close to 90% of the time, according to the Fed’s preferred ACM model⁶. This shift has significant implications for capital market assumptions and subsequent asset allocations. While still not the belle of the ball, at least long-term Treasuries edge out cash again, all else equal.

Second, an increase in the Term Premium could impact the relationship between stocks and bonds. Before the Global Financial Crisis (GFC), a rise in the Term Premium has been associated with an ensuing rise in stock / bond correlations. This relationship weakened post-GFC following global central bank interventions, a time when the Term Premium was driven less by market forces. But the pullback in central bank activity, and the size of the recent positive move, places this historical relationship back into focus. The most recent change in the Term Premium places it in its top historical quartile, a period that has corresponded with the largest increases in subsequent stock-bond correlations.

This graph compares the average two-year change in Term Premium with the subsequent average two-year change in stock-bond correlations during the pre-GFC period. The largest quartile of Term Premium changes averaged +1.3 standard deviations. During these periods, the subsequent average change in two-year stock/bond correlation was +0.4 standard deviations.

Now that quantitative easing efforts are ending (looking at even you Japan!), should we expect a more normalized relationship to return? How would increasing stock-bond correlations impact the defensiveness of bonds in a diversified portfolio? While recent memories of correlations have been negative, historically they have been positive a majority of the time since Term Premium estimates began. With this recent rise, a closer look at diversification assumptions might be warranted⁷.

¹ Fisher (1930), Macaulay (1938), Lutz (1940)

² https://www.cfm.com/causing-a-stir-why-short-term-rate-forecasts-often-miss-the-mark-and-why-that-matters/

³ Adrian, Crump, and Moench (2013)

⁴ Gallmeyer, Hollifield, Palomino, and Zin (2007)

⁵ Greenwood, Hanson, Stein, and Sunderam (2023)

⁶ https://www.newyorkfed.org/research/data_indicators/term-premia-tabs#/overview

⁷ https://www.cfm.com/a-systematic-path-to-true-diversification/

Disclaimer:

All opinions and estimates included in this document constitute judgments of CFM as at the date of this document and are subject to change without notice. Future evidence and actual results could differ materially from those set forth, contemplated by or underlying these statements. CFM does not give any representation or warranty as to the reliability or accuracy of the information contained in this document. CFM accepts no liability for any inaccurate, incomplete or omitted information of any kind or any losses caused by using this information. All figures are sourced from CFM.