The "duration" effect

Much of this year's broader market performance has been a reflection of what is happening in fixed income. In a year where geopolitical tension in Europe has reached levels not seen since the end of the Cold War, and where energy prices have risen by double digits, the performance of major risk premia and of the dollar cannot entirely be explained by these extreme phenomena. While the recovery of 2020 was largely fuelled by extremely accommodative monetary conditions, 2022 is seeing the exact opposite effect: demand-driven inflation can only be countered by a recession, so the tightening monetary conditions being orchestrated by the central banks are leading to a global rise in interest rates and, in their wake, equities are falling. This decline in equities has raised the question across the investment community of whether this decline in equity prices reflects the risk of recession? Absolutely not. In our view, the decline merely stems from the good old-fashioned "duration" effect, which has also been responsible for the negative performance rarely seen in bond markets. To understand this, a trip back in time is necessary to replay the 1970s and early 1980s and gauge the extent of the damage and as well all the mechanics of an inflation shock.

 

Total portfolio duration

The majority of those working in portfolio management today were only growing up in the 1970s and early 1980s, and therefore can’t benefit from any personal experience gained during that period. And yet, it is this period that we should look to for inspiration. The current situation bears many similarities to the late 1970s: demand-driven inflation and a price / wage loop accelerated by rising energy prices; central banks initially lagging behind inflation before later becoming aggressive to the point of generating a recession; while fiscal measures to support weakened demand clash with central bank efforts to slow it down... if the situation were not so serious, it would be comical because of these many similarities. At the time, the community of "quants" or systematic managers addressed the question of whether rising interest rates posed a risk to equities. In a well-known 1986 article entitled "Total Portfolio Duration: A New Perspective on Asset Allocation"¹, Martin Leibowitz analysed the duration of equity markets.

Similar to bonds, equities pay cash flows in the future, so their valuation should move inversely to changes in interest rates, which serve to discount future cash flows. Leibowitz proposed a quantitative way to measure this linkage based on three ingredients:

• The duration of bond indices;
• The correlation between equity and bond returns;
• The ratio of equity to bond volatilities.

Using data spanning the 1980-1985 period, he calculated that a global equity index has a duration of 2.2 when a bond index has a duration of 4.3; a 60-40 portfolio thus has an interest rate sensitivity of about 3.5 years. Better still, a greater equity exposure would make it possible to somewhat diversify the impact of a rate shock by lowering its duration. This is an attractive story, but it suffers from obvious weaknesses: the equity / bond correlation used for the calculation is 30%, whereas this same correlation could have reached 70% or even 80% over the period; with an 80% correlation, the duration of equities increases from 2.2 to 5.14, eliminating this valuable diversification effect. As the author himself notes, this diversification effect depends precisely on the robustness of the equity / bond correlation and that is the problem. Secondly, since dividend yields are higher on average than coupons and companies have a life span that generally exceeds the maturity of bonds, what magic is there in equities having a shorter duration than bonds? These two points rock the “total portfolio duration” boat. And yet, as shown in chart 1, if we apply this model to the period from 1960 to 2022, we can already see that a significant part of the negative performance of equities during 2022 comes from a duration effect.  

 

Chart 1: Decomposition of the year-on-year performance of the S&P 500 between 1960 and 2022 using the Leibowitz model

Source: Bloomberg, LOIM

 

Gordon and Shapiro to the rescue

One way of dealing with the problems associated with the Leibowitz method is to use a structural model of the relationship between rates and stock prices. The simplest version is the Gordon and Shapiro model which states that the price of an equity reflects the level of its next dividend, divided by the difference between the interest rate and the dividend growth rate. One can explicitly calculate the sensitivity of the price of an equity or a group of equities to interest rates mathematically within this framework. And here the conclusion is quite different: the duration of the S&P 500 actually exceeds the duration of bonds. Today, the duration of a standard bond index is around 7 years, while the estimated duration of equities is over 14 years. Chart 2 shows the same decomposition of S&P 500 performance but using the Gordon and Shapiro model.  This time, equities do not reflect the full duration effect, which would have taken them around -10% lower than they are now.

 

Chart 2: Decomposition of the year-on-year performance of the S&P 500 between 1960 and 2022 using the Gordon and Shapiro model

Source: Bloomberg, LOIM

 

What has changed between 1979 and 2022?

These models highlight two key conclusions:

• The two approaches ultimately converge: in both cases an equity index, such as the S&P 500, has indeed been the victim of a duration effect rather than a recession scenario being factored in. Therefore, the question is not whether it is a duration effect, but which of the two models is right? If it is the Gordon and Shapiro model, then we have not yet seen the end of duration’s risk repricing, let alone the impact on corporate earnings. If it is Leibowitz’s, then the duration effect is probably behind us. The truth likely lies somewhere in between.
• The second conclusion relates to interest rate sensitivity today compared to 1979. Both models estimate the duration of equities while the duration of bond indices is known. Chart 3 presents these estimated and calculated values. For both equities and bonds, duration is much higher in 2022 as a consequence of the low interest rate environment we have experienced recently: longer and longer-dated bonds were issued and companies have been able to extend the horizon of their growth models. To put it simply, a 1% rise in long rates in 2022 is probably equivalent to a 2% rise in 1974, given the overall increase in our interest rate sensitivity. What works today for the investment world probably also applies for the economy as a whole: rates will not need to reach 20% to counter inflation but investors should beware of the damage rate hikes may have on our economies and portfolios.

 

Chart 3: Estimated durations between 1979 and today

Source: LOIM, Bloomberg

Simply put, this year's performance has mainly been a reflection of central bank actions. The effects of these actions have been multiplied by our increased dependence on low interest rates in recent years: duration is everywhere.

---

Macro/Nowcasting Corner

The most recent evolution of our proprietary nowcasting indicators for world growth, world inflation surprises and world monetary policy surprises are designed to keep track of the latest macro drivers making markets tick. Along with it, we wrap up the macro news of the week.

Our nowcasting indicators currently point to:

• Worldwide growth is clearly declining. The US and Eurozone are showing signs of decelerating growth momentum while the most recent data shows that this deterioration still has room to go. The US is increasingly showing indications that it is entering into a recession.
• Inflation surprises will remain positive for the Eurozone but are declining elsewhere and are now non-existent in the US.
• Monetary policy is set to remain on the hawkish side: central bankers are likely to be more hawkish than expected.

 

World Growth Nowcaster: Long-Term (left) and Recent Evolution (right)

 

World Inflation Nowcaster: Long-Term (left) and Recent Evolution (right)

 

World Monetary Policy Nowcaster: Long-Term (left) and Recent Evolution (right)

Reading note: LOIM’s nowcasting indicator gather economic indicators in a point-in-time manner in order to measure the likelihood of a given macro risk – growth, inflation surprises and monetary policy surprises. The nowcaster varies between 0% (low growth, low inflation surprises and dovish monetary policy) and 100% (the high growth, high inflation surprises and hawkish monetary policy).