Decomposing Duration Risk Part 2

The final installment in a three-part deep dive into duration risk​

​Cedars Hill Group — September 16, 2025 —​ The markets are worried about inflation and weaker growth at the same time which is creating schizophrenic pricing in the yield curve. Investors expect that the Fed will be forced to lower rates against their will while tariffs keep upward pressure on inflation, and the AI capex boom will support corporate profits but be a drag on employment...

---

​Cedars Hill Group — September 16, 2025 -- The markets are worried about inflation and weaker growth at the same time which is creating schizophrenic pricing in the yield curve. Investors expect that the Fed will be forced to lower rates against their will while tariffs keep upward pressure on inflation, and the AI capex boom will support corporate profits but be a drag on employment.

The common thread across these views is that inflation will be resilient and this has been priced into the short end of the yield curve. When the market is pricing a certain scenario, it means that it is offering odds to take the other side and that is what has incited this now three-week exploration of duration risk. We don’t have a strong view on the direction of rates or the economy (who can) but the shape of the curve has created opportunities where investors are being paid to take duration risk which allows us to create portfolios that can outperform across economic scenarios.
​
In terms of high-conviction views, we do have two that are related to this discussion. First, the 40-year secular bond bull market is over. We’ve broken out above the 40-year secular down trend channel and have been lingering at yield levels we haven’t seen since before the GFC. Second, after ~20-years of disinflation followed by a decade of consumer deleveraging a new inflationary trend has taken hold economically and most importantly psychologically. It’s important to note that neither of these long wavelength forces will play out linearly over the coming years, instead we will see cyclical ups and downs in interest rates and inflation.

These views seem to be held, at least in the short term, by the wider market, which may explain why the market is pricing in a temporary drop in rates followed by a return to 4% by 2031. This pricing makes sense considering the political pressure on the Fed to lower rates combined with a market view that the economy is fine and inflation is a latent problem. However, the future may, and often does, play out differently than may seem likely today so we are going to decompose the market’s expectations of Fed policy, via the SOFR curve, and back out a few scenarios of how Fed policy may unfold.
​
The SOFR curve represents the market’s average expectations for Fed policy which means within it there are investors who are betting that rates will be higher, lower, and some who think there is no change. The SOFR curve therefore reflects the combination of all these individual views and is merely the price at which all those views are balanced out. Over the past two weeks we have looked at forward rates to estimate what bond total returns might be and today we are going to utilize the SOFR curve to estimate what short and intermediate treasury yields would be under different Fed policy rate paths giving us a different way to arrive at total returns. We can do this because a treasury bond can be replicated by buying a strip of SOFR futures and in fact this is what the bond market quants do to arbitrage any differences between the two markets. The SOFR futures are only liquid out to around 2031 which is why we are restricting our analysis to the 5-year and in part of the yield curve. We have decomposed these average expectations into three scenarios that account for possible paths of Fed policy based on the Fed’s own dot plot. What we’ve done in the chart below is show three broad scenarios for the path of short rates: 1) resurgent inflation which would call for the Fed to raise rates 100bps higher than the high-end of their own longer run expectations, 2) a no rate hike scenario which sees them lower rates to their longer run median level of 3% and hold them there, and 3) a lower for longer scenario where they lower rates 100bps lower than the low-end of their longer run expectations.

Over the past three weeks we’ve been trying to figure out what point on the curve we’d want to own if we wanted to be own duration risk. We excluded the long-end of the curve given our views on the secular interest rate and inflation cycle, so we’ve been focusing on 10-year maturities and shorter. Today’s exercise gives us another way to come at this problem by looking at how the 2-year through the 5-year would react to different Fed policy paths. The table below shows the change in yields that we would see under each scenario and interestingly we find that the 5-year outperforms in the two lower rate scenarios and under the resurgent inflation path it is only bested by the 2-year. If we probability weigh these scenarios 40%, 35%, 25% (which approximately ties out to current market yields) the 5-year is the only one that sees price gains.

changes in yield for each tenor (bond prices go up as rates go down)

Once again, despite the conventional wisdom that the belly is rich we find that under several scenarios it would perform better than the other parts of the curve. This supports our earlier intuition that the market expectations of rates returning to 4% by 2031 leaves value in the belly of the curve despite the current inversion.

While this analysis is very rough and far from conclusive, what it shows is that there are deeper ways to look at relative value along the treasury yield curve. A more robust analysis, and one not as prone to bias as this one is, would be overly complicated for our purposes and wouldn’t yield much additional insight. What this analysis allows us to do is isolate different scenarios against current market pricing to discover any opportunities or latent risk exposures in our portfolios.
​
Our current positioning does very well under an inflationary, positive dollar-bond correlation regime: when the dollar falls and rates rise, so naturally we want to look at disinflationary scenarios where the dollar rallies and/or bonds rally as we are exposed there. We have caught a few glimpses of what a stronger dollar would look like, and it is not pretty: it has tended to lead to lower stocks, bonds, gold, and higher crude. The positive correlation between the dollar and crude is counterintuitive, and the intermediate-term, realized correlations are currently negative but in these short-term, reflexive, risk-off dollar moves we do get a sense of what the next correlation regime may look like. This is just another reason to own crude in your portfolio as it seems to be acting as a long dollar proxy.

It’s logical to think that we should see a negative correlation between the dollar and bonds: a stronger dollar could come from a tighter Fed and indeed that is the correlation dynamic we are currently seeing, however the historical experience does not show a consistently negative correlation between the two. (shown as a negative correlation between the dollar % change and change in 10-year yields in the chart below) Correlations are not stable and depend on many other factors, such as relative real rates which incorporate whether the central bank is ahead/behind of inflation and other central banks. The dollar’s strength during ZIRP was widely attributed to it being the “cleanest dirty shirt” amidst a broad DM drive to devalue their currencies. Trump’s trade policy could be seen as a strong dollar policy since it utilizes coercive trade policies to offset any reduced competitiveness of US goods on the global markets. We can easily go too far down this rabbit hole, but it’s safe to conclude that the dollar-rates correlation is not set in stone which means that we could see a stronger dollar with lower US rates (think US exceptionalism as tech benefits the most from lower rates).

negative correlation means dollar up/down corresponds to bond prices up/down

We are long tech, but we own a lot more PM and foreign equities which have been especially vulnerable to a stronger dollar. So, what if lower for longer rates run the US economy hot and the dollar firms? We’ve already bet in the direction the market is pricing: lower dollar, higher inflation, steeper curve; but what if all these reverse? Since tariffs are a tax, they could be disinflationary which would call for lower rates for longer and that could mean a firmer dollar as the economy runs hot on the back of the AI capex boom and dollar inflows keep interest rates low.
​
All these potential scenarios seem unlikely today, but it is always a good exercise to game plan how the future may unfold and be proactive in risk management instead of being reactive and getting caught off-side. Currently the market is pushing back against Trump and most of his policies, however if he gets his way we could see some of these seemingly unlikely scenarios play out.

Source: https://cedarshillgroup.substack.com/p/chg-issue-201-decomposing-duration