Pennies & The Steamroller

18 September 2025

The paradox of the “Sharpe-Ratio” or value at risk approach to managing exposure is that long periods of stability encourage market participants to take on excessive leverage when liquidity is plentiful to boost returns. As a result, market participants become “short convexity.” That is the key reason why long phases of low and stable interest rates lead to episodes of instability. It is the accumulation of leverage. As we noted yesterday, the real world is not linear. Systemic risk tends to increase at an accelerating rate when an event happens, for whatever reason. The more important portfolio risk is correlation and ensuring adequate compensation for risk.

A popular strategy in a low rate and spread environment is to borrow in the low interest rate and invest in the higher interest rate to profit from the differential, or “carry.” In the foreign exchange markets, a carry trade involves borrowing a currency with a low interest rate (the “funding” currency) and using that capital to buy a different currency that has a higher interest rate (the investment” currency). These trades often use leverage to magnify the potential profit/return. Of course, this also amplifies potential losses when rates/price move in the wrong direction. As we often note, if you want to know where the future risk is, follow the leverage.

The key risk in a portfolio is not standard deviation of returns, but diversification or correlation. In a major episode, correlation tends towards one. Historically, the key “diversifier” a typical multi-asset portfolio was sovereign bonds. However, that only worked when the bond-equity correlation was negative. In turn, the bond-equity correlation is only negative if inflation is pro-cyclical. Inflation is only pro-cyclical when demand shocks dominate supply shocks. The challenge since 2021 is that supply shocks have dominated demand shocks. We would also add that interest rates/yields were too low leading into that episode (amplified by portfolio leverage and carry trades).

Looking forward, the “good outcome” for bond markets would be a growth shock or recession. In contrast, the “bad outcome” given fiscal dominance and large deficits is that inflation re-accelerates. The current FOMC might have repeated the Arthur Burns error of the 1970s. Recency bias is a cognitive or behavioural bias that favours recent events over historical ones. Recency bias can skew investors into not evaluating economic cycles objectively, especially when there has been a potential regime shift. Potential structural economic regime shifts are not easy to evaluate even with hindsight, especially when most macro models and analysis are cyclical (mean reverting) in nature. Key macro variables growth and inflation are inter-related and reflexive. However, their causality is not stable. Typically, the relationship is two-way, but this can be influenced by inflation regime shifts. When inflation shifts higher, it starts to become a more important driver of growth.

As we noted yesterday, whatever your view is on the stage of the business cycle, what we do know is that risk compensation in equity and credit is extremely poor. Yesterday we noted that US investment grade spreads are now tighter than the pre-2007 levels. Similarly, USD emerging market spreads have almost compressed to comparable levels (chart 1). To be fair to emerging markets, sovereign leverage and inflation rates are much better than in the developed world. Nevertheless, correlation is likely to remain high with US markets in a leverage and liquidity driven unwind.

Chart 1

Source: Bloomberg

20250918_Pennies_And_The_Steamroller.pdf

Download PDF • 406KB