Blog Posts — February 24, 2021

Worried about inflation? Here’s how it impacts multi-asset portfolio risk…

by Christoph Schon, CFA, CIPM

Investors are getting jittery over inflation, thanks to continued fiscal stimulus, combined with the effects of prolonged monetary easing. This, in turn, has pushed long-term government rates to 12-month highs, while share prices continue to climb. Accordingly, the two major asset classes have resumed their customary countermovement, after being mostly decoupled for most of 2020. In addition, the increased volatility of sovereign yields means that the returns of higher quality corporate bonds are once more being driven by changes in risk-free rates, rather than fluctuations in credit spreads. The net result is a significant reduction in the predicted risk for multi-asset class portfolios.

The chart below shows the 10-year breakeven inflation rate for the United States, derived from the price differences in inflation-linked and nominal US Treasuries, together with the same-maturity benchmark yield and the STOXX® USA 900. As shown, once long-term inflation expectations had risen above the Federal Reserve’s target rate of 2%, the 10-year US Treasury yield immediately surged more than 20 basis points and then proceeded toward levels last observed in February 2020.

US breakeven inflation, sovereign yields, and stock market

Source: Qontigo

That said, Fed Chairman Jay Powell was quick to assure market participants that the central bank was prepared to tolerate “inflation moderately above 2 percent for some time” in order to achieve its revised aim of “inflation that averages 2 percent over time” and that it would maintain its “patiently accommodative monetary policy” stance for the time being. This had the simultaneous effect of keeping short-term interest-rate expectations firmly anchored within the 0-0.25% Fed Funds target range (see the yield-curve chart below) and boosting share prices at the same time. In fact, the close co-movement of stock prices and expected consumer-price growth seems to indicate that moderate increases in the latter are still seen as signs of a recovering economy.

US Treasury yield curves and Fed Funds target corridors: current versus 12 months ago

Source: Qontigo

How did these recent changes in cross-asset correlations affect the predicted volatility of Qontigo’s global multi-asset class model portfolio[1]? The correlation matrices below were generated from daily risk factor returns over the three months preceding the displayed valuation date, using the Axioma Risk™ portfolio analysis platform. The subsequent charts, showing the risk decompositions by risk type and asset class, respectively, were taken from the weekly Multi-Asset Class Risk Monitor report. It is worth noting that the correlations represent the parts of security returns that were driven by respective factors, not the factor timeseries themselves. So, for example, a negative correlation between equities and “interest rates” implies an inverse movement of stock and sovereign-bond prices.

The matrix on the left depicts the environment shortly before the announcement of the first COVID vaccine on November 9. It indicates a strong co-movement (0.72) between equities and foreign exchange, reflecting the dollar’s strong depreciation, while share prices continued to recover. It also highlights the close relationship between share prices and credit spreads. Again, the positive 0.88 indicates that the portion of corporate bond returns driven by opposite changes in credit risk premia moved in unison with the stock market. Sovereign bonds, meanwhile, were decoupled from most other asset classes and factors, except breakeven inflation.

Major risk type correlations

Source: Axioma Risk™

The matrix for the recent environment on the right displays many more red cells, which indicates increased diversification benefits. Especially notable is the renewed countermovement of stock and sovereign bond prices, implied by the -0.51 between equities and interest rates. The near-zero correlation between share prices and FX rates likewise helped reduce overall portfolio risk. It is also worth noting the increased positive interaction of 0.51 between share prices and inflation expectations.

The expanded diversification benefit and reduced risk are also mirrored in the waterfall charts below, which show the volatilities of major risk factor types, weighted by the corresponding exposures within the portfolio. If all of these broad factor types were perfectly positively correlated, the total portfolio risk would be equal to the sum of the blue bars. However, due to imperfect, and sometimes negative, interactions between them, the actual overall volatility of the portfolio is usually lower.

Weighted portfolio volatility contributions by major risk type

Source: Qontigo Multi-Asset Class Risk Monitor, Axioma Risk™

The chart for last fall on the left shows much higher predicted portfolio risk of 16.3%, dominated by equity volatility and with almost no diversification benefit from low or negative correlations. The impact from the latter was, however, much more pronounced for the recent environment on the right, although fluctuations in share prices remained the predominant source of overall volatility. The beneficial effect of the renewed countermovement of stocks and bonds is even more apparent in the bar charts below, which show the percentage risk contributions by asset class, in relation to market-value weights.

Portfolio volatility contributions and market-value weights by asset class

Source: Qontigo Multi-Asset Class Risk Monitor, Axioma Risk™

In the chart on the left, all assets seem to add to overall portfolio risk—with the slight exception of US Treasuries and JPY cash, both of which neither increase nor decrease volatility, as they were decoupled from stock markets. More recently, holding US Treasuries and high-quality USD-denominated corporate debt once more lowered expected volatility. The negative contribution of investment-grade corporates is especially noteworthy, as throughout most of 2020, their returns were largely dominated by credit-spread fluctuations, making them seem positively correlated with share prices. Non-USD denominated fixed income securities, on the other hand, provided less diversification benefit, as their returns were dominated by exchange-rate variations.

The recent close co-movement of breakeven rates and share prices (highlighted also by the 0.51 correlation in the February 19 matrix) demonstrates that the inflation increase has so far been considered “reflation”, which implies the notion that rising consumer prices are a sign of a recovering economy. At the same time, the accompanying surge in sovereign yields provided risk-reducing benefits by holding high-quality fixed income assets alongside equities.

That said, the question now is, how much longer are central banks willing to tolerate above-target inflation expectations before they feel the need to step in and tighten monetary conditions, which could lead to a simultaneous sell-off in both the stock and bond markets? This could create a “perfect storm”—especially for multi-asset class investors based in the United States—as it might also result in a stronger dollar (due to higher interest rates), which would, in turn, amplify losses of non-USD positions. It is, therefore, paramount to keep a close eye on these underlying correlations. We certainly will.

For more regular updates on these topics, we invite you to subscribe to our weekly Multi-Asset Class Risk Monitor Commentary newsletter.

[1] A more detailed and comprehensive analysis can be found in our latest whitepaper on Cross-Asset Correlations in the Corona Crisis.