The optimal portfolio for the next decade
Unlock the Editor’s Digest for free
Roula Khalaf, Editor of the FT, selects her favourite stories in this weekly newsletter.
The writer is head of asset allocation research at Goldman Sachs
Since interest rates started to rise in 2022, investors have been recovering from one of the largest shocks to their portfolios — and to their belief system around multi-asset diversification.
The surge in inflation during the recovery from the Covid-19 crisis resulted in one of the largest losses for multi-asset portfolios in more than a century. But up until then, simple buy-and-hold portfolios that invested 60 per cent in equities, 40 per cent in bonds had been remarkably successful for the current generation of investors. Inflation has now broadly normalised. So, what is the optimal portfolio for the next decade — is it fine to just go back to 60/40?
The modern portfolio theory advanced by Nobel Prize winner Harry Markowitz allows us to find the so-called “optimal portfolio” with the highest return relative to risk, incorporating potential for diversification. This portfolio can then be combined with cash or leveraged up, depending on risk tolerance. In hindsight, the 60/40 portfolio did indeed deliver the highest risk-adjusted returns since 1900, but over 10-year rolling investment horizons the optimal asset mix had large swings and was seldom exactly 60/40.
In fact, the optimal asset mix in the three decades up until the Covid-19 crisis was more like 40/60. Helped by low and anchored inflation, bonds had a strong bull market and provided diversification benefits by buffering equities during episodes when investors adopted a “risk off” approach. But since 2022, long-term bonds have performed poorly. While bond yields are now higher and near long-run averages, rates volatility remains high. The curve of yields on bonds of differing maturities is also flat. These factors suggest little benefit relative to just cash.
As a result of the poor bond performance since 2022 — but also strong equity returns — the optimal portfolio over the past decade has shifted close to 100 per cent equities. Unsurprisingly, the value of long-term bonds in the portfolio remains in question. With more uncertainty on inflation in the coming years, as well as growing risk from fiscal policies and higher government debt to GDP ratios, bonds have also become riskier.
Still, having only equities in the portfolio appears imprudent after the strong rally and given elevated equity valuations, especially in the US. Expected equity risk premia — that is, prospective excess returns for equities versus bonds — are at the lower end of their historical range. This can either reflect lingering concerns on inflation or more long-term growth optimism.
We think the latter is more likely, in part due to technology revolutions such as generative AI and new weight loss drugs, but also due to the unusually high profitability of the US tech sector. During previous periods of high productivity growth, like the 1920s and 1950/60s, equities also outperformed bonds for prolonged periods of time.
You are seeing a snapshot of an interactive graphic. This is most likely due to being offline or JavaScript being disabled in your browser.
Potential adverse trends for equities though include: deglobalisation, both economically and geopolitically; decarbonisation, with more risk of commodity supply shocks and rising costs around climate change; and demographic trends such as lower population growth, higher levels of dependent people and income inequality.
So for the coming years we see value in being more balanced again in multi-asset portfolios. Broader diversification is necessary to diversify structural risks. The optimal portfolio for the next decade could be one-third equities with a bias to “growth” stocks, one-third bonds and one-third real assets.
In that portfolio, growth stocks would provide more targeted exposure to improving productivity and diversify disruption risk. A key challenge is that equities tend to anticipate higher productivity growth before it materialises, resulting in valuation expansion and an increased risk of overpaying. With already elevated growth stock valuations, investors will need to be selective in their hunt for the beneficiaries from future technology revolutions.
Bonds would provide protection in the event of stagnation with higher real yields. They are currently factoring in low inflation but if growth in prices picks up, exposure to real assets in the optimal portfolio can help diversify risks. These include stocks with pricing power in areas such as infrastructure, real estate and commodities. So investors could put an additional 20 per cent of the portfolio in equities on top of the growth stock investments. The rest of the real assets could go into inflation index-linked bonds.
This journey leads us back to a roughly 60/40 portfolio. But the optimal portfolio for the next decade needs to address the potential for pick-up in productivity growth and the risk of higher and more volatile inflation. This means more targeted exposures within equities and bonds.
