Stocks don’t always beat bonds and bonds aren’t always low risk: What matters are regimes (2024)

The lodestar of all modern investing is that over suitably long periods of time, equities return a premium compared to less risky assets such as bonds. The concept of the equity risk premium sits at the heart of pretty much everything most of us do professionally every day. The catch is that although that last statement is largely true – on one measure, 80% of most long-duration periods - it isn’t always true for all decades, and all markets. None of that should come as a surprise to anyone but it’s worth repeating: Equities are worth the extra risk from volatility, most - though not all - of the time.

I dwell on this because we’ve had a slew of academic papers in recent months (three, in fact) that reinforce this element of caution – sure, equities are a great investment, but not always. One way of looking at this is to dig around in what is perhaps the definitive data source on investment history.

Every year, around the end of February, the UK-based investment academics Paul Marsh, Elroy Dimson and Mike Staunton bring out their definitive account of long-term returns from investing, the Global Investment Returns Yearbook, this year debuting from UBS after its acquisition of Credit Suisse. It examines long-run returns since 1900, including 35 developed markets, 23 of which started in 1900 with a wider, less comprehensive sample comprising of 90 markets. Most of its conclusions won’t surprise the informed reader. Take the notion that the historical equity risk premium vs. bonds has been around 3.3% per annum, while the equivalent premium versus bills is about 4.7% per annum. That feels about right and syncs up with most investment narratives. The report also suggests that US real annualized equity returns are running at about 6.5%, with the world at 5.1% pa.