The global equity risk premium (ERP) over US Treasury bills held at a steady 4.3 per cent during the last 115 years, according to the latest Credit Suisse ‘Global investment returns yearbook’.
In the latest iteration of its well-respected annual market statistical analysis, authored by the London Business School’s Elroy Dimson, Paul Marsh and Mike Staunton, the Credit Suisse Yearbook put the global ERP – as measured against a 23-country index – at 4.3 per cent from 1900-2014 and exactly the same rate over 1965-2014.
Annualised real equity returns (in $US terms) on a 23-country global index stood at 5.2 per cent over the 115-year period while bonds returned 1.9 per cent, the study says.
However, excluding the USA, the report found greater volatility in both long-term share returns and the equity premium, which stood at 4.6 per cent over 1965-2014 and 3.6 per cent for the 1900-2014 period.
“… from the perspective of a US-based international investor, the real return on the world ex-US equity index was 4.4% per year, which is 2.1% per year below that for the USA,” the Credit Suisse Yearbook says. “This suggests that, although the USA has not been the most extreme of outliers, it is nevertheless important to look at global returns, rather than just focusing on the USA.”
New Zealand’s average ERP (in $US terms) relative to US Treasury bills varied between 2.8 per cent over the last 50 years and 4.4 per cent for the full 115-year period covered by the Yearbook.
The study also concludes the influence of country allocation relative to industry is weakening.
“To exploit diversification opportunities to the full, investors need to diversify across a wide spread of industries and countries,” the Yearbook says. “Both matter, although there is evidence that globalization has led to a decline in the relative importance of countries, with industries assuming greater importance.”
Dimson et al also warn investors against mean-reversion tactics based on ERP and discount rates.
“Asset allocators and strategists, however, need to take a view on the attractiveness of markets, and can use market-implied discount rates as a signal. Their job is difficult,” the report says.
“… For all intents and purposes, monthly changes in the discount rate and ERP are random walk. The market can remain seemingly irrational for long periods, debunking naïve arguments for near-term mean reversion.”