Are value-weighted global benchmarks optimal for individual investors?
Core argument: The authors estimate the World Portfolio of all investable assets globally at $261tn, or 200% of world GDP, and construct a long-term proxy back to 1950 plus an investable index since 1990.
The World Portfolio is the sum of all investable assets globally, which we currently estimate to be [$261T], or 200% of world GDP. We create two versions of the World Portfolio: (1) a long-term proxy since 1950 using monthly data for a reconstructed, historical universe of global equities and bonds (mostly sovereign) and Gold, and (2) an investable index since 1990 using mostly daily data for a universe comprising benchmark indices for global equities, sovereign bonds, credit, real estate, private markets, cryptocurrencies and Gold. [Editor's Note: All figures shown here, excluding Exhibit 1, use the second measure]. Persistent shifts in asset weights are often closely linked to macro trends, like inflation boosting gold relative to equity and bond markets during the 1970s, or equity bubbles like the Tech Bubble during the late 1990s. Currently there are three major trends visible: (1) the equity weight relative to bonds has increased materially since the GFC but it remains below levels from the 1990s, (2) both in equities and bonds the US has accrued a larger weight and is very dominant and (3) alternatives such as private markets, Gold and cryptocurrencies have grown relative to public equities and bonds (but remain relatively small). Benchmarks like the World Portfolio are a major influence on investors' asset allocations. However, following benchmarks is not necessarily a good idea – even over longer horizons the World Portfolio has seldom been optimal [for any particular investor] and performance varied materially with structural macro regimes. Value-weighted benchmarks miss out on diversification benefits from smaller assets and, while US asset dominance was a tailwind in recent years, regional performance varied historically.




