I recently came across one of the more interesting things I’ve seen in a while: a report published by Goldman Sachs titled Investing in Everything, Everywhere, All at Once. As the name suggests, it’s a sweeping overview of investable assets globally and how the mix has changed over time.

By looking at the outstanding market value of all major assets globally, the report’s authors created two different versions of a neutral market portfolio. The first version is based on monthly data for global stocks and bonds, with weightings going back to 1950. The second version is a more current snapshot based on investable indexes with a broader range of underlying asset classes, and it dates back to 1990. The chart below shows the current version.

The global market portfolio isn’t static. During the 1990s, for example, it was made up of about 60% stocks and 40% bonds, which reversed to 40% stocks and 60% bonds in the wake of the global financial crisis. Since then,stocks have grown significantly as a percentage of the total, although the equity weighting is still below the peak levels from the 1990s. Within the US equity market, the technology sector (which the authors define as technology plus communication services) has ballooned and is now similar in size to the level it reached (more than 35% of the US market) in the 1999 tech-stock bubble. Nontraditional asset classes, including private markets, gold, and cryptocurrency, have grown and now represent a more significant slice of the overall market.

The report’s authors note that simply matching the global benchmark typically doesn’t lead to the best results, and an optimal portfolio can look dramatically different depending on the time period. In the inflationary period of the 1970s, for example, a portfolio with a hefty stake in gold (as high as 70% of assets at some points) would have generated the best results. During the “lost decade” of the 2000s, investors would have fared best by keeping the bulk of their assets in bonds.

The authors discuss numerous strategies for making tactical shifts in a portfolio’s asset weightings versus the neutral market portfolio, such as adjusting the overall mix of stocks, bonds, gold, international assets, and currency exposure. Separately, they found that an “enhanced diversification” portfolio combining various assets such as bitcoin, infrastructure stocks, low-volatility stocks, dividend aristocrats, as well as alternatives (including commodity carry, cross-asset trend-following, hedge funds, and private markets) led to better risk-adjusted returns over most five-year rolling periods going back to 1990.

This approach sounds promising, but there’s a catch: There’s an overwhelming amount of evidence that shifting a portfolio’s asset composition usually doesn’t work in practice. While it would be great to invest in an optimal combination of asset classes, what that mix is can only be known after the fact. So far this year, for example, tactical-asset-allocation funds have continued to lag a static mix of stocks and bonds, such as the basic version of a 60/40 portfolio.

While I’m skeptical about whether investors can identify an optimal asset mix in advance, there are several other important lessons I gleaned from the report.

1) You might have too much in US stocks.

Thanks in part to the weak dollar, international stocks have pulled ahead of their domestic counterparts by a wide margin in 2025, gaining about 28.3% for the year to date through Oct. 24, 2025, versus 16.4% for domestic stocks. But that victory comes on the heels of a long, nearly unabated stretch when US stocks trounced everything else. As a result, the United States now accounts for 64% of the global equity market. That still leaves 36% outside of the United States, though, which would represent a neutral weighting in international stocks. The typical portfolio for a US-based investor is probably well below that level, as international funds currently make up only about 28% of all equity-fund assets.

2) Don’t neglect bonds, either.

Fixed-income securities make up a large portion—nearly 40%—of the global market portfolio. That doesn’t mean that every investor should maintain a 40% weighting in bonds (although it does probably explain part of the enduring success of the 60/40 portfolio). However, most portfolios should probably include some allocation to bonds. The reason is simple: Because the correlation between stocks and bonds is typically relatively low, adding bond exposure can improve a portfolio’s risk-adjusted returns. That lends support to the notion that even young investors saving for retirement many years down the road should probably maintain at least a small allocation to bonds.

3) Limit exposure to alternative assets such as gold and cryptocurrency.

Both assets have been on a tear so far this year. Gold prices have surged by nearly 60%, driven by aggressive buying by central banks looking to diversify their reserve assets away from the US dollar, as well as retail investors buying up gold-focused funds (although there’s been some selling pressure in the past week or two). Cryptocurrencies like bitcoin have gained a more modest 20% or so. Both assets have benefited from weakness in the dollar, which investors fear could continue amid record-high debt levels and additional rate cuts by the Federal Reserve. However, both bitcoin and gold can be extremely volatile, with price trends driven by speculative buyers. They also represent a relatively small fraction of the global market portfolio (particularly for bitcoin).

4) Be wary of overweighting private equity and private debt.

Investors are hearing more and more pitches about the merits of adding private capital to their portfolios. In his 2025 letter to shareholders, for example, BlackRock Chair Larry Fink floated the idea of a portfolio made up of 50% stocks, 30% bonds, and 20% private capital. Based on BlackRock’s projections, adding the 20% stake in private capital could improve a portfolio’s risk-adjusted returns and help investors build more assets for retirement. That may or may not turn out to be true, but a 20% weighting would represent a massive overweighting relative to the neutral market portfolio. Given the additional costs and risks of private capital, I would be cautious about any allocation to these assets, much less one that’s 4 times as large as current market values would suggest.

5) Don’t overdo it on real estate.

Investors frequently get carried away by the above-average yields, tax benefits, and potential diversification benefits of real estate investment trusts. But as the first chart above shows, real estate as a whole only accounts for 2% of the global market portfolio. In addition, the potential diversification value of REITs might be overstated. The Morningstar US REIT Index has had a correlation coefficient as high as 0.90 when measured against a diversified stock benchmark. (To be fair, the correlation for the trailing three years through Sept. 30, 2025, was a bit lower, at 0.76). It’s not uncommon to see portfolio weightings of 5% to 10% in REITs, but that would represent a pretty big bet on the sector given its relatively small size.

Conclusion

Not every investor needs to own the global market portfolio in the same proportions. In particular, adjusting the stock/bond mix is one of the most important levers investors can employ to tailor a portfolio’s risk level to their specific goals. And just because something makes up part of the overall market doesn’t mean you need to invest in it. (For example, I don’t own any gold as part of my portfolio).

That said, I’d argue that the global market portfolio is still an essential reference point for building a portfolio. If you decide to allocate more or less to a given type of investment than current asset values would suggest, it’s important to keep in mind that you’re making an active asset-allocation bet that may or may not pay off.

Get more Morningstar insights in your inbox