Monty Python and the Search for the Ultimate Index


The Financial Times reports that financial index provider MSCI Inc. has joined Sir Galahad, Indiana Jones and Monty Python in the quest for the Holy Grail. In this iteration the goal is not a cup, but the ultimate market portfolio. This modern crusade is based on a misunderstanding of 60-year-old work in quantitative finance, and seeks something that was proved not to exist back in 1977 and then declared dead in 1992.

The idea of a “market portfolio” originated with the Capital Asset Pricing Model in the early 1960s. Researchers realized that if they could identify one portfolio that was mean-variance efficient, meaning a portfolio that has the highest possible expected return for a given level of variance, they could price all capital assets. Several different economic arguments were advanced to claim that the market portfolio of all assets should be efficient. But even if those arguments were correct (they are not, and even at the time they were regarded only as possibly useful fictions) there could be lots of efficient portfolios, and no reason to assume the market portfolio is a good choice for any particular investor.

The CAPM market portfolio got confused with another idea about a decade later, one championed by Vanguard founder John Bogle, that investors should hold broadly diversified index portfolios that weighted assets by market capitalization. The reason for market cap weights is not CAPM theory; it’s that they minimize transactions since if an asset goes up or down in price, the dollar amount you wish to hold of that asset changes in sync, so there’s no need to rebalance.

These two ideas are independent. You can believe CAPM and not like market-cap-weighted index funds, and you can buy Vanguard funds without any trust in CAPM.

In 1977, economist Richard Roll proved that the market portfolio does not exist. In 1992, the father of efficient markets, Eugene Fama, along with co-author Ken French, declared the death of the idea of a single-factor market portfolio ring to rule them all.

For the last 30 years, academic research in finance has shifted to identifying the multiple factors that explain asset returns rather than trying to identify and price all capital assets in the world in the hopes one portfolio can explain everything. This research has led to quantitative equity strategies including smart beta and factor investing. The quest for the optimal mix of all assets has moved on to ideas such as risk parity and minimum variance that consider the statistical properties of asset classes, not their dollar size.

On the portfolio management side, indexing has grown rapidly not by building portfolios with more assets, but by creating index funds for more asset classes. Investors can select among them to build portfolios optimal for their goals, risk tolerance, horizon, liabilities, tax status, currency and other parameters. The practical task for investors is to build close-to-efficient portfolios using assets that are cheap to buy-and-hold, with good liquidity and reliable price information, not to include every possible asset.

MSCI researchers have been leaders in these efforts, so why are they trying to turn back the clock? I think MSCI is putting a dramatic marketing spin on a less-exciting goal. (A spokesperson for MSCI says the firm is not actively developing an “ultimate index.”) John Bogle, mentioned above as the father of index investing, long ago came up with a rule of thumb that investors should be 60% in stocks and 40% in bonds. At the time he said it, it was not unreasonable. These were the only asset classes with good liquidity and their volatilities were similar.

But things changed. Bond volatility fell to the point that over 90% of the risk of 60/40 portfolios came from stocks. Liquidity in other asset classes grew. Nevertheless 60/40 grew from a casual rule of thumb thrown out for average investors to a near-universal benchmark for institutional investors.

The problem with using a narrow and unbalanced benchmark like 60/40 is it makes assets other than stocks look too good, because their correlations with 60/40 are less than their fundamental correlations to global wealth. Another problem, if you are as cynical as I am, is MSCI can’t charge much money for selling a 60/40 benchmark index. Any broader and better-balanced index would be an improvement in both management of institutional portfolios and MSCI’s revenues.

There are institutional reasons for the asset management business to agree on a primary standard. While no one defends 60/40, there are an infinite number of potential replacements. In order to stand out from the pack, you need to make dramatic claims like “the ultimate index,” “the Holy Grail,” and “the true market portfolio.” None of these things are possible, or even meaningful, but you don’t get a bout with the champion — even an over-the-hill one who wasn’t very good in her prime like 60/40 — by being modest. Faint heart ne’er won fair lady, and understated academic research won’t overturn 60/40.

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

Aaron Brown is a former managing director and head of financial market research at AQR Capital Management. He is the author of "The Poker Face of Wall Street." He may have a stake in the areas he writes about.

©2021 Bloomberg L.P.

BQ Install

Bloomberg Quint

Add BloombergQuint App to Home screen.