EquitiesDec 10 2015

Investment monkeys and passive portfolios

twitter-iconfacebook-iconlinkedin-iconmail-iconprint-icon
Search supported by

Imagine for the moment that you had never heard the term ‘passive investing’ before, and that you are approached by a confident, investment manager resplendent in a very sharp suit. The well-dressed investment professional tells you about a new investment strategy that his investment team has been experimenting with.

He explains the strategy. It involves identifying the 100 stocks listed on the London Stock Exchange with the largest market capitalisation. A portfolio is then constructed of these 100 stocks where their portfolio weights are determined by dividing the market capitalisation of each stock by the sum of market capitalisations of all 100 stocks. This portfolio is held for three months.

At the end of the three months the process is repeated: the main adjustments that need to be made to the portfolio at this quarter end, and at all subsequent quarter ends, involves divesting from those stocks that have fallen out of the top 100, and to invest in those that have been elevated into the top 100.

This process is what investors understand to be passive investing. My example describes, more or less, how one could invest on this basis using the FTSE 100 as the benchmark index. But looked at like this, a more appropriate description of this investment strategy is that it is rules-based.

At Cass Business School we have been investigating whether there is anything special about investing on a market cap-weighted basis. Billions and billions of investors’ money is currently invested on this basis. As one way of determining how good market cap-weighted investing is as a rules-based investment strategy, we decided to come up with our own rules-based strategy.

We began by collecting return data on the largest 500 US stocks spanning the period from January 1964 to December 2014, updating this set at the end of each year. Using this data we constructed a market cap-weighted portfolio where we updated the portfolio weights annually. Using the same data we also constructed a portfolio using our own rules-based approach. These rules were based on the popular board game Scrabble. Here is out it works.

Every stock in the dataset that we collected has a ticker symbol, that is, a unique three or four letter code. For example, the ticker for the software group Apple is AAPL and for Exxon Mobil it is XOM. For each company we calculated its ticker Scrabble score based on the points awarded for each letter in the game.

To implement our investment strategy we then summed each company’s ticker score. For example, AAPL scores six while XOM scores 12. We then divided each stock’s score by the total score of all of the stock’s weight in the portfolio. We repeated this process at the end of each year, rebalancing the weights according to the Scrabble rules. We were pleasantly surprised by the results that are shown in the chart.

An investor that had invested US$100 in January 1969 in a strategy where portfolio weights were determined by a stock’s market capitalisation would have seen this initial investment grow to US$7,718 by the end of 2014. However, the same US$100 invested in a portfolio where the weights were determined by the rules of Scrabble would have grown to US$14,108.

We performed another experiment, where the weights were chosen at random by a computer instead of according to the market capitalisation of the stocks, or by Scrabble rules applied to the company’s ticker. In fact, we asked the computer to do this not just once, but 10 million times. Essentially we created 10 million portfolios where the constituent weights may as well have been chosen by a monkey – lots of monkeys.

The results of this random experiment were also very revealing. Only 26 per cent of the “monkeys” – 2.6m – managed to produce a portfolio with a better risk-adjusted performance better than the Scrabble-inspired approach. But the popular market cap approach only managed to beat 0.12 per cent of the monkeys. Put another way: on 9,988,179 occasions out of 10m, investors would have been better off having a monkey design their portfolio than rely on market cap-weighted approach. Please take some time to digest this result, in our view it is quite astonishing.

The idea then that there is this neutral, but ultimately beneficial way of investing which we refer to as passive investing, is clearly wrong. Actually when we invest in what are widely thought of as passive funds, we have made an active choice to trust in one particular rules-based strategy – one that could have been beaten by almost any set of randomly chosen set of rules, and any old monkey.

There are an infinite number of approaches that one could use to determine portfolio weights. Our results suggest that investors need to consider carefully the set of rules that they wish to follow. Indeed, the old debate about ‘passive versus active’ investing needs to be couched in different terms. With a wide and increasing range of rules-based investments strategies available in fund and ETF form, the choice is not between passive and active, it is between rules-based investing, or investing based on discretion.

Professor Andrew Clare is chair in asset management of Cass Business School

Key points

Billions of investors’ money is currently invested in a market cap-weighted basis in passive funds

An experiment was run to adopt a different rules-based approach

The alternative rules-based approach was nearly always more successful than the conventional passive approach