Stock market investors have to decide whether they will be ‘active’ or ‘passive’ investors. The fundamental ideas that frame this choice are poorly understood, even by many investment professionals. So, when I heard (on 16 January) of the passing of the great Jack Bogle – the foremost exponent of passive stock market investing, and the founder of Vanguard – I thought I would write this article on the fundamentals of the active v passive question in his honour. It is the first in a series on active vs passive management issues that I have planned.
Figure 1 Proportion of US equity Mutual Funds and ETFs passively managed
As long as there have been stock markets (since the early 1600s) there have been both passive investors and active investors in those markets. Passive investors are ‘buy-and-hold’ investors who only trade for two reasons:
Liquidity: Buying to turn cash into stocks and selling to return to cash.
Rebalancing: When there is a change in the set of shares that a passive investor has access to, then a passive investor may rebalance their portfolio to match those changes. For example, passive investors may buy the shares of firms that list for the first time on the stock market.
Active investors trade for a third reason:
Private information: They believe they can identify mis-priced stocks. Those investors ‘actively’ seek out shares that are under-priced (to buy before the price rises) and shares that are over-priced (to sell before the price falls).
Financial economists didn’t have much to say about the relative merits of active v passive stock market investing until the early 1960s because of the lack of computing power to analyse large stock market datasets. Once large-scale computation was feasible, they were surprised to find that stock prices evolved essentially as a random walk, meaning that the change in a stock’s price in one period is very nearly independent of the change in the next period.
What does the change in GE share price in January tell you about GE’s price change in February? – nothing. Researchers found that this is true for the shares of every company that is frequently traded. Paul Samuelson and other great economists of the era knew what this finding implied – that new information, which will affect the price of a share, is very quickly incorporated into the share price.
For instance, if some good news about GE shares developed in January, such as higher than expected growth in earnings, and that news only slowly impacted the share price of GE, then it would affect both the January price change and the February price change of GE; thereby causing those monthly changes to be positively correlated. But, in fact, GE’s monthly price changes have near zero correlation which implies that new information that develops in January has no effect on February’s price change because the new information is rapidly incorporated into stock prices (entirely in January).
Competition to acquire information
If the independence of stock returns across years, months and weeks was caused by the rapid incorporation of new information into the stock price, then what caused that speed of incorporation? The answer was evident – the competition among active investors to acquire the information first and act on it.
If an event occurs (new information is created) that means that the GE stock price will rise $1, then that information is fully incorporated into the GE share price once the ‘discovery’ of the information by active investors, and their subsequent buying, drives the stock price up by $1. The intensity of the competition between active investors to be among the first investors to acquire and act on information causes new information to be rapidly incorporated into stock prices.
I should note before we move on that competition is not the only thing that drives the speed of information incorporation – other factors are also important, such as taxes and stock market rules. A particularly important example is bans on short selling, which impede the incorporation of new information because short selling is a crucial channel for negative information to make its way into the market.
Tests of the Efficient Markets Hypothesis
Eugene Fama introduced a formal notion of the ‘information efficiency’ of stock markets to describe the speed with which information is incorporated into stock prices. Fama described three different levels of market efficiency in terms of what type of information becomes ‘immediately’ incorporated into stock prices. Fama’s levels of market efficiency are:
Weak form (all types of past price information are immediately incorporated);
Semi-strong form (all publicly available information are immediately incorporated); &
Strong form (all information is immediately incorporated).
Weak-form efficiency is obviously true (stock market ‘charts’ are all baloney) and Strong-form efficiency is obviously false (insiders know a lot more about firms than outsiders), so economists focused on testing whether stock markets are Semi-strong form efficient.
Tests of any theory focus on the implications of the theory – that is the scientific method. An important test of Semi-strong market efficiency is whether active investors can earn higher returns than passive investors. If all publicly available information becomes immediately incorporated into stock prices, as posited by the Semi-strong version of the Efficient Markets Hypothesis, then there is no scope for active investors to acquire the information first and act on it to earn higher returns.
Researchers began studying the returns of professional portfolio managers to see whether on average they earned higher risk adjusted returns than passive investment. Michael Jensen’s 1968 study of portfolio manager returns was the first to use the newly developed Capital Asset Pricing Model (CAPM) to do the risk adjustment of portfolio managers’ returns. It is from that seminal study that the definitions ‘alpha’ for value added by portfolio managers through stock picking, and ‘beta’ for value added by market timing were first introduced.
A great many studies of portfolio manager performance followed, and that work continues to this day. The studies found that stock markets are not perfectly Semi-strong efficient because active portfolio managers as a whole do earn slightly higher risk adjusted returns than the rest of the stock market. However, the set of portfolio managers who can consistently earn excess returns that are higher than their management fees is very small. And, those managers are not of much benefit to ordinary investors — by the time their exceptional ability becomes apparent, ordinary investors cannot easily place money with them.
Equilibrium between passive and active portfolio management
These research findings lead directly to the creation of low fee, passively managed investment portfolios for investors. In 1973 David Booth and Rex Sinquefield created, for institutional investors, the first passively managed funds that tracked the S&P500 (an ‘index tracking’ fund). In 1975 Jack Bogle created the first passively managed mutual fund for individual investors in US stocks.
As the benefits of low fee, passively managed funds became more widely known, the proportion of passively managed money across all professionally managed share funds grew steadily (see the graph above). But it was soon recognised that passive management can never take over all of investment management.
Passive management and active management must exist in equilibrium. Passive management provides value to investors because stock markets are highly (but not perfectly) efficient. However, it is the competition between active investors, to be the first to acquire and act on new information, that sustains market efficiency.
As money moves from high fee actively managed portfolios to low fee passively managed portfolios the funding of active management falls, which in turn reduces intensity of competition for new information, which ultimately reduces the level of market efficiency. As passive management expands the returns to active management will increase (lower market efficiency) and the fees charged for active management will decline (competitive response to loss of market share) until an equilibrium is reached between passive and active management.
Effect of Technology on the Equilibrium
At what level of passive management that equilibrium will be reached is unknown. Passively managed funds own about 18% of all publicly traded equities, across the globe. The proportion is higher in the US, where the trend toward passive management is more advanced and 80 percent of stocks are held in managed portfolios (as opposed to self-managed, household portfolios).
In any case, the equilibrium between active and passive management will be a dynamic one because of the impact of evolving technologies. An example will illustrate. Companies release vast amounts of information to the public through their financial statements, regulatory filings and analyst briefings. Individual analysts can absorb the information of a small group of companies – all the large mining companies, or telcos, for instance. But no human analyst can draw out information by making comparisons across all the many millions of pages of information that is released in each quarter. But machines can do that with growing effectiveness.
Natural language processing is a form of artificial intelligence that draws out information by digesting the reports of thousands of companies simultaneously and identifying patterns. It is an example of an automated, relatively low cost, method of quickly incorporating new information into prices. It is active acquisition of new information, but at lower and lower cost.
There is a lot to say about passive v active management. In the next article I am going to discuss whether passive investment, and in particular exchange traded funds (ETFs) destabilises markets.
How investors should choose between active and passive management and how to implement passive or active management, through managed funds, listed investment companies, ETFs, superannuation, etc. is covered in my Finance Education for Investors course. Visit windlestone.com.au.