From the September 2003 issue of Investment Advisor • Subscribe!

Paths to Success

In the face of arguments for active or passive management, how about a middle ground where both sides can meet?

Should we choose to make our investments actively or passively? While the arguments are fast and furious on both sides, there is a place for both active and passive--the marketplace for investment management is evidence enough of this.

My goal is to propose a structure for thinking about the arguments and counterarguments, and to enrich the discussion by exposing a middle ground between them.

Let's look at the three main considerations of the debate. The first is that of market efficiency. If we truly believe that the market is informationally efficient, an active manager cannot add value and perhaps we should simply go passive. Most of us believe that the market is not totally efficient, however, so we need to dig more deeply.

The second consideration requires discriminating between skill and luck. To go active, we need to be convinced of our active manager's consistency and skill. We need to do this ahead of time and we need to believe that the skill will persist, or at least that we will be able to identify when it deteriorates. This is the most complex pragmatic issue of active investing.

Having identified the skillful managers, we come to the third consideration: If the active benefits are high enough and the costs are low enough, then go active; if the costs are too high, then go passive. Costs typically include risks, taxes, fees, and transaction and trading costs. Excess performance--alpha--is a primary benefit, but there are others that are not so easily measured. It is harder to evaluate benefits and costs when they are not quantifiable or even explicit.

In the Beginning

There are two objective premises which both sides will probably accept. The first--a formal argument--is the arithmetic of active management. In the January/February 1991 issue of The Financial Analyst's Journal, Bill Sharpe crisply articulated this theorem: "If active and passive management styles are defined in sensible ways, it must be the case that: Before costs, the return on the average actively managed dollar will equal the return on the average passively managed dollar; and after costs, the return on the average actively managed dollar will be less than the return on the average passively managed dollar. These assertions hold for any time period . . . and depend only on the laws of addition, subtraction, multiplication, and division. Nothing else is required."

By definition, passive provides the return of the average dollar invested in the market. Some active investors will do better than this average and some worse. The average active dollar will underperform passive because of costs; the higher the costs, the poorer will be the performance. In a more efficient market, the distribution of active investor returns will be tighter. In a less efficient market, there will more opportunity for a skilled active manager, and there will be a greater downside from a poor manager. This holds in inefficient markets and over any time period.

The second argument uses empirical data. Many researchers work with data on past performance of a set of managers to study the issues. This data is often useful, but must be interpreted with care. For one thing, such data usually has a survivorship bias. It is also not unusual to define the "index" as the S&P 500 and to use this to evaluate managers who select securities outside this universe of stocks, or to equal weight rather than to cap-weight managers when averaging their performance. This is not "sensible" averaging in the words of the theorem. Any study that shows the performance of a set of active managers outperforming a passive average mathematically must have issues with survivorship bias, ill definition of the average, or some other data issue.

But while empirical studies are useful, they will not be the focus of this article. Instead, we will focus on the three main considerations identified earlier.

Informational Efficiency

The concept of market efficiency is a useful construct in economic theory, but is not to be taken literally in real markets. I am after a pragmatic rather than a theoretical discussion.

The use of the term "passive" is perhaps unfortunate as it has the negative connotation of being inert, of offering no resistance. The active advocate sees passive as being "information-less."

This is too extreme for the passive proponent, who observes that prices encapsulate the wisdom, information, and expectations of a vast amount of market intelligence. This is a powerful concept, underlying capital pricing and investment management. A passive investor can exploit the prices that are set by the active investors, even if he believes that market inefficiencies exist. To beat the market, he points out, one needs to add something that is not already contained in these prices.

The active advocate is not comfortable accepting consensus prices, since doing so is too close to believing in market efficiency. He does not trust them. How can a market that drops 5% or 10% in one day with no real news be efficient? A decision to accept consensus prices means giving up his individuality and decision-making capability. Most people he knows are not rational. How then can the market prices be correct?

Market efficiency is a key conceptual discussion as it highlights the fact that opportunities to add value are not easy or obvious. But this discussion does not directly address our question. A belief in market efficiency may be sufficient for going passive, but it is not necessary.

Searching for Alpha Managers

In evaluating manager performance, it is difficult to distinguish skill from luck. Noise and uncertainty are high. The more efficient the market, the harder it is to identify skillful managers and the more costly it is to search for alpha.

Certainly, a large number of bright and eager participants spend time and money searching for opportunities or anomalies. As a result, opportunities in most publicly traded markets are rare. Those that exist quickly become unsustainable as other investors observe and imitate. Finding them becomes harder all the time.

The active advocate's first argument is typically his past performance. This doesn't cut any ice for the passive proponent, who observes the poor historical record of most active managers. In order to succeed, the active manager must be better than the best managers, not better than the average. He points out that in any population of active managers there will always be a winner and that a lucky break is easily spun as evidence of skill.

The honest active advocate counters thusly: Here are my specific ideas; I have found the following opportunities; I believe I have outstanding skill that will persist; and I will convince you of this. Those who are in the business of selecting managers will similarly try and prove their skill. The less honest will, with a smile, presume clairvoyance.

While factual and measurable, past performance is of little relevance to this discussion. Not everything that can be counted counts. There is little empirical evidence that past performance persists.

The winner of the search is the manager with the best active story. To the passive diehard, the active manager can never satisfactorily prove his skill because even the greatest success can be explained by luck. On the other hand, the active champion's promise of skill cannot be absolutely refuted.

For the sake of argument, let us suppose that we have found an active manager and we have a compelling case for where his edge is coming from. We then need to move on to costs.

Evaluating Costs and Benefits

In discussing costs and benefits, the passive proponent argues simply that the costs of active management are high. There are trading costs, management fees, taxes, and the expenses of monitoring an active manager and replacing him when necessary, not to mention added risk. For him, these costs exceed their benefit, which he sees as the unlikely possibility of alpha. A passive portfolio is substantially less costly to manage.

The active advocate can't easily dispute the implementation costs, but he feels there is something missing from the picture, some poetry. Active provides a more complex set of softer benefits that count but cannot be counted. Here are a few examples.

Agency arguments. Investors often work through an advisor who provides services including discipline, perspective, connections to good ideas, and a focus on implementation detail. The agent provides personal attention, a human and symbiotic relationship. The presence of an agent brings additional issues, particularly when his compensation is based on performance. Consider, for example, a corporate pension fund officer. Going active gets him attention, customized research and advice, and a convenient scapegoat if things go wrong; his job may even become enjoyable.

Arguments from psychological benefit. When we make active decisions, we are forced to be on our toes. We never stop questioning, pushing ourselves to the edge. Active management satisfies a need for control, perhaps even drama and risk-taking. In contrast, with passive we are etherized.

Many of us derive solace from a search for patterns in noisy data, particularly if we think we can obtain insight, information, or influence. We like to feel in control of our financial destiny, so going active offers hope when markets are down and when times are tough. The passive proponent sees the market as a Japanese pachinko parlor, a noisy roulette table. He tolerates the market uncertainty and seeks his satisfactions elsewhere.

Appeals to authority, status, and uniqueness. Most of us like to believe that we are independent thinkers, superior to that mass of average people. Evolution demands this.

We like to hear success stories, to be associated with winners and to be winners ourselves. So people like Warren Buffett and Peter Lynch become popular authorities, exemplars of active success. This is ironic to the passive proponent since both Buffett and Lynch have pointed out that passive is appropriate for most investors.

Some of us may regard these emotional benefits with cynical disdain. Economists often deride them as being "non-economic." Or, with the formulated phrases of rational decision-making, they sneer of irrationality. Some money managers and consultants, with their own psychological needs and reasons for doing what they do, disparage them.

But let us not underestimate their importance. Emotion is rational in the sense that it probably has an evolutionary basis. Emotional benefits are a key element of most decisions, and all our purchases involve some notion of emotional well-being. The feeling of control, confidence, and uniqueness that comes with active investment management is a good buy for many investors even at a high price.

Bad Arguments

Because many benefits of active are not quantifiable, its advocates struggle to make their case. In seeking an active manager, we need to avoid the bad arguments.

Arguments from innumeracy discount the key arithmetic of active management. These arguments use data to "prove" successful performance of a population or class of active managers or of the behavior of passive in certain market environments. Arguments have been made that active outperforms in down markets, providing protection when the market crashes, or that active outperforms in periods when value (or growth) outperforms, or that active value managers outperform but that growth managers do not.

Arguments from lack of information claim that passive is blind, insightless, or mediocre. These deny that market prices encapsulate the consensus information and opinion of a large number of informed participants.

Arguments from mean-variance inefficiency express a dislike for the average nature of indexing by proposing another benchmark that is in some way better. They argue, for example, that the market average is not a good place to be, or that market diversification can be improved. A subset are arguments against cap-weighting. Individual active managers often overweight small names at the expense of large names, and their excess performance is negatively correlated with that of large-cap stocks. But these are active notions and need to be justified independently of the active/passive decision.

And then there are arguments that are actually red herrings. There are criticisms of specific index definitions that proceed to the false conclusion that passive is therefore flawed as a strategy. For example, the S&P 500 index has been criticized as posing difficulties, since as a popular subset of the whole market, some exploit its changes and thus affect the movement of its component price. And some active advocates complain that the index is not a good measurement criterion, that their performance should be measured against some other benchmark.

Finally, when appeals to reason fail, there are emotive arguments that criticize passive as a fad, a cop-out, or even anti-American. While there are many good arguments for active, these are not among them.

Wheat Among the Chaff

Many of these arguments contain an implicit plea: I am not average, and neither are you; I have good ideas and they are a better choice. Despite the obvious fallacies in these arguments, we can sometimes find a hidden truth. Calling the index "mediocre" expresses distrust of consensus prices. Complaints of index-relative performance measurement contain a valid kernel: the index benchmark does not evaluate the non-economic and psychological benefits that the active manager provides. If it is an American characteristic to express our independence and always seek the best, then even the anti-American argument may strike near the truth.

However, these bad arguments are disingenuous not least because they distract us from asking key questions of ourselves and of the managers. We should know our reasons for deciding to go active, what we expect to derive in performance and other benefits, and the costs of doing so.

Active management is difficult, and the decision to choose a particular manager needs to be made with care. In interviewing an active manager, we should avoid a focus on past performance. To obtain superior investment returns, the manager should have and be able to clearly articulate an investment philosophy from which flows his superior value. The manager must be able to explain why he is better than the best active managers, not better than the average.

This is not to say that these questions are enough. In deciding to go active we need to do even more work: to evaluate the risks the manager will incur, to consider the investment process, costs, and to perform other duties of diligence.

Yet passive investing is also a difficult concept for many to accept. Its key drivers are a comfort with consensus prices, the difficulty in finding a skilled manager, and the high expense in doing so. We do not need to believe in efficient markets to go passive.

In the end, our question has no general answer; there is a place for both active and passive management. Each must answer the question for ourselves, being honest about our needs and our preferences.

Reprints Discuss this story
This is where the comments go.