Incentives in Investment Management

January 29, 2019

Most of economics can be summarized in four words: ‘People respond to incentives.’ The rest is commentary” – Steven E. Landsburg

We are certainly not the first to observe the misalignment of incentives in long-only equity management. For example, Warren Buffett touched on the subject as early as 1967:

“These investment company performance figures have been regularly reported here to show that the Dow is no patsy as an investment standard. It should again be emphasized that the companies were not selected on the basis of comparability to Buffett Partnership, Ltd. There are important differences including: (1) investment companies operate under both internally and externally imposed restrictions on their investment actions that are not applicable to us; (2) investment companies diversify far more than we do and, in all probability, thereby have less chance for a really bad performance relative to the Dow in a single year; and (3) their managers have considerably less incentive for abnormal performance and greater incentive for conventionality.

– Warren Buffett[1]

The Traditional Approach

Today, as in 1967, most long-only equity managers are paid a flat fee on assets under management (AUM).[2] Thus, the only reliable way these managers can grow their revenue (and correspondingly their personal wealth) is to manage more capital. Hence, for the vast majority of long-only managers, a profit maximizing strategy equates to an asset maximizing strategy. Because clients have historically been slow to fire underperforming managers (as evidenced by their very abundance), a good asset maximizing strategy has been to try to closely track the benchmark hoping for modest outperformance, while avoiding large deviations from the benchmark for fear of drawing their clients’ attention and getting fired. This translates to the following practical consequences of the flat management fee structure:

1. Propensity to run a highly diversified portfolio that will tend to closely mirror the benchmark in composition to avoid meaningful deviations from it. This also means that the manager will be hesitant to invest a significant portion of capital into a single idea, regardless of its merit, as such an investment could lead to short-term underperformance relative to the index. This reinforces the primacy of descriptive statistics (discussed later).

2. Willingness to accommodate external and internal restrictions even if they are likely to hurt performance. A manager who is paid a flat fee on AUM, regardless of the underlying performance of the Fund, is likely to be more accommodating to restrictions that could result in a short-term boost to AUM at the expense of underperformance. For example, an investor who does not believe in the distinction between Value and Growth styles might accept the decision to own, largely, statistically cheap investments because he is told that, “today Value is in favor”.

3. Aversion to closing a fund prior to reaching a point where incremental returns on new capital become dilutive. If growth in AUM is the only road to financial success, closing a fund to new investors would go against a manager’s self-interest. In an industry where achieving outstanding returns becomes more difficult as AUM grows, this eventually translates to a sub-par experience for the clients.

An Alternative Approach

We realize that pointing out deficiencies in an existing model is easy. Developing a superior alternative to the status quo is much harder. Below is an alternative framework that we developed, tested and have now made available to institutional investors:

Pay for Performance

We believe that long-only managers are more likely to succeed if they operate under an incentive structure that reinforces adding value as opposed to adding assets. With that in mind, we opted for an “ETF-like management fee” that covers operating expenses. Additionally, we explicitly restricted the amount of outside capital that we can raise in order to maximize the probability of maintaining optimal performance.

In addition to the low management fee, we also charge an incentive distribution that is a function of value added, a close proxy to alpha, relative to the client’s opportunity cost.

We are not blind to the inherent drawbacks with performance-tilted compensation schemes, namely, the asymmetry of outcomes between the manager and the client. Specifically, such an incentive structure could incentivize the manager to take undue risks in the pursuit of the performance fee because they stand to benefit if they are successful while only the client loses if it fails. Although this risk is certainly real, we believe it can be mitigated, to a large extent, with the following mechanisms:

1. A manager should have a substantial portion of their net worth invested alongside their clients. The amount should be in excess of what the manager generally expects to earn in performance fees in any year and, ideally, should represent the bulk of the manager’s liquid net worth. Therefore, the manager experiences losses personally as a result of risks taken in the fund.

2. The performance fee should be earned on a value-added basis over several years (three years is reasonable) as opposed to a full payout after a single year of good performance.

3. Any shortfall in performance should be carried forward to future years to ensure that the manager only earns an incentive fee if the totality of clients’ experience with the manager has been satisfactory.

When answering the question of how much capital we can accommodate in this strategy, we give an honest answer: we do not know. But there are certain things that we do know: 1) most managers do not find out that they manage too much money until they crossed that line; and 2) there are some like-minded investors who successfully manage portfolios in excess of $1 billion under much more restrictive terms than ours.

Thus, although we are unlikely to reach the point of negative incremental returns until we get well beyond $1 billion in the strategy, we have voluntarily implemented a number of restrictions to limit our capital base.

[1] Buffett, W. (1967) 1967 Letter to Limited Partners.

[2] Ritholtz, B. (2018) A Hedge Fund That Only Charges for Alpha.

« Back To Blog