Commentary: Performance fees: Making fresh lemonade from old lemons

Genuine novelties in our industry are rare, whereas imitation is more than readily available given the surfeit of asset managers. However, just because an idea is not new does not mean it cannot or should not be revisited in the light of current industry thinking or market conditions.

This juxtaposition of new and old comes firmly to mind when I consider recent market noise about performance fees. I have been an advocate of performance fees for 25 years. As my consulting clients in the 1990s may recall, as a consultant I virtually always asked active managers to quote for new business on both a flat and performance basis. My rationale was firstly that it was always beneficial for a client to be able to have a choice. Secondly, a good performance fee structure produces a less volatile net-of-fees return stream, which is what matters most. Thirdly, the Achilles' heel of active managers in our industry has long been a failure to maintain discipline on capacity, and so a structure that hardwired a commercial preference for performance over AUM would (or certainly should) reinforce a strong capacity mindset.

Of course, not every client was comfortable, sometimes for good reasons. For example, the risk of having to pay a performance fee for good relative performance, even if absolute performance was negative, was legitimately a client concern in some settings, even though it would still be an indicator that the manager was being rewarded for adding value. Moreover, there were good and bad performance fee structures too. Nevertheless, I generally thought the arguments were typically very favorable and I recollect that the majority of my clients tended to agree with me.

Thus performance fees are definitely far from a novelty. However, I do believe they could, and arguably should, be central to other key debates that our industry now faces. These arguments don't seem to get the air time they deserve, and I want to try and remedy that.

The first of those is the constant struggle between active and passive and what weight should a client give to each. Given the column inches devoted to this, it's hard to imagine there is an angle that is underexplored, yet performance fees often fail to even enter the discussion, even though the availability and nature of an active performance fee approach should be highly relevant to a client who is considering this debate in a balanced way. For purposes of analysis, I am going to assume that a client has available a base fee that involves the same amount of fixed cost as would be charged for passive management, and that they then pay a proportion of any excess return over the index to a manager (in a way that minimizes adverse optionality for both sides over the lifetime of the mandate).

Of course, a performance fee is not likely to change the minds of those who don't believe in the possibility of investment skill. For everyone else, however, I would stereotype the arguments against active along the following lines:

• The fees are disproportionate relative to the expected added value.
• While skill exists, diagnosing it successfully is not automatic, and there is a risk of picking a bad active manager.
• The governance burdens of selection and monitoring are not justified by the expected reward.

The presence of a performance fee option of the type above materially impacts all three. By design, the first point is addressed directly: The client will pay the passive fee unless there is outperformance and then only in direct proportion to that outperformance. This argument can be extended because whatever an investor's probabilistic belief about the manager's future gross return stream, the resulting net-of-fee return stream in the presence of this performance fee design will have less volatility to the client.

So as a good quant, I thought we should build a model illustrating the impact of performance fees. Specifically, I conceived of a two-state model to demonstrate how performance fees would lower the level of conviction required to appoint an active manager. But as I was running this theory by other members of our team, we hit upon a very different construction, based on a separate old idea, George Akerlof's famous "The Market for Lemons," written in 1970.

In Mr. Akerlof's version (which helped win him the 2001 Nobel Prize, shared with Michael Spence and Joseph Stiglitz), the argument goes loosely as follows: In any market characterized by asymmetric information, where the seller has a large information advantage over the buyer, the pricing has to be set in such a way as to protect the buyer from unwittingly getting stuck with what is popularly referred to as a lemon.

In the asymmetric market for active equity strategies, using Mr. Akerlof's construct, performance fees conceivably offer a way to bridge the information gap that no amount of backward-looking checking under the hood provides. Rather than simply accept the average fee, the manager, by offering a low-base, fee-plus performance percentage, is implicitly communicating to the client a high level of confidence that its skills are highly above average. The client, in turn, requires a lower level of conviction to ensure that the manager isn't pulling a fast one or simply misguided about its abilities.

Extending Mr. Akerlof's "lemon" analysis thus provides a further elegant refinement of the significance of performance fees in the active-passive discussion. But there is also another novel angle I want us to consider, which is where the availability of performance fees should enter in the whole investment strategy process.

Our industry — maybe because of the principal agent nature of its relationships — typically approaches decision-making in a hierarchical fashion. For example, we consider first how much should be allocated to equity, then how much to each market, then whether to use active or passive, then which manager, and finally which fee scale (if you embrace my choice at the start). In fact, given the arguments above, as a client or adviser, you should actually bring the availability of a performance fee of this sort into the earlier decisions. In almost any statistical framework — given the better overall net-of-fee return shape, the lower cost of being wrong and the lower governance threshold theoretically required — the presence of the performance fee will have a bearing on the optimal weighting to active vs. passive and indeed may well tip the scales on a yes-no question.

So what's the message here? Performance fees are not a new concept; in fact, they are probably much older than most industry commentators realize. However, don't let the lack of novelty put you off. They potentially have new relevance to our current industry debates, whether through genuinely fresh thinking about how and where they apply or looking at them through a different lens.

Andrew Dyson is CEO of Quantitative Management Associates LLC, Newark, N.J. This content represents the views of the authors. It was submitted and edited under P&I guidelines but is not a product of P&I's editorial team.