Decisions with Other People’s Money

The investment industry is driven by the need for intermediation.  Individuals and organizations seek to invest funds and feel like they lack the expertise or time to do so on their own.

That intermediation sets up potential principal-agent problems, where the interests of a client and the professional(s) she has hired are not aligned.

Academic research

A good survey of the applicable research, The Principal-Agent Problem in Finance, was published in 2014 by the CFA Institute Research Foundation.  The author, Sunit Shah, described the main areas of study and provided summaries or excerpts from many of the notable publications on the topic.

Much of the focus regarding asset management firms relates to fee structures:

[I]f an investor’s incentives are not aligned with those of the manager, the manager often has both an incentive to act counter to the investor’s best interests and the ability to do so undetected.  Given the magnitude of payment generally involved in asset management contracts, misaligned incentives have significant potential to override a manager’s fiduciary responsibility to his or her clients.  Structuring such contracts optimally is, therefore, of the utmost importance.

A section regarding compensation for asset managers says that:

The optimal contract balances the trade-off between the motivations in the equity-like compensation of the management fee and those in the option-like compensation of the incentive fee.

However, when it comes to contracts, most investors are price-takers.  There is some haggling on the level of fees — with larger asset owners being able to get better terms — but fee structures have been resilient.  “Optimal” contracts remain a pipe dream, despite a good deal of focus on the shortcomings of current approaches.

The research regarding incentives primarily pertains to the asset management firms as a whole rather than the incentives for individuals within them, which are often quite diverse; those incentives vary depending on the respective roles of the employees.

Because that detailed information is not publicly available, there has been little examination of the effect of incentive structures on decision making by individuals — although those within asset management firms can attest to behaviors they have witnessed that have been induced by them.

Experimental studies

One way that academics are trying to get a sense of the impact of incentives is through experimental studies; experiments are created to judge behavioral responses to certain conditions.

An example is “Delegated Investment Decisions and Rankings,” by Michael Kirchler, Florian Lindner, and Utz Weitzel.  The abstract reads (in part):

Two aspects of social context are central to the finance industry.  First, financial professionals usually make investment decisions on behalf of third parties.   Second, social competition, in the form of performance rankings, is pervasive.   Therefore, we investigate professionals’ risk-taking behavior under social competition when investing for others.

It fits with other work detailed in “Investing Other People’s Money,” from Sascha Füllbrunn, Ola Kvaløy, and Wolfgang Luhan, a chapter in the upcoming Handbook of Experimental Finance.  That piece “aims to consider and summarize the existing experimental evidence on investment decisions taken for others.”

The concept of experimental finance will turn some readers off, since there is a debate about whether the results of such experiments are applicable to decision making in the real world.  (For example, some foundational elements of behavioral finance have been challenged on that basis.)  And the small stakes at play in the experiments seem laughable in comparison to the compensation available to real money managers.

Nevertheless, many of the questions asked in the studies are worthy of consideration by the leaders of asset management organizations who would like to think about whether actions taken by portfolio managers fit with the expectations of those “other people” whose money they manage.

Social distance

One question of interest is the degree of contact that exists between a principal and an agent, and how that affects decision making.  While some professionals (such as investment advisors) might have fairly regular contact with their clients, asset managers are often at great remove from the owners of the assets they manage.

Do their decision processes (and therefore the resulting portfolios) differ based upon the degree of interaction with clients, even when the same investment guidelines are in force?  For example, does managing an institutional separate account portfolio for which the portfolio manager has periodic review meetings directly with the client — or a mutual fund subadvisory relationship with a similar degree of interaction — lead to different choices than if there is no interpersonal communication with a client?

In other words, is there a “social distance effect”?  Given the way that investments are often marketed — selling what is billed as the same strategy, within different kinds of vehicles to different kinds of clients who have different levels of access — the discrepancies among them may give clues to whether the extent of communication is a factor in portfolio composition.  (Since some vehicles may have cash flows in or out and others not, that can provide further insights as to preferences at the margin by a portfolio manager.)

Incentive structures

Across those vehicles said to be using the same strategy, there might be a mix of incentive structures.  When that’s the case, what can be learned by studying the choices made in the respective portfolios?

Comparisons between strategies that aren’t advertised as being the same (but which have a manager in common) can provide evidence as to relative choices as well.  Which of those choices make sense based upon the actual differences in mandates — and which stem from incentives or unrelated preferences (such as favoring one portfolio over another for some reason)?

Risk choices

In seeking to evaluate decision making on behalf of others, academicians face a problem:

“Most theoretical and empirical work on incentives focuses on effort choices . . . while investment management is about risk choices.”  (Füllbrunn, et al.)

Therefore, studies about managing other people’s money try to compare the riskiness of choices made by investment professionals for themselves versus those made for others.  The body of evidence is mixed — and fragmentary.

There have been empirical studies that looked at how life events affect performance, and in a previous posting, Paul Marshall was quoted as saying that “the reddest flags for underperformance . . . are problems in people’s personal lives — the three Ds of death, divorce and disease.”

Age is another factor — how does risk-taking change over time?  Does that scrappy young manager become less scrappy?  How about wealth?  There are those who keep playing the game the same way even though they have a lot of chips, some who feel free to take even more risk, and others who start to take risk management more seriously than they had before.

And then there is this:

The central truth of the investment business is that investment behavior is driven by career risk.  In the professional investment business we are all agents, managing other people’s money.  The prime directive, as Keynes knew so well, is first and last to keep your job.  To do this, he explained that you must never, ever be wrong on your own.  To prevent this calamity, professional investors pay ruthless attention to what other investors in general are doing.

That is the start of an oft-quoted 2012 piece from Jeremy Grantham of GMO, whose firm has been on the wrong side of the herding effect more than once (and for long periods), resulting in substantial declines in assets under management (thus proving the point).

Clients

To return to the notion of proximity to clients:  They often are their own worst enemies and can exert pressure on portfolio managers at just the wrong time.

Their expectations might be out of whack because the marketing effort to win their business misrepresented what they would get — or they just didn’t understand it.  As a result, a mutual fund shareholder may decide to sell if a fund is underperforming, while an asset owner may put a firm on a “watch list,” as if to signal to the manager that they should try harder.

Whether the pressure comes from outflows or suasion, a manager may feel compelled to respond to it.  There are two sides to the cycle, of course; a portfolio manager gets lauded and encouraged to keep up what he is doing when things are going well, and then is pressed to change when the the flip side occurs.  Each phase of client pressure spawns procyclical behavior rather than clear-eyed assessments.

According to Füllbrunn, et al., “Traditional assumptions in economics . . . do not allow for other-regarding preferences.”  They also don’t account for the fact that those preferences migrate, making all of this even more difficult to figure out.

To analyze — or not?

If the goal of an asset management firm is to provide attractive performance that meets the expectations of its clients, then the impediments to that happening should be top of mind for the leaders of an organization.

Given the near-universal pull of increased assets under management, marketing-induced expectation gaps likely have the biggest impact.  But which of the other factors mentioned above are important determinants of inconsistent approaches by portfolio managers?

The information needed to examine the question is buried within asset management firms themselves.  For example, think of those large multi-product firms that have been around for a time; they are sitting on a mountain of behavioral information.

But they may be uninterested in it.  Philosophically, many leaders think that portfolio managers should be left alone to live and die on their performance — as long as they don’t violate mandate parameters.

Other firms are more proactive in evaluating speculative questions like these, trying to determine whether there are factors that are interfering with investment judgment when decisions are being made.

Which do you think is the right approach?  Why?