Social Forces and Sell-Side Analysts

Sell-side analysts have been the subject of much academic and practitioner research over the years.  Certain parts of their work — earnings estimates, target prices, and recommendations — are easy to track, leading to conclusions about whether individual analysts, groups of them, or analysts overall add value by means of those outputs.

Left out of such reviews are the other aspects of an analyst’s job, which surveys show are more highly valued by institutional investors — and the social forces that affect their actions.

The last posting in this series reviewed a paper by Millo, Spence, and Valentine about beliefs in active management during an era defined by the rise of passive management.  Another paper from the authors, “The Field of Investment Advice: The Social Forces that Govern Equity Analysts,” is the topic now at hand.

The same set of interviews was used as the basis for this paper as for the previous one.  Of note is that James Valentine, one of the authors,

had spent a considerable amount of time in the field as a practitioner and continued to pursue commercial interests in that space while simultaneously holding an academic position.  This represented both an opportunity and a challenge for the data analysis.

Valentine (the author of Best Practices for Equity Research Analysts, reviewed here) was able to help with “the jargon and tone of respondents,” but his experience presented a challenge in that the goal of such studies is to glean ideas from the research subjects, not from a member of the research team.  Often academic research misses important considerations in interpreting results because of a lack of direct knowledge of how things work.  Having a practitioner involved can minimize that risk, but the potential pitfalls of doing so must be mitigated.

This paper is representative of the other social science research examples that have been used in this ongoing series of postings.  Each of the studies has gotten to the heart of the particular investment subculture being studied, in the process revealing implications that go beyond the narrow confines of those subcultures, since “the way things are done” within them affects investment discourse and practice in general.

A conundrum

The abstract for the paper indicates that sell-side analysts present “a conundrum,” in that they “are seen as influential market participants, yet researchers widely criticize them for their bias and inaccuracy.”  Studies that focus on easy-to-track estimates and ratings not only leave out the other activities that investors prize but ignore the social forces that shape analysts’ work.

“Webs of relationships” characterize the environments within which analysts operate, including those with the companies that they analyze, the buy-side investors who tap them for information (and “concierge services”), others in the firms at which they work, and their analyst counterparts at other firms, with whom they compete.  The dynamics of those relationships are overlooked in most evaluations of analysts, which thereby “underappreciate the social sensitivities and social pressures that analysts have to navigate on a recurring basis.”

Among the “other” responsibilities of analysts (in addition to writing reports that contain their predictions):  keeping the lines of communication open with the company managements that they follow; arranging for access to those managements by buy-side investors (at conferences and smaller private meetings); serving as a go-to source of information about the companies and industries they follow; developing and maintaining relationships with buy-side analysts and portfolio managers; and supporting their firm’s corporate finance and trading businesses (in more of a wink-and-nod way than used to be the case).  Plus, the work product of an analyst is “often used ceremoniously” by companies and the analyst’s own firm for promotional reasons.  In short, the assessments of covered companies are affected by social considerations as well as by financial analysis.

Field theory

The authors use field theory in their diagnosis, which “expands the motivations relevant to actors’ behaviour beyond those associated directly with economic utility maximization.”  The theory also recognizes the need for actors “to maintain and strengthen the social order,” even though they are in competition with others within the field.  Everyone involved is aware of the “rules of the game.”

Patterns of behavior are not just economic, but take into account “the whole structure and history of the surrounding field,” resulting in two important insights related to the work of analysts:

First, [field theory] expands the repertoire of relevant social skills in the field of investment advice and implies that sell-side analysts, as actors in a social field, aim to improve their situation using all resources they perceive they have at their disposal; economic resources as well as social (e.g. connections) and cultural (e.g. skills, expertise, professional authority) resources.  Second, the “rules of the game” that prevail in a given field are shaped and learned gradually through recurring interactions and are the product of experience, habit, and routine which, to some extent, evade conscious consideration as they become part of the taken-for-granted worldviews of these actors.  This habituation to the implied rules of the game is also expressed in the belief that the established social order in the field represents an objective and natural truth; that the way things are is the way they should be.

The notion that “the way things are is the way they should be” is not limited to the world of sell-side analysts, although they serve as a good example of a broader principle.  “The interdependencies and relationships that have built up over time between sell-side analysts and buy-side actors that may have become habitual, taken-for-granted features of investment decision making” have close cousins in other parts of the ecosystem.  (What are the taken-for-granted features in your realm?)

The rules of the game

The environment in which sell-side analysts operate is characterized by social inertia, a product of the “inter-personal and inter-organizational interdependencies that maintain the structure of the field, despite regulatory and economic changes aimed at disrupting these.”  The social order is protected in a variety of ways:

~ Relationships are valuable and analysts “hone their trade through the cementing of social ties over time.”  Therefore,

maintaining the social order may be so important for actors that they would protect their social ties even if this would imply rejecting potentially innovative ideas or new opinions, as the actors’ worldviews are embedded into the existing social structure of the field.

~ “The buy-side is slow to move from an existing sell-side relationship and there is not enough bandwidth to constantly cultivate new relationships.”

~ “The continued existence of social ties between different financial intermediaries was evoked as a reason to explain the persistence of underperforming sell-side analysts in the marketplace.”

~ “Members of buy-side firms reinforced the view that buy-side/sell-side relationships outlive their useful economic lives and that payments for sell-side services do not rely solely on the quality of their analysts’ output.”  (At some buy-side firms, people weren’t even sure how the payments to firms were calculated or how their assessments of analysts factored in.)

~ Buy-side interviewees mentioned the need to have good relationships with analysts at firms that can provide deal flow; the effect of “reciprocity generated by gifts and entertainment;” the “cultural matching” power of friendships that develop; and even sympathy for someone who is a “nice guy” or has a family to support.  None of these would be a part of an objective assessment of an analyst’s forecasting ability.

~ Another factor:  bigger firms need to field a large analyst staff that covers a broad universe, at times “privileging coverage over quality.”

Consensus estimates

The paper uses earnings estimates as a way to illustrate the competitive relationships among analysts.  The dynamic among them is similar to that found in other fields, described in this way by a sociology book cited by the authors:

Actors make moves and other actors have to interpret them, consider their options, and act in response.

Because data platforms publish “consensus” earnings estimates, which average the views of all of the analysts covering a company, individual analysts can “position themselves vis-à-vis an aggregated version of views in the wider field.”  That leads to patterns of convergence and divergence:

Consensus numbers and the positioning practices that surround them are thus indicative of a social and informational infrastructure that may help provide a richer explanation for phenomena such as herding, or conversely of boldness whereby some analysts try to diverge from the herd in abrupt ways.

Consensus estimates serve as a benchmark, framing “how actors perceive the field and how they differentiate between different categories of actors.”  The attention paid to the consensus “has the unintended consequence of amplifying the impact of the average consensus number and the opinions associated with it.”  In practice, “the consensus number serves as an infrastructure for developing opinions rather than a visible building block in the discourse,” embedded and weighty in the investment processes of most practitioners.

The consensus also allows those willing to stray from the crowd to be easily identified, while everyone else blends in.  The authors describe the stay-safe career calculus that causes a clustering around the consensus as fulfilling “the normative demand from analysts to avoid erroneous predictions,” since being wrong when you’re with the consensus is not the same as being wrong when you are away from it.

While clinging to consensus is the norm, “in a social field that rewards alpha generation and distinctiveness, heterogeneity was extolled as a virtue by actors who saw diverging from consensus numbers as beneficial.”  This disparity in motivations and actions among analysts — the desire for most to converge to consensus and a few to diverge from it — is missed in analyses that ignore the social element and gameplaying inherent in analysts’ roles.  (A similar dynamic exists among those on the buy-side; see the “Real active or faux active?” section in the previous posting.)

According to the authors:

The diverging patterns of reactions to consensus numbers may be understood as different interpretations of the rules of the game in the field.  The differentiating factor is the type of resources the followers and challengers assume they have.  In other words, actors work within the realm of what they perceive as possible, having internalized the structures of the surrounding field.  Our research adds a deeper understanding to existing findings in the analyst literature that indicate sell-side analysts who have a higher status or are more courageous and want to improve their situation by gaining more attention are more likely to challenge the consensus.  In contrast, others who see the best course of action as not attracting attention, tend to comply with these numbers.

(Another aspect of the estimate game is summarized in a Bloomberg article about the “overly conservative guidance” companies provide regarding earnings, so that they can be credited with “beating” consensus.  “That is how the game is played,” said the researcher who authored the study the article was based on.)

Broader implications

Sell-side analysts are key players in the investment chain.  Despite perceptions that they are biased and that their predictions are inaccurate, their influence is undeniable and multifaceted.

Consider just the earnings estimates that those analysts produce.  The aggregated consensus estimates are the core feature of much investment discourse.  Buy-side analysts and portfolio managers are constantly forming their own estimates and judging the work of others based upon the consensus and the array of estimates around it.  (As the paper indicates, generalist analysts are prone to relying on the consensus more than specialists.)  In addition, all kinds of quantitative strategies work off of the spread of estimates versus consensus, the migration of the consensus over time, earnings surprises away from consensus, the patterns of individual analyst predictions versus the group, etc.

The estimates are products of a complex social environment.  Despite the inherent competition in their world, sell-side analysts and those who interact with them generally “resist change and stick to existing practices,” those rules of the game ingrained in that web of relationships.  What seems to outsiders to be a rational economic endeavor is something much more.