Commentary: Tackling the identity crisis in institutional investing

Many investors struggle to gain an accurate, comprehensive view of "who they are" in terms of their organizational capabilities, and the inherent constraints on those capabilities; yet it's those same capabilities and constraints that are the foundational ingredients of their identities.

It's hard to overstate the severity of this identity crisis. As a casual proof, consider one of the most central questions in institutional investing: What's the key driver of long-term investment returns? Many claim it's the act of portfolio construction (that is, the choice of which asset classes to invest in, and the selection of particular assets within those classes). Others say it's the act of portfolio management (i.e., the choice of when to buy and sell specific assets). Still others argue that risk management is most important. While all of these activities do indeed matter for long-run performance, our multidecade exploration into the behavior of institutional investors finds a different explanation: An investor's long-run performance is dictated by the extent to which its approach to portfolio construction, portfolio management and risk management (in industry parlance, its "model") matches its organizational identity. Put slightly differently, it's the business model — the combination of all of these factors — that drives the investment performance of a fund. And yet, most investors struggle to understand their models and whether they are even appropriate.

Many investors try to adopt models that other funds have famously used to achieve long-term outperformance. Examples of these "role models" include the Yale model, the Norway model and the Canadian model, among others. Sadly, these attempted adoptions frequently fail for the simple fact that institutional investors are all snowflakes: no two are identical in terms of their capabilities and constraints, and so copying a business model from a peer is unlikely to produce the same outcomes. The model that made Canadian pension funds a household name can readily fail under a slightly different set of circumstances. This isn't to say that an existing model can't be adapted to fit a particular investor's identity; in many cases that's very much possible. The problem lies in identifying what needs to be adapted to achieve a good fit. Historically, that identification has been challenging because institutional investors have lacked a mirror with which to inspect their own identities.

In a recently released paper ("The Investor Identity': The Ultimate Driver of Returns") we've attempted to build the missing mirror, by way of a framework for investors to analyze their own capabilities and constraints in order to clarify how those characteristics map to suitable models. Our framework rests on a straightforward (and, we think, poetic) observation that's arisen from many years of closely studying institutional investors worldwide: All investors are snowflakes, but all snowflakes are made of water. Said differently, all investors share the same "production function" for creating investment returns because all returns generation boils down to the nature of four "inputs" and how they're combined:

• Capital: Unsurprisingly, investors need capital to produce any returns at all. But not all capital is the same. Pension fund capital, for instance, comes with an obligation to pay retiree benefits at prescribed future dates, whereas the capital of many sovereign wealth funds comes with an obligation to invest some fraction of it domestically to promote economic development.
• People: Despite rapid improvements in artificial intelligence, there's still a need for many investment decisions to be made by humans, and the quality of those decisions is a direct result of collective expertise, fastidiousness and personal networks.
• Process: To stave off chaos, decision-making in organizations requires infrastructure to allow it to be coordinated, coherent and consistent. This infrastructure involves things like organizational structure, due-diligence checklists, prescribed valuation methods and rules for benchmarking performance, risk budgets and more. We refer to this infrastructure and how it works as an investor's process for decision-making.
• Information: In financial markets, sound decision-making relies on fresh and accurate information. But there are other characteristics of information that also affect decision quality, including the granularity, novelty and accessibility of information by others.

The quality of these inputs, and the manner in which they're combined, are the key determinants of an investment organization's long-term performance, and they're also core features of every investor's particular identity. But they aren't the only features; in producing investment returns, investors also depend on a trio of enablers — governance, culture and technology — to shape and support the four inputs. By analogy, if the four inputs are like personality or physical traits, then enablers are like tools for improving (or at least maintaining) those traits, with things like fitness regimens, education or improv classes — all of which get melded into a person's identity via the experiences they create. Figure 2 represents the layers of an investor's identity, with outer layers being more changeable than inner ones.

Notice that, inside the input layer, there's yet another facet of investor identity: the thumbprint. This layer of identity is effectively unchangeable, and includes existential characteristics, such as geography and organizational type, for example, a state-level pension fund in the Midwestern U.S. or a sovereign wealth fund in the Middle East. Generally, thumbprints come with constraints on what an investor can and cannot do. Yet, thumbprints shouldn't just be seen as obstacles: being a county-level pension fund in Silicon Valley can bestow unique access to venture capital funds; and the A$70 billion ($46 billion) Cbus Super, an Australian pension fund for construction workers, has used the specialist knowledge of its board members (many of whom come from the construction industry) to build a best-in-class property investment program. In short, thumbprints — and identities in general — can be parlayed into competitive advantages, which is a key reason why investors should strive to fit models to their identities, rather than the other way around. Another (related) reason for investors to pick models that square with their identities (as opposed to trying to change their identities to conform with a specific model) is that every model is like a combination lock: it's got a specific set of inputs and enablers (and in some cases, thumbprints) that are needed to make it work. But if any one of these is missing, the model won't "unlock" for the investor, no matter how excellent its other inputs and enablers are.

For role models, one or two focal inputs or enablers are the linchpins of the model, and it's these linchpins that get accentuated when the model is referred to by other investors. Yet, in each instance, all other inputs and enablers must be "just so" in order to support those linchpins, and make the model work. Moreover, many investors treat the set of role models as an exhaustive and mutually exclusive menu of models to be emulated — as if these are the only worthwhile "career paths" available. This is partly understandable, because these models have demonstrably performed well in the past, albeit through configurations of inputs and enablers that might be difficult (if not impossible) for other investors to replicate. For most investors, it's best to use role models as inspiration in designing their own form-fit models, rather than using role models as paint-by-the-numbers recipes to be followed exactly.

A final complication in the investor identity crisis is the presumption that role models are static, and don't change or merge over time. The fact is, the investors who pioneered these role models are in part so successful because they haven't stuck doggedly to their models: For example, at times the Yale endowment embraces components of the Collaborative model, and Australian superannuation funds sometimes emulate parts of the Norway model. By analogy, it can be a good idea for funds to look for hybridizations, by selecting elements of known models that can be fused together (and that jibe with their identity). But it's vital to make sure these elements are actually compatible: It'd be mighty tricky to be both a rock star and a surgeon (which would give an altogether new meaning to the phrase "rock-star surgeon"); but it's altogether plausible to be both a scientist and astronaut — in some instances, being both may be superior to being merely one or the other.

Yet, no matter what model an investor chooses — whether a hybrid of existing role models or something altogether new — there's an inalienable need to have that model align with its identity. Otherwise, the investor is flirting with failure. And that failure might not come straight away; it might occur when some future shock hits. That's one of the most pernicious things about the identity crisis: its gravity may become clear only during other types of crisis (e.g., a fireball of rising inflation and cross-market drawdowns… like now). But that's a negative surprise that's avoidable, so long as investors are willing to take a long look in the mirror and deeply analyze their inputs, enablers and thumbprints. We hope they make use of our framework to do just that.

Ashby Monk is the executive and research director of the Stanford Research Initiative on Long-Term Investing and Dane Rook is a research engineer, also at Stanford. Both authors are based in Palo Alto, Calif. 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.