You’re probably not a long-term investor. But that’s okay.

At a recent conference, a speaker defined themselves as a long-term investor with absolute certainty to the audience. As the sole voice of dissent, I thought I would risk my invitation by challenging the scientific basis of that assertion. In other words: prove it.

The ensuing conversation was uncomfortable to say the least. There are few topics that elicit such a strong sense of self-affirmation, but equally provide so little in the way of hard evidence as to what is a long-term investor. After all, long term comes from the investment horizon of the retirees/citizens, so it feels natural to adopt it as a centre piece for the investment process. But this is a misleading assumption; the investment agent (asset owner) is hired to optimise returns for their clients, through matching the liability structure and focusing their efforts on the information edge (‘skill’) they can access and deploy to produce returns. How they build their ‘house’ of returns, either brick-by-brick in a gradual methodical way, or through large, spanning metal beams is up to the architect, as both ways can achieve the outcome.

With this niggling observation, I undertook some effort to search for empirical proof or definition for what it is to be a long-term investor. And this is what emerged.

Defining ‘long term’ was the first challenge. The numbers ranged from one year to around five years, to multiple decades and a few that couldn’t quantify it at all. I’ve settled on ten years as a useful benchmark and this aligns to the strategic asset allocation benchmarks that most asset owners manage to. This is still a very short time relative to ideas of ‘equilibrium’ returns to risk premia (such as equities) which are typically measured in fifty years and more. Already I began to see how difficult this challenge would be.

I needed to find a mechanism to uniquely discern a long-term investor from all the others, without reference to their self-identification. In other words, objective proof. I used the idea of ‘signalling theory’ where an agent is able to uniquely identify itself by creating an action that is not replicated by others. If I am an excellent student, I can signal that by acing my exams, something that an average student is not able to do. In other words, proof lies in the actions that other investors can-not mimic. Therein lies the opportunity to add value, for example, where others can’t.

The first red herring I found was the holding period.

Long-term holding period is not a defining trait of a long-run investor even though its often mistaken for one. For public securities with liquid prices, most investors will adjust their holdings based on the risk and valuation of the securities, which means that they will buy or sell based on changes to these measurements. If prices move rapidly and extremely, investors will buy or sell, even if their intention had been for the long-term. This is because, even in the presence of physical cash-flows like dividends, expected returns are still largely contingent on changes in prices as the main driver of value, and so investment horizon is endogenous, and not fixed. It is worthwhile reflecting on this point. I can hold a factor in my portfolio, such as value, for the long-run, all the meanwhile buying and selling stocks in my portfolios at a very high frequency. I am holding the factor based on a long-term fundamental conviction, but judging by my turnover numbers only, you may be forgiven for judging me something else. In specific market environments, there are many examples where rational, long-run investors will buy or sell within a short period of time the same position because of market pricing and a significant change in market conditions. Restricting the holding periods and turnover artificially is only a constraint, not an enabler for asset owners.

If not holding period, then where does the signalling lie? Perhaps it is in the nature of the information that I pay attention to, focusing on long-term trends, and ignoring short-term fluctuation? The convergence of these information sets is where long-run investors live, so maybe I would need to investigate how forecasting is undertaken differently based on the investment horizon?

Geoff Warren from Australian National University notes that it’s really a mindset of the investor and the kind of information that they focus on. It certainly sounds credible, though if that mindset leads to the same actions as a short-term investor, then it still fails to meet the signalling criteria I need. Long-term investors need to perform actions that liquidity-bound short-term investors can-not, such as providing liquidity during market sell-offs, and stepping in to provide capital for illiquid long-run projects. Locking up capital for long-run projects is certainly an action that is hard to replicate, but practically, constitutes a very small part of their overall portfolio by weight.

Puzzled, I have failed to find empirical proof (so far). But it also occurred to me that the narrative of the long-term investor was largely unnecessary to serve clients interests. If you could demonstrate investor skill making short-term allocations, avoiding pitfalls, and making small but positive incremental improvements, you could build successfully to a long-term outcome. It does not need to be ‘grand’ in its ambitions to discover the future of industries, and global economies, it can be as simple as making the best decision for this month. While our clients are long-term savers, our role as fiduciary and expert investors may be best fulfilled as using whatever skill and means we have to build to that long-term target.

Leave a comment