Finding the Right Speed

Across the investment ecosystem, the pace at which decisions are made and implemented varies widely.  That raises the question of what the optimal speed should be for any particular investment activity.

Which decisions ought to happen quickly and which ones should be allowed to — or required to — play out over longer periods of time?

The human element

One factor to consider:  By nature, people operate at different speeds.  Some are quick in their responses to developments, at the ready with ideas on how to proceed and prepared to pounce into action.  Others take time to process.

For example, Jim Simons stressed that when people are interviewed for jobs, “there’s too much emphasis on a person’s being able to answer questions quickly.”  He relayed a story of overruling others at his firm who didn’t want to hire someone who “gave slow answers.”  (When hired, that person turned out to be “very good indeed.”)  Then Simons — who was an expert mathematician and one of the greatest investors of all time — said, “I would never have done well at a math competition.  I’m not an extremely fast thinker myself; I just work hard.”  Imagine not tapping into his expertise because you’re in a rush.

Knowing the processing speeds of those involved can help you think about the best time frame for a particular kind of decision.

Organizational impediments

Going too fast can be an issue, but so can going too slow.  Especially in bigger organizations the wheels of decision making can get bogged down.

To avoid that, The Friction Project, a book by Robert Sutton and Huggy Rao emphasizes making “the right things easier and the wrong things harder.”  That means first determining what those right and wrong things are before applying the kinds of tactical recommendations the authors suggest on how to add or reduce friction where needed.

While investment organizations are thought to be less bureaucratic than other kinds of organizations (which is not necessarily the case), one particular area of attention should be the layers of oversight that exist.  A decision that requires a series of approvals is slower — and not necessarily better.

Slowing down time

The environment for investment decision making is problematic, in that the constant hum of the markets and the seeming need to be at the ready with hot takes on the events of the day can distract individuals and organizations from their ultimate mission.

A 2021 paper from NZS Capital, “Slowing Down Time in Organizations,” considered that problem by way of an evaluation of networks, culture, and organizational structure.  Some excerpts from it:

How is it that organizations spend so much time on mission, vision, and culture, yet most employees experience chaos in the day-to-day environment?

At NZS, we’re somewhat obsessed by the idea of time dilation.  More specifically, how can we slow down time compared to everyone else running around as if their hair is on fire?

Through understanding the key network governors – nodes, style bias, and connections – companies can architect structures that provide time for creativity, innovation, and thoughtful decision making.

One size does not fit all, however.  The paper includes sections on different kinds of organizations.

The goal should be to have the right balance of decision quality and decision speed, and that varies by investment philosophy and mandate.  In general, if you’re always deferring decisions in want of more information, opportunity will pass you by, while being too quick on the draw can lead to different kinds of errors.  While not specific to investments, a posting from Trigger Strategy (“How much research is just enough research?  How much is too much?”) outlines and illustrates the point.

Examples

Some situations from across the landscape of the markets where these ideas come into play:

~ Most research on public equities shows that being early on a stock is better, since alpha decays over time (except perhaps for those luminous compounders that seem to go on and on).  On the contrary, analysts covering stocks often ease into ideas, raising recommendations over time as they get increasingly comfortable — just as many portfolio managers will initiate a starter position before becoming more aggressive.  (Even more contrary, there are some firms who say that a required part of their process is studying a prospective holding for a given length of time — even more than a year — before buying.)

~ One aspect of the Dimensional Fund Advisors investment approach is to avoid the trap of using a pre-defined trading game plan that has you transacting at times when everyone else does.  For example, a recent paper from the firm shows the benefits of “avoiding the costs of demanding immediacy” when trading in response to index reconstitutions.

~ Three decades ago, when processing capabilities weren’t what they are today, most quantitative managers traded on a monthly basis.  Some still do.  The general principle involves comparing the quality and frequency of the decision signals to the cost of transacting.  (A footnote in a recent Cliff Asness paper said that “the move from minutes-old information to nanosecond-old information in the last 30 years is the height of irrelevancy for low-turnover strategies.”)

~ A key reason for the current popularity of the OCIO model with asset owners is that the traditional approach of using a consultant for manager selection could lead to six or nine months (or even more) elapsing from the initial discussions of a change in strategy to its implementation.  In most cases, OCIOs have the discretion to make manager changes within the current asset allocation structure, virtually eliminating that delay.

~ Speaking of manager selection, allocators of all types tend to look at three-to-five year windows of past performance for public strategies, both when initially selecting a manager and when deciding whether to retain it.  Unfortunately, most studies show that time frame is a poor decision window.  (A similar issue plays out in a different way in private markets.)

~ In the most recent upcycle in venture capital, some investors changed the game by aggressively investing in new deals with little or no due diligence, no requirements for a board seat, etc.  The increased cadence allowed money to be put to work quickly and early success led to others copying that approach (and arguably intensifying the blowoff top).  Given the disappointing results in venture since then, that speedy strategy doesn’t look as wise as it once did.

In every situation, searching for the right speed is important, as is considering when things ought to be sped up or slowed down from normal — or if they should be varied at all.  Schematics of investment process rarely show how quickly things are supposed to happen.  An internal exercise to describe what normal looks like — and what kind of variability is to be expected (and under what conditions) — can unveil differences of opinion that lead to a more clear examination of beliefs and operating procedures.

Faster and faster

With each passing decade, the computer age has changed the availability of information and the nature of investment practice.  We need to pause and think about the effects of that “sheer speed-up” — a term which Marshall McLuhan used in 1964! — and assess its implications, just as he tried to do back then.

A 2013 posting in the original Research Puzzle blog said:

It might seem that speed is always of the essence, even though in today’s world you’re not likely to add much value through speed.  You are propelled along at the pace of the business, whether that results in good decisions or not.

For the leaders of investment organizations, those realities raise a question:  Does your strategy really require you to play the game as everyone else does, to run at the same speed at which the market runs?  The chances are parts of your process are vestiges of a time gone by.  Rethinking them could be liberating and fruitful.

The need for such rethinking has intensified since then.  Taking a broader perspective, a recent essay by Rory Sutherland (derived from a Nudgestock presentation) asks the question, “Are We Too Impatient to Be Intelligent?”

The general assumption driven by these optimization models is always that faster is better.  I think there are things we need to deliberately and consciously slow down for our own sanity and for our own productivity.  If we don’t ask that question about what those things are, I think we’ll get things terribly, terribly wrong.

Given the advent of artificial intelligence applications, we not only need to ask which tasks are best done by machines and which ones are best done by humans, we must also judge how fast we should try to make decisions using that combination.