Quantitative investment management has been around for quite a long time.  Three decades ago, changes were often made at a monthly frequency, based upon attributes identified by leading academics and developed into strategies by new units set up at Wall Street firms.

Today, according to Bank of America Merrill Lynch, “A seismic shift in assets and resources toward data-driven, systematic strategies and shorter-term investment strategies, which tend to rely on access to better, faster and larger stores of data is underway.”  The quant rush is definitely on.

There are a host of questions for investors to consider, some of which are outlined below.  Most of them have been covered in extensive depth elsewhere; this, then, is not meant to be a comprehensive list of questions or a detailed analysis of any of them.  Instead, it represents what seem to be some open questions of importance, based upon where we are right now and how quantitative investing has been embraced in theory and practice.

So what is this “quantitative investing”?  We might as well start off with a hard question.  In fact, it’s a lot of different things.  At one level, it’s applying fundamental analysis in a systematic framework (that’s what was done in the old days).  But factor investing has really become the dominant quantitative approach, high-frequency trading falls into the broad category too, and there are the emergent areas of artificial intelligence and machine learning.  As with a lot of investment classifications (think “hedge funds”), the name often is applied too broadly in an attempt to lump things together (as I am admittedly doing here).  Precise definition and categorization can especially be lacking once the specialists give way to the marketers, who try to simplify an idea for asset owners.  A lot can be lost in the wash.

What is the “science” behind quantitative investing?  This is the most important question, since the sellers of financial products try to make them sound scientific and the buyers generally fall for the pitch.  (For an entertaining look at the problem, check out this video.YouTube | Posted by Mathematical Investor.)  Backtesting is dressed up as proof of concept, even though it’s notAlpha Architect | This piece is called, “The Future for Factor Investing May Be Different Than its Backtested Past.” and major errors in backtests have been discoveredAdvisorHub | Transamerica is the latest to be nabbed; F-Squared the most famous. (and probably many more are out there that haven’t been).  Despite all that, it is especially easy to be swayed by the visual evidence of the “beautifully monotonic” (to use the words in one quantitative presentation I was at) results of a strategy.Google | Here is a random but representative example (not the one I heard) off of Google Images.  Despite the fact that many math and physics PhDs have come into the investment realm, finance remains a social science.SSRN | The mismatch is pointed out in this brief Journal of Portfolio Management piece.

What does that mean in practice?  The results that you are shown may or may not be representative of the future in direction or degree.  Which is not to say that there might not be some value to them; they just can’t be taken as having a high degree of certainty.AQR | Notably, “there is much less statistical evidence of long-horizon return predictability than implied by existing research.  Investor behavior isn’t stable (just like human behavior isn’t stableMarket Fox | Here’s an analogy between psychological traits on investment factors.), investment characteristics (returns, volatilities, correlations, growth rates, discount rates, etc.) migrate over time, and investment “edges” of a particular kind tend to get worn down, especially as assets balloon, as they have in many quantitative strategies.  Plus, despite the fact that we think we have a lot of data at our disposal, we’ve seen precisely one course of history out of many possible ones.  Others can have much different characteristics.

But there are some legendary quant funds that have seemed to generate good returns on a regular basis, right?  That’s true, although they might not let you into the fund with the great numbers; you can get into one with lower returns (but lots of potential conflicts).Wall Street Journal | This article covers the topic.  Plus, on what basis will you make your choice?  Performance alone?

Does it make sense that different standards of due diligence are applied to quantitative investing versus other strategies?  I ask this question of attendees at my due diligence and manager selection workshops.tjb research | This page has the information on the next workshop.  In many cases there is no independent analysis/verification/attribution of a strategy, since important aspects of models/algorithms aren’t shared for proprietary reasons.  That has kept some investors away, but many view that as a worthwhile trade-off (as long as performance stays good).

What are the ramifications for the fiduciary responsibility of decision makers given the existence of those different standards of due diligence?  This is a question that has received very little attention, and the lack of discourse is troubling.  If understanding is the goal of due diligence in the process of strategy and manager selection, what are the ultimate implications of the lack of transparency that often exists?

Are manager research analysts prepared to deal with the nuances of quantitative management?  Probably not.  I know I’m not.  I found that out when observing a discussion among the quantitatively astute; the acronyms and jargon were flying fast and I was lost.  A behavioral trap among those doing due diligence (just as it is with all humans):  pretending to understand something when they don’t, because they don’t what to demonstrate their ignorance.  There are basic questions that you can ask,Vanguard | This is “a quick guide to evaluating back-tested strategies.” but I now think that if you’re investing in quantitative products, you need someone who not only speaks the language but understands the issues in great detail.  That’s not the case for many today.

Are some firms actually downplaying the degree to which their investment processes are quantitative in nature?  Someone mentioned this the other day; I never would have thought of it.  Essentially, the idea is that legal and regulatory risks are greater for quantitative strategies because they are relying on models or algorithms that are supposed to do certain things — and if errors are discovered, there’s a price to pay (whereas bad discretionary results can be chalked up to poor choices).  But if some managers are making their processes sound more discretionary and less quantitative than they actually are, then it’s the responsibility of the person doing due diligence to identify those gaps and judge accordingly.

Where can I go for help?  Those old enough to remember the Good Housekeeping Seal of ApprovalNew York Times | It’s actually still around, although it doesn’t have the same cachet as it did when it was a common advertising hook. might wish for an independent arbiter of product value.  Or perhaps something like Cochrane reviews,Chochrane | Here is a general description of the information found on the site. a crowd-sourced repository of information for health decisions, (apparently) without financial conflicts.  But there’s really nothing like that out there, so you are reliant on consultants and gatekeepers for assistance.  Then the question becomes whether they have the expertise to get at the finer aspects of these strategies (or the due diligence techniques to make the discoveries that you would expect them to make).

Are you saying to stay away?  Not at all.  But an organization allocating assets to these strategies ought to have thought deeply about the questions on the table.

Is there a future for me if I’m a fundamental analyst or portfolio manager?  I’m reminded of a section of a Michael Lewis profile of Tom WolfeVanity Fair | Two very different but gifted writers. (regarding astronauts supplanting test pilots as heroes in The Right Stuff):  “Process had replaced courage.  Engineers had replaced warriors.  A great romantic way of life, a chivalric code, had been trampled by modernity.  Not for the first time!”  It must feel to you like the torch has been passed and your livelihood is at risk.  It is, to an extent.  The game has changed, although not completely.  There will be fewer jobs like the one you have today and professionals will have to reinvent themselves and discover new ways of doing their work to stay relevant.

Are you done yet?  For now.