In his most recent missive, “Investing Without People”, Oaktree’s Howard Marks takes on the subject of quantitative investing, albeit with “trepidation,” because it’s something he admits he is just learning about. We always look forward to reading Mr. Marks’ memos, as they are packed with valuable insights from an unquestionably venerable career. I for one owe him thanks for the investing knowledge he has imparted to me over the years; I’m certain I’m the better for it.
But in this case, the questions/objections that he poses are all too familiar, and in my opinion, exhibit faulty logic. Let’s examine some of the major points.
- “Most quantitative investing is a matter of taking advantage of standard patterns (the factors that have been correlated with outperformance) and normal relationships (like the usual ratio of one stock’s price to another’s or to the market).”
This is absolutely true. But do discretionary managers behave any differently?
Having sat on the manager due diligence side of things for much of my career, I can say without question that all investors attempt to identify factors correlated with outperformance. Discretionary managers may look for companies with high ROIC, low debt/equity, low price to cash flow, or long tenured management teams, because they believe (and hopefully have done the work to confirm) that those characteristics offer excess risk-adjusted returns over time.
Likewise, strategists appear daily on CNBC pontificating about such things as the normal relationship between say dividend yields on stocks and yields on 10-year Treasuries, under the competing asset theory (however misguided it is to compare two assets without regard to risk). Or they may reference the fact that a stock is trading well below its historic average P/E relative to the market.
Investing is clearly about patterns and relationships, and quantitative managers can do exactly the same things that discretionary managers can, but with greater efficiency and with less bias in their approach. A computer doesn’t act differently before its first cup of coffee after a sleepless night. It doesn’t allow a big loss on a stock to influence how it looks at the next stock. It won’t dismiss low ROIC “just this once” because it has a feeling about how management might turn that around. It simply searches for the factors that investment professionals have programmed it to look for, and it does so tirelessly and consistently.
And those investment professionals who programmed those computers (or who instructed their computer science colleagues to do so)?
They are people who in most cases have studied economics, marketing, and finance, and hold CFA designations. We find it quite useful to have both finance and programming knowledge in order to build effective models.
- “Quants invest on the basis of historic data regarding these things. But what will happen if patterns and relationships are different in the future from those of the past?”
Yes, quants do rely on historic data, in exactly the same way as discretionary managers must. Case in point, we are currently enduring another painfully long period in which investors are shunning value characteristics. But we know that historically, valuation has been a powerful long-term predictor of future returns. Quants know this. Discretionary managers know this. Have things changed? Does valuation no longer matter? We doubt this, as I’m sure most discretionary value managers do. But it is a question that all investment professionals must grapple with. Is the past prologue?
Things can change. When they do, who is more likely to have the discipline to change when the data says so, quants or discretionary managers?
- “Is it important that most quantitative investors have operated only in periods when interest rates were declining…”
I have no doubt. But ask the former Bond King about how he made his career on the tailwinds of declining rates, and how things have been working out for him recently, now that that gravy train (or fish cart?) has left the station.
Virtually everyone in the investment business today has only ever worked in a declining rate environment (with relatively minor exceptions). Who will respond better to change, discretionary managers or quants? We’ll see.
- “It seems obvious that a formula’s application and popularization eventually will bring an end to its effectiveness.”
No arguments here. It was true when growth stock investors formulaically piled into tech names in the late 1990’s. It was true when investors formulaically piled into real estate in the 2000’s. And we suspect we’ll be able to look back and say it was true of the way that investors formulaically piled over a trillion dollars of new capital into private debt and private equity in the 2010’s at a time of historically high valuations, in the hope that a Hail Mary would save them from poor expected returns on public equities.
Dr. Philip Zecher, the CIO of Michigan State’s endowment recently said, “Remember “Quant” is not necessarily a black box, the human head is the ultimate black box. If someone can express their ideas in code or an algorithm, it is more transparent than what rambles around in a trader’s mind.”
To that we’d add, quantitative investing is not investing without people, it’s investing without peoples’ biases and inconsistencies. Or more correctly, it’s investing without as many biases and inconsistencies, as the code is only as good as the coder.
Read our last blog, It is Time for the World Cup to Begin: My Picks…