Quant out, qual in

Good morning folks!

J.P. Morgan Asset Management has just published results of its annual European Equity Survey. This survey covered the 128 largest institutional investors in Europe. Of those 128, 102, managing over $3 trillion, said that they were more inclined than a year ago to favor qualitative approaches to investing than quantitative. Only 1 in 10 of the big investors said that they are more favorably inclined to quantitative approaches.

As I am certain most of you know I feel deeply that an equal weighting of quantitative and qualitative approaches is necessary to do well as an investor. However, because data are abundant the majority of research effort has long gone into quantitative methods of investment evaluation. This means that emphasizing this exclusively in your research is unlikely to yield good investment results. Why?

The reason is that there is tremendous competition to crunch numerical data. And many big, big investors employ computers to crunch numerical data. They operate at a far faster rate than human beings can. However, this massive over-emphasis of quantitative approach means that it is possible to consistently beat big, and better equipped, institutional investors with a qualitative, soft approach. Herein we emphasize understanding the data, analyzing the data, analyzing markets for goods and services, evaluating the mood of the consumer, listening to the management teams that head up the businesses we invest in, and making emotional choices when we invest.

So what would have caused so many institutional investors to abandon quant models? Because they stopped working. Why did they stop working? Because the world shifted out from under them so all of the assumptions were proven wrong for a new paradigm world. Ouch!

The shocking thing to me is that quant approaches consistently have proven that they have a very hard time of predicting changes in the direction of markets. As an analogy, one of the very best ways to predict the weather is simply to predict that today’s weather will be the same as yesterday’s. This simple strategy consistently beats other approaches. However, it misses every single change in the weather. Computer investment models use as their inputs past data. If a sea change shift is to be identified by the computer then the tectonics of the new shift must look similar to those of old shifts, otherwise the models miss the shift completely. Double ouch!

Unfortunately, I wish I could say that this will never happen again, but I can’t. The same thing happened after the dot.com bubble im/exploded back in 2000 also. For awhile there was an reliance on qualitative approaches, only to be abandoned in the next bubble: the credit bubble. Triple ouch!

Truly Yours in Quality,

Jason