As I write this, the Financial Sector SPDR (XLF) hit a low of 23.22 from its high in late January 2018 of 30.3255. In early January of 2018, I predicted that the financial sector would experience the equivalent of a bear market in approximately 12 months (i.e., drop more than 20%), with an erratic but largely steady decline over that period. Well, here we are approximately 12 months later, and those who have been following my prediction and the XLF’s development have been asking me “What events did you predict that caused the financial sector decline?”
So here’s the big reveal to the events I predicted would happen to cause a decline in the financial sector. Ready? Ok.
The answer is, “I don’t know.” Seriously. I don’t. Many of you will recall that my prediction was based on my new method for approximating extreme negative returns in large markets published in the National Academy of Science of Poland’s official physics journal, Acta Physica Polonica A. You can access it here (SSRN) or here (official APPA site) if you’d like. The heuristic I explore is not based on traditional mean reversion techniques, but on an entirely new approach. I won’t bore you with the mathematics here, but the net result is that if you use this method to look at certain markets in certain time periods, very clear patterns emerge. That was the case for the weekly historical data of major U.S. bank stocks as I was casually looking at them one day. All the bank stocks had the same pattern, which indicated to me that the entire sector was due for a significant downturn.
But I relied strictly on the mathematics and the geometric patterns they produced. I did not base it on Trump’s tariffs, or the North Korean Summit, or Russian politics, or currency stability, or the death of George H. W. Bush, or any other “event” or combination of “events.” This brings me to my larger point. While I am not necessarily convinced that all large systems have non-linear patterns, I am convinced that some do.
Two things here:
(1) I know financial analysts make their living churning out news and explanations for why a market or index behaves as it does. I understand why they do this. They need to earn their paychecks and investors spend money on answers to the question, “why did that happen?” in a perhaps unconscious effort to figure out what’s going to happen next. It’s natural for all of us as humans to looks for an explanation or story to explain events (just look at our mythologies to explain the movement of the planets). But I do not believe that large, heterogenous systems respond to a single event or single chain of events that way, at least not long-term. Sure, there might be a perturbation like the Brexit vote or the Trump election that sends markets into wild oscillations temporarily, but the markets tend to move back to their attractors rather quickly in these instances. This is all a very long-winded way of me saying stop looking for a logical sequence of events that explain large system behavior. You might be able to construct a narrative after the fact, but it’s fictitious because I don’t know many who can create a narrative forward in time to predict a market’s behavior. It’s more complicated and more complex than that.
(2) Which brings me to my second point: If markets and other large, heterogenous systems are, in fact, non-linear, there is no way our brains are going to be able to wrap our heads around why a market behaves as it does long-term. We do not think non-linearly, and our stories and narratives are not non-linear. So let’s stop trying to tell linear stories about non-linear systems. It just doesn’t work; not predictively, at least. In my opinion, the only language in which we can tell a non-linear story is the language of mathematics, and not strictly analytical mathematics. But mathematics does not tell us about individual events or a chain of events. It just tells us about the system.
So in short, I was able to predict the decline of the financial sector in 2018 because the mathematics told me it would happen, not because of any single or series of human events. This has some very strong philosophical implications. For example, if a financial market, which is a human system, has a deterministic non-linear behavior that is separate from individual human behaviors, how is that possible and what does that tell us about age old questions concerning free will and determinism? There are plenty of other questions, but I will save these for another time (in the meantime, I recommend reading P. W. Anderson’s “More Is Different“).
That’s it for now. Sorry for the long post. I hope it answers some of your questions. Again, I in no way directly or indirectly took a position on the financial sector or any related stock in 2018. While I could have made over 23% in less than a year, I thought it would hurt my credibility and the credibility of the prediction if I was writing these posts while having a financial stake in the outcome.
More to come in 2019! Stand by.
-Jack
Predictions & Predilections 