Investment StrategyMacro PerspectivesMost Recent Stories

The Temporal Problem in Market Forecasting

You can’t talk about money and investing without talking about time.  After all, the two are inherently interconnected.  There’s the time value of money, the erosion of value due to inflation, the linkage between money and interest rates, etc.  The problem is, time is a great unknown for all of us.  It is a concept that we apply in a strict sort of theoretical sense to create structure to our lives.  But there is nothing certain about our lives within this temporal construct.

The concept of time is the epicenter of all the great debates in finance and economics.  I bring this up in reference to a great discussion on Twitter between some very smart people (which I elbowed my way into, pardon me gentlemen, make some room for the newly rotund, thanks Paris, Cullen Roche).

The very smart Barry Ritholtz was highlighting the importance of avoiding market forecasts during the investment process.  He referred to a post earlier this year citing the also very smart Jesse Felder who was saying many things similar to what I often say here.  Basically, anyone who allocates assets is making specific forecasts about the future.  Barry disagreed saying that one need not make forecasts, but only needs to make probabilistic decisions about the future.

This seems to be more a disagreement about taxonomy than anything else.  We all can agree that we have to make decisions about an uncertain future.  But I think we can also agree that making very short-term decisions about the financial markets is very difficult.  The empirical data backs this up.  In the long run, we are all dead, pay taxes, the Earth explodes in a fiery ball, etc.  Those are easy predictions because they are long-term predictions.  Of course, no one is impressed by those predictions because they are useless within the short-term that is our lives.

So, we arrive at this sticking point.  We want to make long-term decisions about the future because those are likely to be higher probability decisions.  But we have a rather finite time period within which to make those decisions.  This is why I caution about those who focus on extreme temporal concepts.  The idea of the “long-term” can be just as dangerous as the idea of the “short-term”.

As Howard Marks says, great investors think about the financial world as a probability distribution.  We all have to make decisions about the future, but the kicker is that there are very smart ways and very stupid ways to make those decisions.  When we bash “forecasters” I think what we’re really bashing is people who think that short-term decisions are likely to be high probability decisions.  That thinking just doesn’t pan out when one considers the data and after tax and fee returns of investors.  Unfortunately, far too many people still focus excessively on the “short-term”, succumb to recency bias or rely on guru worship to make financial decisions.  There’s a good middle ground in my opinion, but it requires a great deal of behavioral fortitude to stick with given all the uncertainties that time creates for us.


The Intertemporal Conundrum

Problems With “The Long-Term”

Problems With “The Short-Term”