When Tail Risk Isn’t Tail Risk

Half of the battle in portfolio management is wading through the sea of BS narratives that we are bombarded with on a daily basis. Knowing which narratives to avoid can be even more important than knowing which narratives are actually useful. Many of the stories I write here involve me thinking out loud about much of this BS. So I had a good chuckle when I saw this list of “tail risks” from the most recent Merrill Lynch Fund Managers Survey (I added the lovely “wrong” column and I was apparently drunk when I did so because it’s pretty crooked):

Tail risks are low probability events so they should be hard to predict. But the interesting thing about this list is that it’s filled with tons of BS narratives. In fact, regulars will recognize many of these because I wrote about many of these in real-time (like the low risk from Fed balance sheet reduction, the probability that Trump was good for the stock market, the myth of the bond bubble, etc). Don’t get me wrong. Some of the things listed here are serious risks like the North Korea risk. That’s a very real tail risk, but we have to be careful separating the BS narratives from the real risks.

Now, the tricky thing with assessing tail risk (and risk of any type) is that it really can’t be quantified. No one knows what the chances of a nuclear strike are. Heck, the data on how poorly active management does shows just how hard it is to predict what the markets will do. Then again, any smart investment manager has to assess potential risks. If the world involved no risk management then we’d all just lever up long and wait for the inevitable portfolio imploding bear market before we were reminded that risk management matters. But how do we separate the BS narratives from the real risks?

1. To begin with, most indexers probably have no need for assessing tail risks. A true buy and hold indexer must accept that the markets have always gone through huge ups and downs. There is no shortage of challenges for mankind to overcome throughout its history and given an appropriate time horizon we do always overcome them.¹