What do we lose when we use a benchmark?
Make everything as simple as possible, but not simpler.
Everything should be made as simple as possible, but not simpler.
Everything must be made as simple as possible. But not simpler.
Everything should be as simple as it is, but not simpler.
Everything should be made as simple as possible, but not one bit simpler.
Benchmarks enable simple analyses. Simple is good. If you don’t know that I’m a big fan of simplicity, then you don’t know me.
But benchmarks are too simple.
A key problem with Value at Risk is that it reduces risk to a single number. Risk is more than one-dimensional. Returns have more than one dimension as well.
In the days when 60 million shares was a heavy day of trading on the New York Stock Exchange, the news reported not only the level of the Dow, but the advance/decline line — how many stocks rose and how many fell. That was — in some ways — a fuller picture than we typically get now.
Whenever a benchmark is used, the proper response is (almost surely): “But it is more complicated — and more interesting — than that.”
When portfolio constraints are in play, then random portfolios are a likely replacement. In other cases some thought may need to be put into it. Thinking is sometimes a good thing.
Any good alternative will certainly involve graphics. Our drive to get a single number is because numbers are hard for us. However graphics — good graphics, that is — are easy for us. (The reason for this is explained in Brain Rules.)