Shared by JohnH

From 1997! Links above to full interview.

DS: What’s going to happen if everybody in the financial system starts using VAR?

NT: VAR players are all dynamic hedgers and need to revise their portfolios at different levels. As such they can make very uncorrelated markets become very correlated. Those who refuse to learn from the portfolio insurance debacle do not belong in risk management.

In 1993 hedge funds were long seemingly independent markets. The first margin call in the bonds led them to liquidate their positions in the Italian, French and German bond markets. Markets therefore became correlated.

VAR is a school for sitting ducks. Find me a dynamic hedger who is a reluctant liquidator and I will front-run him to near-bankruptcy.

DS: So one problem with VAR models is that they don’t account for the fact that the market corrects for the models that trades are based on?

NT: Bingo. Even more: Our perception of what’s going on in the real world can hurt us simply because we have to realize that we are the major players ourselves and we act according to our perceptions. In physics it’s called the Heisenberg uncertainty principle. In the social sciences its even more pronounced.

When people ask me what alternative to VAR I have to offer, my answer is smaller leverage, less naïve diversification, less reliance on dynamic hedging.