Madoff and Renaissance Technologies

A short while back, I blogged
about the OIG report on the SEC investigation of Madoff. One of the
interesting nuggets in this report is about how the leading hedge
fund, Renaissance Technologies, analysed and dealt with their Madoff
exposure way back in 2003. It struck me as a good example of prudent
risk management.

The first internal RenTec email about its Madoff exposure contains
a brief description of the red flags, but what interests me is the
risk analysis:

Committee members,
We at Meritage are concerned about our [Madoff]
investment. …

… you have the risk of some nasty allegations, the
freezing of accounts, etc. To put things in perspective, if
[Madoff] went to zero it would take out 80% of this year’s
profits.

Sure it’s the best risk-adjusted fund in the portfolio, but
on an absolute return basis it’s not that compelling (12.16%
average return over [the] last three years). If one assumes that
there’s more risk than the standard deviation would indicate,
the investment loses it[]s luster in a hurry. It’s high season
on money managers, and Madoff’s head would look pretty good
above Elliot Spitzer’s mantle. I propose that unless we can
figure out a way to get comfortable with the regulatory tail risk in a
hurry, we get out. The risk-reward on this bet is not in our
favor.

In one short email, you have several lessons in risk analysis:

• Worst case scenario: Madoff goes to zero
• Risk sizing relative to risk appetite: 80% of profits.
• Analysis of tail risk, separately from the historical
  volatility.
• Analysis of liquidity risk (freezing of accounts).
• Concern about regulatory risk (Spitzer).

What is interesting is that this email led to a flurry of emails
analysing the red flags in Madoff at great length, collecting data
from published sources and from conversations with market
participants. At the end of it all, there was disagreement about the
course of action between those who wanted to exit the position
completely and those who drew comfort from the fact that Madoff had
survived an SEC investigation. Finally, they decided to reduce the
exposure by 50% (perhaps as a hedge fund they had the risk appetite to
lose 40% of profits in a worst case scenario, when the investment
looked attractive otherwise).

What is also interesting is that these smart hedge fund managers
thought that the one regulator who was likely to catch Madoff was the
New York Attorney General, Spitzer. Markopolos also thought that the
New York Attorney General was the best financial regulator in the
country (see my blog post here).

Of course, the RenTec people come across as having a self
confidence bordering on hubris. At one point, they analysed
Madoff’s stock trading and determined that “the prices
were just too good from any mode of execution that we were aware of
that was legitimate. … And we would have loved to figure out how he
did it so we could do it ourselves. And so that was very
suspicious.” They finally decided that Madoff could not be doing
what they were not able to do themselves: “Well, I knew it
wasn’t possible because of what we do.”

I can quite imagine the RenTec people thinking that there was no
way Madoff
with his AS400 could do what RenTec could not do with the 60th
largest supercomputer in the world.

Yet, there is no reason we should not learn from a bunch of
arrogant people.

As an aside, I thought that the internal RenTec emails were the
best leads that the SEC got. These were not complaints and were not
even intended to be read by SEC – they just got picked up during
an SEC examination of RenTec. There was clearly no motive, no hidden
agenda. The SEC was peering into the unedited thinking of some of the
smartest hedge fund managers in the world.

As another aside, the very fact that these internal emails got
picked up as a lead for investigation of another entity conflicts with
the idea that the SEC is so badly incompetent. My
Hanlon’s Razor is taking some dents.