On Position Sizing in Equity Long-Short Hedge Funds

This article is prompted by the following article by John Hempton of Bronte Capital.  This is not meant as a criticism of him; I have nothing but respect for him.  The article triggered memories of my own experiences with position sizing at a hedge fund.

The hedge fund I once worked for had great expertise with financial companies, and I worked for them in the boom years of the 2000s.  Our leader was bearish on depositary financials, a view that would eventually be right.  Of course “eventually right” is another way to say “wrong in the short run.”

Let me describe the problem from another angle.  When I was a corporate bond manager, I would mentally set three levels with the bonds that I held.

  • Spread necessary for an ordinary-sized position.
  • Spread necessary for a big position.
  • Spread necessary for a maximum position.

These spreads I would adjust for premium vs discount, optionality, and a bunch of other things.  The point is that I would always have a schedule for where I would be willing to buy more, or lighten up (sell some).  I often dealt in some of the least liquid corporate bonds, and I was patient, and even willing to break rules by holding more than 20% of a given issue.  My analysts almost always did good work, and I trusted them.

When markets are illiquid, they “trade by appointment.”  If you have a balance sheet behind you that is not worried about liquidity, you can do interesting things by buying assets that most ordinary managers won’t touch, because the issue is too small.

I came to the hedge fund after I managed corporate bonds.  In one sense, I had managed a far more complex long-only portfolio.  But being able to short creates complexities of its own.

I can’t tell you how many times at meetings at the hedge fund we had tough discussion on position sizing, more frequently on short positions. We were perpetually long quality, short market capitalization, long insurers, short banks, and long value.  Great idea, if too early. This would be an extreme example:

Boss: “This short position is killing us, it is up 50% from where we shorted it, and now we have a 6% short position, what do we do?”

Others answered in front of me, essentially suggesting no change.  He asked me personally and I said:

David: “If you had no position, and you were approaching this company today, what would you do?”

Boss: “I would short the maximum — 4%.”

David: “Then buy in 2%.”

Boss: “But that locks in the loss.”

David: “Do you want to risk locking in a bigger loss?”

The boss once said to me that I was the only one on his team that was natively a portfolio manager rather than an analyst.  (That said, I remained an analyst, while an analyst was made an assistant portfolio manager.  I think it would have been too difficult to have the insurance guy to manage the portfolio of what was a banking shop.  That said, as a corporate bond manager, I managed the financials, which were mostly banks.)

Setting position sizes on shorts is always harder than longs.  When your thesis goes wrong on a short, your risk increases, as the position size gets larger.  When it goes wrong on a long position your risk decreases, as the position size gets smaller.

As I have often said, being short is not the opposite of being long, it is the opposite of being leveraged long.  When you are short, or leveraged long, you do not fully control your trade.  The margin desk can take you out of your trade if the equity in the account gets small enough.  They are ruthless in doing so, because the margin desks at brokerages do not want to take losses.

That makes it all the more important to set a schedule of sizes on short positions.  The first question should be: at what price would I put my maximum position on?  That would help in sizing introductory and normal positions.  They would be far smaller than what most hedge funds do.

Again, the same exercise is easier in a long-only format, but the protocol is the same.  Establish introductory, normal and maximum position sizes, and hold to them.  Also put into effect the idea that analysts must give greater scrutiny to large positions.

All That Said

This is a reason I am not a fan of most hedge funds.  I believe in the funds of my former employer and those of Mr. Hempton.  But the difficulties of dealing with bad decisions with a weak balance sheet kills a lot of hedge funds.  Long only — it might survive.  But when you go long and short with leverage, the risk arises of total loss.

So don’t think you are a “cool kid” because you invest in hedge funds.  Long only does better over the long haul, because it is less risky, and compounds value.

 

2 Comments

  • hodedofome says:

    To me the only reason to be short-on a regular basis-is because you want uncorrelated returns to long only. I think if people were honest, they would see that their shorts by themselves hardly ever make much money over the long haul, if they make anything at all. This is true of CTAs as much as it is for long/short equity funds. There’s just so many issues with shorting that it’s not worth it outside of special situations.

3 Trackbacks