Internal Hedge Fund Platforms: Get the Model Right or GET OUT!

Trashing in-house hedge fund platforms is currently all the rage. And why not? They had a bad year, no? This from today's Institutional Investor's emii.com:

Poor performance by in-house hedge funds launched by investment banks and the like may lead some big Wall Street names to exit the field, Dow Jones reports. While many a top hedge fund enjoyed returns that outperformed the standard indices, a number of in-house HF offerings by the likes of Goldman Sachs, Bear Stearns and Barclays have not.

And there were several pieces in the past week that were in this same vein. Bottom line - this is not new news. Most banks have gotten this business wrong for quite some time, in pursuit of its promise of great riches. And last year's weak performance does not begin to illustrate the core problems of most internal hedge fund platforms. Recently stories are trying to sell longer-term business model issues as near-term news. Sorry to disappoint, but these writers don't really get what's going on. Fortunately, I do.

Most pure quants and trend followers had a bad year in 2007. They got whipsawed, statistically durable relationships broke down and they generally got hurt. Luminaries as bright as Jim Simons, Cliff Asness and the Goldman and BGI quant groups got hurt, and they were by no means alone. Whether the teams in this area were internal or external, the markets brought harsh reality to the lot. And let's talk about hedge funds investing in structured mortgage paper. Unless you were shorting like a Paulsen you were getting murdered. Again, it didn't matter whether your were internal or external. Someone ripped your face off, pushed you in front of the mirror and you screamed for the last five months of the year. This is just the way it was.

So Goldman (quant), Bear (mortgages and mis-management) and Barclay's (quant) suffered the same as off-platform funds, so this this whole line of argument is a bunch of garbage (pronounced gar-baaaaage). It makes for a good story, but this isn't the real story. The REAL story is why most on-platform funds suffer from adverse selection, where the best of the best either start their own hedge funds or get a sizable book at an established hedge fund, while the remainder trade in-house. Here are some of the issues:

• Compensation
• Autonomy
• Stability of investment capital
• Compounding of capital
• Branding
• Ego

Compensation is an easy one. Most in-house operations don't pay nearly as well as hedge funds do. Because it is just not the bank culture to pay like hedge funds. If you want to be successful running an in-house platform you need to pay a much more competitive wage. And if you have truly great traders, it is worth it. And the more liquid the portfolios they run (and therefore attract a lower amount of regulatory capital), the better the deal for all parties involved. I could talk about this for days but this is the gig. Pay fair, create a separate culture for the in-house group versus the traders in the broker/dealer and get on with it. Or don't even try.

Autonomy is what it is. Some top traders just want their own show. And they have to go; because if you contort yourself to keep them they will kill the culture and damage the platform. However, if what they really want is to do their thing and be left alone then agree to clear parameters for book size, portfolio construction and risk parameters. And stay on it. But a little light mentoring and strong risk oversight can give the bank the security and comfort it needs while giving the trading team the perceived autonomy it wants. Again, a culturally challenging thing to do but absolutely necessary for building and running a successful in-house operation.

The biggest fear traders have is getting their capital yanked away from them and precisely the wrong time (or at any time, for that matter). If you run and hedge fund with some locked-up capital, you don't deal with this risk. This is a tricky point, because as head of an in-house platform you want to balance favorable capital treatment (which generally connotes the ability to liquidate positions at a moments notice) with giving the trader a measure of comfort (that you won't arbitrarily make them cut the book). Given these opposing needs, a pretty big measure of trust needs to exist between the trader and trading management, because the manager can't get the security they really want without killing the capital treatment (which the bank won't accept). It is what it is. The answer to this one isn't perfect.

Compounding of capital is an interesting point. When you run a hedge fund that starts the year at 100 and you generate 20% gains, you begin the next year at 120. In a bank where you are trading prop capital outside of a fund structure (i.e., in a managed account with the assets held on the bank's balance sheet), that 20 gets swept out and the trader begins year 2 at 100. Over time, the loss of the value of compounding can be stunning. Once an on-platform team begins trading client capital, this issue is ameliorated through a melding of manager and client capital which both benefit from compounding. But when a trader is running purely prop funds, they generally lose the benefit of compounding. Just something to be aware of when negotiating a deal.

Branding means more to some than to others, especially once client capital is being managed. The trader generally wants to own their brand and to have portability, which I think they need to have if you want to attract and retain (for a period) the true rock stars. Inevitably the rock stars are going to want to set up their own fund, and the strong in-house platform should be supportive of that move subject to terms in the agreement that provide a package of benefits to the bank, i.e., capacity guarantees, fee discounts, distribution agreements, etc. If a bank tries to hard to keep the intangible assets (the brand, the track record, the models) all to itself, it will lose out on the best teams out there. Because the best will want to be in-house at the right time and for the right reasons, and if they feel like they are being held up they simply won't do it, or will be resentful and cost the bank in other ways down the road.

Ego really relates to branding as well as portability. Many platforms don't anticipate teams being able to leave in a pre-agreed manner, and end up fighting bitterly with the team when the time comes for them to externalize. This just dumb. Put it all on the table upfront, let the team keep the brand, track record, and models should they leave, with the proviso that the bank gets the benefits mentioned above. it become a win-win, where only the best teams are able to spin-out because of strong performance and investor demand, which creates enormous value for the bank through their capacity option, discounted fees and distribution rights. This is the natural evolution of things. Don't fight it.

In sum, the in-house platform isn't dead. It has been dead for most for a long, long time. It is precious few firms that have ever done it well, but those that have created tremendous value for shareholders, traders and fund investors.

Comments

[Greg Battle]

Great article Roger.

Can you be clear on the difference between in-house prop desks and in-house hedge funds because, as I've always understood it, there is a fundamental difference. The most important one being liquidity vs. being forced to trade a fixed sum of capital. A prop trader who hits some liquidity constraint like a huge margin call has the option of being bailed out by the bank - an infinite checkbook. Hedge funds proper don't have that security.

The other reason why the best talent leaves the bank is because, more than ever, anyone looking to start his or her own shop needs an audited track record. Here again lies the difference between in-house prop and in-house hedge funds. From my own experience, it's particularly difficult to get verification on a prop guy's track record, especially anything you can successfully market.

So, you are right, if you want to retain your top talent, yes, give the great prop trader a platform resulting in an audited, exportable track record. The price for this are all the benefits you mentioned (fee discounts, capacity, piece of the management company, etc.).

I do know of some hedge funds that don't do the compounding on a trader by trader basis, depending upon the capacity of the trader's strategy.

[Yaser Anwar]

Good point Philip.

The bank who starts an internal platform, even external ones by non-banks, has to make sure that the CFO/CEO/non-trader personnel of the fund get compensated on the overall performance and not for asset gathering.

An interesting technique for traders could be their P&L reset every 6 months. While at year end this isn't any different from full-year, it has the psychological effect of not swinging for the fences just because one is up or down for past 6 months. However, I'm not sure if the compounding process you allude to above can be done while maintaining the 6-month reset plan.

Another interesting technique to package and sell the work already being done by traders/research staff is to create different product lines.

For example- if the GSPS fund is long/short value investing (which it is) then having different extensions like long-only; long-enhanced fund (150/50); leveraged market-neutral fund (200 long/short) etc.

Would very much appreciate a comment, Sir.

Great post overall. Thanks.

[philip]

The main problem with in-house hedge funds marketed to outsiders is that the marketers run the show. The marketing capabilities of the firm allow funds that would not garner assets on their performance to raise too much money too quickly. An investment bank can generate greater revenues for more businesses by growing assets (as opposed to managing fund size and producing higher returns and revenues for a select group). I have the utmost respect for Ranaan Agus of Goldman Sachs...but he should not have launched with $7 billion.