It is too simplistic to blame the allocator model for the failure of certain opportunity strategies. Investors should be more concerned about funds becoming too large, says Roger Barris

Although the real estate opportunity fund industry has gone through a near-death experience from 2007 to the present, the cause is yet to be ascertained. This is a problem because, while there has been a limited amount of detailed diagnosis, there has been no shortage of proposed cures. And although I am a huge fan of Hugh Laurie’s character in the TV series ‘House’, I am not sure I want to see his brand of ‘ready, fire, aim’ treatment applied to the industry I like to call home.

Take one example: the ‘allocator model’, which, in the eyes of many investors, is a leading candidate for the villain’s role. Now, it is certainly true that some ‘arch allocators’ - such as the defunct Warburg Pincus, which modelled itself very closely on the traditional corporate private equity company by maintaining a very small staff and placing its money with ‘platforms’ - excelled at losing money during the last cycle. This worked well with the rising tide, but when things got tough and the platform started to wobble, there was no one to step in and preserve value.

However, some of the other entrants in the reality show, Real Estate’s Biggest Losers, were in fact ‘operators’ - although strangely they are not generally perceived this way. Goldman Sachs’s Whitehall funds, for example, have taken swingeing loses, yet almost all of their asset management services are provided by the Archon Group, an in-house asset manager that numbered over 1,000 people at its peak. Let there be no mistake about it, in the spectrum of allocators to operators, the Whitehall funds are firmly in the latter camp. In fact, Whitehall has not been an allocator since its early - and profitable - first funds.

Another example of a stealth operator that has suffered is the Morgan Stanley Real Estate Funds (MSREF). Until it was recently forced to sell it for US regulatory reasons, Morgan Stanley made extensive use of its German asset manager Argoneo. At its peak, Argoneo employed over 100 people, but despite this, MSREF has had to ‘hand back the keys’ on many of these deals, doubtless faring much worse in this part of its portfolio than in its investments with its other, earlier, operating partner, DIC.

And now a counter-example: Blackstone. Blackstone is doubtless a survivor of the real estate Permian Mass Extinction (thanks to John Forbes of PWC for this terminology). And yet, where does Blackstone fit on the operator-allocator spectrum? Blackstone is predominantly an allocator, although it will often manage assets in-house, as it has to be to achieve its global reach. Yet, strangely, I think that many investors perceive Blackstone to be a pure operator.

Enough. I think the point is made. It is simplistic to blame the allocator model and praise the operator model since time and again firms have shown that you can apply both models successfully or unsuccessfully.

But if the allocator model is not to blame, what is? Although there is never a one-size-fits-all explanation for something this complicated, for me the most obvious cause is that funds became too big. This fact, especially when combined with a fee structure that was designed for a very different scale, created the perverse incentives that were the root cause of the debacle.

The opportunity fund industry, and its fee model, was created at a time when $300-500m (€227-378m) was a big fund. As the manager of a fund in this size range, I can tell you that every penny of management fee goes towards keeping the heat on and retaining the best junior people in their seats. The only way the partners and the senior employees make any serious money is through the carry; in other words, they only make money when their investors do, which tends to focus the mind.

Contrast this situation with a mega-fund of, say, $8.8bn. This fund generates well over $100m of investment management fees every year just for turning on the lights, not to mention all the ancillary acquisition, advisory, financing and other fees its investment banking parent can manufacture. This is a lot of money, particularly when you can put it in the ‘recurring’ part of the income statement, where it attracts a hefty multiple - even better if you can raise this kind of money, spend it quickly and then go out and do it all over again. And this is precisely what happened in the glory days of 2006 and 2007.

But how do you spend $8.8bn fast enough to justify a rapid reload? Particularly in an environment of huge leverage, which is the best way - combined with fantasy operating assumptions - to pretend that you are generating opportunistic returns out of competitively auctioned portfolios? Think about the numbers: at an average loan-to-value ratio of 75%, to invest $8.8bn in 18 months requires buying almost $500m of real estate per week (assuming some time off for Christmas and the summer). Even for a Master of the Universe, that’s a lot of site visits, particularly if you actually want to take the time to understand what you are buying. The answer to the question is that you ensure you don’t lose a big auction. Ever. At least not on the basis of price.

Does this sound familiar? I think it does and I certainly think it provides a better explanation for the recent failures of the industry than a simple allocator-operator dichotomy. It also suggests that investors need to be selective in the way they adjust the model: the fact that the model is ‘broken’ for a mega-fund does not mean it is broken at the scale for which it was originally intended.

Investors say they want a broad selection of small, independent and focused managers whose incentives are fully aligned with their own. Although it needs some tweaks, the traditional model actually does a pretty good job of achieving this. To return to the medical theme, let’s all remember the Hippocratic Oath: First, do no harm.