Diversified portfolios of hedge funds for small and large asset allocators

Hedge funds — and any other asset or investment strategy for that matter — should be included in a portfolio until the marginal benefits of additional investment equal the marginal costs of additional investment.

An optimized portfolio is achieved when additional investments in each asset or strategy are equally attractive. In other words, exposure to each hedge fund strategy and manager, and to hedge funds overall, should be added until the net benefits have diminished to the point that allocations to other investments are equally attractive.

Using this simple framework, it is rather common to argue that small investors shouldn't consider direct investments in hedge funds and, instead, should use a fund-of-funds platform if they want exposure to hedge fund strategies. The reason is that creating a diversified portfolio of hedge funds costs money. Of course, having a concentrated portfolio of hedge funds costs money as well. The investor is exposed to strategy as well as manager risks.

On the other hand, it is highly unusual to argue that very large investors might have a difficult time justifying large absolute allocations to hedge funds. Interestingly, similar economic forces make investments in hedge funds challenging for both small and large asset allocators.

Suppose a small endowment decides to dedicate a portion of its portfolio to hedge funds. The initial cost of finding and evaluating a hedge fund is around $100,000. The annual monitoring cost will be around $50,000 per year. Of course, one may decide to cut some corners and spend less on due diligence, but there will be associated costs. Studies have shown that more than 50% of fund failures are due to operational risks. In addition, if enough resources are not spent on the evaluation of fund managers, there is a greater chance that a poor performing fund is selected. Even if one spends the full amount on fund evaluation, there is a reasonable chance that the fund will underperform its peers. After all, alpha generation is close to a zero-sum game and not all active managers can be above average.

To create a diversified portfolio of hedge funds where the portfolio covers more than one strategy, and each strategy has more than one manager, the endowment should consider a portfolio of about 15 managers. Suppose the hedge fund investments will generate about 2% alpha on an annual basis, and there is 20% turnover every four years. Under these simple assumptions, the allocation to hedge funds should not be less than $50 million. Otherwise, the benefits of investments in hedge funds will be less than the cost of creating the hedge fund portfolio. Finally, if we assume that 10% of the entire portfolio is allocated to hedge funds, then the entire portfolio should be around $500 million, a very large figure. If we assume the average alpha is 4% per year, the minimum portfolio size will be $250 million, which is still rather large.

There is a simple solution to the above problem. Our hypothetical endowment could choose to invest in a fund of funds, which could provide access to a diversified portfolio of hedge funds in exchange for another layer of fees. With some modification, the above calculations can be applied to a fund of funds. That is, how large should a fund of funds' AUM be so it can perform rigorous evaluation and monitoring of its managers and remain profitable? It appears that one should be careful before investing in a single diversified fund of funds that has less than $1 billion in assets under management. It might be difficult for smaller funds of funds to create diversified portfolios of hedge funds while performing a rigorous evaluation of their managers.

Suppose our endowment continues to grow and has no problem justifying its direct investments in hedge funds. Can this endowment reach a size where increased absolute (constant relative) allocations to hedge funds might not be economically justifiable?

As this endowment continues to grow, it will have a number of options when deciding to increase the absolute size of its allocation to hedge funds: (a) increase allocation to the same funds that have helped it grow; (b) allocate to new funds launched by the same fund management companies; (c) seek new managers; and (d) create internal hedge funds.

There are two scenarios under the first option. Under the first one, the endowment increases its allocation to the original funds while these funds are not growing as rapidly. This means the endowment soon will become the largest investor in many of its funds — a very risky position. For example, the endowment will have too much manager risk and will find it challenging to withdraw some of its capital from these managers. Under the second scenario, the funds grow at the same rate that the endowment is growing, and, therefore, the above risks are avoided. However, another problem arises here: There is strong evidence that performance suffers as funds become large and old. These funds tend to lower raw returns and lower alphas. It seems that option (a) could be a dead end.

Option (b) seems to be quite appealing. The endowment sticks with the same fund companies but invests in newly launched funds by these successful fund companies. Because these new funds are launched by known fund families, due diligence and monitoring costs could be smaller as well. While this option largely avoids the problems associated with option (a), it might not be economically viable. There is strong evidence that new funds launched by successful fund families significantly underperform their flagship fund.

How about the third option? The endowment can continue to discover new and highly skilled fund managers, and, thereby, avoid too much concentration and the drag on performance. Of course, due diligence costs will be high for these new managers, and some might not have the capacity to absorb large allocations by our endowment. There are a few other problems to consider. First, empirical evidence shows that returns on hedge fund portfolios consisting of a large number of funds tend to have higher correlations with respect to traditional asset classes and lower skewness (i.e., higher tail risk). In addition, it will be challenging to continue finding new managers with sufficiently high expected alphas to justify the increased absolute allocations.

Option (d) is quite viable for large, sophisticated investors. In fact, the economics of this option are exceedingly appealing. A large asset allocator will be able to hire highly skilled managers and allocate optimal amounts of capital to each manager such that the highest level of risk-adjust return is achieved. However, non-economic reasons might make this option less than optimal for some large asset allocators. For instance, it might be politically impossible for some asset allocators to provide the same level of compensation to their internal managers that they can obtain in the open market.

There is one last option that was not discussed above. The fund might decide to keep the absolute size of its allocation to hedge funds rather constant, thus reducing its weight as the fund grows. Will the relative size of the allocation become so small that the portfolio manager might decide it no longer makes economic sense to spend significant resources and bear exposures to some unusual risks (e.g., reputation risk) for an investment that cannot make a meaningful impact on the overall performance of the portfolio? Can we see a significant number of very large asset allocators liquidating their hedge fund positions? That is, are we going to see more cases like the California Public Employees' Retirement System, where a large asset allocator decides to liquidate its hedge fund positions?

We might see a few. However, given the size of the hedge fund industry, there are few pools of capital that might have outgrown the hedge fund industry in the above sense. Further, new investment opportunities are created each day, and astute asset allocators can still find emerging managers with compelling value propositions. As a result, we are not likely to see a significant number of large asset allocators reducing the relative size of their hedge fund positions anytime soon. More importantly, by the time this becomes an important issue for large asset allocators, we are likely to see industry innovations designed to address the needs of this class of investors — the same way funds of funds address the needs of small investors.

Hossein Kazemi is the senior adviser to the CAIA Association.