From the period of their earliest development by A.W. Jones in 1949, hedge funds have concentrated on listed securities. The reasons for this are simple. As conceived of by A.W. Jones, a hedge fund's purpose was the pursuit of alpha, of returns uncorrelated with those of the market. In order to operate in this way it is first necessary to define the "market". The "market" should be large, liquid, and display a high degree of inefficiency.

The definition of the market which fell most easily to hand in those days was "US equities". The US equity market remains, large, liquid and transparent. Note that A.W. Jones was not arguing for the US market to offer superior returns but rather that it offered a large pool of alpha in which he might fish. However, it may be argued-and not without support from academic research-that since 1949 the US equity market has grown less inefficient. This is probably true of all traditional sources of hedge fund alpha. Partly as a result of this decline in the scale of the pool of potential alpha offered by traditional markets, hedge funds have been constrained to explore new frontiers.

In addition, in recent years the growing scale of hedge fund assets, the range of "take-private" opportunities offered by depressed public equity markets, and the strength of returns from private equity in the late 1990s, have conspired to provoke a convergence of hedge funds focused on the public markets with private equity investment pools. Such investment pools have traditionally seen public markets only as elements in their exit strategy.

Hedge funds have therefore begun to converge with other funds aiming at absolute returns. There is of course a large difference between a classically constructed hedge fund and an absolute return fund, but given this convergence it should not be a surprise that real estate, which in some forms presents itself in a private equity-like form yet at other times appears as an archetypal area of hedge fund, long/short, opportunity, has also been drawn into this net.

Leaving aside the question of non-hedged absolute return funds for a moment, a backward filling and forward expansion of hedge funds is at work in this connection. The backward filling is towards ever more micro definitions of the "market". A.W. Jones's "market" was the US equity market. Now we have hedge funds concentrating on intra-sector investing, where the "market" is the sector. Property - or at least listed property securities-is one such market. The forward expansion is towards ever wider definitions of the "market"-not just US stocks, but all developed country equities; not just listed securities but unlisted securities and related physical assets.

Property investing through hedge funds encompasses both the backward filling and forward expansion elements. Most property-based hedge funds attempt both to extract alpha from intra-sector investments, and to derive non-correlated returns from a range of assets outside this definition, principal among them physical property.

We now have two "market returns" rather than one against which a fund can attempt to extract alpha. The first "market" is a narrow definition of the listed sector. The second "market" is the broadest definition of the sector, expanded to include not only listed stocks but unlisted securities and physical assets.

The possibility of superior returns from the asset class as a whole would be an argument for a property-based absolute return fund. By contrast, the real argument for property as a hedge fund asset class is not that property as a whole might offer better index returns than stocks as a whole, but that a broad definition of the "market" allows inclusion of areas of such significant mis-pricing that the pool of alpha available for extraction by managers is greater than in more efficient, more liquid, markets. However, there are important complications.

Liquidity and Imperfect Hedges

The most important difference between real estate and listed equities is the lumpiness of liquidity in the physical property market. While increasing securitisation has improved the liquidity and depth of the market in real estate related claims, the small number of transactions in physical property (small relative to the number of daily trades in listed securities) is a natural consequence of the cumbersome nature of real estate lending. It is not really possible to trade everyday when a banker is involved.

This difficulty is compounded by the fact that it is not yet possible to "borrow" and short a building in most markets. (In fact, several very rare cases of transactions which come close to the effect of physical property shorts have been observed; they usually take the form of the sale of call options.) In effect, therefore, one is constrained to run a portfolio in which long positions in physical property and private equity-like exposure are hedged by short positions in publicly quoted stocks. Implicitly this balance only works if private market values are less demanding than the valuations attached by public markets to a significant fraction-though not in fact all-listed securities.

To the extent, therefore, the necessity that this should be the case mitigates against property hedge funds actually being the pure pursuit of alpha that hedge funds should be, or are often marketed as being. To some degree, unless very carefully negated by the manager, hedge funds focusing on real estate are likely to become directional plays on the property market-in other words, they risk turning into absolute return funds, not hedge funds. In our opinion, this is not necessarily a problem, but it should be made clear when investors invest that, in addition to the alpha arguments above, a projection of positive market returns may also enter the picture.

Furthermore, should private market values exceed public market values then the approach which is now being taken by real estate related hedge funds would be invalidated in the area of the forward expansion referred to above. What would remain would be the exposure to relative value plays within the listed sector. Again, to an investor who had entertained hopes that he was entering a real estate fund because the real estate market offered superior potential appreciation the discovery that he (she) is invested in fact in a pure long-short intra-sector strategy might not be a congenial development.

It is especially important, therefore, in the context of a hedge fund structure built around illiquid, often physical, assets, that the degree to which the fund will be a directional, absolute return fund, versus the degree to which it will remain a hedge fund, seeking-in the traditional manner-to extract alpha from fully hedged positions, is fully explored and thoroughly ventilated.

Time Frames/Lock-ups/Fees: The Difficulty of Aligning All Interests

Liquidity also causes difficulties in terms of the structure of the fund. In contrast to funds purely focused on listed securities, any fund invested in illiquid instruments will be unable to offer investors liquidity on highly frequent basis.

These liquidity questions prompt, of course, fund managers to ask investors to accept long lock-up periods. This in turn raised the issue of whether lock-ups should be structured on a rolling basis or on the basis of a single period common to all investors. If the lock-up is common to all investors, irrespective of when they actually invested, the ending of the lock up can be followed either by a transition to free redemptions on a monthly or quarterly basis, or to a staged redemption on a percentage basis (ie 25% in the first year of free redemptions, and so on).

The disadvantage of this approach is the initial lock-up period has to be sufficiently long so as to allow the manager opportunity to start harvesting profits before the lock-up expires. The advantages of a rolling lock-up are a much flatter profile to possible maximum redemptions and so a shorter necessary lock-up period. Investors rightly question why they should pay management fees in the earliest stages, when any fund mainly focused on physical property will still hold un-invested cash balances, but are somewhat less concerned by the similar phenomenon which develops towards the end of a "one shot" lock-up period as the manager raises cash to satisfy possible redemptions. The problem of management fees on un-invested balances can be finessed by drawing down investments already committed in tranches, but only at the expense of further complicating the timing of eventual redemptions.

The greatest difficulty with allowing for rolling lock-up periods is that it implies, when the lock-up period of the earliest investor approaches expiry, a degree of dissonance between the interests of investors who have subscribed at different times. While it may be in the interests of an early investor that the manager harvest gains on a particular property, it may not be in the interests of the later investors, for whom continuing to hold the position might be advantageous. Of course, all these conflicts exist in any hedge fund format, but they are magnified when situated in a low liquidity environment such as physical real estate. Our view is that these difficulties demand an additional effort from the manager to keep all investors fully informed at all times as to what the fund is doing, and also where potential conflicts of interest may reside.

There are no perfect answers in this area. The same is true of fee structures. The REITs tend to have the ability to charge three sets of fees. The first is an acquisition fee, the second a fee related to total assets, the third a fee related to Funds From Operations (FFO). Increasingly a fourth, performance, fee linked to the performance of the REIT's stock relative to a representative index is in evidence.

This is typical of long-only real estate structures. Once in the hedge fund area, however, the structure of fees is of necessity modelled on that of other hedge funds. This is scarcely a problem. But valuation of portfolios containing unrealised gains in lumpy, illiquid, physical property or private equity-like exposures is potentially problematic, and raises issues not faced by hedge funds operating in the highly liquid area of listed securities. Again there are no perfect answers. The compromise we have reached is to invite representatives of the largest investors in our fund to be members of a valuation committee. This committee values the fund's positions.

One can see immediately that it is in the interests of investors to delay upward revaluation of difficult to value assets until close to the date at which gains can be realised. But this seems a reasonable compromise when one considers the liquidity penalty that investors have accepted in the form of the lock-up.

Conclusion

Our view, as might be deduced from the above discussion, is that liquidity characteristics and the compromises inherent in enclosing real estate investing in a hedge fund structure, mean that such funds are likely to remain niche strategies.

They appeal to those who, looking across their whole portfolio, are searching for investments likely to produce returns entirely uncorrelated with any others. Japanese real estate-as a class-is likely to prove uncorrelated with anything other than, perhaps, small capitalisation Japanese equities. But this implies that such investors already hold a highly diversified portfolio of investments that already includes all the major asset classes-driving them towards less traditional areas such as property.

Real estate focused hedge funds, like all such micro orientated funds, will also appeal to the more academically inclined, who will appreciate the greater possibilities for producing alpha when the "market return" is derived from a pool of assets as relatively inefficient as, for instance, Japanese property and property-related securities.

Our experience supports the foregoing discussion. Real estate focused hedge funds appear to be of appeal to a relatively small number of relatively large scale investors. Real estate within a hedge fund format has an important role, but it unlikely to become entirely "mainstream".

Alexander Kinmont has followed the Japanese equity and real-estate markets since 1985. Prospect Asset Management manages Japanese real estate-related funds, including a hedge fund, as well as a range of long-only and long-short equity portfolios on behalf of a diverse range of clients. Prospect Asset Management and its real estate affiliates do not invest outside Japan. Prospect Asset Management is registered with the SEC. Nothing in this discussion should be construed as constituting a solicitation or offering of investment advice or of investment management services.