As a general rule redemption terms should reflect the investment strategy of a fund. Having acted as an advisor to both sellers and buyers of hedge funds, what this means in practice is much harder to define.

1. Have redemption terms changed over time?
2. Are redemption structures different between US and European hedge funds?
3. Do investors get compensated for liquidity risk?

Lock-ups and redemption penalties
Of the four main parts of the redemption terms (frequency, notice period, lock-up and “gate”) we find most of the discussion tends to relate to lock-up provisions. All hedge fund managers wish to pursue so-called “sticky money”. With few exceptions, where a hedge fund investor is actually widely known to have a high portfolio turnover, it is very difficult to know beforehand what qualifies as such. In addition, those types of investors which are viewed as providing “sticky money” (for instance foundations) are in general in the minority, and hence hedge funds will in reality have to deal with a wide range of funds of hedge funds and other institutional buyers of hedge funds. In order to “protect” themselves from investors going in –and out- of the fund, lock-ups are often set.

In general, there tend to be two reasons why investors redeem; past returns have not been as strongexpected, or the investor itself is facing redemptions and needs to sell hedge fund holdings. Few hedge fund investors operate in a counter cyclical manner, selling those funds which performed best and buying into lesser performing funds. Even where there is an issue of hedge fund strategy allocation decisions and the FoHF wants to reduce its exposure to a certain strategy, it still tends to be the lesser performing funds within that strategy, which will be redeemed first. Hence, if a hedge fund manager performs well, there is hardly a need for a lock-up.

Starting hedge fund managers often underestimate the fact that the decision making process to invest in a hedge fund tends to be quite long, taking multiple meetings often over many months. Hence, a certain level of understanding of the manager’s stated investment strategy is built up prior to investing. It also means that from an operational perspective many of these FoHFs would simply be ill-equipped to handle a very high portfolio turnover, while still undertaking the same level of due diligence. In short, most investors are given their own investment in research time, focused on long-term investments. The investor and hedge fund managers’ minimum investment horizons are not necessarily too different.

The approach to lock-ups does tend to differ between the US and Europe. Using the Lipper Tass database we looked at the length of lock-ups displayed. Unfortunately, this database does not clearly distinguish between hard lock-ups and soft lock-ups (i.e. redemption penalties). With this caveat, we have calculated that the average lock-up period for equity long-short funds managed/advised by US-based managers is seven months, with a median of 12 months. For European-based managers, the average for their equity long-short funds is 2.0 months, with a median of 0. Of course, it may be that the latter simply do not record redemption penalties in this database.

However, from my own experience the above statistics are not surprising at all, with US managers often showing hard locks ups of one year. Whereas such terms are quite common in the US, in many cases they will lead to questions from European investors. Any time a manager needs to spend time on fund structural issues, time is being taken away from talking about the main selling point, i.e. his/her unique investment activities.

Although no specific statistics exist, we believe that the inclusion of redemption penalties is more prevalent for European funds, than for US funds. The two important factors are the period for which these penalties apply and the size thereof. Redemption penalties with a greater maturity than 1 year can cause a negative impact on the marketability of a fund, even if lower in a second or third year. The actual level of redemption penalties is also important. Too low and they become almost meaningless, so why bother? Too high and they will be seen as a hard lock-up. Finally, the recipient of the redemption penalty also matters. If these fees are received by the remaining investors in the fund, they can be seen as a compensation for some forced selling of positions. However, if the fund invests in liquid positions, this clearly cannot be claimed, thus undermining the reason for redemption penalties. Redemption fees which will be received by the investment manager tend to lead to many questions by investors.

When it comes to trends as to lock-up maturities our conclusion is that until 2005, none can be discerned from the statistics derived from the Tass Lipper database. The average lock-up period for equity long/short funds which commenced in 1989 was 4.8 months, while for funds starting in 2005 it was 5.0. The average for the preceding years has been both higher and lower.

Much has been made in the press of hedge funds who wished to escape the need to register with the SEC by changing their lock-up period to two years. For the great majority of hedge funds, making their fund less liquid would significantly impede their chances of raising additional funds. It is only the very largest and most successful funds, many of which are either soft or hard closed, that have been able to lengthen their lock-up periods. Of course these large funds have a higher profile thus creating a false sense of the market. This may be even more the case for the super start-ups, where very long lock-up periods seem to almost be a badge of honour, to put it politely.

We believe a key question is whether investors get compensated for the higher risk of illiquidity. This can be either the illiquidity of the underlying strategy, or the illiquidity of the actual fund structure (which may not necessarily be related to the former).

Redemption Frequency
Looking at redemption frequency, the deciding factor should in theory be the liquidity of the portfolio. However, we again can see that there is quite a difference in the approach taken by US-based managers as opposed to Europe-based managers. In the US the average redemption frequency is 3.7 months and the median is quarterly. In Europe the average redemption frequency is 1.2 months and the median is monthly. This, we believe, is not due to a much greater focus on illiquid stocks in the US than in Europe, but more so because of a greater acceptance by US investors of lesser portfolio liquidity.

It is difficult to clearly identify a trend as to what the average redemption frequency is for equity long-short funds, which have been set up since 1989. If anything, there is a slight decrease in the average redemption frequency towards monthly. However, this may also be due to a relatively greater number of European hedge funds set up in the last decade or so, as opposed to the early 1990s. As to redemption frequency we again have looked at whether investors have been compensated for the greater liquidity risk. Fig.2 shows the annual realised returns for approximately 450 equity long-short funds from January 2003 to December 2005. Not making a distinction between US and Europe-based managers and thus having a larger and hence statistically more reliable database, we can see a clear pattern that funds with the lowest liquidity, generated the highest returns. This pattern holds true for equity long-short funds advised by US-based managers as well as in Europe. For the latter the sample size becomes smaller, and possibly less reliable.Interpreting the above data remains difficult. Between 2003 and 2005, small and mid-cap equity indices outperformed large cap indices, and we would expect hedge funds to show some level of correlation with their relative benchmarks. Although the Lipper Tass database sometimes does provide a description if a fund invests in small, mid or large cap stocks, this information is sporadic and may not be up-to-date. However, in our experience redemption frequency may often not be in-line with actual liquidity of the portfolio.

Redemption Notice Period
When looking at redemption frequency as a measure of liquidity it may be useful to also take into consideration the redemption notice, which needs to be given. Again US managers demand slightly longer redemption notice of, on average, 41 days for “their” equity long-short funds, versus the 28 days demanded by European managers. However, the median in both cases is 30 days. As can be seen from the graph below, there has been a slight increase in the length of the notice period. As in all circumstances, we highlight that the Lipper Tass data may not be fully up-to-date or accurate. In addition, no distinction is made between funds stating a notice period in calendar days versus in business days.

Gates
An often overlooked part of redemption terms by hedge funds is the so-called “gate”, i.e. the maximum level of redemptions allowed at any given redemption date. As hedge fund managers may be afraid that generous liquidity terms will be abused by “hot money”, a gate may provide some protection to the manager and the fund investors to limit outflows. Such gates tend to be 10%, 20%, or 25%. In theory the level of liquidity of the portfolio of the fund determines the level of the gate, given that large sell-offs may in theory lead to the forced selling of illiquid positions and/or leaving a portfolio with a greater portion of illiquid positions as the more liquid positions are sold off first. The counter-argument is that hedge funds get plenty of notice to gradually sell-off illiquid positions and that they usually could borrow modest levels of cash to deal with redemptions and sell off the illiquid positions in a gradual fashion.

Where hedge funds do impose low redemption gates and they do receive redemptions on a particular redemption in excess thereof, it still may not be in their best interest to actually limit these redemptions to the level of the gate. We believe that in such a case a manager should disclose to other investors that a gate has been imposed. If the manager does not disclose this, and other investors do find out, which invariably is the case, the damage to the reputation of the manager may be far greater, than if there is proper communication to the investor base.

Even with such an active investor relations programme, for most investors there is little downside from issuing a redemption notice. In most such cases the hedge fund manager may allow them to cancel the notice and stay in the fund if there is nothing wrong, and if this is not the case, there may be other attractive investment opportunities elsewhere.

Conclusion
As hedge funds’ redemptions terms can form part of the decision making process of investors, it is important to structure these in the right manner. Despite press attention on longer lock-ups, on average we have not been able to identify any trend in changes to any of the terms. However, there is a clear difference between hedge funds run by US and by Europe-based managers/advisors, with the latter providing on average more generous liquidity terms.

Our research does show that for the period January 2003 to December 2005, investors in equity long-short hedge funds of US-based managers were well compensated for the higher liquidity risk, both as to lock-ups and redemption frequency, but that this has not been the case in Europe. In short, the hedge fund market seems to be reasonably efficient.