The final Alternative Investment Fund Managers Directive (AIFMD) essentially contained the same employee/partner remuneration principles as CRD3 but expressed CRD3’s proportionality principle using subtly, but possibly significantly, different wording. Whilst CRD3’s proportionality principle took into account the features of the manager alone, the corresponding provisions of the AIFMD (in Article 13 and Annex II) also take into account the features of the funds that are managed by the manager.
Whilst it is hoped that the European Securities and Markets Authority will apply the AIFMD’s proportionality principle in a similar way to CRD3’s and not apply most of the AIFMD’s remuneration principles in relation to “limited licence/activity” hedge fund managers, there is a risk that hedge fund managers may only enjoy temporary respite from CRD3’s employee/partner remuneration principles until 2013.
I entirely agree that the remuneration principles of CRD3 should have been limited to the banking sector and that the re-insertion of these principles into the AIFMD was ill-thought out and probably politically motivated. What needs to be asked, however, is whether the standard performance fee model should be replaced with a new standard that better aligns the interests of the investor and the manager and, whatever 22nd July 2013 may bring, whether the employee/partner remuneration principles of the AIFMD should actually be applied, not to the remuneration by the manager of its employees and partners, but to the remuneration by the fund of the manager himself.
There are some good arguments for the proposition that the ‘standard’ remuneration model strongly aligns the interests of the manager and investor. The main arguments are:
• The principals of a manager are generally expected to materially co-invest in its funds;
• The performance fee is a percentage (generally 20%) of profits above a high water mark (in other words, performance fees are not payable until the net asset value (NAV) exceeds the highest NAV on which performance fees were paid);
• Hedge fund managers live and die by their performance and are often judged on the most recent set of short-term returns; and
• Subject to the terms of investment, investors can realise their gains, and crystallise the performance fee, by redeeming at any time.
These arguments are convincing but the ‘standard’ performance fee model, like many ‘standards’ in the industry, has been self-fulfilling for a very long time. The fact remains that a fund could have a very good year, and pay a substantial performance fee to its manager, only for performance to be flat or negative in one or more following years. Any investor that chooses to remain in the fund would not have realised any gain but would have effectively paid 20% of an unrealised gain for the privilege. True, no performance fee would be payable to the manager until the high water mark had been reached again, but the investor would need to remain patiently invested in the fund and redeem once this had happened (if it ever did happen) in order to realise the gain for which a fee had already been paid.
Do investors believe that the current performance fee model strongly aligns the interests of the manager and the investor? In a word: no.
CalPERS has suggested that only a portion of performance fees that are accrued in a performance period should be paid to the manager in that period, with the remainder being held in accrual and paid out pro rata over several performance periods. This method would subject the deferred performance fees to a degree of claw back. Similarly, the third core principle of the Alignment of Interests Association (AOI) provides that “performance fees should be deferred and crystallised over a period of years”.
Applying AIFMD principles
I believe that the application of the AIFMD remuneration principles to the performance fee itself would provide a remuneration model that delivers a genuine alignment of interest. Paraphrasing the wording of Annex II of the AIFMD, the key AIFMD principles that could be applied to the performance fee are:
• The assessment of performance should be set in a multi-year framework appropriate to the life-cycle of the relevant fund and be based on longer term performance;
• The actual payment of performance-based components of remuneration should be spread over a period which takes account of the redemption policy of the fund and its investment risks;
• A substantial portion, and in any event at least 50%, of any variable remuneration should consist of units or shares of the relevant fund;
• A substantial portion, and in any event at least 40%, of the variable remuneration component should be deferred over a period which is appropriate in view of the life cycle and redemption policy of the relevant fund and should be correctly aligned with the nature of the risks of that fund. The deferral period should be at least three to five years unless the life cycle of the relevant fund is shorter; and
• Remuneration payable under deferral arrangements should vest no faster than on a pro rata basis.
Applying these principles to the performance fee would meet the requirements of institutional investors (as expressed by CalPERS and AOI above). It would also compel a symbolic and often substantive co-investment by the manager.
Accommodating the principles
In order to accommodate the above principles into a corporate fund structure, one would need to address a number of considerations and issues. Whilst complex, none of the issues should be unsolvable.
The main considerations would be:
• Over what period should the actual payment of the performance fee be spread?
It makes sense to read the first two and last two principles above together. Doing this would seem to suggest that the period for assessing performance and the deferral period should be the same – i.e. at least three years, with option to vest on a pro rata basis.
I could tell you about the life cycle of a frog but it remains to be seen how the life cycle of a fund will be defined. I feel, however, that a deferral period of three years would be too long for a fund with monthly liquidity.
• How should the deferral be effected?
The first assumption is that deferred amounts would consist entirely of shares of the relevant fund. The second assumption is that the manager would not be able to redeem the relevant shares unless sufficient performance had been maintained at the relevant vesting point.
• To what extent, and when, should deferred amounts be clawed back?
Should, for instance, 20% of losses from the year one high water mark be clawed back at each vesting point during the deferral period or only at the end of the deferral period? Should all deferred amounts arising from any performance period be at risk of clawback?
In other words, should losses from the year one high water mark be clawed back only from amounts deferred from the year one performance fee, or should it be possibleto claw back any losses from any deferred amounts?
On a related point, as the value of the relevant shares that are subject to the deferral will have fallen, should clawback be limited to the number of shares that were originally issued at the end of the relevant performance period (and not their value)?
• How should deferred amounts be clawed back and how do you ensure that only the appropriate investors benefit?
A compulsory transfer of shares, followed by a compulsory conversion of shares (as the shares held by the manager would almost certainly not be subject to management and performance fees), could work. This mechanism would also ensure that the clawback is attributed only to those shareholders who, and to the extent that those shareholders, held shares at the end of the relevant performance period.
• How should investors who redeem during a deferral period be treated?
It is true that such an investor would have realised any performance on its investment and that performance can be meaningfully assessed on the date of redemption, but should the investor be able to claw back losses? As the redemption is during a deferral period the manager would have no chance of reversing any losses over the remainder of the period so should any deferred performance fee therefore be forfeited by the investor and paid to the manager?
One can see how an investor could object to this, although if investors want performance to be assessed over a longer term, it is reasonable for the manager to be assessed over that longer term (or at least until the next vesting date during the deferral period).
The main issues for UK managers would be:
• The Investment Manager Exemption (the IME)
A UK manager would need to comply with the conditions of the IME. The deferral of the performance fee itself would not cause any issues under the customary remuneration test. This test, however, may require that the entire performance fee be recognised for UK tax purposes when earned so a UK LLP would carefully need to consider the allocation of the performance fee profit.
The ‘20% test’ would need to be considered. This test broadly prevents the manager and connected persons from being entitled to more than 20% of the fund’s chargeable profit. Whilst the test would normally disregard standard management fees and performance fees, it could take into account deferred performance fees or performance fees that consist of units/shares in the fund.
• Tax would be paid on performance fee earned, not what is actually received
Whilst this is more of an issue at partner/employee level, it could be the case that a UK manager would pay tax on 100% of the performance fee earned in a performance period and yet ultimately could receive only 60% of that performance fee.
Careful tax structuring and planning will therefore be required.
With the AIFMD itself EU member states should be mindful of the potential for performance fees to be deferred in order to ensure that national laws that reflect the AIFMD’s remuneration principles do not require the deferral of partner/employee level remuneration where the performance fee is itself deferred.
On partner/employee level remuneration, a hedge fund manager that does not operate any kind of deferral may, arguably, be able to pay more to attract and retain better talent. I strongly believe, however, that deferral will become (and may have to become) the standard practice, wherever the manager may be based. Already several hedge fund managers are ahead of the curve on this and offer deferred crystallisation of performance fees and/or payment of some or all of the performance fee in locked-up performance fee shares.
There is no doubt that the hedge fund industry is facing a period of rapid change. Notwithstanding the increasing amount of regulation, I believe that the most pressure on the industry to adapt will come from institutional investors. Now is the time for the industry to question whether what was the market standard in the 20th century (where hedge fund investors were predominantly high net worth individuals) should remain so for the 21st century (where hedge fund investors are predominantly institutional investors). The window for hedge fund managers to work with investors to develop better fund structures and business models has certainly not closed.
James Tinworth is Partner in the corporate practice of international law firm Stephenson Harwood. He specialises in investment management and investment funds work with a focus on offshore hedge funds and their management groups.