12

Distribution Waterfall

Risk models for funds can be constructed bottom-up or top-down. In a bottom-up approach the limited partnership agreement's provisions related to the distribution waterfall are often the most complex part to model. The waterfall sets out how distributions from a fund will be split and in which priority and when they will be paid out, i.e. what amount must be distributed to the LPs before the fund managers can take a share from the fund's profits. One immediate reason to model the distribution waterfall is its relationship with the returns of the fund in question. A fund's economics has a significant impact on incentives and, as a consequence, on the behavioural drivers of the fund managers' performance (Mathonet and Meyer, 2007).

The design of the waterfall's terms and conditions is one of few opportunities where LPs can anticipate and manage risk: it will always have effects – sometimes even unintended ones – as it drives motivation and attitude, sense of responsibility, accountability and priorities of fund managers.

Waterfall structures influence incentives significantly and practices can differ significantly between geographies (USA, Europe and Asia) and types of fund (particularly in the case of VC). Notwithstanding these differences, however, this chapter is primarily the modelling of the waterfall in a bottom-up derived fund model. Another application of the principles presented in this chapter could be in the context of different classes of limited partners, such as government investors with subordinated stakes in the fund. This leads to additional refinements which also go beyond the scope of this book.

12.1 IMPORTANCE AS INCENTIVE

The main incentives that align interests between fund managers and their investors are based on management fees, GP investment in the fund, carried interest allocations and distribution provisions.

12.1.1 Waterfall components

These partnership agreement provisions, but also other terms and conditions such as investment limitations, vesting, transfers, withdrawals, indemnification or the handling of conflicts of interest tend to look quite similar between different fund agreements.

• Management fees. The purpose of management fees is to cover the basic costs of running and administering the fund. These costs comprise mainly salaries for investment managers and back-office personnel, expenses related to the development of investments, travel and even entertainment expenses, and office expenses such as rent, furnishings, utilities or supplies. Management fees are nearly always calculated as a percentage – in the case of private equity funds typically between 1% and 2.5% depending on the fund size – of the capital the LPs commit to the fund, but generally taper off after the investment period or when a successor fund is formed. While the management fee's calculation is relatively simple and fairly objective, there are controversies surrounding the finer details.
• GP investment in fund. GPs typically invest a significant amount of capital – typically about 1% – in their funds, which is treated in the same way as that contributed by the limited partners. There are a number of reasons for this, for example that GPs contribute a meaningful amount of capital to ensure their status as a partner of the fund for income tax reasons. More important, however, is putting “skin into the game” to help align the interests between fund managers and their investors.
• Carried interest. Management fees are paid regardless of the fund's performance and therefore fail to provide an incentive to work hard and generate superior returns. Excessive and quasi-guaranteed management fees stimulate tentative and risk-averse behaviour, such as following the herd. Consequently, the carried interest (i.e., the percentage of the profit that goes to the fund managers) is the most powerful incentive to create value. The typical carried interest split is 80/20 and gives the fund managers a share in the fund's net profits that is disproportional to their capital committed and is essential to attracting talented managers.

Somewhat surprisingly, fund terms have been relatively stable across market cycles. One explanation for this phenomenon might be sought in the fact that both fund managers and their investors have sufficient negotiation power to reject “off-market” terms sought by the other side, but not enough leverage to move the market in one direction or the other. In the case of private equity, to some degree the ILPA Private Equity Principles released in September 2009 may have initiated a shift in the power relationship between GPs and LPs but may also lead to further cementing of standardized terms.

12.1.2 Profit and loss

How is a fund's profit figure determined? For instance, the profit and loss can either be aggregated or the GP can be allowed to take a share of the profit on each individual investment. Depending on which approach is taken, it can lead to different amounts of carried interest being paid out to the fund managers.

Participating in every investment's profit can be problematic as the GP can make profits on successful investments but has little exposure to unsuccessful transactions. As the LPs thus cover the bulk of the capital risk, this approach significantly weakens the alignment of interests.

12.1.3 Distribution provisions

The distribution provisions govern the timing and content of payments in respect of the carried interest. While fund terms are by and large stable, the significant exception to this general rule appears to be the set of fund agreement provisions governing the timing and apportionment among the partners, and how to operate distributions.

This multiplicity of approaches arises because no single mechanism can satisfy all the economic goals of both the GP and LPs. Often one party's gain in the arrangement is the other party's loss. As a consequence, negotiations over such distribution provisions for capital and carried interest often are difficult and time-consuming.

12.1.4 Deal-by-deal vs. aggregated returns

Another important parameter is whether returns are aggregated or “deal-by-deal”. The methods we describe in this chapter can be applied in both cases. When aggregating, no distributions are made to the GP until the LPs have received distributions equal to the amount of their overall capital contributions. Only thereafter are distributions made to the LPs and the GP according to the agreed carried interest split. This arrangement provides LPs with the greatest percentage of early distributions and minimizes the possibility that the GP will receive more than its agreed-upon percentage of the fund's cumulative net profits.

The other extreme is the “deal-by-deal” distribution approach where carried interest distributions are made following the return of capital contributions attributable to individual realized investments. From the viewpoint of the fund managers, “deal-by-deal” has a major advantage as it allows them to receive carried interest distributions sooner. It therefore creates perverse incentives for the fund manager to realize successful deals early and to delay the recognition of unsuccessful deals and write-downs of unprofitable investments. As a result, the “deal-by-deal” approach creates a clear possibility of overdistributions to the GP and thus requires a clawback provision.2

12.2 FUND HURDLES

Fund managers cannot take a share in the distributions until the LPs have received aggregate distributions equal to the sum of their capital contributions plus an additional amount determined by the set hurdle.3 Without the hurdle, the GP would receive a “straight carry” and participate in any return of capital in excess of the original investment. The addition of the hurdle provision generally has the effect of further subordinating the GP's right to receive distributions, and is intended to align the interests of the GP and the LPs by giving the fund managers an additional incentive to outperform a traditional investment benchmark.

12.2.1 Hurdle definitions

For the hurdle, only the fund's ultimate performance matters. It does not offer the LPs the power to place the fund in default if it is not paid. From an investor perspective, it protects the return on investment in case of low return and gives the manager the incentive to achieve returns above this threshold. It is a standard term for illiquid funds worldwide, at it ensures that the LPs receive at least as much as they would have made on a safer investment.4

Most partnership agreements foresee that the hurdle rate – typically 8% – is defined on the basis of the compounded interest. In this case the GP has to first return an “amount as is equal to interest at an annual rate of 8% (compounded annually) on the daily amount of the partnership share” to the LPs before receiving carried interest.

A number of partnership agreements foresee that the hurdle rate and the catch-up should be applied just as a simple interest. Here, the preferred return calculation could be defined as a “return of 8% (simple rate) per annum with respect to all unreturned capital contributions made on the partnership shares in excess of the nominal value of such partnership shares”. Compound interest is standard in finance and simple interest is used infrequently. For interest r on a principal P compounded over n periods, the final amount to be returned is A = P* (1 + r)n. For simple interest, i.e. where the interest is not added to the principal, this amount is A = P*r * n. Another approach, which we will discuss later, is to define the hurdle rate not in terms of interest but on the TVPI to be achieved by the fund.

12.2.2 Option character and screening of fund managers

The hurdle gives the limited partnership fund an option-like character. The fund managers as holders of the carried interest, like the holder of a call option, enjoy the possibility of theoretically unlimited upside gain (Rouvinez, 2005). If the fund loses value, the fund managers – unless they invested a significant share of their personal wealth – have neither a gain nor a loss, just like when an option holder declines to exercise an option.

This option-like character is a main argument against hurdles in the case of VC funds. Venture capitalists have to be willing to take large but informed investment risks. Here, a hurdle can give sub-optimal incentives, e.g. large losses early on in a fund's life could put it so far under water that the fund managers are inclined to quit the game altogether or probably worse, fund managers might “swing for the fences” and become overly aggressive with subsequent investments to get back into the zone where they could receive carried interest.

On the other hand, in the case of infrastructure or buyout funds where there is less volatility and the implied cost of capital is different from that of VC funds, setting a hurdle rate is argued for more fervently. Here, doing away with it entirely would create a moral hazard, as fund managers could then receive carried interest for pursuing a low-risk, low-return strategy (Fleischer, 2005).

12.3 BASIC WATERFALL STRUCTURE

It makes a huge difference whether the hurdle is defined as “soft” or “hard”. The “soft” hurdle defines a sharing of all profits if a return of not lower than the hurdle rate is achieved. In this case, once a fund has returned the initial capital plus, say, 8% return, it has cleared the hurdle and thus becomes entitled to take the full carried interest. To achieve this objective, the agreement includes a so-called catch-up provision. Once the hurdle is cleared, profits are then allocated disproportionately to the GP until it catches up to the point where it would have been had it received its carried interest on the entire profit. See Figure 12.1.

The “hard” hurdle, on the other hand, defines a sharing of profits only that are above the hurdle rate. In contrast to the “soft” hurdle, which basically gets “extinguished” once passed, the “hard” hurdle is relevant for all scenarios where the fund's IRR is above the hurdle rate. This arrangement is also sometimes called a “floor” (Fleischer, 2005). In instances where there is a “floor” and therefore no catch-up has been agreed, the carried interest only applies to those net profits that exceed the hurdle.

12.3.1 Soft hurdle

We define a_x to be the amount required as the residual value of a fund's portfolio to give an IRR of x% or a multiple of x before splitting it up according to the waterfall. c is the carried interest due to the general partner, h the hurdle rate and u the catch-up (with u > c). The payout of the portfolio value a attributable to GP and LPs is shown in Table 12.1.

Table 12.1 Payout split

How can one determine the amount to be caught up and the residual amount to be split up as carried interest for u < 100%? In the case of a soft hurdle, the waterfall can be generalized as in Table 12.2.

Table 12.2 Generalized waterfall

The “catch-up” provision allows the fund managers to participate faster in the gains, once the hurdle has been passed, whereas without catch-up they only participate in the proportion of its agreed carried interest. With a 100% catch-up (a “full catch-up”) the hurdle will have no ultimate effect on the carried interest if the fund clearly exceeds its target IRR and does not terminate with returns still in the catch-up zone.

During a fund's life some investments may be exited earlier and distributions can already be made to the GP. However, this may be followed by years of losses, e.g. caused by failures of the underlying projects and portfolio companies or through lack of exit opportunities during prolonged economic downturns. This may mean that the GP receives more than the intended carried interest based on the overall performance of the fund. The GP should not receive profits in excess of the agreed carried interest percentage.5 “Clawback” provisions aim to protect the economic split agreed between the GP and the LPs. The clawback provision is sometimes called a “give-back” or a “look-back”, because it requires a partnership to undergo a final accounting of all its capital and profit distributions at the end of a fund's lifetime.

12.4 EXAMPLES FOR CARRIED INTEREST CALCULATION

We discuss soft and hard hurdle rate-based carried interest calculations for compounded interest and multiples for the example in Table 12.3. In this example, it is assumed that the GP himself holds a 5% stake in the fund.

Table 12.3 Example fund (all amounts in EUR)

The fund's development over its lifetime is depicted in Figure 12.2.

For the purpose of this discussion and for simplicity, we just discuss yearly periods. The approach described bases the waterfall calculation on the previous period and the changes during the year. It aims to determine the split of cash flows for the year and the amount of clawbacks that need to be returned to the LPs in case the fund terminates at year end. Only the shares held by the LPs are affected by the carried interest split and consequently we do not have to consider the GP's own stake in the fund.

12.4.1 Soft hurdle for compounded interest-based carried interest allocation

For this example (Table 12.4) we assume a hurdle rate of 8%, a catch-up of 60% and a carried interest of 20%. Appendix Table 12.A.1 shows the development over the fund's entire lifetime.

Table 12.4 Payout split for example fund (compound interest case)

	IRR range
Hurdle zone	IRR ≤ 8%
Catch-up zone	8% ≤ IRR < 12%
Full carried interest	12% ≤ IRR

How do we determine the amount that the distributions need to exceed the threshold set by the hurdle? To calculate the amount to reach the hurdle at the end of the year we sum up the amount paid-in (i.e., the remaining aggregated contributions until then) in the beginning of the year, the interest payments on this amount and the changes within the year. See Table 12.5.

Table 12.5 Calculation of amount in excess of 8% hurdle (all amounts in EUR)

Table 12.5 warrants the following comments: (6) the interest is paid on the remaining amount paid-in (3)+(4)+(5). If this amount is positive a share of the distributions will be paid-out as carried interest. The paid-in (6) is brought forward to the next period (1). Until year 4 the fund is below its hurdle and the GP is not entitled to carried interest yet. In year 4 there is a significant distribution and the hurdle is exceeded by EUR 276, 408.

In the case of compounded interest considered here there is an alternative way to calculate this amount based on the IRR and the resulting a_8% for the year. The IRR is the discount rate that gives a NPV equal to zero:

where C_n are the contributions and D_n the distributions in period t_n and L the lifetime. The interim IRR is a rough but widely used estimation of IRR performance and in the case of private equity forms the basis of most published comparative performance statistics. For active funds, the IIRR is computed by taking the NAV as the last cash flow at time T:

a_x is the amount required as the residual value of a fund's portfolio to give an IRR of x% before splitting it up according to the waterfall:

This gives

Based on this formula we can calculate the amount to reach the hurdle of 8% as

The net distribution in year 4 is EUR 21, 584, 000 – EUR 15, 390, 950 = EUR 6, 193, 050 exceeding this threshold by EUR 276, 408. As the fund's IIRR in year 4 is 10.4% it is still within the catch-up zone. As a result, 60% of this amount, i.e. 165, 845, is due to the GP. See Figure 12.3.

An alternative way of determining whether the fund is still in its catch-up zone is to determine the amount that exceeds the threshold of 12% in the same manner as for 8% (hurdle and start of catch-up zone). See Table 12.6.

Table 12.6 Calculation of amount in excess of 12% (end of catch-up zone, in EUR)

A complication occurs in year 5, where we have just a contribution but no distribution and the IIRR drops below the hurdle rate (see Figure 12.3). Should the fund end in this year, the carried interest paid in the previous year would have to be clawed back from the fund managers. We assume that clawbacks will only be realized when the fund has come to the end of its lifetime and is wound up. As we see in the example, the fund recovers in later years and eventually no clawbacks are necessary.

In year 8 we have distributions that put the fund again in the carried interest area (see Appendix Table 12.A.1). Should the fund come to the end of its lifetime now, the carried interest of EUR 9, 419, 250 would need to be set off against the clawback of EUR 165, 845 carried interest already paid out and would reduce it to EUR 9, 253, 405. In fact, the fund's repayments put it firmly above the threshold of 12% where the catch-up ends, essentially “extinguishing” the hurdle and giving the fund manager 20% of every repayment in excess of the capital called.

In this case, how do we split the LPs' shares' repayment of EUR 45, 600, 000 between the LPs and the GP? The LPs receive the difference between the LPs' shares' accumulated distributions until year 8 of EUR 121, 129, 750 and the accumulated distributions until year 7 of EUR 84, 949, 000 minus the reduction of clawback from year 7 to year 8 of EUR 165, 845, giving a distribution of EUR 36, 346, 595 in year 8. See Figure 12.4.

Until year 8 the GP has to receive the distribution of its own 5% stake in the fund of EUR 6, 871, 000 plus the carried interest of EUR 9, 419, 250. For year 8 the GP receives the difference between his accumulated distribution until year 8 of EUR 16, 290, 250 and his accumulated distribution until year 7 of EUR 4, 471, 000 minus the change in clawback from year 7 to year 8 of EUR 165, 845, giving a distribution of EUR 11, 653, 405 in year 8.

The fund's lifetime ends in year 10, where overall it has generated a return of 53%. The GP receives the full return of its own 5% stake in the fund of EUR 8, 671, 000 plus a carried interest of EUR 16, 259, 250, which is 20% of the difference between the LPs' shares' accumulated distributions of EUR 164, 749, 000 and the accumulated contributions of EUR 83, 452, 750 (see Table 12.3). The LPs receive EUR 83, 452, 750, i.e. the full amount of the distributions until the accumulated contributions are fully repaid plus 80% of the excess distributions, giving them a total of EUR 148, 489, 750.

12.4.2 Hard hurdle for compounded interest-based carried interest allocation

In the case of the hard hurdle we base the calculation of the carried interest again on the amount by which the accumulated distributions exceed the threshold set by the hurdle of 8%. As in the example before, for year 4 this amount is EUR 276, 408. However, with a hard hurdle there is no catch-up and the GP only participates in distributions above 8%. In other words, the hurdle is never “extinguished”, resulting in an overall lower aggregated carried interest for the GP. See Figure 12.5.

Compared to the aggregated carried interest of EUR 16, 259, 250 in the case of the soft hurdle, here the GP would just receive an aggregated amount of EUR 15, 736, 450. Note that just the amount is reduced compared to the soft hurdle case, but the timing when the GP receives carried interest (and also when it is clawed back) remains the same. Appendix Table 12.A.1 shows the development over the fund's entire lifetime.

12.4.3 Soft hurdle for multiple-based carried interest allocation

Another, albeit less often used, approach is a multiple-based hurdle rate. Here again, in principle, a soft and a hard hurdle are possible. For the example, we assume a hurdle multiple of 1.5 × and again a catch-up of 60% and a carried interest of 20%. Note that it is not intended to draw a comparison between the payoffs of a compound interest and a multiple-based carried interest scheme. The purpose of this example is just to compare the calculation approaches.

How do we determine the end of the catch-up in this case? We follow the same approach as for the compounded interest in the soft hurdle case. For a hurdle multiple m_h = 1.5 × the catch-up is between

See Table 12.7.

Table 12.7 Payout split for example fund (multiple case)

	TVPI range
Hurdle zone	TVPI ≤ 1.5
Catch-up zone	1.5 ≤ TVPI < 1.75
Full carried interest	1.75 ≤ TVPI

How do we determine for period t_n the threshold after which the GP receives carried interest? Accumulated distributions have to exceed

In the example, this does not happen before year 8 when finally the threshold set by m_h is exceeded by EUR 5, 369, 875. With an interim TVPI (ITVPI) of 1.56 this is still within the catch-up, which results in a carried interest payment of EUR 3, 221, 925.

As Figure 12.6 demonstrates, the ITVPIs develop more “steadily” and are less “volatile” than the IIRRs. Consequently, clawbacks are possible but less likely for a multiple-based carried interest.

When the end of the catch-up zone is reached in year 10, the hurdle is “extinguished” and the GP has received the same total amount of carried interest of EUR 16, 259, 250 as in the case of the soft hurdle for compounded interest-based carried interest allocation. See Figure 12.7.

For all figures over the fund's lifetime, please refer to Appendix Table 12.A.3.

12.4.4 Hard hurdle for multiple-based carried interest allocation

In the case of the hard hurdle we base the calculation of the carried interest again on the amount by which the accumulated distributions exceed the threshold set by the hurdle multiple m_h = 1.5 ×. As in the example before, this does not happen before year 8 where the threshold set by m_h is exceeded by EUR 5, 369, 875.

With a hard hurdle there is no catch-up and the GP only participates in distributions above the target multiple and thus just receives EUR 1, 073, 975. See Figure 12.8.

Again in the hard hurdle case the aggregated carried interest of eventually EUR 7, 913, 975 is lower than that of EUR 16, 259, 250 in the soft hurdle case. The timing when the GP receives carried interest remains the same. For all figures over the fund's lifetime, please refer to Appendix Table 12.A.4.

12.5 CONCLUSIONS

In this chapter, we have presented the major principles of the waterfall determining the investment profits of the LPs and the GP. However, over the years, many new layers have been added to the basic approach making it increasingly complex and difficult to model the waterfall. CPEE (2004) found that some “newer GPs (who can tend to have fewer financial staff) might not even understand their own waterfall, let alone their LPs.” Thus, the examples in this chapter can only be a broad description of the key fundamentals.

In determining the waterfall of the portfolio of illiquid fund investments, LPs can in principle follow either a bottom-up or top-down approach. A bottom-up approach requires tailor-made models for each fund to capture their specific terms, conditions and the range of possible variables. For risk management purposes, this may be too cumbersome, however, and in most situations a top-down modelling approach can be the preferred solution.

APPENDIX – EXAMPLES

Table 12.A.1 Example compounded interest-based carry calculation for soft hurdle 8% with 60% catch-up (all amounts in EUR)

Table 12.A.2 Example compounded interest-based carry calculation for hard hurdle 8% (all amounts in EUR)

Table 12.A.3 Example multiple-based carried interest calculation for soft hurdle 1.5× with 60% catch-up (all amounts in EUR)

Table 12.A.4 Example multiple-based carried interest calculation for hard hurdle 1.5× (all amounts in EUR)

1 Available from http://www.evca.eu/toolbox/glossary.aspx?id=982 [accessed 24 July, 2009].

2 The ILPA Private Equity Principles are commonly regarded as a response to the market excesses in the mid-2000s. Following these recommendations, “deal-by-deal” carried interest looks set to further lose in significance. However, it remains to be seen whether the current trend remains intact once a new investment cycle gains momentum.

3 “Hurdle” is often used interchangeably with the term “preferred return”, although CPEE (2004) interprets the preferred return as the limited partners' downside protection only, not as any sort of incentive for the fund managers.

4 Hurdle rates typically range from 5% to 10% and are often tied to a spread over risk-free rates. Whether setting a hurdle makes sense or not ultimately depends on the importance of deal flow relative to deal harvesting. In cases where deal flow incentives are relevant, e.g. in the case of buyout or mezzanine funds, the fund managers should not be rewarded for investments that do not return at least the investor's cost of capital. Setting a hurdle achieves this objective, whereas a straight carried interest creates an incentive to go for low-risk, low-return investments. For VC funds, however, deal harvesting is clearly more relevant, and with the hurdle incentives they are even distorted when the option on the carried interest is “out-of-the-money”. Here, a straight carried interest is more efficient than a hurdle in providing the proper incentives to the fund managers.

5 When talking about clawbacks, usually “GP clawbacks” are being referred to, i.e. corrective payments to prevent a windfall to the fund managers. Nevertheless, there can be, albeit rarely, situations where LPs have received more than their agreed percentage of carried interest (Mathonet and Meyer, 2007). Consequently, some partnership agreements address the question of the so-called “LP clawback” as well. LPs aim to minimize the risk that the GP lacks liquid assets and the clawback right would be unenforceable. The simplest and, from the viewpoint of the LPs, the most desirable solution is that the GP does not receive carried interest until all invested capital has been repaid to investors. But that can take several years before the fund's team sees any gains and it could demotivate the individuals. An accepted compromise for securing the clawback obligation is to put a fixed percentage, e.g. 25%, 30% or 50%, of their carried interest proceeds into an escrow account as a buffer against potential clawback liability.