4

Investing in Illiquid Assets through Limited Partnership Funds

Understanding the key characteristics of limited partnerships is critical for effectively measuring and managing the risks of investing in private equity and real assets. In this chapter, therefore, we explain the basic characteristics of this structure, which has remained the most common form for long-term investing in these asset classes. As we shall see and also discuss in more detail in Chapter 5, the very features of this structure are essential to reap the illiquidity risk premium that private equity and real assets offer to investors. Limited partnerships have not remained uncontroversial, however, which explains why alternative investment forms have emerged, including listed vehicles, direct and co-investments, and deal-by-deal investments. However, notwithstanding its constraints, the limited partnership structure has stood the test of time and provides a superior framework for long-term investing, which is typically associated with a high degree of uncertainty.

4.1 LIMITED PARTNERSHIP FUNDS

Limited partnership funds are unregistered investment vehicles that pool capital for investing in private equity and real assets. These funds are set up by fund management companies – also referred to as “firms” – to attract institutional investors. Funds fulfil a number of functions. They allow the investment process to be delegated to fund managers, who have significant experience in screening, evaluating and selecting investment opportunities with high expected growth potential. This potential is generally reaped through a combination of strategic, tactical and financial measures, requiring particular skills in controlling, coaching and monitoring portfolio companies' management. Finally, fund managers source, exit opportunities and realize capital gains on disposing investments. In managing a fund, managers are usually supported by advisory committees and a network of industry experts, sharing knowledge and other non-financial resources.

4.1.1 Basic setup

A partnership is a contract between two or more individuals who agree to carry on an enterprise, contribute to it by combining property, knowledge or management, and share its profit. The most basic form of partnership is a general partnership. In a general partnership, all partners manage the business and are personally liable for its debts, as every partner is both an agent and a principal of the firm, thus binding the firm and the other partners. As these liabilities can be significant, another “asymmetric” form for investment vehicles has evolved: the limited partnership (see Figure 4.1). In a limited partnership, the “limited partners” relinquish their ability to manage the business. At the same time, however, their liability for the partnership's debts is limited. Additional factors that have played an important role in the emergence of the limited partnership as a dominant investment vehicle for long-term investors lie in tax considerations and regulatory requirements, with a high degree of transparency allowing investors to be treated as investing directly in the underlying assets.

Figure 4.1 Basic limited partnership fund structure.

The limited partnership was expressly designed to be tax effective under US law. Commonly, domestic private equity funds in the United States are limited partnerships under the law of the State of Delaware. A Delaware Limited Partnership is a separate legal entity that continues as such until it is dissolved and winds up its affairs pursuant to the partnership agreement. While limited partnerships organized in Delaware are not generally required to register with any regulatory authority, the management company of the fund, which is normally organized as a separate entity, may be subject to registration as an investment advisor.

Importantly, limited partnerships under the law of the State of Delaware are tax transparent, i.e. “look-through” entities. This allows the limited partners to be taxed on capital gains in their own jurisdiction, which is a major advantage for those who are wholly or partly tax exempt in their home jurisdiction. Because cash flows freely both in and out of the fund as investments are bought and realized, investors do not suffer any tax leakage they may not be able to reclaim. Moreover, limited partnerships satisfy various legal requirements for regulated investors (for example under the Employment Retirement Income Security Act (ERISA) in the United States).

Similar structures exist in other countries. In the United Kingdom, the largest market for private equity investments in Europe, funds organized as limited partnerships must be registered under the Limited Partnerships Act of 1907. Under UK law, the liability of the limited partners is limited to the amount of capital in the partnership as long as they do not take part in its management. UK limited partnerships are similarly tax transparent, a key advantage for investors in such funds.

Consistent with common practice, in this book we use the term “general partner” to refer to the firm as an entity that is legally responsible for managing the fund's investments and with unlimited personal liability for its debts and obligations. “Fund managers” are the individuals involved in its day-to-day management. The group of fund managers forms the fund's “management team” which includes the carried interest holders, i.e. those employees or directors of the general partner that are entitled to share in the “super profit” made by the fund. The term “limited partners” refers to the fund's passive investors.

4.1.2 The limited partnership structure

Whereas terms and conditions, investor rights and obligations are defined in specific non-standard partnership agreements, the limited partnership structure – or comparable structures used in different jurisdictions – has evolved over the last decades into a “quasi-standard”.

• Investors – who are mainly institutions with appropriate liability structures (see Chapter 2) – serve as LPs, committing a certain amount of capital to the fund. As LPs, they have little, if any, influence on the day-to-day management of the fund. However, LPs are given some oversight of the fund, i.e. by inviting them to serve on advisory boards or on special committees while still being protected from general liability.
• Depending on the strategy followed, funds have a contractually limited life ranging from typically 7–10 years up to 15 years. No early redemption rights are granted to the LPs. Moreover, secondary sales of the fund's shares are generally subject to the approval of the GP.1
• The fund manager's objective is to realize all investments before or at the liquidation of the partnership. Limited Partnership Agreements (LPAs) often have provision for an extension of 1 or 2 years (sometimes even longer).
• The main part of the capital is drawn down as “contributions” during the “investment period”, typically 3 to 5 years, when new investment opportunities are identified. Sometimes, there is provision for an extension of 1 year. During the “divestment period”, only the existing and successful investments will be further supported with some follow-on funding in order to maximize the value creation until the final exit. The manager's efforts during this time are concentrated on realizing or selling the investments and “distributing” the proceeds to the LPs.
• Management fees are a function of the size of the fund and the resources required to implement the proposed strategy. They generally range from 2.5% of committed capital for funds of less than €250 million, between 1.0% and 1.5% for large buyout funds, to 50 bps and 150 bps for funds-of-funds. The fees are often scaled down once the investment period has been completed and adjusted according to the proportion of the portfolio that has been divested.
• Furthermore, some GPs charge transaction and monitoring fees to the companies they have acquired. These fees are not easily visible to the LPs as they are taken directly out of the portfolio companies.
• Management fees are supposed to cover the expenses of running a private partnership. The fund manager himself typically earns a relatively low base salary. His main financial incentive results from the “carried interest” (often simply referred to as “carry”) he may earn. Carried interest usually amounts to 20% of the profits realized by the fund, although a small number of GPs (mostly top US VC firms) charge higher carry of up to 30%. Funds-of-funds, by contrast, are paid less carry.
• Carried interest is usually subject to a “hurdle rate”. Only if this hurdle rate is met or exceeded do GPs receive carry. While in the United States carry is usually determined on a deal-by-deal basis, subject to a claw-back clause in the LPA, in Europe carry is not normally paid out until all LPs' capital has been returned and the hurdle rate is met. Thereafter, normally 100% of additional returns go to the GP. This catch-up period ends when the agreed carried interest split is reached. After that point, distributions are shared according to the agreed split.
• Fees and carried interest have a meaningful impact on the net returns for LPs. While carried interest is related to the performance of a fund, management and transaction fees are not. Metrick and Yasuda (2009) estimate that on average about 60% of a GP's compensation is due to non-performance-related management and transaction fees.
• GPs usually invest a significant share of their personal wealth in the fund to ensure a close alignment of interest with the LPs.
• Commitments are drawn down as needed, i.e. “just-in-time” to make investments or to pay costs, expenses or management fees. Funds typically do not retain a pool of uninvested capital, and the timeframe for LPs to respond to the GP's capital call is generally very short, often only 10 days or even less.
• When realizations are made or interest payments and dividends are received, they are distributed to investors as soon as practicable. Thus, the fund is “self-liquidating” as the underlying investments are realized. However, LPs typically have no right to demand that realizations be made, and these returns come mostly in the second half of the fund's life up to its final liquidation (Figure 4.2). Distributions can also be made “in kind” in the form of securities of a portfolio company, which normally requires that these securities are publicly tradable.
• These broad principles generally also apply to funds-of-funds. The manager of a fund-of-funds acts as a GP. The GP earns a management fee and participates in the investment profits, subject to a hurdle rate. At the same time, the fund-of-funds serves as a LP in the partnerships it commits capital to. While funds-of-funds add a cost layer, they may be especially attractive for institutional investors who lack the skills and experience in selecting the best funds, have only limited access to them, lack the resources to monitor different market segments on a global basis and/or lack the necessary (but often overlooked) portfolio and risk management systems.

Figure 4.2 Typical limited partnership fund lifecycle.

The LPA defines the legal framework and the terms and conditions for the limited partnership fund. It mainly addresses the allocation of capital gains or losses among partners; the allocation of interim distributions; management fees to the general partner; possible investment restrictions; and major governance issues. The management company enters into agreements with all employees and with the general partners. One management company can act as a “group”, managing several such partnerships in parallel.

4.1.3 Is “defaulting” an option for limited partners?

Theoretically, as Litvak (2004) argues, LPs have the option to abandon their investment simply by defaulting, in which case the exercise price of the option is the default penalty that should be weighted against the undrawn commitments that can be saved. Litvak concludes that the threat of capital withdrawal is a useful contractual tool to reduce agency costs between investors and low-quality firms. However, Fleischer (2004) points out that default involves a “reputational” penalty investors suffer when they exercise this “walk-away” option. This penalty may be substantial as a defaulting investor might not be allowed to invest in other capital funds. In practice, the repercussions of becoming a defaulting investor go beyond the penalties described in the limited partnership agreement. Fleischer (2004) argues that the reputational costs of a default may vary among the different investor types within a fund. Pension funds, university endowments and other repeat players care deeply about their reputation, while some individuals and corporate investors might be indifferent. The private asset market has been described as relying heavily on informal relationships and the reputation of its participants. Repeatedly defaulting damages a LP's reputation as a reliable investor and almost certainly undermines future attempts to participate in partnership funds, especially those that are oversubscribed. The negative consequences for defaulting investors in this networked industry can thus substantially outweigh the potential savings that may arise from simply walking away from an underperforming fund. Instead, as we discuss in Chapter 6, a LP is likely to prefer to sell its stake in a fund in the secondary market.

4.2 LIMITED PARTNERSHIPS AS STRUCTURES TO ADDRESS UNCERTAINTY AND ENSURE CONTROL

Despite its virtues, the limited partnership structure has often been criticized. Importantly, it has been argued that in a typical partnership managers need to show quick results to prove they deserve commitments to a new fund they plan to raise. However, this works against the raison d'être of investments in areas where patient capital is particularly needed. One frequently cited example is VC, where deals are sometimes found to be exited prematurely relative to what would have been the optimal holding period of the investment. In principle, the solution may be sought in “evergreen funds”; that is, funds which do not mature but instead have an – at least theoretically – infinite lifetime. But why does the alternative asset industry still use what Love (2009) calls “archaic LP structures that are spectacularly ill-suited” to long-term investing? One answer may be sought in the liability structure of DB pension plans and insurance firms, the largest investors in limited partnerships. Their horizon typically involves 10–15 years, which more or less coincides with the lifetime of a limited partnership fund. More generally, however, it is the particular way the limited partnership addresses uncertainty and grants control rights to investors.

4.2.1 Addressing uncertainty

The question “why are funds not evergreen?” could be compared to the question “why do publicly quoted companies pay dividends?” According to the dividend irrelevance theory, dividend payments should have little, if any, impact on the stock price as investors can always liquidate a portion of their portfolio of equities if they prefer cash.2 For instance, Apple, the world's most valuable company in terms of market capitalization (third quarter 2012), did not pay dividends for 17 years until 16 August 2012, when a new dividend programme entered into force. However, dividends may serve an important signalling function with respect to the financial well-being of a company.

Similarly, there is a strong argument to be made for self-liquidating funds in private equity and real assets. Take VC, for example, where valuations are notoriously difficult and are often viewed by investors with considerable doubt. To be sure, it is not that the fund managers are not seen as trustworthy; rather, it is the recognition that in any appraised asset class valuations are highly judgmental. The only way to give investors confidence is by exiting portfolio companies, thus liquidating investments and showing investors that valuations are “real”.

In coping with the extreme uncertainty of innovation or with a rapidly changing economic environment, the fund's limited lifetime forces its managers to regularly return to the capital markets and ask existing or new investors to back the next fund. Rather than having a “magic formula” to guarantee success, fund managers periodically need to convince investors that they provide superior investment opportunities. This results in high evolutionary pressure and requires fund managers to listen closely to their investors and adapt rapidly.

4.2.2 Control from the limited partner perspective

While fund managers may prefer evergreen funds as such structures give them the opportunity to raise fresh capital on a permanent basis, investors often shy away from being locked up under such conditions. In contrast to evergreen funds, limited partnerships follow a fundraising cycle in the sense that GPs return to the market at – more or less – regular intervals to raise capital for a new fund. Typically, this interval is around 4 years, although it became significantly shorter during the last investment boom in the mid-2000s.

The future of the GP depends essentially on the success of his fundraising efforts. To the extent that funds become smaller, management fees decline, requiring the general partner to reduce its cost base. In the extreme case, where no new funds are raised at all, full liquidation eventually becomes inevitable as management fees on old funds under management dry up. Heikkilä (2004) views this periodic liquidation of private equity funds as essential from the viewpoint of their limited partners, because the exit and reinvestment cycle allows them to withdraw capital from less competent fund managers or managers whose industry expertise has become obsolete. It also allows setting back the clock for new investors, who do not need to value and pay for an existing portfolio. The fund management team's track record and reputation are critical for the successful closure of follow-on funds. Typically, limited partnership agreements do not allow follow-on funds with the same strategy by the same manager before the end of the investment period or before a high percentage of the active fund is invested.

LPs can exercise “evolutionary pressure” through the selection and monitoring process, which is an essential component of the “alternative asset ecosystem”. Obviously, this pressure would not exist if investors in funds also had management control. The separation of the GP and LP functions in a limited partnership is therefore a critical governance characteristic of such structures.

4.3 THE LIMITED PARTNERSHIP FUND'S ILLIQUIDITY

The liquidity of an asset – or lack of it – is generally viewed as its ability to be sold and converted into cash with a minimum loss in value. This loss in value corresponds to transaction costs. There is no specific amount attached to these transaction costs that would draw a line between liquid and illiquid assets. Therefore, assets are in relative terms more or less liquid than others. As we discussed in the previous chapter, illiquidity can occur even in capital markets that may be considered “deep”. One example is the corporate debt market. But even in the public stock market, where trades typically occur on a high-frequency basis, there may be situations where positions for a specific security are too large compared to its daily trading volumes. Apart from this “blockage factor”, some assets have very specific characteristics, making it difficult for the seller to find an acquirer.

Limited partnership funds belong to the most illiquid assets. As explained above, their illiquidity is known ex ante, as opposed to asset classes whose liquidity may dry up ex post in periods of financial turmoil. Investors in limited partnerships deliberately accept illiquidity in order to harvest an illiquidity risk premium (Pastor and Stambaugh, 2003; Acharya and Pedersen, 2005). With the required return being inversely related to the liquidity of an asset, the risk premium for limited partnerships tends to be relatively high (see Chapter 5). Importantly, the high degree of illiquidity of limited partnerships is not just an unintended consequence of its general contractual structure. Instead, illiquidity is the purpose of such investment vehicles, preventing the emergence of a secondary market as an accounting standard setter.3

4.3.1 Illiquidity as the source of the expected upside

Thanks to the institutional capital provided outside listed markets, unquoted portfolio companies are able to sustain difficulties or successes without having to release information that could be detrimental to the success of the investment due to the adverse behaviour of competitors, suppliers or clients. Information is provided to the LPs in strict confidence relating to specific aspects of the development of underlying investments. In fact, in private equity a key purpose of the fund structure is to shield fledgling portfolio companies in their early stages and those that are being restructured in turnaround situations from disruptive market influences. Similar considerations apply to explorations and green field projects that are particularly susceptible to sudden changes in investor sentiment and with a high risk of being cut off from sources of funding. To allow the value of such investments to increase, financial resources need to be provided on a sustained and predictable basis within a reasonably long timeframe. From this perspective, it is conceptually problematic to look at the market value of a fund's underlying investments, especially at an early investment stage. This implies that market values come to existence and thus risk are controlled at the level of the fund rather than the level of the underlying investments.

4.3.2 The market for lemons

From a strictly legal viewpoint, limited partnership shares are illiquid, although – as we discuss in Chapter 6 – secondary transactions do take place in practice. However, the secondary market may to some degree be subject to what is known as the “lemons problem”. Coined by Akerlof (1970) more than four decades ago, the “lemons problem” exists because of asymmetric information between the buyer and the seller. In the used car market – the example Akerlof gives – the buyer usually does not know beforehand whether it is a good car or a bad one (a “lemon”). Therefore, his best guess is that the car is of average quality, and he will be willing to pay for it only the price of a car of known average quality. This means that the seller of a good used car will be unable to get a high enough price to make selling that car worthwhile. As a result, he will not place his car on the used car market.

The same dynamic can be observed in other markets, which are subject to asymmetric information. In asset markets, where there are doubts about asset quality, the exit of the highest-quality seller leads to a reduction in the market price. His exit triggers another exit wave by sellers with somewhat lower-quality assets, resulting in a further decline in the market price. Eventually, and sometimes quickly, the market may turn from an efficient, high-volume one to a transaction-less market (Tirole, 2011).

Investors who are not a LP in a limited partnership fund are faced with the lemons problem, because they have less information on the quality of the fund. This lack of information makes them less willing to pay what may actually be a fair price for a fund share and under certain market conditions they might not be willing to buy a share at all. Importantly, this does not mean that the fund is actually a “lemon”. But information asymmetries can result in significant spreads between the asking price and the bid price, potentially resulting in few or even no transactions.

The extent to which the “lemons problem” plays a role in secondary transactions in the private equity market remains unclear. Some specialized players maintain proprietary databases with detailed intelligence on funds and pro-actively seek to source deals. Information asymmetries, to the extent they exist, should be relatively small. In fact, in individual cases potential buyers may even possess superior information about a fund or a portfolio of funds a LP wants to sell.

A related issue concerns the potentially adverse signal a secondary transaction may send to the investor community in terms of the perceived quality of the fund and its manager. However, this issue can be addressed by selling only a part of a stake in the fund, ensuring that the seller maintains his role as a LP. Signalling his confidence in the fund manager and maintaining his relationship with the GP, the LP retains the option to commit to new partnerships formed in the future.

4.3.3 Contractual illiquidity

Notwithstanding the restrictions regarding the transferability of shares stipulated in the limited partnership agreement, GPs usually give their consent to secondary transactions. However, Lerner and Schoar (2002) argue that private equity fund managers, by choosing the degree of illiquidity of the security, can influence the type of investors the firm will attract. This allows them to screen for “deep-pocket” investors who have a comparatively low probability of facing a liquidity shock. Such investors can ease the GP's fundraising efforts in future fund-raising rounds. Importantly, GPs also face a lemons problem, in the sense that information about a LP's quality is asymmetric between existing LPs in their funds and the rest of the investor community (Lerner and Schoar, 2002).

4.3.4 Inability to value properly

Settlement prices in the secondary market are determined by a host of different factors, including asymmetric information, negotiation skills of the seller and the buyer, the pressure to close a deal, to name just a few. Furthermore, there may be different preferences. For example, whereas the seller may place a higher premium on short-term liquidity, the buyer may attach greater importance to the long-term performance, irrespective of holding period. Since the true economic price for an underlying asset cannot be observed, market participants have to rely on particular appraisal techniques. Indeed, many alternative assets may be described as an appraised asset class, where the market value is determined by a small number of experts rather than a large number of sellers and buyers.

4.3.5 Endowment effect

The “endowment effect”, also known as the “divestiture aversion effect” or “status quo bias”, may also contribute to the illiquidity of funds. In behavioural economics, the endowment effect refers to the experimental observation that people generally attach a considerably higher value to a good that they own than the price they are prepared to pay to acquire that good (Thaler, 1980). Thus, subjective considerations often stand in the way of trading an asset. A frequently cited example of the endowment effect is a family's reluctance to sell an old painting, which has been in their possession for a considerable amount of time, (more or less) regardless of the price that is being offered.

While the endowment effect has been shown in various behavioural experiments, it is less clear to what extent this effect exists in financial markets where traders are motivated by the objective to maximize profits (Arlen et al., 2002). Instead, behavioural biases may be attributed to “loss aversion”, i.e. investors' tendency to attach greater importance to losses than to gains. In private equity and real assets, however, it is important to take into account that interactions between LPs in a fund and the fund manager are particularly close, which differentiates these asset classes from publicly traded assets. With access to funds sometimes being restricted and mutual trust being built up over several years, investors in a fund may be reluctant to terminate a relationship – unless there are overwhelmingly clear reasons to do so. Thus, the endowment effect may well play a role in illiquid asset classes, given their particular market characteristics (Blake, 2008).

4.4 CRITICISMS OF THE LIMITED PARTNERSHIP STRUCTURE

Investors in private equity and real assets are generally aware of the illiquid nature of limited partnerships. They deliberately accept illiquidity to harvest a premium, with the limited partnership providing an adequate framework. Illiquidity in private equity and real assets thus does not represent a market failure but is an essential characteristic to generate excess returns. Nevertheless, it is sometimes argued – especially in market downturns – that the limited partnership model is broken and destined to disappear (e.g., Scott, 2012).

Furthermore, some fund managers have argued that periodic fundraising requires a substantial amount of time and resources, which could be better spent on sourcing and making investments. In fact, this was an important motivation behind the public listing of investment vehicles at the peak of the last cycle, with the capital raised by the private equity firms being invested in their funds. Moreover, in many situations – notably early-stage VC transactions – the usual lifetime of limited partnership funds may not be sufficient to develop the underlying portfolio companies owned by the fund and reap their full potential. Therefore, some GPs have called for a longer fund life or have advocated, as discussed before, evergreen structures.

LPs have also expressed concern about the limited partnership as the dominant investment vehicle in private equity and real assets, albeit from the opposite angle. Locking up capital for a period of 10 years or more plays a key role in this regard, given investors' reduced ability to rebalance their portfolios. This consideration is particularly important in periods of heightened economic uncertainty.

Another factor emphasized by some LPs, such as David Swensen of the Yale University Endowment, is the less-than-perfect alignment of interest between LPs and GPs. As mentioned earlier, Metrick and Yasuda (2009) and Phalippou (2009) find that non-performance-related management fees account for around 60% of the total compensation of fund managers. Among other things, this compensation structure, which has come under increased pressure from LPs after the recent downturn, has been identified as a possible disincentive for GPs to divest underperforming portfolio companies.

Finally, the limited lifespan of limited partnerships does not allow an investor to build up reserves. As Achleitner and Albrecht (2011) point out, listed stocks can be held over very long time periods, allowing them to build up considerable reserves. These reserves, called unrealized capital gains and losses (UCGL), are linked to inflation and the overall growth of the economy. They are of considerable importance for life insurance companies, which can smooth returns and help achieve the predictable and stable investment results expected by both shareholders and life insurance holders by realizing UCGLs.

4.5 COMPETING APPROACHES TO INVESTING IN PRIVATE EQUITY AND REAL ASSETS

Consequently, investors are continuously looking for ways to avoid or mitigate the limited partnership's existing or perceived shortcomings.4 In addition to the evergreen funds discussed before, a range of competing approaches have been marketed to investors over the years. However, these approaches have generally produced mixed results.

4.5.1 Listed vehicles

Listed vehicles appear to be a way to combine the advantages of alternative assets with relatively high liquidity. Resembling investment-trust structures, listed vehicles address the hurdles of regulatory requirements, tax efficiency and transparency. Furthermore, they provide immediate access to a diversified portfolio and management expertise. Moreover, institutional investors that are new to alternative assets often use such products to gain insights into the market before deciding to set up their own investment programme.

However, liquidity may be strictly limited for listed vehicles as well, given that share price discounts can make it difficult, if not impossible, to divest without incurring significant losses. While the lack of liquidity should generally be reflected in the market price, extreme discounts are not unusual, with thin markets resulting in high bid–ask spreads. Potential buyers who are interested in acquiring relatively large positions will typically find it difficult to find sellers. This may take a prolonged period of time, possibly pushing up prices significantly. Likewise, potential sellers may incur considerable losses as the market is able to absorb the increased supply only at considerable discounts.

Furthermore, while listed vehicles increase investors' flexibility in tactical asset allocations and permit them to accelerate their exposure to alternative investments, interests are less well aligned compared with traditional partnerships where investors exercise tight control over the use of funding. Essentially, listed vehicles display the same characteristics as public stocks, where control is much less direct than in limited partnerships.

Finally, a distinction has to be drawn between investments in listed vehicles and investments in publicly listed alternative asset managers themselves. As far as the latter are concerned, investors hold shares in the management company, which may serve as a GP in limited partnerships. In fact, several of the largest private equity firms, which have transformed themselves into alternative asset managers, have gone public (e.g., Blackstone Group, Carlyle Group, KKR). However, to the extent that investors acquire shares in the management company, they are exposed to all asset classes the management company is involved in, such as private equity, infrastructure, real estate, credit and hedge funds. This may or may not be in the interest of the investor. The same applies to investments in benchmarks of publicly listed private equity, such as the LPX 50 index, which includes both listed vehicles and listed management companies.

4.5.2 Direct investments

Private equity and real assets are relatively expensive asset classes, with the compensation structures in traditional limited partnerships having a profound impact on returns for LPs. Thus, some investors have looked for alternative ways to get exposed to alternative assets while avoiding paying high management fees and carried interest. However, cost savings may not be the only reason for pursuing direct investments. Other factors, such as better market timing, may also play an important and sometimes even the primary role (Fang et al., 2012). A prominent example of a limited partner going direct is OMERS PE, the private equity investing arm of the Ontario Municipal Employees Retirement System, a CD 55 billion pension system. In 2004, OMERS PE initiated a major shift from private equity fund investments to direct investments, with the latter targeted to ultimately represent 85% of their total exposure to private equity (Witkowsky, 2012). This strategy has found followers especially in other Canadian pension funds, which increasingly act as investment companies, spearheading an approach that pushes for more control and cost reductions.

Fang et al. (2012) find that direct investments have actually significantly outperformed standard fund investment benchmarks. Nevertheless, important questions remain as to the applicability of this approach within a wider investor community. In fact, it has been tried before: such “captive” structures represented an important category of investors until the end of the 1990s in Europe. However, the share of these players in the market decreased significantly because of a number of challenges namely: (i) to create a governance structure that gives investment authority and decision-making power to the investment professionals; (ii) to attract other investors in the structure, given the potential conflicts of interest with the parent company; (iii) to build and retain a team of talented individuals; and (iv) to manage the difficult interaction of these entities with the rest of the parent company. However, addressing these challenges in an appropriate way may also entail significant costs, and it is not clear whether running an in-house direct investment programme that can reasonably expect to achieve the same gross returns will actually be cheaper than investing through funds as intermediaries.

4.5.3 Deal-by-deal

Some firms offer their investors the option to come in on a deal-by-deal basis, an approach that shares certain characteristics with the direct investment approach. At first glance, this seems highly attractive from a LP's standpoint as it gives them the possibility to: (i) opt out, for example, in situations where they are liquidity-constrained; (ii) reduce management fees; and (iii) build up a portfolio that is more in line with their targeted asset composition. However, following a deal-by-deal investment approach requires fast decision-making processes and investment professionals with direct investment experience, something which is beyond most institutional investors' capabilities. From the fund manager's perspective, the deal-by-deal model is also subject to considerable challenges. Importantly, fund managers need to ensure that their investors who participate in such non-binding arrangements are not just strategic “deal-flow watchers” but are sufficiently committed. Overall, deal-by-deal arrangements have failed to gain traction in most markets and there are few signs that this will change in the foreseeable future (Romaine, 2012). Thus, the Middle East, where deal-by-deal arrangements have traditionally played a more prominent role, looks set to remain an exception.

4.5.4 Co-investments

Co-investing has emerged as an increasingly accepted way of getting around the shortcomings of the limited partnership structure. Co-investing entails the syndication of financing between a fund and one or more of its LPs. In principle, this brings complementary capabilities of GPs and LPs together. A fund may be too small to acquire an asset whose risk-adjusted returns are viewed to be particularly attractive. At the same time, its LPs may lack the required industry-specific knowledge and the networking capabilities to source such deals.

In contrast to direct investing, co-investing involves transactions that are sourced, pre-screened, structured, and priced by the fund manager. While LPs generally do not pay fees and carry for co-investments, the role such investments play in investors' allocation approaches can vary significantly. Some LPs follow a passive and nearly automatic approach and rely more or less entirely on the fund managers. Others, however, behave nearly like a direct investor, with co-investments being best described as a form of syndication. In any case, invitations to co-investments are often confined to the largest LPs in a fund, and running a meaningful co-investment programme typically requires a significant primary fund franchise. At the same time, selecting among the universe of invited deals requires experience and skills that are akin to those needed for direct investments, which explains why co-investors in private equity transactions and similar deals in real assets are often large and sophisticated institutions.

4.6 A TIME-PROVEN STRUCTURE

The limited partnership has resulted from the extreme information asymmetries and incentive problems that arise in the market for many alternative assets. Although this partnership structure is often seen as an innovation that has been developed for VC investing and subsequently for other forms of private equity and real assets more generally, its origins date back to ancient times. Partnerships have existed throughout most of recorded history. Indeed, for unusually large or especially risky enterprises it is natural to pool resources to be better able to exploit investment opportunities. Combining resources helps partners achieve an outcome that is greater than the sum of the individual investments. Importantly, this insight cuts across cultures and legal systems, prompting Borden (2009) to suggest that humans have a natural tendency to form partnerships to conduct business. His analysis reveals that ancient and modern partnerships, in all their forms, have numerous common characteristics.

• One example is the relationship between the Sumerian “damgar” merchants and their agents, the “shamallu”, which existed in ancient Babylon as early as 2750 BC. The arrangement between these parties provided for independent action by the shamallu, as well as profit sharing and liability allocation between the shamallu and the damgar, and in several ways has parallels to the medieval commenda and modern partnerships. Documents show that as early as 1947 BC trade partnerships, so-called “tapputum” existed that required partners for some defined business project to make equal capital investments and to use the contributed capital for purchasing merchandise and for reselling at a profit. The structure allowed for debt financing of the operations and provided for joint liability or joint and individual liability of the partners. This Babylonian tapputum possessed characteristics of the medieval commenda.
• According to Hansmann et al. (2005), ancient Rome failed to develop general-purpose commercial entities, but found a type of multi-owner firm known as the “societas publicanorum”. This fourth-century BC type of partnership consisted of groups of investors, known as “publicani”, who bid on state contracts for projects or collection of taxes. Upon accepting the bid, the state paid a portion of the contract and the rest when the contract was completed. Investments could not be withdrawn, nor could the firm's assets be liquidated before the contract with the state was fulfilled. Hansmann et al. (2005) argue that multi-owner structures for amassing capital in ancient Rome were absent because they were not needed, as significant wealth was already concentrated in certain families. Only a few undertakings – such as the construction of public works or the manufacture of armaments – were too capital intensive even for a wealthy Roman family, with the multi-owner societas publicanorum filling the gap.
• Although partnerships can be traced to ancient history, many scholars attribute the origins of modern partnership law to either Byzantine or Arabic origins. Partnerships were based on a statute in the Code of Justinian and on the Rhodian Sea Law, a seventh-century body of regulations governing commercial trade and navigation in the Byzantine Empire. This “Lex Rhodia” focused on the liability for lost or damaged cargo and divided the cost of the losses among the ship's owner, the owners of the cargo and the passengers, thus serving as a form of insurance against storms and piracy. It was effective until the twelfth century and greatly influenced the maritime law of the Italian cities from the eleventh century onwards.
• It appears that Arabian contract law has provided at least one of the bases for the partnership structures widely used today. The “mudāraba” is a form of commercial contract whereby an investor – or a group of investors – entrust capital to a “mudārib”, i.e. an agent, who trades in it and then returns the principal along with a predetermined share of the profits to the investors (Brunnhuber, 2007). The mudāraba form of financing a venture has developed in the context of the pre-Islamic Arabian caravan trade. As such, it appears that its roots are indigenous to the Arabian Peninsula as a critical part in the long-distance caravan trade for the Hejaz region.
• Consequently, the mudāraba's conceptual approach appears to be particularly relevant in the context of long-distance international trade. According to Brunnhuber (2007), Islamic contract law has had a long tradition of defining and sharing investment risks among the relevant parties to a business venture. One conceptual underpinning of Islamic finance is commonly perceived to be the prohibition of interest taking, not unlike other belief systems like Christianity or Hinduism that at some point in their history have disallowed the charging of interest. The common perception is that interest in a standard loan is quasi risk-free, because the lender is guaranteed a return, independent of whether the underlying business transaction is successful or not. In contrast to this, Islam stresses the need to share rather than to transfer risk.
• With the Arab conquest, the mudāraba spread to Northern Africa and the Near East, and ultimately to Southern Europe. Partnerships were introduced into Europe as Italian merchants increasingly traded in the Eastern Mediterranean, thus becoming familiar with business practices in this region and adopting the “commenda”. Indeed, researchers have demonstrated a very strong correlation between the structures of the mudāraba and the commenda. The commenda contract had a sedentary investor, known as the “commendator”, who advances capital to a travelling associate, known as a “tractator”.
• The commenda both combined financing with insurance, as the recipient of the financing was freed of any obligation to the provider of the financing if ship or cargo was lost. The contract ended when profits were distributed after the merchant returned. The commendator received a portion of the profits, but had no liability for losses, was not guilty of usury and, in turn, could diversify risk by entering into commenda contracts with many different merchants. A commenda was not a common form of long-term business venture as most long-term businesses were still expected to be secured against the assets of their individual proprietors. Instead, the commenda was “self-liquidating”, i.e. it ended upon the completion of a venture or the death of the tractator.
• Colbert's Ordinance on Commerce of 1673 and the Napoleonic Code of 1807 reinforced the commenda's limited partnership concept in European law. Britain enacted its first limited partnership statute in 1907. In the USA, limited partnerships became widely available in the early 1800s.

While a number of legal restrictions made them unpopular for businesses, in the early twentieth century limited partnerships were increasingly used to raise capital for prospecting new oil fields. The establishment of the first VC limited partnerships in the USA dates back to the late 1950s and 1960s. In 1959 Draper, Gaither and Anderson adopted this structure and raised what, in all likelihood, was the first limited partnership in the VC industry.5 This model for VC investments is arguably the most successful worldwide and is followed in many international markets. Indeed, many see the limited partnership as an ideal vehicle for investing in private equity and real assets.

4.7 CONCLUSION

Limited partnerships formed by GPs and LPs are a particular form of financial intermediation that entails the asymmetric sharing of risks and rewards and represents a time-proven response to dealing with uncertainty. The efficiency, scalability and strong controls associated with modern large corporations are traded off against the flexibility of and the communication within small teams that allow a much quicker adaptation to changing market conditions and newly arising opportunities. While the limited partnership has frequently been pronounced dead, such pronouncements have thus far proved premature. To be sure, the limited partnership may have important shortcomings, which lie not least in their self-liquidation and their limited lifespan. However, no superior structure has evolved yet. To paraphrase Sir Winston Churchill, the limited partnership may be the worst form of investing in private equity and real assets except all those other forms that have been tried from time to time. Limited partnerships have been around for thousands of years in one form or another. Surprisingly, however, they are still poorly understood in terms of their risk properties from the standpoint of a diversified investment portfolio. Instead, investments in limited partnerships are often treated in the same way as investments in traditional asset classes. As we discuss in the remainder of this book, such treatment may have unintended consequences. Instead, a risk management approach is needed that reflects the specific characteristics of the limited partnership.

1 Note that the Carlyle Group has recently decided to pre-approve the sale of stakes by the LPs in their most recent buyout fund to five secondary fund managers in an effort to address possible liquidity concerns by their investors.

2 In fact, little or no dividend payout is more favourable for investors, as taxation on a dividend is higher than on a capital gain.

3 According to the IASB, an active secondary market is characterized by quoted prices that are readily and regularly available from an exchange, dealer, broker, industry group, pricing service or regulatory agency, with those prices representing actual and regularly occurring market transactions on an arm's length basis.

4 For the purpose of this discussion we focus on institutional financing and ignore the continuous innovation and experimentation with vehicles such as business angel networks, pledge funds, crowd funding, etc. mainly found at the boundary between formal and informal capital markets in early-stage entrepreneurial investing.

5 The long-term success of this firm, later renamed Draper Fisher Jurvetson, contrasts with the demise of the VC firm ARD (American Research and Development). General George Doriot, the pre-WWII Harvard Business School professor, organized ARD as a publicly traded, closed-end investment company subject to the Investment Company Act of 1940. This closed-end structure was, according to Hsu and Kenney (2004), plagued by three main problems:

ARD was a pioneering organization whose business model ultimately failed while the limited partnership had a better fit with the business environment.