8
Limited Partnership Fund Exposure to Financial Risks
Over the past few decades, investors have increased substantially their exposure to private equity and real assets, predominantly through commitments to limited partnership funds. At the same time, important efforts have been made to improve financial risk management. Arguably, these efforts were to a large extent motivated by repeated financial crises, such as the stock market crash in October 1987; the financial turmoil associated with the balance of payments crises in several emerging markets in the late 1990s and early 2000s; the bursting of the tech bubble in 2000; the sub-prime debacle in the USA that led to the collapse of Lehman Brothers and the Great Recession; and, most recently, the European sovereign debt and banking crisis. However, as Bongaerts and Charlier (2009) argue, the academic literature at the intersection of financial risk management and private equity has remained surprisingly close to an empty set. Similarly, contributions by practitioners have remained rare, with Weidig and Mathonet (2004) and Diller and Herger (2008) representing notable exceptions.
Similar observations can be made on the regulatory side. While the last couple of decades have seen important changes in bank regulation, which have been accompanied by a rapidly expanding literature on the subject, there is very little on the regulatory treatment of private equity and real assets. The Basel Committee on Banking Supervision (BIS, 2001) mentions that investing in private equity usually takes the form of commitments to such partnerships, but remains vague as to the implications for risk management. As far as the insurance industry is concerned, the European Insurance and Occupational Pensions Authority (EIOPA, 2012) does propose particular capital requirements for private equity under Solvency II, but fails to recognize the particular characteristics of limited partnership funds. However, ignoring the specific characteristics of such structures is bound to result in inappropriate capital requirements, which has motivated the EVCA to establish a working group to develop guidelines as to how limited partners in closed-end funds may measure their risk exposure.
Building upon this work, 1 the present chapter identifies the various types of financial risk LPs in limited partnership funds are exposed to. Our emphasis is on those risks that are specific to the particular characteristics of such funds. Thus, we do not discuss risks that are of a general nature, regardless of the particular way of investing. For example, operational risk (including legal and reputational risk) is common to all activities and not specific to investments in funds.2 Instead, we concentrate on capital risk and liquidity/funding risk as the two major components of financial risk LPs face. As far as funding risk is concerned, we propose a funding test as a critical tool for LPs to manage their liquidity positions in light of undrawn commitments. Finally, given the increased amount of cross-border investment flows in private equity and real assets, we discuss the significance of foreign exchange risk.
8.1 EXPOSURE AND RISK COMPONENTS
In standard risk models, financial risk is measured by two variables: (i) the probability of a (negative) event and (ii) the loss an institution may suffer due to that event relative to the institution's total exposure. In credit risk models, for instance, the expected loss is the product of the probability of default and the loss given default (LGD). In quantifying financial risk, it is therefore imperative to determine the institution's exposure and the risks to which its exposed capital is subject.
8.1.1 Defining exposure and identifying financial risks
Turning to investing in private equity and real assets, let us first focus on the question of measuring an LP's exposure to such asset classes. This question is far less trivial than it might seem. Generally speaking, the risk exposure of limited partners is determined by their share in a fund or portfolio of funds (EVCA, 2011). However, given the particular characteristics of self-liquidating partnerships, an LP's exposure may change continuously, depending on how the exposure is actually measured. Importantly, this does not even require that the LP buys or sells shares in the fund. In practice, there is no common policy among LPs, who use one or several of the following measures to determine their exposure: the fund's NAV, its net paid-in capital, the NAV plus the undrawn commitments, the net paid-in capital plus the undrawn commitments or just the fund's commitment. All these measures have shortcomings and do not give a full picture of the risks. A fund's NAV or its net paid-in capital, for instance, does not capture the funding/liquidity risk caused by its undrawn commitments. If (undrawn) commitments are taken into account, the total exposure may exceed the investor's available resources due to overcommitments (Section 8.3).
This confusion comes out clearly from an online conversation among risk managers, as summarized in Box 8.1.
- One participant thought that quantitative modelling would not be of much value. In fact, there was no point in trying to quantify all risks. Instead, it was more important to rely on the experience and intuition of the investor in assessing risk in illiquid funds.
- This view was supported by another blogger who believed that such assets do not lend themselves to the quantification of risk. In his view, there are insurmountable challenges arising from the absence of an organized market for these investments: daily pricing, standard performance metrics or disclosure requirements. Furthermore, he stressed that mark-to-market events were rare.
- Given the long investment horizon of commitments to limited partnership funds and the substantial challenges in calculating risk-adjusted returns, the best – and, according to one blogger, perhaps the only – thing risk management could do was to choose investments and fund managers wisely. “After that, about all you can do is monitor [sic] IRR by vintage year and track it against whatever benchmarks you set … and hope for the best.”
- The literature known to the blogger who had posed the original question was in line with these responses, confirming that one theoretically would need to forecast future cash flows. However, the blogger recognized that every approach to it would be flawed or biased and it would be “almost impossible to have a quantitative measure for the risk.” Unfortunately, this was of little help to him as “the banks need some kind of pseudo measurement” even if it makes little sense.4 Would there be an approach to mitigate the identified shortcomings and generate a model for risk measurement that is as good as possible?
- One respondent proposed to throw “together some garbage spreadsheet and tell them it is based on extreme value theory.” Another suggested using rules of thumb of the kind that if the private equity portfolio generates a return of less than 6 percentage points per year below the Wilshire 5000 or Russell 3000, the portfolio is subject to “high” risk. If it generates a return of 9+ percentage points greater than the index, so the argument put forward in the blog, it has “low” risk. Anything in between represents a “moderate” level of risk.
The frustration with measuring risk in limited partnerships that is expressed in the blogs seems to be shared more widely. Interviewing a consultant, Kreutzer (2008) reports that managing risk in private equity was essentially a function of proper due diligence and relationship management: “You really want to know people that you're doing business with. You can quantify leverage on earning streams but it can be monkeyed with. It's not an easy thing to approach scientifically.” In the consultant's view, the attempt to quantify risk in a systematic way was probably “a waste of time”.
In Chapter 10, we address the issue of undrawn commitments in greater detail. While undrawn commitments are usually treated as an asset whose NPV is zero and hence are ignored, the experience of many investors during the recent global financial crisis suggests that undrawn commitments do matter in determining an investor's exposure to financial risks associated with investments in limited partnership funds in private equity and real assets.
Let us now turn to the second component of financial risk management, the identification and quantification of financial risks, given an LP's exposure to illiquid investments. As we discuss in the following, we can distinguish between two main types of risk in fund investments, i.e. (i) capital risk – the risk that their invested capital plus an expected return is not returned and (ii) liquidity risk – the risk that the investor's liquidity position does not allow him to respond to capital calls (funding risk) as well as the risk that he is unable to liquidate his shares in the secondary market (market liquidity risk), disabling permanent portfolio rebalancing.
8.1.2 Capital risk
In assessing the capital risk of fund investments, it is important to distinguish between the LP's commitment to a self-liquidating partnership and the GP's acquisition of assets. As far as private equity is concerned, the acquisition of a portfolio company obviously represents an equity investment in the company's capital structure. However, it is less clear how fund investments by LPs should be treated. Arguably, they share certain credit characteristics in the sense that investors provide capital that may or may not be returned to them after a period defined in the LPA. Thus, shares in a fund, which are highly illiquid, are subject to what may be considered as default risk (Meyer and Mathonet, 2005). Consistent with this view, rating-like approaches have been developed for evaluating the risks of limited partnership funds (see Chapter 13). Such approaches, which group funds into different risk categories, are widely used in the industry and have been endorsed by the Basel Committee (BIS, 2001).
However, in the context of limited partnership funds, there is no common definition of a default event. Theoretically, a default for a fund could be defined as the failure to pay back capital to the LPs. Alternatively, a GP could already be in default if he fails to meet a certain hurdle rate of return. The problem with such definitions is, though, that the default event can only be determined at the end of the fund's lifetime and therefore does not form an “annualizable event” in the true sense.5
Nevertheless, a number of practitioners and researchers have tried to apply existing credit portfolio models to private equity. Krohmer and Man (2007), for example, use a dataset provided by the Center of Private Equity Research (CEPRES). Their sample includes 252 private equity funds with 16, 097 investments in 12, 008 portfolio companies in 1971–2006. They employ a default model that is based on Wilson's CreditPortfolioView™. In their model, a default is defined as a total loss with an IRR of –100%, with the PD/LGD of funds being determined by simulating the development of portfolio companies. As far as venture capital is concerned, Krohmer and Man (2007) find a default rate of 30%, significantly higher than for buyouts (11.5%).
These estimates appear high, especially when compared with studies by Weidig and Mathonet (2004) and Diller (2007). A key issue that limits the applicability of credit risk models to limited partnership funds lies in the fact that such models only reflect downside risk. Aggregating the PD/LGD figures for individual funds, without taking into account the upside potential of other funds, is bound to lead to excessive overall risk weights for portfolios of funds. Diversified portfolios of funds are significantly less risky than a single fund as the upside of well-performing funds compensates for the losses from “defaulting” funds – implying that economic capital should be allocated to fund portfolios rather than individual partnerships. Indeed, the Basel Committee has long accepted that unrecognized and unrealized gains (or latent revaluation gains) on equity investments can act as a buffer against losses (BIS, 2001). We shall return to this issue in greater detail in Chapter 9.
8.1.3 Liquidity risk
The financial crisis beginning in 2008 highlighted again the importance of managing liquidity risk. In response to this market turmoil, the Basel Committee on Banking Supervision issued principles for sound liquidity risk management (BIS, 2008):
“Liquidity is the ability of a bank to fund increases in assets and meet obligations as they come due, without incurring unacceptable losses. […] effective liquidity risk management helps ensure a bank's ability to meet cash flow obligations, which are uncertain as they are affected by external events and other agents' behaviour.”
In fact, regulators now generally see liquidity risk management as of paramount importance and require that regulated investors have a sound process for identifying, measuring, monitoring and controlling liquidity risk:
“This process should include a robust framework for comprehensively projecting cash flows arising from assets, liabilities and off-balance sheet items over an appropriate set of time horizons.”
Liquidity risk can generally be defined as the potential loss due to time-varying liquidity costs. Investors in limited partnership funds take substantial exposure to liquidity risk: undrawn commitments can be seen as economic obligations, and LPs have to apply liquidity management processes along the lines described by the Basel Committee. In general, “liquidity” comprises market liquidity (Stange and Kaserer, 2009) and funding liquidity.
Market liquidity
Market liquidity risk arises from situations in which a party interested in trading an asset cannot do it because nobody in the market wants to trade that asset or only at a price that is far away from its – or its “last observed” – fair value (Dowd, 2001; Buhl, 2004; Amihud and Mendelson, 2006). Market liquidity risk can be reflected, for example, in steep discounts to valuations, in the need to hold liquid cash reserves or by the need to hold on to assets over longer periods than desired. Thus, investors in limited partnership funds will usually require an illiquidity premium that compensates them for the inability to constantly rebalance their portfolios – a critical assumption in standard asset pricing models.
Funding liquidity
Funding liquidity can be defined as the ability to settle obligations immediately as they come due. Funding liquidity risk relates to the possibility that, over a specific horizon, investors are unable to honour their obligations or can only meet them at an uneconomically high price (Drehmann and Nikolaou, 2008). Typically, LPs have to respond to a fund's capital call within a short timeframe of just 10 days or even less. This requires them to either keep sufficient cash reserves at any point in time or to liquidate other assets at short notice. As already discussed, a LP who is unable to meet his obligations by providing funding in response to a capital call would become a defaulting investor and, in the extreme, might lose his entire invested capital in the fund. The LP's ability to either hold on to his commitment to a fund and thus extract its full long-term value or to conduct an orderly secondary transaction at a fair price depends on keeping this funding risk under control.
Typically, LPs use the distributions from mature funds to respond to capital calls from less mature funds within their portfolio (see Box 8.2). This works well for portfolios of funds that are well diversified over time and under “normal” market conditions. However, as experience during the recent global financial crisis has shown, cash flow models may break down in periods of severe market dislocations, unexpectedly forcing investors to seek additional financing to cover the shortfall.
8.1.4 Market risk and illiquidity
Capital risk and liquidity risk in fund investments have received surprisingly little attention in the literature, which we have reviewed briefly in Chapter 5. Instead, to the extent that academic studies have attempted to measure risk in private equity, they have focused on market risk in the framework of standard asset-pricing models, such as the CAPM. Much of this debate has concentrated on the empirical problems of estimating alphas and betas (infrequent observations of valuations, potential sample biases in available databases) and their interpretation, given the restrictive assumptions of such models (transparent, liquid and low-friction markets).
From a more fundamental standpoint, the question arises as to how relevant market risk really is for investors in limited partnership funds. Implicitly, it is assumed that the short-term variation of valuations of illiquid fund investments – however measured – matter for LPs as much as the variation of prices of their portfolios of marketable instruments. This view may be disputed on the grounds that long-term investors in limited partnership funds deliberately decide to lock in capital for a period of 10 years and more. At the time when LPs make commitments to such funds, they have usually no intention of selling their stakes before maturity and should therefore be less concerned about quarterly changes in the NAV of their fund investments.
However, as we have seen in Chapter 6, LPs do sell their stakes in the secondary market. Sometimes, such sales are motivated by a strategic reorientation, but in the majority of cases divestment decisions have been driven by liquidity problems. Interim valuations therefore do matter, even if LPs intend to hold their fund investments until the end of the partnership's life. Market risk and liquidity risk are therefore closely intertwined, which has encouraged researchers to examine the impact of illiquidity on portfolio construction (e.g., Ang et al., 2011), and more specifically on the optimal holdings of private equity. As regards the latter, Ang and Sorensen (2011) argue that “(o)ne way of interpreting the risks and returns of PE investments, especially for illiquidity risk, is for an investor to consider PE from an investor-specific asset allocation context”. This is exactly what we propose in the following section – to conduct a funding test to ensure that the LP's specific funding needs are always met.
8.2 FUNDING TEST
As we discussed in Chapter 6, the secondary market is an imperfect price discovery mechanism for a limited partnership fund's fair market value. In the absence of observable market prices, risk managers have to turn to model-based approaches. Models, however, are generally based on the critical assumption that the LP is able to hold on to his assets or, alternatively, is able to sell his assets under normal market conditions. However, the question arises as to how a fair market value can be determined in a distressed sale. Unfortunately, industry valuation guidelines provide little guidance on this important question.6 Against this background, it is critical for LPs to ensure that they do not have to sell in a situation of distress and always have sufficient resources available to respond to capital calls for the remaining undrawn commitments through a so-called “funding test”. Although the total amount of undrawn commitments is known, the amount and timing of individual drawdowns are not.
There are various approaches to a funding test, ranging from the simple monitoring of key ratios to a more sophisticated scenario analysis for future cash flows. Ratio analysis focuses on setting key accounting variables in relation to each other and monitoring their behaviour over time in an effort to extract risk-relevant information. It is a quantitative technique for comparing a firm on a relative basis to other firms or to the market in general. Changes in ratios can help signal in which direction the portfolio is developing. Key ratios that are indicative of a LP's funding risk are the current ratio (CR), the adjusted current ratio (ACR), the overcommitment ratio (OCR) and the outstanding commitment level (OCL) (Mathonet and Meyer, 2007).
The CR is a standard tool to analyse a firm's capacity to finance its short-term liabilities by its short-term resources:
The CR can be interpreted as a measure of a firm's ability to survive over the near term by being able to meet obligations with the available liquidity. A CR of 100% means that even without doing any business, the firm could theoretically survive for 1 year. A low CR is a signal that the available liquidity would be insufficient. However, over the long term a high CR could also be an indication of inefficient use of the firm's resources. In any case, it is obviously risky to overcommit for a CR below 100%.
However, for portfolios of limited partnership funds the CR is of limited explanatory power as it ignores the undrawn commitments.
- In the short term, a LP will have to pay a part of its current undrawn commitment, which can be considered as a short-term liability (percentage times undrawn commitments).
- Furthermore, the percentage of the distributions that will be received by the LP has to be taken into consideration (percentage times paid-in).
These items are reflected in the ACR:
The projected amount of distributions as a percentage of the current paid-in depends on the maturity of the portfolio of funds. This figure, and the share of the undrawn commitments to be disbursed in the short term based on the current cumulative outstanding commitment, can be determined based on the analysis of historical data.
The OCR is defined as the ratio of undrawn commitments of limited partnership funds relative to available resources. Undrawn commitments are the aggregated commitments for individual funds, whereas the definition of resources available depends on the individual investor. For funds-of-funds, the resources available are simply the commitments from their investors. For insurance companies, banks and other institutional investors, “resources available” are liquid assets – or sometimes even clearly predefined and fixed future income streams – that have been allocated for fund investments in private equity and real assets.
The calculation of the OCR thus depends on the definition of “resources available”. The most conservative case for an investor would be to avoid all risk and cover all undrawn commitments through available resources in the form of cash. However, in reality only very few LPs invest their undrawn capital risk-free. Therefore, the challenge for investors is to find the appropriate OCR. To some degree, overcommitments are unavoidable as the committed capital is only called over several years and for some funds not even the full commitment is drawn.
An OCR below 100% typically suggests an inefficient use of resources. Above 100%, the ability of an overcommitted investor to honour its commitments and to avoid default decreases proportionally to an increase in the OCR. Opinions vary in terms of the degree of overcommitments that would still be seen as prudent. Generally speaking, OCRs in the range of 105–115% would still be seen as relatively prudent. In practice, however, OCRs of between 125% and 150% are not untypical (Mathonet and Meyer, 2007), a level that warrants increasing attention. As long as there is no additional funding, the expected distributions of the existing portfolio put a natural limit on commitments to new funds given the availability of funding. Consequently, the long-term average return of the asset class puts a ceiling on pursuing overcommitment strategies.
Financial analysis typically differentiates between short-term, medium-term and long-term liquidity ratios. We suggest monitoring three different ratios:
- OCR1, mirroring immediately available cash; i.e. through cash holdings or liquidity facilities.
- OCR2 as the sum of OCR1 assets and potential proceeds from selling highly liquid assets, for example, high-grade sovereign bonds and equities.
- OCR3 as the sum of OCR2 assets and proceeds from selling assets that can be liquidated over a timeframe of several weeks, such as corporate bonds.
Therefore:
Comparing the undrawn commitments to the total commitments provides information on the maturity of the portfolio. The OCL is defined as:
Assuming that there is no change in the investment and commitment strategies, the current OCL can be indicative of the capital expected to be drawn down over the following financial year. It is important to maintain a reasonable OCL over time as it represents a future liability.
Figure 8.1 Cash flow projection for a portfolio of 10 private equity funds.
Obviously, the monitoring of these ratios can only be a first line of defence (see Box 8.3). A proper funding test should be based on the assessment of several cash flow scenarios (Figure 8.1). Without being able to honour all capital calls over the portfolio of funds' entire lifetime, the investor would either default on investments or sell parts of his portfolio, and in the extreme case the entire portfolio, on the secondary market. Depending on the market cycle, selling a stake in a fund or a portfolio of funds may be possible only at substantial discounts, putting the LP in a different risk position.
- Investor A is an insurance company that invests 4.4% of its total assets in illiquid assets. The insurance company receives quarterly net inflows from its operative businesses of EUR 75 million, which will be directed to all asset classes. Overall, the insurance company has total assets of EUR 2 billion with an allocation of 75.6% to bonds, 15% to equity and 4.4% to illiquid assets. The remainder, 5%, is held in cash.
- Investor B is an endowment, which has a total available cash volume of EUR 55 million. In addition, a liquidity facility of EUR 10 million is arranged.
- Investor A has an OCR1 of 100% as the cash holdings amount to EUR 100 million. In addition, a continuous net cash inflow decreases OCR2 to more than 57% (and even lower when taking all highly liquid bonds and equities into account). Hence, investor A has no liquidity issues and it can be expected that he will be able to hold its portfolio through various market phases without any problems.
- Investor B has an OCR1 of 154% and is hence running an overcommitment strategy. As there are currently no other sources of income other than the cash account and expected private equity distributions, it is a significant risk that the investor B (endowment) runs out of liquidity.
8.3 CROSS-BORDER TRANSACTIONS AND FOREIGN EXCHANGE RISK
The preceding discussion has taken a general view of financial risks investors in limited partnership funds are exposed to. An additional risk dimension arises, however, if investments involve cross-border transactions which are subject to foreign exchange risk. As long as national markets for private equity and real assets were largely separated, this type of investment risk did not matter much. However, as markets have become more integrated and LPs have committed capital to foreign currency-denominated funds, which in turn make cross-border acquisitions, investors have increasingly become exposed to currency fluctuations. Thus, we dedicate a special section to this particular type of risk, which interacts with both capital and liquidity risk.
8.3.1 Limited partner exposure to foreign exchange risk
Exchange rate changes are particularly relevant over longer time spans. Thus, LPs who want to diversify their alternative investment portfolios internationally have a keen interest in monitoring and managing their foreign exchange exposure.
Foreign exchange risk is often thought to lie in sudden jumps in bilateral rates in the wake of balance of payments crises. These crises typically occur if a country pursues a fixed exchange rate policy that can no longer be defended as reserves are being depleted amid unsustainable current account deficits and capital outflows. Reinhardt and Rogoff (2009) give a comprehensive overview of balance of payments crises, most of which, at least in the last couple of decades, have occurred in emerging economies (e.g., Mexico, 1994; Asia, 1997; Russia, 1998; Brazil, 1999; Argentina, 2001. A notable exception is the foreign exchange crisis in the United Kingdom in 1992, which forced the country to leave the European exchange rate mechanism). These currency crashes have significantly reduced, or entirely wiped out, foreign investors' returns.
However, foreign exchange risk is by no means limited to investments in emerging economies. While virtually all advanced countries today maintain flexible exchange rate regimes that make a sudden collapse much less likely, their foreign exchange rates do move significantly, reflecting inflation differentials, divergent monetary policies as well as speculative capital flows, for example, associated with carry trades. Take, for example, the USD/EUR rate, which links the world's two most important markets for private equity and real assets. In 2001, 2 years after the introduction of Europe's single currency, a US investor had to pay 0.82 US dollars for 1 euro. In 2008, the exchange rate had nearly doubled to almost 1.60. Between mid-2002 and mid-2003, the value of the US dollar fell by 26% relative to the euro, a movement which would have qualified as a currency crash according to Frankel and Rose's (1996) definition.
8.3.2 Dimensions of foreign exchange risk
Foreign exchange risk has several dimensions for investors in limited partnership funds. To begin with, an investor who commits capital to a foreign currency-denominated fund is exposed to currency movements between the point in time when he makes a commitment in a currency that is not his home currency and when the GP draws down the capital. Suppose that the currency in which the fund is denominated has appreciated against the investor's home currency. As a result, the contributions in the investor's home currency exceed the original commitment to the fund, which may cause liquidity problems. Conversely, the LP may find himself underexposed to private equity and real assets relative to his target exposure.
Second, exchange rate movements affect the performance of a foreign fund when the returns are converted into the investor's home currency. An investor based in the Eurozone is interested in returns in euros, just as much as a US-based investor is interested in returns in US dollars rather than in local currency returns. To the extent that the home currency has appreciated over the lifespan of the fund, the impact will be negative, other things being equal. However, the impact may also be positive if the home currency depreciates during the lifespan of the limited partnership.
At a third level, investors are exposed to foreign exchange risk to the extent that the fund itself makes acquisitions in foreign currencies. Suppose, for example, a Eurozone pension fund commits capital to a euro-based partnership, which uses part of its capital to acquire sterling-denominated assets in the United Kingdom and dollar-denominated assets in the United States. To make things even more complex, consider now the following situation where the European pension fund commits capital to a US-based private equity fund, which in turn acquires assets in Latin America in local currency.
The foreign exchange risk LPs take when committing to foreign currency-denominated funds may be mitigated by cross-border acquisitions undertaken by the partnerships they have invested in. For instance, several global USD-denominated funds backed by European limited partners have actually acquired European assets. However, foreign exchange risk at the fund investment level may also be amplified by foreign exchange risk at the portfolio company level, depending on the timing and direction of investments.
There is even a fourth level of foreign exchange risk, to the extent that portfolio companies are engaged in foreign transactions through trade and treasury operations. However, this risk is not different from the foreign exchange risk publicly listed companies may be exposed to. Furthermore, at the company level, foreign exchange risk is typically hedged and will therefore not be considered specifically in the rest of the book.
8.3.3 Impact on fund returns
How relevant is foreign exchange risk for LPs in terms of the fund returns in local currency? Using quarterly cash flows from a large number of partnerships of a Eurozone-based fund-of-funds investor, Cornelius (2011) estimates that the foreign exchange effect is indeed sizable. For example, IRR returns on investments made in USD-denominated funds in 2002 were reduced by more than 5 percentage points as of 30 September 2009 thanks to the depreciation of the US dollar against the euro during this period. Given that the long-term average of private equity returns (net of fees) has been around 15%, this implies that foreign exchange rate movements may have a significant impact on portfolio performance in the investor's home currency. Admittedly, some of the losses calculated by Cornelius (2011) have probably been recouped more recently as the US dollar regained ground in 2011–2012. However, the overall picture is clear – foreign exchange risk matters, and not only for cross-border commitments to funds in emerging economies.
As Cornelius (2011) shows further, foreign exchange rate risk also matters for GPs who invest abroad. The returns of the fund in the fund's currency may be boosted by favourable currency movements – or vice versa. For example, a US-based fund, which acquired a European company in 2001 and sold this company in 2006, would have made a significant foreign exchange gain thanks to the appreciation of the euro during this period. This raises a number of fundamental questions. How should a LP benchmark funds in their due diligence? Should returns due to currency movements be considered as sheer luck and hence irrelevant for the GP's track record? And how should currency gains (and losses) be treated from the point of view of carried interest?
While it is relatively straightforward to calculate ex post the impact of foreign exchange rate changes on the performance of funds, it is far more difficult to measure the exact degree of foreign exchange risk a given portfolio is exposed to. The LP knows, of course, the share of capital he has committed to foreign currency-denominated funds. And he also knows on a look-through basis the foreign exchange exposure that arises from cross-border investments made by the funds. What he doesn't know, however, is the foreign exchange risk his unfunded commitments are subject to. Although the limited partnership agreements typically provide some guidance as to how much a fund can invest outside their home markets, ceilings have become increasingly flexible to allow GPs to chase attractive deals across different markets.
8.3.4 Hedging against foreign exchange risk?
As far as limited partners are concerned, the particular characteristics of investments in illiquid funds render traditional hedging instruments largely irrelevant.7 Using forward contracts, futures, currency swaps and options requires knowing the timing and the amount of cash flows that are subject to foreign exchange risk. This is generally not the case, however. Cash flow libraries can only provide an approximation of expected contributions and distributions, which is not sufficient to employ standard hedging instruments. This challenge applies especially to relatively young portfolios that consist of a limited number of funds. Furthermore, foreign exchange hedging is expensive, adding to the challenges LPs face when making commitments to international funds, which themselves operate across borders.
Thus, long-term investors are left with two options. The first option is just to “go naked”. As Froot (1993) has argued, currency hedges have very different properties at long horizons compared with short horizons. In fact, at long horizons fully hedged international investments may actually have greater return variance than their unhedged counterparts. Suppose, for example, a US real estate fund makes an acquisition in the Eurozone. Suppose further that the euro depreciates in the short term due to unanticipated disturbances that lead to currency-induced losses in the value of the investment for the US-based investor. With the price level remaining unchanged in the short run, the euro also depreciates in real terms. However, if purchasing power parity holds in the long run, the price level in the Eurozone should increase relative to the United States, resulting in a real appreciation of the euro. To the extent that the value of the investment is linked to the national price level, the price effect should offset the currency effect in the long term. Thus, the investment is “naturally hedged” over long horizons.
The second option is a currency overlay. This option is usually pursued by large international investors who employ currency overlays for their entire portfolios, not just their portfolio of alternative investments. Currency overlay strategies are typically delegated to an external or in-house specialized manager who decides on the positions taken in currencies and manages currency risk. With currencies regarded as financial prices, the client usually determines a benchmark hedge ratio that reflects the investor's desired neutral currency exposure. How currency overlay strategies can be designed is discussed, for example, in Solnik and McLeavey (2009).
8.3.5 Foreign exchange exposure as a potential portfolio diversifier
Modelling cash flows for limited partnership funds, as we discuss in Chapter 11, enables LPs to undertake a risk component analysis that offers additional insights regarding foreign exchange risk. A simple way to show the impact of exchange rate changes on capital risk lies in the recalculation of cash flows at a historically fixed exchange rate. Importantly, investing internationally can increase the volatility of cash flows but may also decrease the capital risk of a portfolio of funds thanks to potential diversification effects. Whether an investment in a fund that is raised in a foreign currency reduces overall risk for the LP essentially depends on the behaviour of exchange rate movements relative to the fund's cash flows. In this context, it is important to note that exchange rate expectations are typically an important variable in the GP's investment and divestment decisions. While limited partnership agreements tend to be relatively vague in terms of a fund's foreign currency exposure and its management – beyond certain ceilings for investments in currencies other than the fund's currency – it is important for LPs to monitor closely their foreign exchange risk at the portfolio level with a view to their liquidity needs.
8.4 CONCLUSIONS
At the most basic level, the proper management of financial risks in illiquid assets requires (i) knowing an investor's exposure to such assets and (ii) identifying the risks this exposure is subject to. As trivial as this seems, there remains substantial confusion about both dimensions.
While we have emphasized the importance of undrawn commitments, an issue we return to in Chapter 10 in greater detail, much of the debate in the present chapter has focused on the definition of financial risks. As we have argued, LPs in limited partnership funds are subject to two key risks – capital risk and liquidity risk. Capital risk refers to the risk that a fund fails to return the investor's capital, but as we have discussed this risk should be seen from a portfolio perspective instead of taking a fund-specific view that is akin to a credit risk assessment.
Liquidity risk arises from the fact that shares in a fund cannot easily be liquidated in the secondary market (market liquidity risk) and that capital calls and distributions are uncertain with respect to their timing. This does not mean that market risk, the focus of academic research in the context of standard asset pricing models as well as from a regulatory standpoint, is unimportant. In fact, market risk, capital risk and liquidity risk are closely intertwined. Arguably, however, long-term investors are less concerned about short-term fluctuations in asset valuations in asset classes where they deliberately lock in capital for 10 years and more.
If there had been a need to stress the importance of liquidity risk, the recent global financial crisis would have been the perfect proof. Against this background, this chapter has emphasized the role of funding tests based on key financial ratios and scenario analysis. This discussion will be taken up again in Chapter 12, where we focus on cash flow projections.
Finally, we have focused on foreign exchange risk, a special risk that is associated with cross-border transactions in different currencies. Foreign exchange rate risk has several dimensions, and each dimension has important implications for the performance of fund investments, the underlying cash flows, and hence capital and liquidity risk. While a LP's ability to hedge foreign exchange rate risk is strictly limited, it is critical for him to monitor this exposure closely from a portfolio standpoint with regard to the impact of exchange rate movements on the portfolio's overall risk level.
1 For disclosure reasons, three authors of the present book (Cornelius, Diller and Meyer) were members of the EVCA working group.
2 As far as Basel III is concerned, regulatory capital is calculated for three major components of risk that a bank faces: credit risk, operational risk and market risk. Other risks are not considered to be fully quantifiable at this stage. Solvency II considers market, credit, liquidity, insurance and operational risk, and for the AIFM Directive market, credit and liquidity risk are viewed as “financial risk”.
3 Blog on “private equity – risk measurement”, posted by Tim Hellmann, 21 February 2005. See http://www.riskarchive.com/link/ar05-1.htm [accessed 14 October 2008].
4 This view resembles an anecdote in Danielsson (2008): “A well-known American economist, drafted during World War II to work in the US Army meteorological service in England, got a phone call from a general in May 1944 asking for the weather forecast for Normandy in early June. The economist replied that it was impossible to forecast weather that far into the future. The general wholeheartedly agreed but nevertheless needed the number now for planning purposes.” To have numbers seems to be more important than whether the numbers are meaningful and useful.
5 At the fund level it is difficult to obtain a sufficiently broad and unbiased sample to estimate a probability of default and loss given default (PD/LGD). Publicly available databases on private equity are typically subject to an important survivor bias and most datasets provide only aggregated figures that are of limited use. BIS (2001) discusses a PD/LGD approach but in a conversation with some of the authors in the early 2000s the Basel Committee explained that they did not see this as applicable to funds. They did agree that a PD/LGD approach could potentially be applied to debt-funded direct investments, as followed by Bongaerts and Charlier (2009).
6 See, for example, EVCA (2005). The International Private Equity and Venture Capital Valuation Guidelines establish an industry standard for determining a fair value for these specific illiquid assets.
7 From a GP's perspective, hedging may be a more feasible option under certain circumstances, given that the timing of deals and their exits are determined by the fund manager. In fact, the timing may be influenced by the GP's view on possible exchange rate misalignments. For instance, if the GP is of the view that the currency in which the foreign investment is made is overvalued and is likely to mean-revert against the GP's home currency, he is likely to bring forward the divestment, other things being equal.