Executive Summary
This white paper addresses the importance of being diligent on asset manager due diligence for the Endowment Approach in order to avoid investing in fraudulent asset managers and analyzes over one hundred past asset manager fraud cases from 1998 to 2014 to understand patterns and techniques of fraudulent managers. The study found that (1) there were a far more hedge fund-like asset manager cases than private equity funds or mutual funds, (2) there were five main fraud techniques used frequently by asset manager frauds and (3) frauds are becoming more sophisticated while more allocators are conducting operational due diligence. The Endowment Approach has embedded advantage to avoid asset manager frauds, however, proper and careful due diligence is still important to achieve the objectives of the Endowment Approach.
In summary:
For the endowment approach to work, there are two sources of outperformance; (1) asset allocation and (2) manager selection.
In achieving good manager selection, many allocators’ main focus is finding a manager who can generate “the best” returns. Much less attention is paid to “avoiding” a bad manager, who appears to generate good returns but is either carrying too much risk or is simply a fraud.
One of the reasons the endowment approach is difficult for families is that they do not often have a proper procedure to avoid bad managers, such as blow-ups or frauds, as they were lured by stable or extremely high returns.
The endowment approach’s equity focus and long-term orientation helps to avoid those bad actors, but allocators still need to pay meticulous attention.
Five asset manager case studies (Bayou, Petters, AIJ, Weavering, Platinum) are enclosed in Appendix.
Introduction
While we have spoken a great deal on the merits of an endowment approach in our previous white papers (An Endowment Approach for Asian Families and An Endowment Approach for Sustainable Philanthropy), it is also worth considering some of its costs and how best to mitigate such risks. When using external asset managers to generate strong returns, two vital parts of the endowment approach to consider are: (1) the allocation to different asset classes and strategies, or Asset Allocation and (2) selection of external asset managers, or Manager Selection and, interestingly, during the period from 1994 to 2013, Manager Selection generated much greater Value Add than Asset Allocation (2.9% p.a. vs 2.2% p.a., Exhibit 1). While the risk associated with Asset Allocation is relatively more empirical and testable, that with Manager Selection has one unexpected, less scientific downside risk: fraud.
Exhibit 1: Value Add Factors for Yale Endowment
The continuous effort to find intelligent and skillful managers who can generate superior returns over the long term is critical for the success of the Endowment Approach. However, in an analogous fashion to our own personal lives, investing can be like falling in love. When you invest you can become blinded by your passion and conviction for a particular manager. And like a compelling romance, a bad investment can ruin your life.
External asset managers can lose money for many different reasons such as lack of expertise, market conditions, or simply bad luck. However, the most catastrophic way for external managers to lose our money is through fraudulent activities. One of most infamous examples is Bernard Madoff, who raised tens of billion dollars and subsequently fabricated investment returns over 20 years. In an ironic but logical sense, fraudulent asset managers tend to have affable personalities. Together with a compelling long-term track record, it may be easy to “fall in love” with them.
When considering which manager to allocate with, it goes without saying that seeking managers that produce the best returns is a high priority. Additionally, choosing managers with attractive risk-adjusted returns, such as a high Sharpe ratio, are also frequently considered. But nonetheless, we would suggest another factor of manager selection should be at least equally weighted—if not more—with presumed ability to generate the highest returns: this being a manager’s integrity.
This aspect is often overlooked and many allocators pay substantially less attention to “avoiding” a bad manager who may appear to generate strong returns with decent risk scenarios. Indeed, with the recent emphasis on risk-adjusted performance, allocators have looked for the unattainable manager with a Sharpe ratio of above 2.0 and generating consistently 10%+ returns with few down periods. The formulization of the hedge fund space with increasing emphasis on such ratios has provided temptation for managers to be the dream manager of consistently high returns with allocators often paying less attention to the mechanics required for investment success and thesis scrutiny.
Thus from an allocator’s perspective, a conundrum is presented. For smaller and nimbler family-office type allocators, it can be impractical to conduct extensive due diligence exercises that larger institutions can afford to conduct. And also, it is rational that these same allocators still hope to gain attractive risk-reward scenarios as many other allocators can, and thus may avoid more ‘volatile’, unhedged equity exposure that has traditionally been the bread-and-butter of hedge funds. Instead, current allocators gravitate towards more obscure, less transparent, less liquid and less public strategies that while can claim to be more ‘consistent’, are also more prone to frauds.
As such, the endowment approach’s traditional equity focus and long-term horizon has helped to avoid some of the most egregious cases of hedge fund fraud—but nonetheless, it is important to stay vigilant. The goal of this white paper is thus to explore and learn from investment management related frauds so as to avoid the next hedge fund fraud.
Thus to summarize:
For the endowment approach to work, there are two sources of outperformance; (1) asset allocation and (2) manager selection.
In achieving good manager selection, many allocators’ main focus is finding a manager who can generate “the best” returns. Much less attention is paid to “avoiding” a bad manager, who appears to generate good returns but is either carrying too much risk or is simply a fraud.
One of the reasons the endowment approach is difficult for families is that they do not often have a proper procedure to avoid bad managers, such as blow-ups or frauds, as they were lured by stable or extremely high returns.
The endowment approach’s equity focus and long-term orientation helps to avoid those bad actors, but allocators still need to pay meticulous attention.
Scope of Analysis
Over the last few years, we have conducted an analysis of asset manager fraud cases and written a dozen case studies published on our website – Present Value of the Future. In order to expand our scope and deepen our understanding of the landscape of the asset manager fraud, we decided to review over a hundred hedge fund fraud cases in the United States that occurred between 1998 and 2014, including several prominent cases such as Madoff and Refco. (Exhibit 2)
Exhibit 2: Largest Hedge Fund Fraud Cases between 1998 and 2014
Philip Bernett, CEO of Refco Inc., one of the world’s largest futures brokerage firms, used a hedge fund vehicle called Liberty Corner Capital Strategy for accounting manipulation while there is no evidence that Liberty Corner Capital Strategy independently raised capital from external investors. On the other hand, Madoff was a pure Ponzi scheme and used hedge fund vehicles to raise capital from external investors. Whether we can readily compare these two cases is up for debate, but for this analysis, we thus used a relatively loose definition of asset managers. In this research, we found that predominantly large number of asset managers called themselves some types of “hedge funds” rather than “private equity funds” or “mutual funds”. Therefore, both private equity funds and mutual funds are classified as “Non-hedge fund collective investment scheme”. (Exhibit 3)
Exhibit 3: Definition of Type of Asset Managers
In our database, we have summarized 102 notable cases from 1998 until 2014. This does not include Platinum Partners which has only come to light in the past year. (Exhibit 4)
Exhibit 4: Sample Statistics
Findings: Characteristics of Asset Manager Fraud Cases
We can see that asset manager frauds peak in 2008 and 2009; this is vital to keep in mind as we can only ascertain asset manager fraud discoveries rather than asset manager fraud themselves. Such frauds only come to light when they become unsustainable and investors begin redeeming at a rate in which the fund cannot handle with new subscriptions. It is important to note that Bernie Madoff Investment Securities LLC came to light in late 2008—December 10 to be exact. The spike in average size should thus be evened out. (Exhibit 5 and 6)
Exhibit 5: Total Alleged AUM by End Year
Exhibit 6: Average Size and Number of Cases by End Year
There is a tricky dilemma that occurs with asset manager frauds: the more successful they become, the more trustworthy they appear. Firms that have a long track record and high number of assets are likely to benefit from confirmation bias—and thus become harder to identify. At the same time, asset manager frauds that have been exposed have, on average, become persistently more durable—with the average fraud being uncovered in 2013 and 2014 lasting well over seven years. In periods without significant asset dislocation, such as our current cycle since 2008, asset manager frauds can become more and more sustainable. When market conditions falter, however, it becomes difficult to rely on new subscriptions to fund outgoing redemptions—and that is when it becomes quite difficult. (Exhibit 7)
Exhibit 7: Average Length By End Year
Asset based lending (ABL) is particularly important to take note of for its propensity to be fraudulent—particularly as a private based strategy, there is less information available but without lockups of private equity funds but with mark-to-market performance updates prevalent in hedge funds. As such, ABL strategies encourages managers to record consistent, mark-to-market returns in asset classes with little publically available information. These two factors put together greatly incentivize hedge fund fraud. (Exhibit 8)
Exhibit 8: Total Alleged AUM by Strategy
With the exceptions of option arbitration and private investment in public equity strategies, the more illiquid an asset class is, the more sustainable a fraud in that asset class is. If a manager were to make a series of faulty bets in credit—knowledge of defaults is newsworthy and thus harder to hide. Additionally, credit investments are finite with set maturity dates—the party cannot go on forever. For fund-of-funds investments or private equity deals, the investment is subject either to lockups or incredibly illiquid volumes. This also disincentivizes scrutiny and fraud discovery.
Surprisingly, there are very few “private equity” related frauds. There were far more asset manager frauds targeting wealthy families and less sophisticated institutional investors. This can be explained by a few reasons: (1) private equity funds set long lock-up periods which means that managers do not have to worry about impending redemptions, (2) the traditional investor base for private equity funds tends to include more institutional investors with often more systematic due diligence processes, and (3) it is more acceptable to assume complete write-downs in private strategies as investors have more realistic expectations of risk. Investors in private equity funds tend to understand that investing in companies entails substantial amount of risks—including the entire loss of capital. This expectation is normalized further through private equity funds’ frequent use of leverage which, if misused, leads to the seizure of underlying assets by creditors. With hedge funds, by contrast, it is never “okay” to have substantial down periods of -20% or -30% which would reflect a similar write-down of a core holding in a portfolio as concentrated as a private equity fund. This, combined with the very real fear of redemptions, encourages hedge fund managers to conceal their losses.
Findings: Patterns of Asset Manager Fraud Cases
We classified causal factors of asset manager frauds into five main categories.
Theft
Fictitious assets
Misvaluation of assets
Outright Concealment of losses
Misrepresentation
The first of which is outright theft. This is when an asset manager misappropriates fund assets for personal use, and often, intends to have done such from the beginning. Next is the creation of fictitious assets and/or the operation of a Ponzi scheme where such reported assets do not exist. The third is the misvaluation of fund assets. Securities here are recorded at a price far exceeding fair market value in the NAV calculation, and thus can conceal further losses. Such happens less in more liquid, transparent securities, and more often in private strategies with poor visibility and poor liquidity. Outright concealment of losses is when managers issue fictitious statements to investors showing positive returns when either those trades never occurred or are loss-making. Lastly, misrepresentation constitutes either (1) managers providing falsified information as to their trading activity and portfolio holdings, or (2) managers deliberately misrepresenting issues such as assets under management, past performance, infrastructure, or other background information of the fund. Within our study of 102 asset manager fraud cases, we categorized funds by their breaches in such areas. In particular, we found that theft was extremely common, as was the concealment of losses and misrepresentation. Classic Ponzi scheme cases and those encompassing more sophisticated strategies of misvaluing assets, were less common. (Exhibit 9)
Exhibit 9: Fraud Causal Factors (% of Total Cases)
It is important to note that the creation of fictitious assets, although somewhat less common in most hedge fund strategies, is in fact predominant within frauds of fund-of-funds and to some extent in asset-based lending. With these two strategies, one should also be careful of the misvaluing of assets—made all the more possible by the lack of information and accurate mark-to-market prices in illiquid, non-public secondary markets. (Exhibit 10)
Exhibit 10: Fraud Causal Factors (% of Total Cases) by Strategy
Misrepresentation is fairly common in all strategies. This is why it is important to understand managers’ investment strategies, their theses, and expect to hold them to account when such is no longer the case. When a manager deviates from what he intended to originally do, this should be a red-flag. Additionally, strategic pivots tend to occur more often when managers encounter high losses. Switching to another strategy or engaging in a “strategic shift” is a logical way for fraudulent managers to conceal losses.
In terms of the ways used by asset managers to go about with their frauds, we identified five central techniques:
Use of related or fictitious of service providers
Lack of third-party supervision
Existence of structural loopholes
Weak internal control mechanisms
Inappropriate marketing and misrepresentation
Among five techniques listed above, inappropriate marketing and misrepresentation is the most frequently used methodology from 1998 and 2009 as more than 7 fraud cases out of 10 used this technique. As this is one of the easiest techniques for allocators to detect through relatively simple due diligence, not surprisingly this technique became less common after 2009. There are two noteworthy changes between pre-2009 and post-2009 periods. Firstly, all techniques but structural loopholes show a declining frequency. As discussed below, asset manager frauds using the structural loopholes are most difficult to detect because either complexity or nature of investment strategy allows asset managers to default allocators. After 2009, it is encouraging that allocators are conducting more thorough due diligence before investing, however, it is also important to notice that frauds are becoming more sophisticated. Secondly, the frequency of fraud cases using service providers and weak internal control declined more than 50%. We think this is a result of more allocators conducting operational due diligence and it became difficult for asset managers to use the techniques to defraud. As discussed below, it is also difficult to conduct due diligence to identify weak internal control as allocators do not monitor asset managers’ activities on a daily basis, therefore, it is encouraging that allocators’ preventive actions (i.e., conducting more thorough operational due diligence) is having a good influence on the overall industry. (Exhibit 11)
Exhibit 11: Number of Fraud Techniques per Case
Fraud Technique: Service Providers
The use of related or fictitious service providers occurs when managers use affiliated or fictitious service providers to prevent legitimate service providers from reporting net asset value accurately. Related service providers or those that are affiliated with the hedge fund in some way, and are thus prevented from conducting thorough checks on the assets of the managements. This is somewhat more difficult to spot as many allocators are unfamiliar with service providers and administrators. However, verifying the independence and proper functioning of the service provider is easily conducted through active due diligence checks. (Exhibit 12)
Exhibit 12: Service Providers by Different Fraud Sub-Technique
Fraud Technique: Lack of Supervision
This occurs when funds either have no auditor, or no administrator, to check and verify the assets of their strategies. This is an egregious and easy-to-spot red flag although it happened a surprising frequency. (Exhibit 13)
Exhibit 13: Lack of Supervision by Different Fraud Sub-Technique
Perhaps the most complicated technique used by hedge funds is through the complicated structuring of investment vehicles. Many large, institutionalized and legitimate funds with an international investor base require almost Byzantine structures for tax efficiency or the targeting of a diverse investor base. However, when funds become too complicated, it becomes harder to spot errors and easier to conceal fraudulent activity. (Exhibit 14)
Exhibit 14: Structural Loopholes by Different Fraud Sub-Technique
Fraud Technique: Weak Internal Control
This is the hardest method for due diligence activities undertaken by allocators to uncover. If firms fabricate bank or expense statements, it is difficult for allocators to go about and verify their veracity. As such, allocators should look out for the separation of duties within investment firms and look for independent operational staff with high levels of personal integrity. (Exhibit 15)
Exhibit 15: Weak Internal Control by Different Fraud Sub-Technique
Fraud Technique: Improper Marketing and Misrepresentation
In recent years with the advent of LinkedIn and increasingly connected investment communities, it has become much more difficult to deceive allocators with regards to managers’ personal backgrounds. Forms of inappropriate marketing, however, can still persist. Many allocators, particularly newer ones, have unrealistic expectations for managers to produce remarkably consistent returns with little down-side risk. Thus, firms may continue to market themselves with a “guaranteed return” when in the investment management space, such is nearly impossible. As such, it is important for LPs to have an open-mind and reasonable expectations for their managers. For example, Bernard Madoff’s four anchor allocators would demand for returns of 18% after receiving performance updates of 16%--which would propel Madoff aide Annette Bongiorno to redraft new customer statements with the new desired return (Source: Fishman, S (2017-01-31). Ponzi Supernova. Amazon Audible). As such, far from being a strength for a fund, a guaranteed return is a likely indication of a fantasy and its corresponding fraud created to prop it up. (Exhibit 16)
Exhibit 16: Improper Marketing by Different Fraud Sub-Technique
Since 2009, due diligence activities verifying the backgrounds and legitimacy of managers and their service providers have become easier to complete and more reliable. However, (1) an increasingly internationalized investor base, (2) more complicated and onerous regulatory regime, and (3) a proclivity for larger, more established funds, have led to managers manipulating their enhanced complexity to create structural loopholes that aide fraudulent activity. The size of hedge fund frauds has become larger and the length of their operations have become longer. Resultantly, each case has become more complex.
Conclusions
Our study finds that almost all asset manager frauds used one or more of five fraud techniques. Some techniques are more sophisticated than others, but it is not impossible to find red flags, or at least many inconsistencies, as shown in our Asset Manager Fraud Case Studies in Appendix. Most of frauds are different, but similar, and even for the fraudulent asset managers, it is difficult to design a perfect fraud. By understanding both Fraud Causal Factors and Fraud Techniques in past fraud cases, we can allocate our limited resources more effectively to detect potential frauds in the future.
There are two types of asset manager frauds: those with the original intention to steal money, and those that have normal motivations, but due to poor performance, hides losses and begin on a slippery slope towards fraud and outright theft. Oddly enough, there were seemingly more frauds that had legitimate original intentions, than those which started out maliciously, and ended up in a fraudulent scenario after a series of bad bets in their investments and decisions in their business conducts. In these cases, we should be able to find many traces of inconsistencies through carefully reviewing documents and related parties and identify red flags before investing (or, even after investing). And, those that start with legitimate intentions but obfuscate poor performance, it can be even easier to detect. With Madoff Investment Securities LLC for example, Bernard Madoff was unwilling to go into details about his investment strategy and would visibly get angry with those investors that probed him into any detail. More sophisticated frauds such as Platinum (see Appendix: Asset Manager Fraud Case Studies), however, did allow allocators some insights into how they made money and what strategies they used. In an odd sense, the firm’s shady reputation provided an alibi as to why they were somewhat secretive. Therefore, it is vital to adopt a realistic mindset with regards to what an asset manager can and cannot deliver. It is perhaps needless to say, but allocators should not expect a perfect track record. Returns that are attractive, remarkably consistent, with some down months—but none high enough to spark redemptions—are likely to be too good to be true. Frauds can “engineer” their returns to look too optimal. Similarly, an allocator should not invest out of emotional proximity or solely from a strong personal relationship. This can derail any due diligence process and give too much free-rein for asset managers to go as they please.
All allocators should have a strong preference for understandable investment strategies with the transparent communication of investment theses and holdings (even if such information is delayed due to market sensitivity). The Endowment Approach’s preference for streamlined and understandable investment strategies, whose volatility may be high due to their long-term, fundamentally-driven and concentrated approach, mean that asset managers selected by the allocators of the Endowment Approach are relatively shielded from frauds. We believe in the merits of specialization. We believe in the benefits of diversification, but also understand the dangers of diversification for the sake of diversification. A search for uncorrelated returns may lead allocators to the strategies, which are vulnerable to frauds, such as asset based lending. It is still impossible to keep ourselves perfectly from frauds, however, proper due diligence and reasonable mindset should help us not blindly falling in love with fraudulent asset managers and ruining our portfolios. That is why it’s important to be diligent.
Our key takeaways from our study are:
Allocators should not easily believe in past performance, which appears to be a straight line even though explanation to achieve such returns sound reasonable (either trading or short-term lending). In reality, a straight line performance rarely exists and an asset manager with such track records won’t come to you for more money.
Exceptionally sophisticated asset manager frauds are hard to come by, but it is worth noting that those who have originally intended to defraud are likely better at it than those who ended up defrauding to conceal their poor performance. Remember it is difficult to design a perfect crime. Almost all frauds we have studied carried discernable red-flags.
Allocators should always conduct basic operational due diligence—it can in itself rule out the majority of fraud cases. Such entails: (1) the reading of all legal documents such as the Private Placement Memorandum and Limited Partnership Agreement, (2) the verifying of assets under management independently, (3) confirmation of the auditor, administrator and legal counsel, (4) reference checks based off of independent sources, (5) the review of audited financial statements, (6) conducting a background check on all the principals of the firm, including matters such dealings with regulators or legal authorities, and (7) having on-site visits and interviewing multiple members of the firm.
We can never be diligent enough, so keep conducting due diligence.
Originally published in 2017-02
Appendix: Asset Manager Fraud Case Studies
Case Study: Petters
Case Study: Weavering
Case Study: Platinum
Evaluating integrity - it's everything . . . and likely a skill AI can't replace!