Allocator Insights: Part III. Alan Snyder, Shinnecock Partners
About Alan Snyder
Alan C. Snyder is the Managing General Partner of Shinnecock Partners, a 28-year-old family office investment boutique.
Present – General Partner, Shinnecock Group; Chairman and Board Member, Western Los Angeles Boy Scout
Formerly – Founder, Snap Insurance LLC; Founder, Chairman, CEO, Answer Financial (largest non-carrier seller of auto/home insurance in U.S.); CEO, Aurora National Life Assurance; Executive-in-Charge, Executive Life; President/COO First Executive Life ($20bn in assets); E.V.P./Board Member, Dean Witter (predecessor to Morgan Stanley), launched Discover card as part of a three-member team; CEO Comspec (film production company); Baker Scholar, Harvard Business School; Wall St. Journal Award, Georgetown University; Boy Scout’s Americanism Award/Silver Beaver
Reluctance to withdraw from an investment manager versus “Damn the torpedoes, full speed ahead.”
Please find Part III of our III part series. This section discusses why Shinnecock, as well as other allocators, exit a fund. Once an allocator does all of the discovery and DD, it is frustrating to have to divest from a manager.
Counterintuitively, withdrawing from an investment manager is hard work with many psychological barriers. However, as we learned from Admiral David Farragut’s famous order to his squadron and its reluctant officers during the Civil War Battle of Mobile Bay, sometimes bold action must be taken.
At first blush departing may seem easy. “Just do it,” say the hardcore, but….
1) Sunk cost. After intense due diligence performed prior to hiring the manager, costs incurred, and often a personal relationship created, we must invalidate our original decision. And hope, aided by encouraging words from the manager, springs eternal.
2) Recency bias. When the Concorde supersonic jet was conceived, huge teams of dedicated professionals were galvanized to action with a singular sense of purpose. At every step of the way, it was apparent that even if built, operational costs would doom the pioneering. Yet, the project inexorably went forward. It was simply asking too much from the people and the governments involved to acknowledge this reality; same here in parting company with a manager.
3) Retrospective foolishness. What happens to our self esteem if manager performance improves after we leave? Oops, we were dumb and too impatient.
4) Timing questions. The most frequent trigger for our departure is unexpected losses. Shouldn’t we wait ‘til we get even? We won’t pay any incentive fees until a rebound, meaning it will be that much easier to get back to even.
5) Frictional costs of departure. If we leave, there may be a redemption charge and usually, an audit holdback. We must wait until the account is trued up based on the next audit, which may be months away. The audit holdback is dead capital during this interregnum period and can amount to 5% – 10% of the capital. Furthermore, do we have an alternative place for the capital to be deployed, i.e., are we ready or must the capital be idle until we have identified the “better” alternative?
Admiral Farragut surveyed his scene of battle carefully, determined the risk of hesitancy and weighed his alternatives. His answer was to proceed with dispatch. We, too, must do the same if after careful evaluation, leaving is the proper course of action.
Formally separating exit reasons into categories can distill the process. Some are totally independent of the manager.
1) Changing goals. For example, an insurance company may have invested with a long duration debt manager as a match against its long duration liabilities from long-lived insureds only to experience a shifting liability mix of now shorter duration policies. No fault of the manager if this happened, yet the insurer’s portfolio must change. Alternatively, a manager may have performed well for years with substantial volatility, but their investor may be older and now unable to ride out the peaks and valleys of this volatility.
2) Cash needs. Following the great recession and market collapse of 2008 – 2009, many institutional investors needed additional liquidity to fund their own investor withdrawals. Others may have had ongoing funding commitments for private equity transactions (which, if not met, would result in harsh haircuts to overall value), or unexpected living expenses from a job loss, reduced income due to the poor economy, etc. Hereto, a manager withdrawal is triggered by events unrelated to its performance or operations.
The above scenarios are easy to determine because they are within our control. The next easiest to decide are those from a clear-cut failure on the part of the manager.
1) Operational. A manager’s business may deteriorate from the loss of a large investor, jeopardizing a going concern possibly triggering an expense allocation burden, key personnel departures, regulatory action, or failure to adhere to its own policies. We saw an example of this last failure in a New York based long/short equity manager. They had a superb track record for several years but became enamored of a single position, attempting to turn a mere homerun into a grand slam. That investment grew from 5% to over 90% of the portfolio from averaging down as the stock declined to near zero. The manager, his portfolio, and those investors who stayed were wiped out.
He contravened all of his own guidelines. For those tracking the portfolio’s changes, there was time to get out, albeit with modest losses.
2) Severe losses. Actual losses, or the possibility of losses, are endemic to almost all investment strategies and their practitioners. Negative performance that dwarfs historical experience and is unbalanced against the possibility of future gains usually indicates a failure of process, particularly if a peer group of managers have not had similar challenges in the same time period. Additionally, if severe enough, the risk is that the remaining positions in the portfolio may be toxic waste if cash and “good” positions are liquidated to fund withdrawals. Finally, large declines may trigger increased and unwarranted risk as the manager attempts to regain their “high water mark” to earn additional performance fees. In such cases, time to go.
The last category of exit triggers are the hardest on which to act. The decision is not black or white, but shades of gray.
1) Strategy drift. An actual manager example is illustrative. This talented practitioner created a portfolio of compelling small business loans generating attractive yields. However, over time the portfolio morphed from positions currently paying interest to ones only accruing interest, which resulted in more of a private equity cast to the portfolio. Despite the manager’s strong capabilities and track record, the opportunity set markedly changed. The risks and rewards attributed to the original portfolio were sharply different from those associated with the manager’s new strategy. Here, the investor had to decide if the new portfolio still fit his or her objectives.
2) Sector weakness. Investment cycles can persist for extended timeframes. Managed futures had been a solid performer for decades but the last four or five years has been disappointing. Almost every manager in the space has suffered no matter whether in the top quartile of performers or not. We believe the only answer is to evaluate the size of the position. Do the original arguments about non-correlation, the “insurance policy” nature of managed futures and likely reversion to the mean (buoying performance in the future), still apply? A similar thought process holds for convertibles, master limited partnerships, real estate investment trusts, etc.
3) Better alternative. After all of the pre-investment qualitative and quantitative analytics, maybe our investor has found a superior manager allocation. Two special caveats are in order. No selection should be made in isolation, i.e., evaluation of a manager as a standalone proposition. Each should be viewed through the lens of how they might fit within the portfolio. Consider how each quantitative metric might change viz. drawdowns, volatility, return, correlation, etc. The last
variable, correlation, demands special consideration. A single number for a point in time is inadequate. Sliding-window cross correlation examines this diversifying metric with time as the variant or as an ensemble average. The key is to ascertain how correlation may have changed over different time periods.
4) Efficient frontier. One other consideration is Harry Markowitz’s efficient frontier. Theoretically, it determines the correct mix of strategies/managers to generate the highest expected return with the same standard deviation or level of risk versus alternative blends. It clearly favors diversification yet, in our opinion, has several glaring shortcomings. It assumes a normal (bell curve) distribution pattern (not typical in financial returns) and the capacity for unlimited borrowing (unlikely for most investors). Moreover, a particularly high return alternative can overwhelm other risk measures and distort the comparison. Efficient frontier only works best over extended time periods, as long as 45 years, at a more macro level. Typically, most investors do not have the luxury of such long time horizons, nor do most managers have such a lengthy performance history.
Lastly, operational considerations should not be overlooked. If the allocation is not readily liquid, i.e., a hedge fund, interval fund, etc., what is the notice period required to trigger a withdrawal and at what intervals (monthly, quarterly, yearly)? From the effective withdrawal date, how long will it be until the funds are disbursed? As noted earlier, is there an audit holdback that can help determine the ideal time to withdraw? Specifically, if the withdrawal is not urgent, a year-end exit minimizes the amount of time until the holdback is returned.
Surprises can abound. When an investor makes the final decision to withdraw, he or she might arrive at this conclusion at the same time as other investors. Are there limits on how many investors, or total capital, can withdraw simultaneously, or is any rule based on the individual investor’s account? Furthermore, the legal documentation associated with most alternative investments contains the right to modify withdrawal privileges based on external events loosely determined by the sponsor. This right to recast withdrawal scenarios has also been embraced by many mutual funds, which we witnessed recently with a high yield bond fund. This last stopper can be particularly nettlesome. Liquidity becomes determined by manager discretion.
With over 30 years experience investing in money managers, we have successfully navigated all of the above scenarios. In other words, we have found that the returns are worth the effort. Sometimes, it truly must be “Damn the torpedoes, full speed ahead.”
For the PDF version of this interview, click here