Product and Strategy Notes: New Products (CBOE - Skew Index), Citi's CLX, UK JP Morgan UCITS; New Research for Advisers and Investors; Implications of the Borg

From many directions come signs of progress toward the eventual listing of interesting new tail risk management products.  The Chicago Board Options Exchange recently announced that it is developing a new market skew index, based on the skew implied by the pricing of S&P 500 Index options. As the CBOE notes in its announcement, "historically, investors have purchased out-of-the-money puts to hedge their equity positions, and sold out-of-the-money calls for premium income. If market participants expect a crisis, they would be more likely to buy put protection, which, all things being equal, would contribute to the increase in the skew [implied by option prices. Statistically, skewness refers to the "tiltedness" of a distribution of possible outcomes, compared to the symmetrical distribution that characterizes the normal "bell curve" distribution]. Market participants could therefore take long positions in the skew index to hedge against future expectations of a tail risk event. Those who feel expectations of a crisis are overplayed could decide to short the index." CBOE also said that its analysis shows that the skew index has no correlation with market volatility. Elsewhere, we have previously written about Citibank's development of a market liquidity index (termed CLX), which one day may be used as the basis for products that enable investors to hedge their exposure to illiquidity risk (which was central to the rapid escalation of the 2008 crisis). Overall, we are enthusiastic about the creation of more investable tail risk hedging products that should rise in value when extreme events occur. Products based on skew and liquidity will be valuable additions to tail risk hedging products that are already available today, including short term treasury funds, volatility products, gold products and currency products (e.g., Swiss Franc ETFs).

On the uncorrelated alpha front, in the UK JP Morgan has launched a new UCITS fund that will invest in 20 separate underlying absolute return strategies, that cover four basic styles: momentum, carry (long high yielding assets and short low yielding assets), mean reversion, and selling volatility (i.e., selling insurance). Another wrinkle in the product is a targeted maximum annual volatility level of 10%, which is to be achieved by dynamically varying fund allocations in accordance with current and forecast market conditions. We have mixed feelings about this product. On the one hand, we strongly support the proposition that giving investors better access to liquid, reasonably priced uncorrelated alpha strategies can raise the probability of achieving long term portfolio return goals. On the other hand, however, we are highly suspicious of bundled products that promise (for higher fees, of course) to achieve something above and beyond what an investor could achieve by herself simply by investing in a number of separate uncorrelated alpha funds. Too often, bundled funds fall short of their intended goals, for the very simple reason that achieving them requires more things to go right (e.g., forecasting, minimizing the cost of more frequent trading, etc.) than have to go right in order to achieve the goals of a simpler strategy. As in so many other areas of life, there is an advantage in KISS -- keep it simple, stupid.

Interesting New Research for Advisers and Investors

• In "Confirmation Bias, Overconfidence, and Investment Performance: Evidence from Stock Message Boards", Park, Konana, Gu, Kumar, and Raghunathan analyze South Korean data to test their "conjecture that investors would use message boards to seek information that confirms their prior belief" and that "this confirmation bias would make these investors more overconfident and adversely affect their investment performance." They find that, as expected, the data support this theory, concluding that "these results suggest that participation in virtual communities increases investors' propensity to commit investment mistakes and is likely to be detrimental to their investment performance."
  
  
•  In "But Will It Make You Happy?" (New York Times 7 Aug 2010), Stephanie Rosenbloom reviews a range of recent research on what really makes us happy, and "emotional efficiency: how to reap the most happiness for your dollar." As we have previously written, Rosenbloom also concludes that "one major finding is that spending money for an experience produces longer-lasting satisfaction than spending money on plain old stuff...Spending on experiences typically strengthens social bonds, which in turn helps amplify happiness." We have previously noted another channel from experience to heightened happiness: The ability to both repeat experiences through memory, and also to re-interpret them in a more positive light. Rosenbloom also notes that "according to retailers and analysts, consumers have gravitated more toward experiences than possessions over the last couple of years...Many retailing professionals think that this is not a fad, but rather the new normal...Industry professionals say they have difficulty identifying any retailer that is managing [to do an excellent job of adjusting to this trend], with one notable exception: Apple, which offers an interactive retail experience, including classes." As we have previously noted, we think this research has important implications for practice development efforts by financial advisers. Specifically, it points to the importance of providing potential and current clients with a series of experiences (e.g., educational newsletters and seminars), ideally in a context that enables them to form stronger social bonds with other people who share their interests, while not costing the adviser too much money to deliver (e.g., a seminar series, a hike, or a social service project). Put differently, it is not just the technical expertise of an adviser that counts, but also the quality and frequency of the experiences that he or she offers clients. 
  
  
• In "Presenting Investment Results Asset by Asset Lowers Risk Taking" Anagol and Gamble confirm what many advisers already know from experience. The authors found that “presenting the results of participants' investment choices segregated by asset lowers their risk taking by 9%." Moreover, "among those participants who spent more than average time analyzing historical results, those who viewed results segregated by asset take on 13% less risk than participants who view historic results from efficient portfolio combinations of these assets."
  
  
• In "The Causal Impact of Media in Financial Markets", Engleberg and Parsons seek to "disentangle the causal impact of media from the impact of the events being reported." They take a very innovative approach to analyzing this problem, "by comparing the behaviors of investors with access to different media coverage of the same event -- earnings announcements of S&P 500 Index companies." They find that, 'local media' coverage strongly predicts local trading, after controlling for [a wide range of other factors].

The Implications of the Borg

On August 6, 2010, Tom Lauricella and Scott Patterson wrote an excellent summary in The Wall Street Journal of what is known so far about the causes of the May 6, 2010 "flash crash" in the U.S. equity market. Their key point is one we have previously made: "Three months later, many market veterans have arrived at a disquieting conclusion: a flash crash could happen again, because today's computer-driven stock market is much more fragile than many believed. Many investors, still gun-shy, have been pulling money out of stocks." As we have also previously noted, we do not believe that the majority of investors have fully absorbed the implications of changes that have occurred in financial markets over the past few years. We live today in a world characterized by multiple pools of liquidity (many of which, known colloquially as "dark pools" are only visible to some investors), in which the majority of trading volume is driven by ruthlessly efficient and constantly adapting algorithms, rather than human cognition and emotion. We have often called this the battle of the trading bots; in light of the implications of the flash crash, perhaps a better term would be a battle of the trading borg (for the Star Trek fans in our audience). In our view, the existence of ever more sophisticated and adaptive trading programs (algorithms) poses a fundamental challenge to those who believe that active management (provided you can identify a skilled active manager) can outperform a relevant index over a long-period of time, on a risk adjusted basis, after all fees and expenses are taken into account.

Algorithms that systematically search for and seek to exploit anomalies and complex interrelationships between securities and securities markets have important implications. First, obvious anomalies and relationships (e.g., systematic "factor exposures") will quickly be discovered and their advantage competed away as algorithms seek to exploit them (to be sure, these anomalies and relationships will remain visible in raw data series, but not data series that take "real world" trading costs and expenses into account.

Second, this will shift the true battle ground of active management to the creation of new algorithms that can discover ever more obscure relationships (or "new factors"), and keep their existence secret for as long as possible (e.g., many have argued that this is the true secret behind Renaissance Technologies Medallion Fund). We have noted before about how this development builds on the "novel intelligence from massive datasets" or NIMD initiative that has been underway in the intelligence community, and received very substantial funding, since the 9/11/2001 terrorist attacks. As a practical matter, the critical question that any investor must ask an active manager whose process is based on taking different "factor exposures" (e.g., momentum, value, small cap, etc.) is "why should I expect you to succeed in the face of competition from algorithmic managers?"

Third, another practical consideration is that if, as we believe, the true high ground in active management based on taking systematic factor exposures has shifted to the quality of managers' learning and discovery algorithms, then most investors are ill-equipped to identify truly skilled managers. Indeed, the managers themselves are probably ill-equipped to make this judgment, except within very short time-frames. While some will point to Renaissance as the exception to this conclusion, our response is that the Medallion fund has been closed for years, and that Renaissance's more recent funds, that were structured to offer more liquidity to institutional investors, have failed to match Medallion’s track record. In other words, even Renaissance isn't perfect.

Fourth, as active management will always be a zero sum game, this raises the awkward question of out of whose pocket are coming the impressive returns being earned by the best algorithms. To be sure, some must come at the expense of market capitalization based index funds. As far as his argument goes, we agree with Rob Arnott that in a world where markets' pricing of assets can wander far from efficiency, market cap weighting systematically must overweight overvalued securities and underweight undervalued securities. That is the logical basis of the alternative index weighting schemes (e.g., Fundamental Indexing) he has done so much to popularize. Where we disagree with Rob is that he chooses not to take his argument to its final conclusion: If everyone adopted fundamental indexing, then it would transform into the market cap weighted indexing. Yet even fundamental indexing, as a systematic strategy whose logic is transparent can and undoubtedly is being exploited by algorithms, in the same manner that every other factor-based active strategy is being attacked and assimilated by the Borg.

Finally, as Star Trek also showed, there are important exceptions to the Borg's claim that "resistance is futile".  We will focus on three we believe are critical. The first are active managers who employ a global macro strategy, forecasting changes in the returns to, and relationships between, broad global asset classes. In the complex adaptive system that encompasses politics, economics and the financial markets, causal factors are frequently complex, evolving, and non-linear in their impact. This makes it extremely hard for any human or algorithm to accurately forecast their effects; indeed, most writers about complex adaptive systems conclude that the best most of us can hope to achieve is a "coarse grained" understanding of their dynamics. However, evidence has also shown that a very few human beings have a combination of cognitive capacity and instinct that enables them to develop an understanding that goes beyond this (see, for example, The Logic of Failure by Dietrich Dorner). Investment managers with this rare skill, who apply it to the world of macro-strategy, have less to fear from the Borg than most of their peers.

The second group with less to fear is active managers who focus on company-specific rather than systematic factors. Call this good old fashioned value analysis. To be sure, the Borg’s discover algorithms are constantly searching for new factors that progressively shrink the size of the company-specific, unique information space in which security-analysis driven active managers can compete. But even the Borg cannot completely eliminate this space, nor assimilate the active managers (or at least their profits) who compete in it. On the other hand, this presents a second order problem, because the investment consultants who evaluate active managers seem much more comfortable opining about "factor-based" managers and the sources of their returns than they are by with the sometimes inscrutable and evolving processes used by old fashioned company and security-focused value managers (of course, this observation also suggests that the Borg are a threat to the existence of the consulting profession -- but that's a story for another time). In other words, the existence of the Borg has probably made it even more difficult to identify in advance the active managers who are truly skilled and therefore could, with a probability beyond sheer luck, deliver superior risk adjusted returns, after expenses and taxes, over a long period of time.

The third group of players who should be able to successfully resist the Borg are market capitalization based index funds (regardless of their form -- e.g., mutual, UCITS, ETF, etc.). As noted above, it is impossible for fund managers, in aggregate, to beat the market, because they are the market. If you believe our assessment of the impact of the Borg, most of their superior active returns will come at the expense of other active managers, rather than market capitalization weighted index funds within a given asset class. At a portfolio level, the answer is less clear, because it is hard to establish just what constitutes a market capitalization weighted portfolio of multiple asset classes (e.g., asset classes like property and timber, with large percentages of their assets not traded present a problem, as do asset classes like commodities, volatility and bonds, where market capitalization is hard to define and/or calculate). Even at this level, however, we would expect that the majority of the Borg’s gains would come at the expense of the most active managers.

A deep understanding of the Borg also leads to deeper worries that are outside the realm of risk and return. When people come to fully understand that the Borg are capturing an ever greater share of active management profits, and that only an elite few institutions and individuals can access the Borg's funds, what happens to people's perceptions of financial markets, the institutions that regulate them, and the political and industry leaders who are in charge of those institutions? In the best case, more and more of them start to invest in broadly based asset class index funds. In the worst case, the majority's perception of financial markets as a rigged casino that is making it impossible for them to realize their long-term financial goals becomes even worse than it is today. And it seems inevitable that the loss of faith and trust in financial markets as an institution would have very dangerous consequences that are hard to envision at this point. But we're betting they wouldn't be pretty. We apologize if you are sitting in a beach chair somewhere while reading this, and we have made your holiday less restful. However, we don't think that the full implications of the Borg are widely (or even narrowly) understood at this point, that a lot is at stake in how this issue evolves, and that it is one that investors much carefully monitor in the months ahead.

| Product and Strategy Notes: New Products (CBOE - Skew Index), Citi's CLX, UK JP Morgan UCITS; New Research for Advisers and Investors; Implications of the Borg | Table: Fundamental Asset Class Valuation and Recent Return Momentum | August 2010 Issue: Key Points | This Month's Letters to the Editor: Who Would You Invite to Dinner?; How Should One Evaluate II's Asset Class Valuation and Economic Updates?; How Did You Identify the Three Regimes You Use in Your Analysis - High Inflation, High Uncertainty and Normal Times? | August 2010 Economic Update: Too Much Leverage and Too Little Demand | Investor Herding Risk Analysis | Overview of Our Valuation Methodology | Uncorrelated Alpha Strategies Detail | Feature Article: The Risk of Deflation and Its Impact on Asset Class Returns | Global Asset Class Returns | Table: Market Implied Regime Expectations and Three Year Return Forecast | Global Asset Class Valuation Updates Detail through July 30, 2010 |