Product and Strategy Notes: Carbon Update; Will This Crisis Finally Change Investor Behavior?; Private Equity Update; Interesting Research; Product Reviews: JBGOUA.SW, ZGLD.SW, USO, USL, ADANX, BWZ, ISHG, IGov, PSAIX; Changes to Index Investor's Uncorrelated Alpha Allocation

Carbon Update

In the past, we have noted both the possibility of investing in carbon emission credits (via the GRN exchange traded note), as well as our initial skepticism about whether doing so would provide much in the way of portfolio diversification benefits (e.g., in the last quarter of 2008, their value fell in line with the FTSE All Country World equity index). The correlation between the two will only fall if the emissions trading system is widely adopted as a means of limiting CO2 emissions, and if supply of emissions certificates is severely curtailed - that is, if the effective price of CO2 emissions is driven very high. In the short term, that would drive down economic growth and equity values, as well as the correlation. Of course, given the current condition of the world economy, it might also trigger widespread political unrest, as has already been the case with Australia's announced but not yet implemented cap-and-trade system. In the near term, the imposition of tighter carbon emissions limits seem likely to be delayed.

However, this is not to say that there has not been interesting news on the carbon front - on the contrary, we've seen a number of very interesting developments. To begin with, the conventional wisdom that cap-and-trade systems are preferable to straightforward carbon taxes may have been fatally undermined by the financial crisis and spike in populist anger at Wall Street. As John Dizard cynically noted in a recent Financial Times column, "Wall Street and Chicago always like the creation of trading markets for new assets, especially if they can be inefficiently priced by the professionals." Canada has already seen an attempt to launch such a tax - the Liberal Party's "green shift' proposal in the last election would have instituted a revenue neutral carbon tax, whose revenues would have been offset by income tax reductions. However, it went down with the Liberals' overall electoral performance.

New research from the McKinsey Global Institute ("The Carbon Productivity Challenge") suggests that we may have not yet heard the end of carbon tax proposals. McKinsey notes that achieving a 10x increase in GDP/metric ton of carbon emissions would require investment of only $570 billion per year between 2010 and 2030 -- just 2 to 4 percent of the total capital expenditures that McKinsey forecasts will be made over this twenty year period. The report further concludes that 70 percent of the technologies required are either available now or on the horizon (including, we note, recognizing the full environmental value of forests). The remaining 30 percent of technological progress can only be achieved if "we put a clear price on carbon." From a corporate investment point of view, this makes sense, as anyone who has presented a discounted cash flow analysis to a board well knows. It is hard to see how the level of certainty required to support higher investments in carbon productivity can be achieved with a cap and trade system - a carbon tax seems a much more efficient approach. We have no doubt that this is why Rex Tillerson, CEO of ExxonMobil, said in January that he favored this approach over a cap and trade system. Yet the proper level at which such a tax should be set remains subject to great uncertainty, as evidenced by another recent report, this one from the The MIT Joint Program on the Science and Policy of Global Change. In "Uncertainty in Greenhouse Gas Emissions and Cost of Atmospheric Stabilization," Webter, Palsev, Parsons, Reilly and Jacoby use Monte Carlo simulation to explore the impact of uncertainty on one hundred parameters in the MIT Emissions Prediction and Policy Analysis model. They "find considerable uncertainty in emissions prices" under various stabilization targets for aggregate CO2 levels in the atmosphere. "For example, the CO2 price in 2060 under an emissions constraint targeted to achieve stabilization at 650 parts per million [of CO2 in the atmosphere, versus about 385 today] has a 90% confidence range of US$14/ton to $88/ton. A 450ppm target in 2060 has a range of $241/ton to $758/ton." As to what is driving these wide ranges of outcomes, the authors conclude that "despite significant uncertainty in future energy supply technologies, the largest drivers of uncertainty in costs of atmospheric stabilization are energy demand parameters, including elasticities of substitution [between the energy, capital and labor needed for a given level of output, and between different sources of energy] and energy efficiency trends." A final uncertainty, much noted by others, is the extent to which developing countries, particularly China, will agree to any type of emissions pricing. In so far as such pricing could restrain economic growth, and thereby threaten political stability, the answer thus far has been "no." Indeed, one suspects that the environmental consequences of higher CO2 levels, and their social impact, would have to become significantly more severe before developing country governments are willing to change this position.

Beyond these political obstacles to successful emissions reductions, another report, just published in the Proceedings of the National Academy of Science, suggests that we may soon see a fundamental rethinking of the division of resources between efforts to stabilize emissions levels and efforts to either mitigate or reverse their climate impact. In "Irreversible Climate Change Due to Carbon Dioxide Emissions", Solomon, Plattner, Knutti and Friedlingstein show "that the climate change that takes place due to increases in carbon dioxide concentration is largely irreversible for 1,000 years after emissions stop...Among irreversible impacts that should be expected if atmospheric carbon dioxide concentrations increase from current levels near 385 ppm to a peak of 450 - 600 ppm over the coming century are irreversible dry-season rainfall reductions in several regions [e.g., around the Mediterranean, the U.S. Southwest and Mexico] comparable to those of the 'dust bowl' era, and inexorable sea level rise."

At the current rate of increase of 2.5 ppm/year, we will reach 450 ppm in just 26 years, and that assumes the rate of increase does not become higher due to non-linear feedback effects. That is an extremely short time window for emissions reductions to have an impact, particularly given political opposition to carbon pricing in most developing countries. Hence, we are beginning to see more discussion not only of increased spending on preparation for a significantly different climate, but also of so-called "geoengineering" to offset the impact of higher CO2 levels on planetary temperature. These fall into two broad categories - "albedo management" which increases the amount of sunlight reflected back from the earth, and "carbon management", which uses largescale techniques to sequester larger amounts of CO2, including aggressive forestation and "bio-char burial" (using more efficient production of charcoal to sequester carbon and simultaneously enrich soils). In a recent report on geoengineering ("The Radiative Forcing Potential of Different Climate Geoengineering Options" by Lenton and Vaughan), the authors conclude that albedo management has the most short term potential. These techniques are basically aimed at increasing the brightness and reflectivity of clouds, either by seeding them with chemicals or by pumping seawater into the air over the oceans to increase the rate of cloud formation. Of course, geoengineering also raises issues, including unforeseen consequences and the political issue of who pays for the negative impacts of the resulting changes in global weather patterns.

In sum, a number of emerging trends in the climate debate have yet to receive significant coverage in the mainstream media. In our view, these trends suggest that emissions credits are unlikely to become a new asset class that can provide investors with a significant new source of returns that have a low correlation with those on existing asset classes. On the other hand, these trends also seem likely to give rise to a much larger set of investment opportunities that could allow skilled managers who take a structured approach (e.g., going long specific stocks and short the overall equity market) to generate the uncorrelated returns that are so valued by wise investors. While we are not there quite yet, this is clearly an area to watch.

Will This Crisis Finally Change Investor Behavior?

A lot has happened since we started The Index Investor in 1997. Through it all, we have held fast to a core set of beliefs: (1) that investors should primarily focus on the returns needed to fund their liabilities, and not external benchmarks; (2) that consistently successful active management (or picking active manages who will succeed in the future) is beyond the skills of most investors; (3) that as a result, only a small portion of a portfolio should be allocated to active management, and that focused on uncorrelated alpha strategies; and (4) that the bulk of investor's efforts should be focused on allocating wisely between different broad asset classes, rebalancing systematically, minimizing expenses and taxes, and maintaining a constant watch for the substantial overvaluations whose subsequent crash can severely affect the chances of achieving one's long-term goals. To be sure, there have been others who have conveyed the same or similar message - for example, see Mark Kritzman's excellent recent article "Rules of Prudence for Individual Investors." Unfortunately, in aggregate, our collective voice has been overwhelmed by the money spent by the active management community to communicate a different message. As Bob Shiller pointed out in a recent New York Times column ("How About a Stimulus for Financial Advice?"), "most people get financial advice only from sales representatives of one sort or another" who "face competitive pressures to promote products that exploit to the hilt [people's tendency to behave irrationally and make judgment errors]". Advice from people with a fiduciary obligation to protect their clients' best interest still remains a luxury to which too few people have access.

In 2001, we wondered how much the bursting of the technology bubble would change this. The subsequent seven years provided an answer that was, at best, only mildly encouraging for people who share our views. So it is only logical to ask whether the aftermath of the current crisis will be any different.

Some of the available evidence is encouraging. The past few months have seen big outflows out of actively managed funds and into index mutual and exchange traded funds. The extent to which this reflects greater awareness - either explicit or instinctive - of recent research on broker sold funds remains unknown. For example, in "Assessing the Costs and Benefits of Brokers in the Mutual Fund Industry", Bergstresser, Chalmers and Tufano found that, "relative to direct-sold funds, broker-sold funds deliver lower risk-adjusted returns, even before subtracting distribution costs." They conclude that their results "are consistent with either substantial non-tangible benefits delivered by the broker-distributed sector [e.g., planning and tax services, education and counseling, etc.] or with conflicts of interest between brokers and their clients." Regarding the latter, another paper ("The Use and Abuse of Mutual Fund Expenses" by Houge and Wellman) concludes that "as the mutual fund industry becomes more adept at segmenting customers by level of investment sophistication...load mutual fund companies take advantage of this ability and charge higher expenses to their target customer: the less-knowledgeable investor. No-load companies, which tend to attract the more sophisticated investor, offer lower expenses."

Elsewhere, Pensions Age reported that ninety six percent of pensions surveyed reported that they get good value for their money from passive investments, while sixty percent thought Funds of Hedge Funds (with their fees on top of fees structure) were a poor value. Yale's David Swensen was even harsher on fund-of-funds in a recent interview with The Wall Street Journal: "[they] are a cancer on the institutional investor world. They facilitate the flow of ignorant capital. If an investor can't make an intelligent decision about picking [hedge] fund managers, how can he make an intelligent decision about picking a fund-of-funds manager who will be selecting hedge funds? There are also more fees on top of fees. And the best managers don't want fund-of-funds money because it is unreliable. You need to be in the top 10% of hedge funds to succeed. In a fund-of-funds, you will likely be excluded from the best managers."

Following another recent survey, Spectrem Group reported that nearly two thirds of U.S. millionaires said their investment advisers had failed them, after watching their portfolios decline by an average of thirty percent last year (17 percent lost 40 percent or more). To put that in perspective, the following table shows the Economist Intelligence Unit's estimate of the number of households that had net wealth of US $1 million or more in 2007:

Country	Households (in millions)
United States	16.6
United Kingdom	4.1
Japan	3.6
France	3.0
Italy	2.8
Germany	2.4
Canada	1.1
Spain	1.0
Australia	0.9
Switzerland	0.7

Not surprisingly, another recent survey, this one by Oppenheimer Funds, found that 66% of U.S. financial advisers planned to spend the most time in 2009 addressing wealth preservation and asset allocation with their clients. And a recent survey by Met Life and Greenwich Associates found that asset allocation was ranked as the most important "risk factor affecting your pension plan" by 54%, almost the same as the 47% who ranked underfunding liabilities as most important - and only 44% of respondents rated themselves as being "very successful" in managing their top-ranked risk factor.

Clearly, there are some reasons to hope that, to use a phrase that is sadly familiar to anyone with a sense of history, "this time it will be different." Above all, the flow of funds data - with the relatively large inflows into indexed ETFs - suggests that a fundamental change may be underway. This is consistent with a body of research that finds personal experience - which generates both cognitive and emotional responses - is a more powerful motivator of behavioral change than information received about the experience of others, which generally lacks the same emotional impact (see, for example, "The Tree of Experience in the Forest of Information" by Simonsohn, Karlsson, Loewenstein and Ariely).

On the other hand, some deeply ingrained aspects of human nature seem to work in favor of actively managed products. For example, studies have repeatedly found that when they have lost money relative to a reference point, human beings tend to become less loss averse and more willing to take risks. In the United States, rising levels of both consumer borrowing and gambling industry revenues in the face of a growing retirement income security seem to provide strong evidence of this phenomenon. This undoubtedly accounts for "performance chasing", and the overall susceptibility to active funds' promise of high returns that will almost magically fix a scary financial situation without the pain of reduced consumption. Repeated surveys have also found that human beings tend to be overoptimistic about their future portfolio returns and overconfident about their relative skills as investors (though we noted this is much more true of men than women). Hence the perennial appeal of the active management community's question to investors who buy index products: "Who wants to earn only average returns?"

When only 18% of the U.S. adult population (about 43 million people) is willing to join a health club (at an average membership cost of between $360 and $400 per year), we are skeptical that a greater percentage would be willing to pay similarly explicit (and possibly greater) average annual fees to a fiduciary adviser who could help ensure their financial well-being. In this regard, both Citibank's "myfi.com" initiative (which involves Jonathan Clements, who wrote so many great columns for the Wall Street Journal) and Financial Engines continue to be experiments well worth watching. On balance, rather than a widespread change in a willingness to pay for unbiased financial advice from a true fiduciary, we suspect that the most recent blow to families' finances will instead result in much greater political pressure to expand public pension systems. At best, this might include a more widespread move towards Australian style mandatory individual pension contributions, with a choice of index fund investment options, and the requirement that a substantial portion of the plan balance be annuitized at retirement. At worst, it will result in demands for higher benefit payments from already strapped "pay-as-you-go" social security systems.

That said, there are also some encouraging early indicators that a significant shift is underway among affluent investors, and among institutional pension plans whose trustees have a fiduciary obligation to invest prudently. Only time will tell whether this trend will sustained, and, if it is, whether it will have a significant impact further down the wealth distribution.

Private Equity Update

A number of new reports were recently issued on various issues related to private equity. The British Private Equity and Venture Capital Association published a report on the sources of returns realized on fourteen of the biggest private equity deal exits that occurred between 2005 and 2007. In aggregate, these deals realized a return 330% higher than a matched sample of publicly listed companies. Of this amount, 100% was due to rising stock market multiples, 167% came from the use of leverage, and 62% (i.e., 19% of the total incremental return) came from strategic and operational improvements. While the sample size was criticized as too small, the finding that only about 20% of the total investment return was due to operating improvements is in line with previous studies. This is significant for investors in private equity funds, since, in the foreseeable future market multiples and the availability of debt financing are both likely to contract - perhaps severely.

No doubt with this in mind, the Boston Consulting Group has published a study titled, "Get Ready for the Private Equity Shakeout." Their conclusion is stark: "Private equity firms have enjoyed extraordinary growth and returns over the past five years, but the collapse of the world's debt markets and the deepening economic crisis have brought this boom to an abrupt end, with potentially severe consequences for private-equity firms, the companies they own, and the real economy." They assert that "private equity is in the middle of the perfect storm", since, "the debt bubble has burst, company earnings have dropped, multiples have collapsed, institutional investors are reducing their private equity asset allocation, and most private equity portfolio companies are expected to default." BCG wonders whether portfolio company debt constitutes "a hidden time bomb", since it is not clear who, after repeated securitizations (via Collateralized Debt Obligations) holds the ultimate risk, and how much capital supports it. In the end, BCG believes that "the biggest impact of the perfect storm will be on the private equity firms themselves - around 20 to 40 percent will disappear, at least 30 percent will survive, and the fate of the rest hangs in the balance." When the smoke clears, BCG expects that "pure debt and multiple players will disappear", leaving those firms with the strongest "operational value creation" capabilities. In a previous report ("The Advantage of Persistence"), BCG concluded that two factors lay at the heart of these capabilities: (1) superior industry expertise, which benefited both access to deals and the ability to generate value creating strategic insights, and (2) "the ability to come in and turn around the operations of a portfolio company."

As someone who spent a significant part of his career advising and later leading corporate turnarounds, this second capability is a subject near and dear to my heart. Over the years, I realized that there were two aspects to it. The first I called "stop losing", or getting the basics right to increase current cash flow. It required a fairly directive approach, and could usually be counted on to move an organization from the left tail into the middle of the corporate performance distribution. However, moving into the right tail - delivering superior performance - was always a far greater challenge. From a valuation perspective, it meant a combination of higher expected growth and/or lower perceived risk. This required not only insightful strategy, but, far more important, a talented and inspired organization that was able to work together to share important information and quickly and effectively adapt as change undermined the original strategy's assumptions. My major disagreement with private equity proponents is their belief that the best way to generate this behavior is by providing management teams with the possibility of large monetary rewards. The first issue is the obvious one: if that upside potential isn't widely shared, it inevitably creates a destructive wedge between the leaders and the led. However, the second issue is more subtle, and in my experience more important: the available evidence suggests that the possibility of high rewards does not lead to better outcomes - in fact, it can lead to worse ones (see, for example, "Large Stakes and Big Mistakes" by Ariely, Gneezy, Loewenstein, and Mazar, and "Doing Good or Doing Well? Image Motivation and Monetary Incentives in Behaving Prosocially" by Ariely, Bracha, and Meier). From what I have seen over the years, organizations are far more likely to achieve superior performance when their members believe they are pursuing a shared and valued purpose (e.g., "improving the environment", "creating great customer experiences") than when they are simply trying to become rich - or, to use the frequently heard term, "maximize shareholder value." Organizations as diverse as high performing schools, special forces teams, and orchestras are filled with people who are inspired not by a desire to get rich, by rather the shared pursuit of a worthy purpose. If the realization of that purpose coincides with great monetary rewards, all the better. But on its own, I've never seen the desire to get rich produce a great company that has stood the test of time. Most human beings don't work that way. Unfortunately I'm not sure how many private equity partners fully grasp this. Maybe that is why  a significant portion of the industry's historical returns seem to be based on a formula of making quick operational improvements to raise cash flow, leveraging up, and selling out a higher multiple. And maybe that is also why, now that the old game is over, so many of their firms are currently trading at a deep discount to their stated net asset values.

Interesting New Research

As regular readers know, one of our core assumptions is that financial markets are a complex adaptive system, in which a wide variety of investment strategies (e.g., value, momentum, indexing, and market making) constantly compete and evolve as investors make decisions on the basis of imperfect information, time pressure, emotions and social considerations. As a result, while markets are attracted to equilibrium and efficient pricing, they are seldom in this state; the supply of and demand for returns on different asset classes are not always in balance, and substantial overvaluations and undervaluations can occur. We have frequently noted that an important area of research is developing a better understanding of the agent/investor level behaviors that cause these macro-level effects to emerge. Coates and Herbert recently published a fascinating paper in the Proceedings of the National Academy of Sciences of the U.S.A. on this issue. "Endogenous Steroids and Financial Risk Taking on a London Trading Floor" reviews the impact of two hormones on traders' behavior and performance. "Testosterone mediates sexual behavior and competitive encounters. It rises, for example, in athletes preparing for a competition and rises even further in the winning athlete, while falling in the losing one. This priming of the winner can increase confidence and risk taking and improve chances of winning again, leading to a positive-feedback loop termed the 'winner effect.' Cortisol plays a central role in the physiological and behavioral response to physical challenge or psychological stressors. It is particularly sensitive to situations of uncontrollability, novelty and uncertainty." The authors found "a significant relationship between testosterone and financial return, and between cortisol and financial uncertainty, with the latter measured by both the variance of returns and of the overall market....When traders experienced acutely raised testosterone, they made higher profits, perhaps because testosterone has been found, in both animal and human studies, to increase search persistence, appetite for risk, and fearlessness in the face of novelty." However, prolonged exposure to elevated testosterone levels can have harmful effects, including "impulsivity and sensation seeking, harmful risk taking, euphoria and mania."

In contrast, "traders experienced elevated cortisol in anticipation of higher volatility." As with testosterone, while a temporary increase in cortisol can be beneficial, prolonged exposure to elevated levels can have harmful effects. Elevated cortisol "can increase motivation and promote more focused attention, as well as aid the consolidation and retrieval of important memories." However, "during periods of chronic stress, prolonged high cortisol levels can promote selective attention to mostly negative precedents, stimulated feelings of anxiety, and produce a tendency to find threat and risk where none exist." The authors note that testosterone and cortisol, "as they fluctuate with risk and return, may alter a trader's ability to make optimal decisions." At the macro level, "cortisol levels are likely to rise during a market crash and, by increasing risk aversion, to exaggerate the market's downward movement. Testosterone, on the other hand, is likely to rise in a bubble and, by increasing risk taking, to exaggerate the market's upward movement." Together, "these endocrine system feedback loops may help explain why people caught up in bubbles and crashes often find it difficult to make rational choices." The only thing we would add to this is that fact that other research has found that social relationships and reactions, as well as returns, can trigger heightened feelings of euphoria and fear. So the feedback loops involved are even more complex than those described by Coates and Herbert.

Still, the paper's findings align closely - and help to explain - findings in two other recent papers. In "Momentum Traders in the Housing Market", Piazzesi and Schneider study the evolution of consumer beliefs about future house prices during the recent boom, using the Michigan Survey of Consumer Attitudes. Their analysis "find that a small cluster of households always believes it is a good time to buy a house because prices will rise further." They also find that the size of this cluster had doubled by the end of the boom, reaching a 25 year high in the second quarter of 2005. Finally, they show how, in a market where volumes are low and prices are set in part based on references to recent transactions, the actions of even a small group of overoptimistic buyers can have an outsized impact on average house prices.

The second paper is "Expectations of Risk and Return Among Household Investors: Are Their Sharpe Ratios Countercyclical?" by Amromin and Sharpe. After also analyzing data from the Michigan Survey of Consumer Attitudes, they find that "expected future returns appear to be extrapolated from past returns" and that "expected risk and return are strongly influenced by respondents' perception of economic prospects. [For example] when investors believe macroeconomic conditions are expansionary, they tend to expect both higher returns and lower volatility." In other words, household investors' expectations are procyclical and wrong - a look at the historical data show that the highest returns are generally realized by investors who buy at the bottom of a market when, for example, dividend yields are highest. Yet this is exactly when household investors have the lowest expectation of future returns, and perceived risks are at their highest.

Another recent paper reinforces this point. In "Forecasting Stock Market Returns: The Sum of the Parts is More than the Whole", Ferreira and Santa-Clara start with the conclusions reached by Goyal and Welch in their paper "A Comprehensive Look at the Empirical Performance of Equity Premium Prediction", who test a wide range of predictors and find they do a poor job of forecasting future equity returns. Ferreira and Santa-Clara attempt to improve on this performance by using a simple dividend discount valuation model, and separately forecasting each of its components (the dividend yield, earnings growth, and changes in the price/earnings ratio). They forecast the future dividend yield using the currently observed dividend yield, earnings growth using its twenty-year moving average, and test a number of different approaches for predicting changes in the earnings multiple. We found it interesting that they use shrinkage estimators to improve their forecasting results, as that is the same approach we have advocated and use in constructing our model portfolios. Just as important, the authors find that "a considerable improvement in the out-of-sample forecasting performance comes just from forecasting the dividend yield and earnings growth separately" - the same approach that underlies our monthly asset class valuation updates. Finally, with respect to market multiples, and in line with the Coates and Herbert paper, the authors conclude that "it is remarkable how little of the time variation in the price-earnings ratio is captured by [changes in economic variables]. It seems the changes in the market multiple over time have little to do with the state of the economy." Hence, "the reversion of market multiples [to long term averages] is quite slow and at times takes almost ten years; as a result, the expected return coming from multiple reversion varies substantially over time and takes both positive and negative values."

Another paper we found interesting was "Predictability and Good Deals in Currency Markets" by Levich and Poti. They find that, between 1971 and 2006, periods of high and low predictability tend to alternate, which is consistent with the adaptive markets hypothesis. Another timely paper is "Wages and Human Capital in the U.S. Financial Industry: 1909 - 2006" by Phlippon and Reshef. They conclude that "wages in finance were excessively high around 1930 and from the mid 1990s until 2006 [the end of their sample]. However, "from the mid-1920s to the mid 1930s, and from the mid-1990s to 2006, the compensation of employees in the financial industry appears to be too high to be consistent with a sustainable labor market equilibrium." For the latter period, the authors "estimate that rents accounted for 30% to 50% of the wage differential between the financial sector and the rest of the economy." They also note that their "investigation reveals a very tight link between deregulation and human capital in the financial sector. Highly skilled labor left the financial sector in the wake of Depression era regulations, and started flowing back precisely when these regulations were removed. Deregulation apparently unleashed creativity and innovation [they cite aggregate IPO activity and credit risk] and increased the demand for skilled workers." In sum, the authors conclude that compensation in the financial services industry has been driven by a combination of deregulation, "creative destruction in the corporate sector, triggered by technological revolutions and the financing needs that result" and employees' ability to extract rents. Given these conclusions, it will be interesting to see how financial sector compensation evolves in the face of two competing forces: on the one hand, a wave of new regulation, and on the other, the increased financing and investing needs that will be triggered by the growing pace of change in nanotechnology, materials science, biotechnology and environmental technology.

We were also fascinated by "Fundamental Value Investors: Characteristics and Performance" by Gray and Kern, as it was like reading the findings of foreign anthropologists who have just studied a community you know well. The motivation of their research is that "real world value investors presumably drive asset prices to fundamental values, in contrast to technical traders or index investors. Therefore, studying fundamental value investors' thought process can help researchers understand why and how assets are priced empirically." A noble and common sense, goal, no doubt about that. The data set used by Gray and Kern is composed of all investment reports submitted to the Value Investors Club between 2000 and mid-2008. And what did our investigators find? "In our sample, value investors overwhelmingly focus on measures of intrinsic value: they examine valuation models based on discounted free cash flows, use various earnings multiple measures, and often search for growth at a reasonable price...They also favor the analysis of open market repurchases, net operating losses, spin-offs, turnarounds and activist involvement." However, "none of the investment theses [the authors] analyzed made use of the statistical asset pricing models found in the academic literature." In other words, "value investors are not focused on high book-to-market stocks, but instead on intrinsic value and signaling factors in the market...They tend to favor smaller stocks with a value bias for long positions and small growth stocks for short positions...[The authors] also find evidence that value investors reliably outperform the market" in spite of the fact that their thinking seems "fairly one dimensional" and that they "utilize only a few tools when making their investment decision." One can only imagine the reaction of Warren Buffett and Charlie Munger (or Ben Graham's ghost) if they ever read this paper.

It was only after we read the paper on value investors that we came across a recent gem by Berger, Kabiller, and Crowell, titled "Is Alpha Just Beta Waiting to Be Discovered? What the Rise of Hedge Fund Betas Means for Investors" (it is published by AQR Capital Management). The authors begin by accurately noting that "colloquially, alpha has come to mean the excess returns from active management. But in truth, the concepts of alpha and beta have their roots in portfolio theory. Empirical analysis uses linear regression to decompose the returns of an asset or portfolio into two components. One component is beta, the portion of returns that can be attributed to one or more systematic risk factors. [At the portfolio level], most common risk factors (betas) have historically been traditional investments, like equity and bond markets. More recently, investors have broadened their portfolio analysis to include "exotic" betas, such as emerging market equities, commodities and real estate. The remaining component is alpha, the portion of returns that cannot be attributed to these various risk factors. At the level of a given equity, return beta factors might include not only exposure to the overall market, but also exposure to factors based on size, value, and momentum.

In other words, from a "statistical perspective, alpha is not returns from active management, but rather returns than cannot be explained by exposure to recognized risk factors. This in turn means that, as new risk factors are discovered and popularized, the portion of returns attributed to alpha decline, as part of alpha is reclassified as beta." In practice, what the discovery of a new risk factor often means is the creation of an index to measure and track it, and the creation of new low cost index investment products (e.g., futures, ETFs and mutual funds) that enable investors to gain exposure to it at a much lower cost than before (when it was considered part of more expensive alpha).

With this background, the authors proceed to describe how alpha can be further reduced by the use of new "hedge fund betas", which capture some of the basic risk exposures that underlie common hedge fund strategies. One example they use is merger arbitrage, a strategy that attempts to profit by going long the stock of the target of an announced but not yet closed deal, and short the stock of the acquirer. There are two schools of thought about why this strategy should be profitable. The first, noted by the authors, is that, not being sure the deal will close, some holders of the target company's stock will sell at less than the price of the deal. The second, not mentioned by the authors, is the fact that buyers are perceived to have overpaid, which drives down the price of their own stock. As the authors note, "the beta of merger arbitrage comes from capturing the risk premium that exists in the aggregate of all investable deals." The authors stress that, at this point, creating and capturing (via trading) these hedge fund betas requires quite a bit of skill.

For example, they note that "a key distinction between hedge fund betas and traditional betas relates to capacity. With any risk exposure, investors must understand the premium they expect to earn for bearing that risk. Any beta - from the most traditional to the most exotic - can become overcrowded [a synonym for "overvalued" or "likely to produce lower returns than you expect, or even losses"]." That said, the authors note that "hedge fund betas, which seek to exploit anomalies in global markets, inherently have limited capacity. If too much money seeks to exploit an anomaly, the anomaly will disappear and the expected returns from exploiting it will fall. Conversely, when capital moves out of a strategy, expected return will rise." As they note, "money tends to flow to different strategies not based on their expected risk premium, but rather on how well they have done in the recent past....This can lead to [situations] where strategies with little or no expected risk premium (due to recent strong performance) nonetheless attract the most capital, shrinking their expected return further, while strategies offering more risk premium (perhaps due to poor recent performance) see their investors flee" because they believe the strategy "is no longer working." In the authors view, this type of investor behavior makes rebalancing one's exposure to hedge fund betas critical, and an important part of the value added by firms like AQR.

As far as it goes, this is an excellent article, which is well worth reading. However, from our point of view, its initial dismissal of the "colloquial" meaning of alpha - returns derived from active management - glosses over some very important points. Whatever you choose to call it, there is a critical distinction between the returns that can be earned passively - that is, without having to make any forecasts - from simply investing in index products that track the performance of broadly defined asset classes, such as U.S. or developed market equities, fixed income, property, commodities and the like. To be sure, asset allocation at this level is not wholly passive, as we have frequently noted in these pages. For example, there are questions about how to define broad asset classes (i.e., the extent to which they represent overlapping exposures to underlying return generating processes, or, if you will, risk factors - say, liquidity, to use a current example). There are also important questions about how to properly measure the returns on these asset classes - for example, the right way to construct an index that tracks the economic performance of bond or commodity markets remains a controversial topic on which reasonable people can and do disagree. There are also questions about how much of each asset class to include in a portfolio, which is a function not only of an investor's goals, but also about the extent to which you believe it is possible to forecast - with a degree of accuracy beyond luck - either returns and/or risk (and we also note that there are plenty of arguments about how to best define the latter). Finally, there are questions about whether even a broad asset class can become so overvalued that a prudent investor should temporarily move out of it, and into either cash or another broad asset class that is fully or undervalued.

Given this, the colloquial distinction between beta and alpha - between returns from basically "passive" (though not completely decision and forecast free) allocations versus returns from "active" exposures that require more, and often more frequent, forecasts and decisions - is far from trivial. When you broaden the range of forecasts and decisions you have to make, you cannot help but increase the chance of error as you move further from the core of your expertise. Some will claim that this problem can be avoided by hiring an expert to provide advice on, or make, some or all of these decisions. Yet that simply shifts the nature of the forecasting and decision problem (which expert should we hire?), and does not eliminate it. Moreover, it also seems likely that the more decisions one has to make in a given period of time, the more likely limited attention and cognitive resources will lead to greater use of mental short cuts (heuristics) and (particularly where the stakes are high) greater influence of affective and hormonal factors that collectively raise the probability of error. In contrast to Grinold and Khan's "fundamental law of active management" (which links active returns to the accuracy of forecasts and the number of bets made), we believe that all the findings of decision research point towards an unavoidable trade-off between the two factors (a point also made by Richard and Robert Michaud in their article "The Fundamental Law of Mismanagement"). Moreover, we have also long noted that the superior models and/or information sources that underlie superior forecasts will inevitably be undermined by the ongoing evolution of the complex adaptive system that we call the economy, society, and financial market, which should also make one skeptical about the likely long term efficacy of an active management strategy. To cite an example we've used before, in their recent paper "False Discoveries in Mutual Fund Performance: Measuring Luck in Estimated Alphas", Barras, Scaillet, and Wermers show that only 0.6% of the 2,076 managers they studied produced skill-based alpha after costs were taken into account. And this does not include the negative impact front-end loads and taxes, which further worsen the performance of many actively managed funds.

On the other hand, we also recognize the substantial potential benefits that returns from a successful uncorrelated alpha strategy can provide to a portfolio. It is for that reason that we include an allocation to these strategies in most of our model portfolios. However, because of our well-founded doubts about the chances of successfully picking truly skilled active managers, and, even if we pick them right, our doubts about those managers retaining their edge over time, we limit the size of our uncorrelated alpha allocation and try to diversify it across a range of strategies.

New Products

The start of the year has seen a bumper crop of new product launches that caught our eye. Barclays has launched two new iPath Exchange Traded Notes (ETNs) that track futures contracts on the VIX index (which tracks the implied volatility of the S&P 500 Index). Both ETNs have .89% annual expense charges. The return on VXX comes from continuously rolling over the two closest month VIX futures contracts. The return on VXZ comes from rolling over contracts on the fourth, fifth, sixth and seventh month out VIX futures contracts. If this sounds complicated, it is. In some ways, these products are like the beauty contest described by John Maynard Keynes, where the object is not to pick the prettiest contestant, but rather the contestant that the greatest number of other people will judge most beautiful. As Keynes noted, "It is not a case of choosing those [faces] which, to the best of one's judgment, are really the prettiest, nor even those which average opinion genuinely thinks the prettiest. We have reached the third degree where we devote our intelligences to anticipating what average opinion expects the average opinion to be. And there are some, I believe, who practice the fourth, fifth and higher degrees" (Keynes, General Theory of Employment Interest and Money, 1936). Let's put this in more practical terms. We know that the VIX index rises when investors become more uncertain about the future. For example, in January 2009, the VIX rose by 11.21%. However, this is not the same thing as the change in investors' expectations for the future level of the VIX. In January, this increased by either 6.60% (for VXX, which measures expected changes in the VIX over the next two months) or 1.89% (for VXZ, which measures expected changes in the VIX during the period from four to seven months in the future).

Beyond the aforementioned cognitive challenge of using the VIX, there is also the fact that, over time, its expected return - that is, the change in implied volatility should be close to zero, leaving an average annual return equal to the implied expenses on these products - that is, an annual loss of 0.89%. For that reason, a decision as to whether to make a permanent allocation to these products or to instead use them tactically really comes down to one's expectations for the frequency of future uncertainty shocks. Finally, if one chooses to only invest in the VIX on a tactical basis (e.g., to hedge the risk of a jump in uncertainty), a more sophisticated investor would have to consider direct investment in VIX options and futures as an alternative to these products, which are debt obligations of Barclays Bank at a time when banks' creditworthiness is, shall we say, more open to question than is traditional in these matters. On balance, while we are very excited that these products have been introduced, and made investing in volatility much easier for retail investors, we also caution that a permanent allocation to them (and to volatility as an asset class) may not make sense for many investors.

We have been asked more times than we can count about our position on gold. Until now, our position has been that (a) gold as a financial investment is already included in most commodity index funds; and (b) gold as means of diversifying one's liquid reserves is better held in physical form, ideally in coins that, in the worst case scenario - e.g., if financial markets are closed - can be used as a means of exchange. Hence, we have been big fans of Maple Leafs, Eagles, Buffaloes, Kangaroos and Krugerrands (also known as the menagerie in the safety deposit box). However, Bank Julius Baer has recently introduced a series of ETFs that have, shall we say, broadened our thinking on this issue. These ETFs have very reasonable annual expenses (.40%), trade on the Swiss Exchange (www.swx.com), are based on the price of one ounce of gold, and are denominated in Euro, Swiss Francs, and U.S. Dollars to avoid exposure to both currency and commodity risk. Most interesting of all, this ETF offers the option of being redeemed not just in cash, but also in physical gold. The ticker for the USD redeemable ETF (note that they also come in a version that is not redeemable in physicals) is JBGOUA.SW. It goes without saying that the physical redemption process is bound to be more complicated than simply opening up your safety deposit box. But we are still very impressed with this product, which, for example, could be used to implement a tilt within one's allocation to commodities, as well as to implement diversification of one's precautionary savings. While Zurcher Kantonalbank has offered a similar ETF product (ZGLD.SW) for a number of years, it was largely aimed at institutions, and only traded in Swiss Franc denominated shares. The Julius Baer product is the first that makes this strategy available to retail investors.

While we are on the subject of tilts within one's allocation to commodities, we should also briefly touch on oil. As we note in this month's feature article, there are very strong reasons to believe that current oil prices represent a low point that will quickly reverse once economic demand picks up again. The large recent inflows into ETFs like USO show that we aren't the only ones holding this view. However, we have some very big reservations about USO as a way to take an active view on future oil prices. The underlying return generating mechanism for USO is the rolling over of near month oil futures contracts. As we have noted before, the sharp fall in demand for oil and the filling up of storage tanks (and tankers used for storage), along with producers continuing need to generate cash has combined to drive spot prices down to a significant discount versus futures. In technical terms, this has created a steep contango, which creates significant losses for commodity funds that are based on selling maturing contracts and using the proceeds to buy contracts with longer maturities. However, unlike USO, another oil futures based fund (USL) is based on taking and rolling positions not just in the near month contract, but in contracts for each of the next twelve months. As we noted in the case of the new VIX products, the latter structure appears to moderate price swings. For example, in January, USO was down by (11.72%), while USL lost only (2.71%). If you assume that the future upward move in oil prices will affect every contract, then it makes more sense to use USL to limit downside losses until that upward move materializes than it does to use USO. While the annual expenses on USL are slightly higher than those on USO (.60% versus .50%), it seems a small price to pay for the additional downside risk protection you receive.

In an earlier product and strategy note, we reviewed a paper on hedge fund beta that was recently published by AQR Capital. This firm has also recently introduced a new mutual fund, the AQR Diversified Arbitrage Fund (ticker ADANX, $5,000 minimum investment, 2.32% annual expense charge, which is temporarily capped at 1.50%). The fund will invest in a range of arbitrage strategies, including merger, convertible, dual class, stub, when-issued, and distressed securities. In theory, the fund should produce returns with a relatively low correlation of returns with those on major asset classes, and for this reason represents a very interesting new investment alternative for retail investors. Hence, for 2009 we are including it in the group of funds we will use to track the performance of our model portfolios' allocation to uncorrelated alpha strategies.

A long time ago, in a galaxy far, far away, people used to write to us and ask why on earth we had included an allocation to foreign currency bonds in our model portfolios. Judging from the number of new product launches in this area in recent months, it would appear that more people are coming around to our view on this issue. For example, SSGA has launched a new ETF (BWZ, annual expenses .35%) that invests in bonds with maturities of one to three years (recent duration is 1.66 years) issued by a mix of developed and emerging market governments. This complements the previously issued BWX, which invests in longer maturity bonds from the same issuers (annual expense .50%, recent duration 6.21 years). Not to be outdone, Barclays has launched two new ETFs that directly compete with the ones from SSGA. ISHG (.35% expenses, recent duration 1.8 years) invests in government bonds with a maturity of one to three years issued by a mix of developed country issuers tracked by the S&P/Citigroup 1 - 3 Year International Treasury Index. Major currency exposures are 56% Euro, 25% Japanese Yen, 5% each UK Pounds and Canadian Dollars. IGOV (.35% expenses, recent duration 6.3 years) invests in longer maturity bonds from the same mix of issuers. On balance, we prefer ISHG, because of its short duration (with money supply growth exploding, which seems likely to lead to higher future inflation, this doesn't strike us as a great time to be taking duration risk) and because it does not include emerging market bonds, which, as we have noted in the past, we believe offer an inferior risk/return tradeoff in comparison to emerging markets equities. We also believe that these new products offer superior value to international bond mutual funds, such as RPIBX (expenses of .82% and recent duration of 6.6 years) and PFUAX (expenses of 1.23% and recent duration of 7.7 years).

The biggest news on the fixed income front, however, was PIMCO's launch of a new global bond index and a mutual fund that tracks it. Way back in our December 2004 issue, we wrote about the shortcomings of many of the then existing indexes that claimed to measure the performance of fixed income markets. With its new "Global Advantage Bond Index" ("GLADI"), PIMCO has addressed many of our concerns. First, the GLADI uses GDP weighting, which avoids giving too much emphasis to potentially irresponsible borrowers who are issuing large amounts of debt. Moreover, if one assumes that changes in bond prices and yields lead changes in GDP (e.g., with yields falling and prices rising in advance of faster GDP growth), then the construction of the GLADI may generate incremental returns from the "buying low and selling high" over the course of national business cycles. Within currency zone allocations, the GLADI is divided between real return bonds (where available), government bonds, corporate bonds, and asset backed bonds. This enables it to capture the full range of risk premia, including those that compensate for uncertainties about duration risk (a function of both real rates and inflation), default risk, and prepayment risk. PIMCO, notes, rightly in our view, that "in the absence of a robust theoretical or empirical rationale for departing from a parsimonious equal weighting of each of these factors, this is adopted as the preferred weighing scheme within a given currency zone allocation." Finally, when it comes to security selection, the GLADI uses another set of criteria, including the requirement that all included bonds must have an investment grade rating and that the mix of bonds reflects the maturity and industry mix of issuers in a given market. Overall, this methodology results in an index with global coverage, an average rating of AA-, and average duration of 4.5 years. In terms of currency exposures, the GLADI is approximately 34% US Dollar, 26% Euro, 10% Yen, 12% other industrial countries (e.g., Australia, 1%, Canada, 4%, Denmark .6%, Norway,.2%, New Zealand, .1%, Sweden, 1%, Switzerland, 1% and the U.K, 4%.), and 18% in emerging markets. In sum, we believe that the GLADI is a better measure of the performance of the global fixed income markets than existing indices. The new fund that tracks the GLADI is known as the PIMCO Global Advantage Strategy Bond Fund, or PSAIX. Annual expenses are a rather hefty 1.45%, and, depending on where you buy it, you may also have to pay a front end load of up to 3.75%. While we believe the GLADI index will provide a good benchmark, we are less enthusiastic about this fund, as the underlying mix of fixed income allocations may not be optimal for many investors. On the other hand, the introduction of this fund now makes it possible for an investor to achieve a very widely diversified portfolio with only three or four investments, including PSAIX, VT (the Vanguard ETF that tracks global equity markets, and charges only .25% of annual expenses), RWO (the SSGA ETF that tracks the global property market, with annual expenses of .50%) and LSC, the Elements S&P Commodity Trends Indicator ETN (annual expenses .75%). To roughly line up with global market weights, the portfolio proportions would be PSAIX 50%, VT, 30%, RWO, 12% and LSC, 8%.

Changes to our Uncorrelated Alpha Allocation

We have recently undertaken a review of the list of funds we have been using to implement the allocations to uncorrelated alpha strategies in our model portfolios, in light of the growing number of retail products available in this space. To cite but one example, PIMCO recently introduced its Fundamental Advantage Total Return Strategy Fund (PTFAX, expenses 1.29% per year, and a front end load of up to 3.75%). What makes this fund particularly interesting is its underlying construction. Using futures and swaps, it takes a long position in the RAFI 1000 Fundamental Index and an offsetting short position in the S&P 500. Because these derivative positions cost a fraction of the face value, the remaining funds are invested in an actively managed portfolio of low to medium duration bonds. In our view, this represents a clear and welcome sign that a new age is arriving for retail investors, in which they will be able to pay low prices for passively obtained returns on broad asset classes, while only paying higher fees in exchange for the promise of uncorrelated returns. Granted, this does not eliminate, and in fact may only aggravate the issues raised in the AQR paper noted above with respect to the "overcrowding" of some strategies, and the resulting reduction in their expected returns. When it comes to active strategies, there is no escaping the need to make forecasts and the chance they could be wrong. Yet the clear split between passive and uncorrelated active returns is a very welcome development.

To implement our model portfolios' allocation to uncorrelated alpha in 2009, we will be using ten funds, with two each in five uncorrelated alpha strategies. While it may not be possible for an investor to efficiently invest in all of them (given the fund minimums in relation to the size of an overall portfolio), we believe that it will be of interest to our readers to see how each of them performs. Two funds are based on an equity market neutral strategy. In the past, we have used OGNAX and JAMNX. After a review of the alternatives, this year we will replace JAMNX with the JP Morgan Highbridge Statistical Market Neutral fund (HSKAX, 1.95% expenses, average three year return 5.10%, three year correlation with S&P 500, .04). The minimum investment in HSKAX is high, at $10,000, while OGNAX is lower, at $1,000 minimum, and annual expenses of 1.75%. The next two funds are based on variety of arbitrate strategies, as noted earlier in our description of ADANX, the new fund from AQR ($5,000 minimum, 1.50% annual expenses), which is one of the funds we will use. The other is ARBFX, which has a longer track record (average three year return of 4.05%), and a higher correlation with the S&P 500 (.33 over the past three years). However, it is a no-load fund, with a $2,000 minimum investment and annual expenses of 1.90%.

In addition to DBV, we will be adding ICI, and iPath Exchange Traded Note as our second currency carry strategy. At .65%, its annual expenses are lower than those on DBV (.81%). However, as a debt obligation of Barclays Bank, ICI also carry's more credit risk than DBV. In the long/short equity category (which may have a higher correlation with equity market returns than strategies that are explicitly market neutral), we will continue to use HSGFX (no load, 1.11% annual expenses, -.06 three year average return, .14 correlation with S&P 500) as well as PTFAX.

Our last strategy is global tactical asset allocation, which aims to generate uncorrelated returns by well-timed shifts between asset classes. The first fund we will use in this category is the BlackRock Asset Allocation Fund, (MDLOX). Over the past three years, it had an average return of 2.4% and a .75 correlation with the S&P 500. Annual expenses are 1.18% and the minimum investment is $1,000. The second fund is PIMCO's All Asset (PASAX). Over the past three years its average annual return was (1.7%), with a .75 correlation with the S&P 500. Minimum investment is $1,000, with annual expenses of 1.58%. PIMCO also offers another fund, All Asset All Authority (PAUAX) that uses leverage to further increase the returns from its tactical asset allocation strategy. Over the past three years, its average return was .01%, with a .51 correlation with the S&P 500. However, its annual expenses are much higher, at 3.58% per year.

And a Change to Our Commodities Allocation

After further review and consideration, we are changing the product we use to implement our allocation to commodities. In the past, we have used "long only" products that track the DJAIG index - either DJP (an exchange traded note that has Barclays Bank credit risk) or PCRDX, a mutual fund from PIMCO. However, we are now convinced that a systematic long/short approach is a more logical way to invest in commodities markets (as we described in last month's product and strategy notes). Hence, in 2009 we will use LSC, an exchange traded note that tracks the S&P Commodity Trends Indicator Index. Annual expenses are .75%. There is also a mutual fund (DSCTX) that tracks this index; however, with a minimum investment of $25,000 and annual expenses of 1.84%, LSC seems to offer more value. Some readers will undoubtedly raise the issue of the underlying HSBC credit risk exposure. On 19 January 09, HSBC issued a press release that included the following language: "HSBC has not sought capital support from the UK government and cannot envisage circumstances where such action would be necessary. HSBC has long been one of the world's most strongly capitalised banks and is committed to maintaining this position." Given, this, and the potential benefits of the long/short approach to commodity investing, we are comfortable with the LSC product.

| Global Asset Class Returns | Uncorrelated Alpha Strategies Detail | Asset Class Valuation Update | This Month's Letters to the Editor: oanda.com Currency Valuation and 2008 Crisis - Do You Still Believe in Diversification? | Product and Strategy Notes: Carbon Update; Will This Crisis Finally Change Investor Behavior?; Private Equity Update; Interesting Research; Product Reviews: JBGOUA.SW, ZGLD.SW, USO, USL, ADANX, BWZ, ISHG, IGov, PSAIX; Changes to Index Investor's Uncorrelated Alpha Allocation | February 2009 Issue: Key Points | Keeping an Eye on Long Term Trends | Economic Situation Update | 2008-2009 Benchmark Portfolios - All Currencies |