|
About IndexInvestor.com |
Privacy Policy |
Transaction Policy |
Legal Disclaimers |
Contact Us |
My Account |
Home
|
|||||
|
|
|
|
|
||
Analysis of Risk/Return Regimes for AUD, CAD, CHF and GBP Based Investors
This month, we continue the analysis of different risk/return regimes over the 2001-2008 period that we began in April with our overview of U.S. dollar returns. These new analyses can be found in the respective currency editions of this month's journal. While there are some local differences, all four analyses have much in common. Asset class returns substantially differ across the high volatility, high inflation and normal regimes across all four currencies. As was true for USD returns, the relative ranking of asset class risk is much more stable across regimes than the ranking of their returns. In all four currencies, volatility offers substantial potential for limiting downside risk. In terms of our principal components analysis, over the 2006 -- 2008 period, the most important principal component in all four cases is highly correlated with the rise in volatility and the negative impact this had on property and equity returns. There is also clear evidence of the impact of the 2006 -- 2008 commodity price cycle. Over the 1991 -- 2005 period, the principal components analysis repeatedly shows the impact of changes in real interest rates, changes in inflation and inflation expectations, changes in commodity prices, and changes in volatility, as well as the prolonged period of strong, low inflation GDP growth once known as "the great moderation."
Sentiment, Trend Following, Momentum, and the Inefficient Market Argument for Index Investing
John Kay had an outstanding column in the June 9, 2009 Financial Times that serves as an excellent starting point for a discussion of a number of other recent developments. As we have also written many times, Kay began by noting that "John Maynard Keynes famously likened the processes of stock exchanges to a newspaper beauty contest, in which the objective was not to choose the most beautiful face, but to choose the one you thought others would find most beautiful." Kay went on to describe how Keynes believed that "two approaches to investment followed from this metaphor. One -- speculation -- required careful study of the fads and fancies of the other contestants. The alternative -- enterprise -- believed that real beauty would always shine through. Speculation involved forecasting the psychology of the market, enterprise the prospective yield [return] of assets over their whole life." Kay then stressed one of Keynes' key conclusions: "Perceptively, Keynes anticipated the development of a paradox. Professionalization of markets would drive out analysts who focused on fundamental value...since it was better [for a professional investment manager's career] to be conventionally wrong than unconventionally right." Finally, Kay noted how the dominance of a market by speculators has spread beyond equities: "In the past two decades, securitization and other financial innovations [e.g., credit default swaps] brought the same phenomenon to credit markets. When loans remained on the balance sheet to maturity, there was no alternative to an assessment of their fundamental value. Once loans could be bought and sold, what mattered was not their soundness but their price -- with the predictable consequences of instability and price fluctuations far in excess of any reasonable assessment of any underlying change in fundamental value."
Other writers have used different words to describe this change in the nature of financial markets, noting that they are now seem more heavily dominated by short-term traders rather than long-term investors, or by trend-followers (whether human or algorithmically based) rather than investors who analyze fundamental security, sector, and asset class valuations. Of course, as Keynes noted, this change began long ago. It was also highlighted in the works of Benjamin Graham, who famously said that "In the short run, the market is a voting machine but in the long run it is a weighing machine" (which is close to our core assumption that, while markets are attracted to equilibrium and prices close to fundamental values, practice they rarely achieve it, and are usually in a state of disequilibrium). However, we also believe that, due to improved communications technology that facilitates greater interconnection between investors, faster spread of information, and easier development of price trends (i.e., herding), the market impact of speculation has undoubtedly become stronger in recent years.
This raises an interesting question -- has the growing power of speculative forces paradoxically increased the probability that active investment management strategies will meet with success? There is a growing body of evidence that a range of emotional and social forces give rise to coordinated human behavior across a wide range of contexts. In previous issues, we have described how neurobiology causes humans’ fear of social isolation to rise in the face of increased uncertainty. Other research has found that human beings differ in the intensity of these responses, and that variations in cognitive ability affect humans tendency towards overconfidence, impatience, and errors of judgment (see, for example, "Cognitive Abilities and Behavioral Biases" by Oechssler, Roider, and Schmitz). Other researchers have noted that "when agents are sensitive to the wealth of others, they tend to herd on the same information, trying to mimic each others' trading strategies" (see "Relative Wealth Concerns and Complementarities in Information Acquisition" by Garcia and Strobl). These tendencies are reinforced by technological changes, including faster access to information and the increasing use of "most popular stories this week" aggregators by a wide range of media that further focus people's limited attention on a few key themes (or memes).
There is more controversy about the investment implications of these findings. On the one hand, there is evidence that a focus on investor sentiment leads to pricing errors and lower investment returns. For example, a recent paper found that "analysts whose stock recommendations are positively correlated with recent or future investor sentiment tend to issue relatively less profitable recommendations" (see "The Profitability of Analysts' Stock Recommendations: What Role Does Investor Sentiment Play?" by Bagnoli, Clement, Crawley and Watts). Another recent paper finds that following these sentiment based recommendations causes short sellers to lose money, while use of fundamental valuation indicators leads to higher returns ("False Prophets: How Analysts Lead Short Sellers Astray During Periods of High Investor Sentiment" by Seybert and Wang). Finally, in another paper, Baker, Wurgler and Yuan conclude that "both global and local sentiment are contrarian predictors of major markets' returns" and that "sentiment appears to be contagious across markets" ("Global, Local and Contagious Investor Sentiment").
Let's examine this issue in more detail. Before proceeding, we note that what follows reflects an argument first put forward by Steve Thorley in his outstanding (and underappreciated) 1999 paper, "The Inefficient Market Argument for Passive Investing." We will start with the following assumptions:
What outcomes do we expect this market to produce? First, because the majority of investors lack accurate forecasting skills, during the year, market prices for securities should diverge from their true values -- perhaps by a large amount. Second, at the end of the year, the returns of the skilled and unskilled investors net out to the returns generated by changes in true fundamental values. Third, because unskilled investors are in the majority, the median investor should earn a negative return for the year.
Now let's introduce an index fund. Assume 50% of the unskilled investors become index investors, leaving 25% skilled and 25% unskilled. What changes would we expect to see versus the previous scenario? First, asset prices should stay closer to their fundamental values, since there are fewer unskilled active investors making valuation mistakes. Second, the median active investor's return should be higher (since skilled and unskilled are now balanced), even though all active returns will still net out to the overall market return. Third, those unskilled investors who became index investors will also enjoy higher returns, since the index fund's return tracks the overall market return, and they are no longer systematically losing money to the skilled investors. Finally, life for the skilled investors will become more difficult -- and their excess return above the overall market return should decline -- because there are fewer unskilled investors for them to exploit.
In our third scenario, let us assume that 25% of the investors who were in the index fund decide that trend following is the best way to earn high returns. And let us assume that 5% discover they are actually skilled at forecasting the future behavior of other investors. We now have a more complicated market, composed of 25% skilled active fundamental investors, 25% unskilled active fundamental investors; 5% skilled active trend following investors, 20% unskilled trend following active investors, and 25% passive index investors. What results is this new mix likely to produce? Once again, all the active returns will net out to the market return, which is what the index investors will earn. However, rather than the previous mix of 25% skilled and 25% unskilled active investors, we now have 30% skilled and 45% unskilled active investors. So the median active investor's return should be lower than it was in the previous scenario. Moreover, since trend followers are not concerned with fundamental values, the gap between the average returns of skilled versus unskilled active investors may be much wider than it was even when 75% of investors were unskilled but focused on fundamental valuation, for the simple reason that trend followers can accentuate the price impact of security valuation mistakes. On the other hand, skilled active investors should earn higher returns in this scenario that in the previous one.
Hopefully, this short example clarifies a number of critical points: (1) even in a market where security prices can substantially deviate from their true values, indexing makes sense for investors who lack active management skills. (2) Skilled active investors have a very strong interest in convincing unskilled investors that they are, in fact, skilled. (3) Skilled investors have perhaps an even stronger vested interest in convincing unskilled investors to pursue a trend following strategy (as operators of boiler room "pump and dump" schemes have known for years). Keep that in mind the next time someone tells you to buy a hot stock because it is "breaking out" and "really starting to move" -- or when someone tries to argue against adding more index funds to your 401(k), superannuation, or other defined contribution pension plan.
Careful readers will note that there are four factors missing from this example that make the challenge faced by active investors much more difficult. First, true prices are not revealed at the end of each year; a more accurate description of markets is that while over time prices are attracted to true values, they usually deviate from them, sometimes by large amounts. This makes it much more difficult to separate skilled from unskilled investors -- even in their own minds. Second, the underlying processes driving true valuation actually evolve over time, and are not stationary. The same is true for the underlying processes governing trend following behavior are also not stationary (e.g., rates of information dissemination and its perceived importance, the degree of uncertainty about overall economic conditions, and the current state of investors' individual emotions and social network connections). Finally, in real markets luck, or randomness, plays an important (if usually underappreciated) role. For example, an unskilled fundamental active investor whose forecast is correct three times in a row can become a market guru, and the focal point for the actions of trend following investors. The extraordinary difficulty involved in being a skilled investor -- whether based on accurate forecasts of future fundamental value or investor behavior -- may also help to explain the sharp increase in the volume of so-called "high frequency" trading in recent years. Essentially, high frequency trading is based on computer executed trading algorithms (i.e., customized software programs) that attempt to exploit -- over extremely short term intervals -- price discrepancies caused by market microstructure factors (e.g., the timing and pricing of submissions to an electronic limit order book) and/or very short-term trends and/or cross market (e.g., futures vs. stocks) valuation inconsistencies. Thanks to the advent of huge databases with tick-by-tick trading records, as well as cheaper and more powerful computing power, these short-term price discrepancies are now much easier to identify and exploit. Moreover, since at least some of them seem to be a function of the structure of the markets themselves, they may be less likely to be arbitraged away. Hence, it should come as no surprise that many active investment managers looking for an edge to justify their high fees have turned to high frequency trading.
Let us now move on to a closer examination of the issue of momentum. Momentum is not the same as trend following. While the latter is generally assumed to be investing in a previous period's winners, momentum technically refers to a market neutral investment position that goes long stocks with the highest recent increase in earnings or prices (say the top 10% or 20%) while selling short an equivalent amount of stocks at the other end of the spectrum. Multiple studies have shown that, at least in theory, momentum can be a profitable strategy (see, for example, "The Global Investment Returns Yearbook, 2008" by Dimson, Marsh and Staunton; "The Case for Momentum Investing" by Berger, Israel, and Moskowitz; and "Global Momentum Strategies: A Portfolio Perspective" by Griffing, Ji, and Martin). More recently, researchers have introduced new indices intended to make it easier for investors to either implement either a trend following or a momentum strategy in their equity allocations (see "Momentum and Contrarian Stock Market Indices" by Eggins and Hill, "The Efficient Replication of Factor Returns" by Melas, Suryanarayanan and Cavaglia of MSCI Barra, and AQR Capital's new momentum indexes).
However, other writers have cautioned that in practice, true momentum strategies may be only marginally profitable (or actually unprofitable), because they often involve high transaction costs (due to monthly rebalancing, expensive short sales, and often times the presence of small, hard to trade stocks among both the momentum winners and losers). Others have argued that momentum is suffering the same declining returns as other strategies that received large cash inflows after they were popularized (see, for example, "The Fading Abnormal Returns of Momentum Strategies" by Henker, Matens and Huynh).
These are all points to keep in mind when evaluating the new "momentum" funds that have recently been introduced in the United States by AQR Capital. These include strategies focused on the Russell 1000 (AMOMX, no load,.49% expense ratio), small cap (ASMOX, no load, .65% expense ratio), and international equities (AIMOX, no load, .65% expense ratio). In light of our discussion, we would say these funds are mislabeled, since AQR defines its "momentum" strategy as investing in shares with the top 33% performance over the previous twelve months, rebalanced no more than quarterly. As they take no offsetting short positions, we think these are better described as "trend following" funds. Thus, rather than being an interesting new uncorrelated alpha strategy, we regard the new AQR momentum funds as tilts within an overall equity allocation. As with all other tilts, achieving superior risk adjusted returns from this one necessarily depends on making a superior forecast of future equity market conditions.
More broadly, however, it will be interesting to see where this increasing emphasis on momentum and trend following investing leads. One likely consequence is that the appearance of low cost tend following index funds will make it even harder for many active managers to justify their fees (or at the very least it will force a clearer distinction between trend following and momentum). In the mutual fund world, papers co-authored by Russ Wermers have shown how momentum has a very strong impact on the returns of successful active funds ("Is Money Really "Smart"? New Evidence on the Relation Between Mutual Fund Flows, Manager Behavior, and Performance Persistence"), and that momentum oriented strategies have become more predominant in recent years ("Analyst Recommendations, Mutual Fund Herding, and Overreaction in Stock Prices"). A more recent paper reviews the performance of 1,448 institutional domestic equity investment management firms between 1991 and 2008 ("Performance and Persistence in Institutional Investment Management" by Busse, Goyal and Wahal). They find that after taking momentum into account, remaining alpha's are not statistically different from zero, net of fees. Moreover, momentum accounts for virtually all the persistence in top performers' returns over a one year time horizon.
In sum, we remain intrigued by the prospect of someone introducing a true quantitatively based long/short momentum product, which would be similar to other uncorrelated alpha strategies. We suspect, however, that if this could have been done profitably, AQR would have taken this approach, as they are known to be a very smart team of quants.
Private Equity, Again
We just read another interesting new paper on one of our favorite subjects -- whether (and if so, how) private equity funds really create value for investors in their limited partnerships. In "Managerial Incentives and Value Creation: Evidence from Private Equity", Leslie and Oyer compare companies owned by private equity funds with similar public companies. They find that the PE owned funds have much stronger incentives for their top executives and use higher levels of debt. "The highest-paid executive at a PE-owned firm owns approximately twice as large a share of the firm, earns about 12% less in base pay, and receives a substantially larger share of his cash compensation through variable pay" relative to his counterpart at a publicly owned company. However, the authors "find little evidence that the PE-owned firms outperform public firms in profitability or operational efficiency." This will undoubtedly come as no surprise to anyone who has been a manager in one of those public firms, and struggled to successfully compete in the face of rapidly evolving customer needs, competitor offerings, technological possibilities, and economic conditions. Rarely have I encountered anyone in this role who believed that stronger incentives would instantly endow them with greater ability to pierce the uncertainty they confront, or magically eliminate the implementation challenges posed by human nature (in fact, widening the compensation gap between senior managers and everyone else often has the opposite effect). Nor have I ever encountered an experienced operating manager who believed that former investment bankers now running private equity firms had any unique wisdom to add when it came to improving profitability and operational efficiency.
Leslie and Oyer also "find that the compensation and debt differences between PE-owned companies and public companies disappear over a very short period (one to two years) after the PE-owned firm goes public." No surprise there either -- returning to a more conservative capital structure makes sense, since it increases financial flexibility, and therefore the company's ability to adapt to unexpected change (which is critical to survival in a highly competitive and uncertain environment). Similarly, reducing compensation disparities also removes barriers to both the free flow of ideas and the rapid implementation of change -- factors that are also critical to a company's ability to adapt.
Unfortunately, these fundamental truths seem to have escaped the notice of the many bankers who loaned enormous sums to PE firms to back their highly priced, highly leveraged bids, as well as the equity investors who bought the PE firms' shares when the highly leveraged portfolio companies were taken public again. In sum, as more research is produced into the dynamics of private equity over the past decade, evidence accumulates that a very substantial portion of recent PE fund returns was derived from a combination of skill in exploiting lenders' ignorance and timing equity investors' tendency to overestimate future growth rates in different sectors of the economy. Of course, that may also be why LP units in many PE funds are trading in gray and secondary markets at 50% or less of their stated values, and so many former investment bankers have struggled to raise their second PE funds.
Four Interesting Research Papers
In "Carry Trades and Global Foreign Exchange Volatility", Menkhoff, Sarno, Schmeling, and Schrimpf provide new insights into the uncorrelated alpha foreign exchange tstrategy known as the "carry trade", where an investor borrows in a low interest rate currency, and invests in a high interest rate currency. According to the theory of uncovered interest rate parity, changes in the exchange rate should offset the interest rate difference, leaving zero profit. However, as the very substantial amount of money invested in this strategy has shown, this has often proven not to be the case. However, the authors show why the carry trade is not the free lunch it first appears (as investors who were long Icelandic Krona can attest). "We find a significant negative return comovement of high interest rate currencies with global volatility, whereas low interest rate currencies provide a hedge against volatility shocks." Bottom line: the carry trade worked well as long as world financial markets were relatively stable. However, a substantial portion of the return earned on the carry trade represented compensation for bearing the risk of very negative outcomes if volatility and illiquidity sharply increased.
In "Performance Maximization of Actively Managed Funds", Guasoni, Huberman, and Wang show a new way that active managers can game performance measurement systems in order to show higher levels of alpha -- in this case, by taking a long position in the benchmark against which the fund's performance is measured, and then writing call options on them. Provided the implied volatility of the call options (or similar derivative) is higher than the realized volatility of the benchmark (which is usually the case), the manager will appear to have generated significant alpha, even in the absence of superior information or skill.
Along somewhat similar lines (gaming the system), Richard Evans has written an interesting paper on the practice of "Mutual Fund Incubation", "which is a strategy for initiating new funds, where multiple funds are started privately, and, at the end of an evaluation period, some are opened to the public." He finds that "funds incubation outperform non-incubated funds by 3.5% risk adjusted, and when they are opened to the public they attract higher net dollar inflows. Post-incubation, however, this outperformance disappears. This performance reversal imparts an upward bias to equal weighted, but not value weighted, return indexes." Interestingly, he finds a higher probability of incubation at fund families whose products are primarily sold through brokers, since fund flows through this channel have been shown to be more sensitive to past returns. Evans concludes that "overall, the evidence suggests that incubation is used by fund families to speciously enhance performance and thereby increase inflows, and it is an effective tool in this regard."
Last but not least, we call your attention to a new piece of research from Morgan Stanley on "The Renaissance of Global Macro Investing." The author, Henry McVey, concludes that the "Great Recession" will lead to a renewed emphasis on a "top-down, global macro investing approach", and that "investors within the asset allocation community will be required to enhance their analytical tools in an effort to find out which investment will actually serve as diversification instruments for large portfolios, especially during periods of market stress." This is not too different from something Jack Bogle said in his recent Financial Analysts Journal interview: "We should build a model based on a winning strategy whereby incentives are not based on trading volume, but on personal financial service, asset allocation, broad diversification and control of the risks and the costs." We couldn’t agree more that this is the way the financial advisory and planning industry must move if higher levels of consumer savings are to result in adequate levels of post-retirement income.
| Global Asset Class Returns | Table: Asset Class Valuation Conclusions and 3 Month Return Momentum | Table: Market Implied Expectation of Most Likely Economic Regime | Uncorrelated Alpha Strategies Detail | August 2009 Economic Update | Feature Article: A Letter to the New Graduate | Global Asset Class Valuation Updates Detail | Product and Strategy Notes: Analysis Risk/Return AUD,CAD,CHF and GBP Based Investors; Trends, Momentum, Inefficient Markets' Argument for Index Investing, AQR Capital Products; Private Equity - Again and Four New Research Papers | This Month's Letters to the Editor: Country vs Industry Factors; Management of Client Expectations; Sold in 2007/08 What Now? | August 2009 Issue: Key Points |