Seeking Alpha

Michael Dever's  Instablog

Michael Dever
Send Message
Founder, CEO & Director of Research for Brandywine Asset Management and author of "Jackass Investing: Don't do it. Profit from it." I have been a professional investor/trader since 1979 and have experience in stocks, managed futures, commodities, mutual fund arbitrage, market... More
My company:
Brandywine Asset Management
My blog:
Mike Dever
My book:
Jackass Investing: Don't do it. Profit from it.
View Michael Dever's Instablogs on:
  • Performance Analysis

    In the first 38 months of trading in Brandywine's Symphony Program, investors earned a cumulative 27% (and investors in our aggressively-traded Brandywine Symphony Preferred Fund gained an explosive 127%). The Sharpe ratio on both programs exceeded 1.0. What made this performance stand out even more was the fact that most other futures traders posted losses throughout this period. In contrast, the past 11 months have been the most difficult on record for Brandywine Symphony, which has posted a drawdown of 16%, while other futures managers have thrived.

    In this report we'll take a look at the two periods with the intent of understanding the differences between the strong performance of the first 38 months and the more recent drawdown.

    The 38 Month Rally

    An evaluation of Brandywine's performance during the positive first 38 months of trading reveals that profit contributions came from a wide range of investment strategies and markets. Moreover, these contributors varied over time. No single market, sector or investment strategy dominated over the entire period. However, the profits were not the result of one continuous move higher in performance. As you might expect, there was an ebb and flow of performance over that period. These can be shown as three distinct positive periods interspersed by two drawdown periods (prior to our most recent drawdown). The positive periods were simply the result of a majority of our investment strategies being in sync with the markets and producing profits, while the opposite was true of the drawdown periods.

    The 11 Month Drawdown

    The current drawdown, which began 11 months ago, can best be summarized by first categorizing our trading strategies into four groups:

    · Value strategies, which include strategies based on production costs and relative value

    · Alpha Hedge strategies, which capture trends

    · Fundamental strategies

    · Sentiment strategies

    The Value group encompasses a number of different strategies, each based on a distinct Return Driver. But they share the common belief that markets will revert to their true value over the longer-term. We realize that good value can turn into great value if markets trend lower. Indeed, a key characteristic of these strategies is that they often accumulate losses before their value is realized (but often, the greater the shorter-term downside, the greater the ultimate upside). Because of our understanding that these strategies can be overwhelmed and generate losses in the shorter-term, we include a separate group of strategies in the portfolio that are designed to thrive in such conditions. These are Brandywine's "Alpha Hedge" strategies.

    Alpha Hedge strategies profit from extended moves in markets (this is how they generate their "alpha"). Their performances tend to be roughly correlated with trend following or momentum strategies. One environment in which they are designed to prosper is exactly the sort of emotionally-trending markets that can be dangerous to the Value strategies. As such, they provide a great complement to Brandywine's Value strategies (which is why we label them Alpha "Hedge"). This was seen during the drawdown. As Brandywine's value strategies produced losses, Brandywine's Alpha Hedge strategies worked as planned and captured profits from the very trends that caused those losses. Through June, profits from Alpha Hedge more than offset the losses from Value. Unfortunately, both groups of strategies lost during July.

    Throughout the strong performance over the first 38 months, the Fundamental and Sentiment strategies tended to be Brandywine's most consistent performers. Their performances not only complemented each other, but often the strategies categorized within these groups would profit when Value or Alpha Hedge were losing. This is the basis of Brandywine's portfolio allocation model, which endeavors to maintain balance among the strategies and markets in the portfolio with the intent of smoothing out overall performance. This favorable characteristic was turned on its head over the last 11 months, as Brandywine's Fundamental and Sentiment strategies contributed to the drawdown. Had the strategies simply performed as they had over the first 38 months (which was in line with expectations), there would have been no drawdown.

    July's Performance

    Whereas the first 10 months of the current drawdown can be attributed to a slight profit in the Value-Alpha Hedge combination being offset by stark underperformance in Brandywine's Fundamental and Sentiment strategies, July's loss was fully caused by the selloff in commodities (crude was down more than 20% and all major commodity indexes fell in the double digits), which hurt the performance of the Value strategies. Unfortunately, Alpha Hedge did not serve as a hedge during the month, but actually added to the losses. The Fundamental and Sentiment strategies were essentially flat on the month. Although the final loss was in the range of what we would consider "extreme," it was achieved by a gradual erosion of performance throughout the month, rather than by one or a few sizable down days. In fact, Brandywine's value-at-risk (NYSE:VAR) declined to its lowest levels in more than a year and our margin-to-equity ratio (another measure of market exposure), fell into the lowest decile of the past year. In other words, the portfolio allocation model did what it could do to contain losses, but pervasive losses across all strategy groups led to cumulative losses, without respite.

    While we obviously cannot predict the future, past drawdowns have each been followed by strong recoveries-with the largest drawdowns being followed by the largest recoveries. And while we are disappointed in the performance of our strategies over the past year, we are confident they remain valid and are based on sound, logical Return Drivers. It is our belief that the first 38 months of Brandywine Symphony's performance is more representative of what to expect going forward than is the past 11.

    Aug 07 1:59 PM | Link | Comment!
  • Where’s The Alpha?

    At a dinner in late 2013, Brandywine's principals spent some time talking with the alternatives head at one of the large research firms. She made the comment that it appeared to be getting more and more difficult to capture alpha (excess returns over those earned simply by buying the "market"). We've continued to hear this refrain from others more recently as well. For readers of these reports, you understand that alpha is simply the term people use to describe Return Drivers that are not yet widely disseminated or accepted in the public domain. Which means that, almost by definition, alpha should always be difficult to find, as it's only alpha if its not commonly known!

    But that's not the reason people state alpha is hard to find. The primary reason for those comments is that most people do not yet embrace a Return Driver based approach to investing and instead try to uncover alpha by looking in the same places that others are looking. Unfortunately, it's over-tilled ground and unlikely to be fertile territory for new discoveries.

    In contrast, it's our belief that there are numerous sources of alpha available to be exploited. Brandywine has looked at hundreds of potential Return Drivers and we have been comfortable enough with a few dozen of them to incorporate them into the investment strategies used in Brandywine's Symphony Program. Without getting into specifics (obviously, we need to be careful not to expose our sources of alpha lest we turn them into 'smart beta'), let's look at a recent example of two that have contributed to Brandywine's positive performance this year.

    A number of Brandywine's investment strategies are based on Return Drivers designed to exploit people's behavior. Over the past few decades, due to the excellent research conducted by people like Amos Tversky and Daniel Kahneman, what many previously suspected has been proven. For a variety of reasons, the average person is a terrible investor. The vast majority of people underperform the very funds into which they invest- by as much as 5% per year on average. This indicates the potential for an investment strategy based on a Return Driver designed to exploit this behavior by 'fading' the crowd. In particular, one of Brandywine's strategies captures returns by looking at money flows into and out of bond and stock market ETFs. When the flows indicate irrational exuberance, the strategy takes short positions and in periods of despair, the strategy potentially enters into long positions. While many other futures managers have struggled in 2015, this approach proved profitable throughout the first half of this year.

    A second strategy, also designed to exploit people's behavior, uses measures that indicate people's expectations for future price levels in a broad range of markets. It uses this information to selectively enter into positions in deferred futures contracts. Investment strategies based on this Return Driver have been Brandywine's strongest performers to date in 2015.

    Because these strategies are based on Return Drivers that exploit human behavior, we don't expect them to be negatively affected by some of the 'usual' excuses assigned to the apparent decline in other sources of alpha, such as central bank intervention, high frequency trading or the globalization of markets.

    But these are only a couple of the dozens of Return Driver based investment strategies employed by Brandywine. There are dozens, if not hundreds, of other relevant Return Drivers that can be developed into strategies to profit from the movement of hundreds of other markets. This provides the fuel for Brandywine's ongoing research.

    Jul 07 11:36 AM | Link | Comment!
  • Return Drivers, Correlation And Diversification

    In last month's report we provided a brief review of Return Drivers and how some of the new fads in the investment industry, such as smart beta (or strategic beta) as well as the well-entrenched terms alpha and beta, are simply clever (or maybe not so clever!) terms that describe the level of public awareness associated with what is actually a small sub-set of available Return Driver based investment strategies.

    As we said in that report:

    "There is nothing special about smart beta. Once we understand that every valid investment strategy (such as buying-and-holding stocks or value investing) is based on a relevant Return Driver, we realize that beta, smart beta and alpha are merely terms used to differentiate the level of public acceptance of each Return Driver. In short:

    • Beta describes strategies (such as buy-and-hold) that are based on Return Drivers that are widely exposed and accepted as being valid in the public domain.
    • Smart Beta refers to Return Drivers that have been exposed but are not yet as widely employed or accepted in the public domain.
    • Alpha is the term used to describe Return Drivers that are not yet widely disseminated or accepted in the public domain.

    It's as simple as that."

    We're repeating this explanation here because Return Drivers not only provide an elegant framework to help resolve some of the most important investment debates (such as that between "passive" and "active" investing), but understanding them is also at the heart of an investor's ability to construct a truly diversified portfolio.

    Here's an example:

    It has been common over the past year or so to hear people make the comment "with interest rates so low, stocks are the only game in town." It is easy to understand why people have this view. If their investment beliefs are constrained by the use of asset classes, alpha and beta, their resultant investment process will lack the flexibility to allow them to create a truly diversified portfolio. For example, with interest rates on 10-year U.S. government bonds at less than 2.5%, a pension plan that requires a 7% return on its capital will lose by putting their money into bonds. If they view the investment world in terms of asset classes, as legions have been taught to do since the invention of asset classes in the 1960s, they are essentially constrained to owning stocks, bonds, commodities, real estate or cash. For many funds in this predicament, holding long stock positions appears to be the only way to achieve their required return.

    The investment world is completely different for the Return Driver based investor however. They see that owning stocks is just a simple investment strategy based on one combination of a valid Return Driver (rising corporate earnings) coupled with a relevant market (common stock). They see that there are dozens, if not hundreds, of other relevant Return Drivers that can be developed into strategies to profit from the movement of dozens (or hundreds) of other markets.

    One such approach that has started to become popular is that provided by trend following CTAs. They exploit a proven Return Driver (market momentum) to capture profits over time from price movements in a wide range of global markets. This strategy is no less valid or relevant than buying stocks based on the expectation of earnings growth. This is an important-perhaps the most important-statement. Just because an investment strategy is more or less accepted by the financial world, doesn't mean it is more or less valid. The validity of an investment strategy is based on the validity of its underlying Return Driver, not the popularity of its use.

    But buying stocks and trend following in global markets are just two out of the many investment strategies that can be employed to create a diversified portfolio. And despite the preponderance of long stock positions held in most people's portfolios, one is no less important than the other. But these two approaches (along with the other long-only strategies defined by the other "asset classes") are not the only games in town. Once an investor accepts the validity of owning stocks in anticipation of rising earnings, or selling crude oil based on the strong downside momentum in its price, the opportunity exists to create a truly diversified portfolio by developing - or investing with managers that have developed - other investment strategies based on other, equally valid, Return Drivers.

    Brandywine has developed and employs dozens of investment strategies based on dozens of disparate Return Drivers. Because of this, as we showed in last month's report, there is no correlation between the returns produced by Brandywine and those of virtually any other investment index. An additional benefit is that we are not limited to what appears to others to be "the only game in town." There are always numerous other games to play.

    While we won't reveal the specifics of each of the investment strategies we employ, as it would be a disservice to our investors, who benefit from us keeping them out of the public domain, we continue to be very open with providing some examples of the Return Drivers we exploit. If you are a serious investor and would like to discuss this with us, please feel free to contact Rob Proctor or Joe Gabor and we will schedule a presentation.

    Jun 09 1:44 PM | Link | Comment!
Full index of posts »
Latest Followers


More »
Posts by Themes
Instablogs are Seeking Alpha's free blogging platform customized for finance, with instant set up and exposure to millions of readers interested in the financial markets. Publish your own instablog in minutes.