Originally published in the Brandywine Asset Management Monthly Report.

The investment world lives by benchmarks. Every institutional investor benchmarks the performance of the individual components of their portfolio to various indexes. Individual investors do the same. That’s why the financial news constantly refreshes the current performance of the Dow Industrials or the S&P 500. This is partly due to regulatory mandate. In 1998 the SEC began to require mutual funds to show their performance relative to a benchmark. They felt this would better enable investors to determine whether a fund was performing well because of investment decisions made by its manager or simply because of factors outside their control. But mostly, benchmarking is seen as a necessary exercise to determine the effectiveness of a fund’s management team.

As a result of this “benchmark-itis,” we often get asked by investors, “against which benchmark should we compare your performance?” Suggestions include the S&P 500 (because everyone watches that), hedge fund indexes (because our return driver based approach is “non-traditional”), or maybe the CTA indexes (since we execute our strategies in the futures markets).

But none of these are a proper benchmark for Brandywine’s Symphony Program. That’s because Brandywine doesn’t employ trading strategies based on common, “public domain” return drivers. As a result, our returns won’t necessarily have any relationship to any index return. For example, some of Brandywine Symphony’s best-performing periods have occurred when stocks dropped significantly (such as during Q3 2011). And Brandywine shows no correlation of returns to any of the few dozen hedge fund indexes. With regards to CTA indexes, well, this year provides a great example of how uncorrelated Brandywine is. While most of the CTA indexes exhibited losses from January through July, Brandywine produced solid profits. And although Brandywine gained strongly along with the CTA indexes in August, our sharp loss in September was in stark contrast to gains for the CTA indexes. Whether we over- or under-perform relative to any of the standard benchmarks over any period of time is essentially random and irrelevant.

Brandywine’s Benchmark
So what is the best index with which to compare Brandywine Symphony’s performance? It’s our own past performance.

The reason is because Brandywine’s Symphony Program follows a return driver based approach that incorporates dozens of independent trading strategies, each based on a sound logical, non-public domain return driver, to trade across more than 100 global financial and commodity markets. Our goal is to achieve the highest possible return for any given level of risk with a reasonable probability that our future performance will match our past performance. We do not employ a single style that can be reflected in any single index.

Which leads to the question, “how is Brandywine’s Symphony Program performing relative to its benchmark (its own past performance)?” Since September posted the worst monthly loss since Brandywine’s Symphony Program began actual trading in July 2011, we’ll focus on the downside behavior.

Brandywine’s walk-forward back-test for our Symphony Program runs from January 1999 up to the start of our actual trading in July 2011(1). Virtually all performance tracking services publish an investment program’s maximum drawdown on a month-end basis (measuring the drop in performance across one or more months from the highest month-end peak to a subsequent month-end low). We’ll look at our drawdowns on that basis as well as gross end-of-day performance (which will almost always be larger as it captures intra-month losses and does not receive any benefit from potential incentive fee give-backs).

Brandywine Symphony’s Drawdowns in Line With Benchmark
In the first chart we display the 16 largest month-end peak-to-trough drawdowns (on a net basis) over the past 16 years (this includes tested performance (dark bars) from 1999 through June 2011 and actual performance (light bars) starting in July 2011). We would expect to incur a drawdown within the range of this chart on average once per year, and that is exactly what has happened. Since the start of actual trading in July 2011 we have had three drawdowns make it onto the chart. The most recent drawdown, which has been almost perfectly bracketed by the beginning and end of September (making the end-of-day performance match the month-end performance quite closely), is right in the middle of the range at -6.46% (this also indicates that it could extend an additional 3% and still remain within expectations).

Now let’s look at the 16 largest drawdowns as measured on a gross, end-of-day basis. We would expect a drawdown to fall within this range on average once per year. As you can see in the following chart, the current drawdown, at -7.14%, just made it onto the chart. Since the start of actual trading in July 2011, two drawdowns made the “bottom 16” and neither registered among the worst. So our performance in actual trading has been slightly less “risky” than our benchmark would project.

One final way to view the current drawdown is to compare this past month’s performance to other 22-day periods (the number of trading days there were in September) in our history. In that regard, this is the 5th deepest 22-day drawdown (measuring the worst 22-day period within all past drawdowns) over the past 16 years. This indicates that a drawdown of this size should occur over a 22-day period once every 3 years. Since we just wrapped up 3 ¼ years of trading, this drawdown is right in line with expectations.

Drawdowns are just as much a part of the performance of Brandywine’s Symphony Program (as well as any investment) as are positive returns. The fact that the drawdowns we have incurred in our actual trading so closely match our Benchmark expectations is confirmation that our investment model continues to work as expected. In summary, the performance of Brandywine’s Symphony Program, as measured by the drawdowns we have incurred since the start of actual trading in July 2011, has been faithfully tracking our expectations pursuant to our benchmark, which is our past performance.

(1) Because this report includes the results of the tested performance, in addition to the actual performance, of Brandywine’s Symphony program, the following disclaimer is required:

HYPOTHETICAL PERFORMANCE RESULTS HAVE MANY INHERENT LIMITATIONS, SOME OF WHICH ARE DESCRIBED BELOW. NO REPRESENTATION IS BEING MADE THAT ANY ACCOUNT WILL OR IS LIKELY TO ACHIEVE PROFITS OR LOSSES SIMILAR TO THOSE SHOWN. IN FACT, THERE ARE FREQUENTLY SHARP DIFFERENCES BETWEEN HYPOTHETICAL PERFORMANCE RESULTS AND THE ACTUAL RESULTS ACHIEVED BY ANY PARTICULAR TRADING PROGRAM.

ONE OF THE LIMITATIONS OF HYPOTHETICAL PERFORMANCE RESULTS IS THAT THEY ARE GENERALLY PREPARED WITH THE BENEFIT OF HINDSIGHT. IN ADDITION, HYPOTHETICAL TRADING DOES NOT INVOLVE FINANCIAL RISK, AND NO HYPOTHETICAL TRADING RECORD CAN COMPLETELY ACCOUNT FOR THE IMPACT OF FINANCIAL RISK IN ACTUAL TRADING. FOR EXAMPLE, THE ABILITY TO WITHSTAND LOSSES OR TO ADHERE TO A PARTICULAR TRADING PROGRAM IN SPITE OF TRADING LOSSES ARE MATERIAL POINTS WHICH CAN ALSO ADVERSELY AFFECT ACTUAL TRADING RESULTS. THERE ARE NUMEROUS OTHER FACTORS RELATED TO THE MARKETS IN GENERAL OR TO THE IMPLEMENTATION OF ANY SPECIFIC TRADING PROGRAM WHICH CANNOT BE FULLY ACCOUNTED FOR IN THE PREPARATION OF HYPOTHETICAL PERFORMANCE RESULTS AND ALL OF WHICH CAN ADVERSELY AFFECT ACTUAL TRADING RESULTS.

Share this post:

• 
• 
• 

• 
• 
• 
• 
• 
• 
• 
• 
• 
• 
• 
• 
• 
• 
• 

Originally published in the Brandywine Asset Management Monthly Report.

Since the 1960s, asset classes have dominated the investment landscape. Originally “invented” out of necessity, they evolved to serve as the foundation of conventional portfolio diversification. They got their start shortly after Harry Markowitz published his seminal paper “Portfolio Selection.” This paper eventually led to the popular adoption of mean-variance modeling to create “optimally” diversified investment portfolios. But in order to determine an optimal stock portfolio, the Markowitz model required a person to calculate the covariance of every stock in a portfolio in relation to all others. Because computer technology was in its infancy and the “cost” (not just in computer resources but in actual dollars and cents as well) to do this was exorbitant, a simpler method needed to be devised. Enter Bill Sharpe, who developed a simplified system that instead compared each stock to the market as a whole. The “market as a whole” became an asset class. Initially, there were just a few asset classes, such as stocks, bonds and cash.

Over the years, investment professionals have established increasingly varied asset classes (and sub-asset class “categories”). For example, many now consider real estate and commodities to be asset classes. And emerging market stocks and high-yield bonds may still be considered to be part of the stock and bond asset classes, but they are also understood to behave differently and are therefore considered to be sub-categories of their asset classes.

All this would be just an interesting (to some) academic discussion if it weren’t for the fact that trillions of dollars in investment decisions are based on allocating capital across asset classes. This makes asset classes an important and integral part of investing. And that is a problem, because the use of asset classes imposes an unnecessary constraint on a person’s ability to create a truly diversified portfolio. But before we explain why, we’d like to introduce the concept of return drivers.

From Asset Classes to Return Drivers
One of the innovations supporting Brandywine’s investment philosophy is our use of return drivers. As Mike Dever states in his book, “a return driver is the primary underlying condition that drives the price of a market” and “every return driver has a time period (and markets) over which it is relevant.” Realizing that the best way to explain the return driver concept is by example, in the opening chapter of his book Mike displays the result of research that shows the relative influence of the two primary return drivers that power stock prices (which are people’s sentiment towards stocks and the growth of corporate earnings). You can read a complimentary copy of the book’s Introduction and first chapter here: https://www.brandywine.com/pdf/special/JackassInvesting_BookThruMyth1.pdf.

Once those return drivers have been identified, they can be exploited to serve as the basis for trading strategies. Mike demonstrates this in the book’s Action Section, where he shows how to develop a specific trading strategy that uses ETF money flows to exploit short-term sentiment in stock indexes and bond markets. This is an actual trading strategy being used by Brandywine today.

The basic nature and elegance of return drivers becomes apparent when you realize that all an asset class is, is a specifically constrained trading strategy employed against a group of related markets. For example, corporate earnings growth is the dominant return driver of U.S. equity prices over periods of 30 years or more. So the U.S. equity “asset class” is simply the application of a trading strategy (holding naive long positions), applied to U.S. equities, designed to capture that return driver.

But there are potentially dozens of sound, logical return drivers (as equally sound as earnings growth driving long term stock prices) that can be exploited to profit from trading in the U.S. equity markets. And there are potentially hundreds of additional return drivers that can be exploited to profit from trading in the hundreds of other freely-traded global financial and commodity markets. To ignore those and exploit just one creates a logical inconsistency. The stated desire of most institutional investors and their consultants (as well as most individual investors) is to create a diversified investment portfolio. But their dependence on asset classes immediately constrains their ability to do so.

One way to attempt to fix this is to expand the universe of asset classes. In the past few years, firms such as Goldman Sachs have suggested that volatility be considered an asset class. We understand their desire to do this. Volatility trading in equities is based on a sound, logical return driver that produces returns that are uncorrelated to equities. It provides tremendous portfolio diversification. But pigeonholing volatility into the asset class construct is awkward and cosmetic. It’s not a true fix.

Brandywine’s Use of Return Drivers
Return driver based investing provides that true fix. Once you recognize that every asset class is powered by return drivers, and is therefore simply a subset of a single trading strategy, it actually becomes illogical to pursue an asset class approach instead of a diversified return driver based approach to investing. That is why return drivers are one of the key concepts underpinning Brandywine’s investment philosophy and an integral contributor to our performance. In contrast, there is an inherent and sizable disadvantage to being asset class constrained.

We pointed this out in our February 2013 report, where we stated that because of the dependence of the S&P 500 on two primary return drivers and Brandywine’s broad diversification across return drivers that “over time, the S&P 500 TR index will be unable to compete on a risk-adjusted basis with the returns earned by Brandywine.” This continues to be our belief today.

Resources
The myth that portfolio diversification can be achieved by allocating money across asset classes is exposed in Chapter 17 of Mr. Dever’s book. You can read a complimentary copy of that chapter here: https://www.brandywine.com/pdf/special/JackassInvesting_Myth17.pdf.

Share this post:

• 
• 
• 

• 
• 
• 
• 
• 
• 
• 
• 
• 
• 
• 
• 
• 
• 
•