Diamond Hill Long-Short Strategy 2010 Slump in a Five-Year Season

By Ric Dillon, CFA

January 7, 2011

010711_grpah_01

010711_grpah_02

Historical performance for Class C shares and Class I shares prior to their inception is based on the performance of Class A shares. Class C and Class I performance has been adjusted to reflect differences in sales charges and expenses between classes.

The Large Cap Fund invests in equity securities (stocks) that are more volatile and carry more risk than other forms of investments, including investments in high-grade fixed income securities. The net asset value per share of this Fund will fluctuate as the value of the securities in the portfolio changes. The Long-Short Fund uses short selling which incurs significant additional risk. Theoretically, stocks sold short have the risk of unlimited losses. The Large Cap Fund and the Long-Short Fund are two separate funds, with separate investment strategies and results. The Large Cap Fund and Long- Short Fund are not interchangeable.

The Russell 1000 Index is a market capitalization weighted index measuring performance of the largest 1,000 companies, on a market capitalization basis, in the Russell 3000 Index, a market-capitalization weighted index measuring the performance of the 3,000 largest U.S. companies based on total market capitalization. The blended index represents a 50% weighting of the Russell 1000 Index as described above and a 50% weighting of the BofA Merrill Lynch US T-Bill 0-3 Month Index. The BofA Merrill Lynch US T-Bill 0-3 Month Index tracks the performance of US dollar denominated US Treasury Bills publicly issued in the US domestic market with a remaining term to final maturity of less than 3 months. One cannot invest directly in an index. Unlike mutual funds, the index does not incur expenses. If expenses were deducted, the actual returns of this index would be lower.

The performance data quoted represents past performance; past performance does not guarantee future results. The investment return and principal value of an investment will fluctuate so that an investor’s shares, when redeemed, may be worth more or less than their original cost. The Funds’ current performance may be lower or higher than the performance data quoted. Investors may obtain performance information current to the last month-end, within 7 business days, at www.diamond-hill.com.

 

Many sport enthusiasts enjoy discussing a team’s “stats” (statistics). The Major League Baseball (MLB) season consists of 162 games in contrast to the National Football League’s (NFL) season of 16 games. As a result of the significantly higher number of games played, the MLB season is more likely to provide stronger evidence (statistical significance) of relative team success. In the NFL, a key injury or controversial call by an official can be the difference between victory and defeat in one game and consequently, could determine whether a team makes or misses the postseason play-offs.

Similarly, Diamond Hill measures investment results over rolling five-year periods, a length of time that we believe is necessary for statistical significance. [For more discussion on this topic, please refer to “The Importance of Being Long-Term” written by Austin Hawley, CFA in June 2009 and posted to Diamond Hill’s website.] Our investment thesis does not always play out in a twelve month time period, but that does not cause us to abandon our thesis. We also understand that many people focus on shorter time periods when discussing stock market results and moves, and some investors were disappointed with 2010 results in our Long-Short strategy.

What factors impacted the Long-Short strategy’s results in 2010?

Although the total returns for Diamond Hill’s Large Cap and Long-Short Funds were very similar over the past five years, there was a significant difference between each Fund’s results over the past year. The long portfolio in our Long-Short strategy is managed similarly to our long-only Large Cap strategy, thus the difference in returns between the two strategies is explained chiefly by the short portfolio in the Long-Short strategy. Our goal for the short portfolio is to be additive to our long portfolio returns, while reducing the return volatility and correlation with other strategies. This low correlation, especially in comparison to other long-short strategies, reduces overall client portfolio risk. We seek to outperform a long- only benchmark over a five-year period with approximately half the market sensitivity (beta) and lower market volatility. We would expect the Long-Short Fund to outperform in a strong down market and underperform in a strong up market, which is exactly the result achieved in 2008 and 2009, respectively. In 2010, the large negative return on the short portfolio eliminated the cumulative positive return achieved during the first four years of the period (2006-2009). Short positions in the consumer discretionary sector were the primary drivers of underperformance in 2010.

010711_grpah_03

The performance data quoted represents past performance; past performance does not guarantee future results. The investment return and principal value of an investment will fluctuate so that an investor’s shares, when redeemed, may be worth more or less than their original cost. The Funds’ current performance may be lower or higher than the performance data quoted. Investors may obtain performance information current to the last month-end, within 7 business days, at www.diamond-hill.com.

 

Our consumer discretionary position is an example in which our thesis has not played out in the short term; however, our long-term thesis remains intact. The U.S. economy continues to require fiscal and monetary stimulus, pulling economic activity forward at the cost of future growth. In 2010, consumer discretionary spending benefitted from the various government initiatives, offsetting the effects of high unemployment rates and household deleveraging. However, we believe that the consumer will face long-term secular headwinds when the unsustainable government stimulus programs come to an end. We believe this will happen over the next five years, but we cannot predict the exact timing.

This secular thesis provides a framework in which to develop specific company assumptions; however, we rely predominantly on bottom up fundamental analysis for stock selection. We believe there are inherent characteristics of some consumer discretionary companies that make them susceptible to shorting:

  • Lower Barriers to Entry: Low barriers to entry make it difficult for companies to defend their market share. They are constantly being threatened with new, viable competitors. The for-profit education companies are a good example of this competition. As the initial entrants in this industry began to earn outsized returns on their capital, new competitors entered the market. By removing the need for large capital investment in infrastructure, it opened the door for anyone with an accredited curriculum to compete.
  • Market Fragmentation: Market fragmentation makes it difficult for a company to gain pricing power and build economies of scale. For example in the homebuilding industry, builders are price takers and building margins stay reasonably close for similar product offerings. The burdensome carrying costs of inventory and substantial fixed overhead costs make the industry very susceptible to discounting in order to drive volume.
  • Consumer Trends: Consumer trends come and go as many consumers are influenced by the public perception of products they choose to purchase. These characteristics create volatile times of boom and bust for many companies offering the “hot” product. At times, the market will price these firms as if they will grow in perpetuity, though history tells us the consumer is fickle. Many great examples of this lie within the teen retailer space in which sales typically fluctuate with the latest fashion trend.

All of these factors influenced our decision to have significant short positions in the Long-Short strategy in 2010, which admittedly worked against us. We are not pleased with 2010 results; however, we continue to believe that many consumer companies are overvalued and continue to hold significant short positions. Despite the inherent characteristics that favor shorting the consumer discretionary sector, we have also identified companies that we believe are undervalued and hold those names in our long portfolio. These include McDonald’s Corp. (MCD) and Wal-Mart Stores, Inc. (WMT).

What factors influenced results during the past five years?

For the five-year period just ended (2006-2010), the large cap Russell 1000 Index posted an annualized total return of 2.59%. This five year period included a span of seventeen months (October 11, 2007 through March 9, 2009) when the Russell 1000 Index declined 55% peak to trough.

Diamond Hill’s Large Cap Fund (Class I) exceeded the Russell 1000 by a small margin during the five calendar years ended 2010, yet ranked in the top quartile of its peer group of 1,182 funds, based on total return, in the Morningstar Large Cap Value Category for the same time period ended December 31, 2010. This is coincident with a period in which passive (index fund) returns exceeded most actively managed funds. Active managers also had a difficult time beating the index during the 1996-2000 period, in sharp contrast to the following five year period (2001-2005) when the index was beaten by a solid majority of actively managed funds.

What explanation might be offered for such contrasts? During the latter part of the 1990s, U.S. stock market returns were among the highest in our country’s history. As a result, some investors chose to invest in index funds and achieve high returns with lower expenses. At the same time, many managers of actively managed funds became closet indexers. [For more discussion on this topic, please refer to “Why Does Closet Indexing Exist?” written by Tom Schindler, CFA and Bill Zox, CFA in 2001 and posted to Diamond Hill’s website.] At the market peak in the spring of 2000, most investors had become insensitive to valuation. Instead, they relied on the index sector weights for their investment guidance or the seduction of theme-oriented internet revolution investments. Ultimately, this behavior set the stage for valuation sensitive investors (like us) to exceed the indexes over the next five years. This was accomplished by avoiding investments in the most extremely over valued stocks, instead preferring to invest in those stocks that had been overlooked due to their smaller index weights and/or lack of internet-related hype.

This reference to earlier periods begs the question: What happened in the past five years? Rather than returns among the best in our country’s history (1995-1999), the most recent five year period (2006-2010) was among the worst!

First and foremost, the success of valuation sensitive investors has led to a narrower dispersion between the best and the worst results, in sharp contrast to the aforementioned bifurcation. But perhaps, a similar phenomenon to the indexing dominance of the late 1990s is underway with the growing popularity of exchange-traded funds (ETFs), both broad market ETFs as well as sector-specific ETFs. Due to the disappointing returns of actively managed funds (both absolute and relative), ETFs are increasingly used as a substitute for individual stocks in many investors’ portfolios with assets invested in ETFs recently exceeding $1 trillion. As a result, the importance of individual security valuation is de-emphasized, as investors instead choose to look to a particular asset class or theme. However, stock correlations have fallen from the historically high levels reached in May and June of this year, perhaps setting the stage for valuation-sensitive investors to hold sway once again.

Antii Petajisto, visiting assistant finance professor at New York University’s Stern School of Business conducted a study that concluded one third of U.S. stock fund assets are managed by “closet indexers” or managers that claim to be active but instead are closely hugging the benchmark. One metric to illustrate the level of active management is “active share” which measures the percentage of fund assets that are invested differently from the benchmark. According to the Stern School of Business study, funds with active share below 20% are most likely pure index funds, while funds with active share between 20% and 60% generally claim to be active managers but are really closet indexers. Collectively, Diamond Hill’s long-only strategies have ranged from 98% to 78% in the past five years.

What have we learned?

Going back to our sports metaphor (baseball in particular), we are never happy when we experience a “slump” during a season, and we are “studying the film” to see what we may need to adjust. With some analysis, we have learned that our worst results were in shorting “good” companies with high valuations. We define “good” as those companies growing rapidly due to various factors, including superior products or services, improving perceptions, etc. Companies like Chipotle Mexican Grill, Inc. (CMG), Netflix, Inc. (NFLX), Salesforce.com, Inc. (CRM) and Tractor Supply Co. (TSCO) would all be examples of good companies. In each case, the companies exceeded our estimates (and probably consensus estimates) of fundamentals, and valuations became even more stretched. While undoubtedly any or all of these companies could experience stock price declines due to their lofty valuations, in perhaps no case are these valuations stretched to the extremes of 10 years ago.

We prefer being short companies, such as Boeing (BA), Macy’s, Inc. (M) or MGM Resorts (MGM), which face competitor challenges. Boeing faces competitors with labor cost advantages, while Macy’s faces a secular shift away from department stores and MGM faces weakness in Las Vegas. In each case, these companies also face various degrees of stress to their balance sheets.

What will change as a result of recent experiences?

Our fundamental intrinsic value investment philosophy will not and has not changed since our firm’s inception. We continually try to improve our decision-making and learn from our mistakes through an ex-post analysis and review of our investment decisions. During this process, we challenge our assumptions and analysis of the market environment, which could potentially lead to refining certain aspects of our investment process. We will not change the investment process solely based on a short-term period of underperformance.

Going forward, we will continue to reduce short positions in “good” companies while seeking short positions in companies facing competitive challenges. In addition, beginning in 2011 my focus will primarily be on the short portfolio while Chuck Bath, Long-Short Co-Portfolio Manager, will primarily focus on the long portfolio. We believe this division of responsibilities makes sense, because Chuck also manages the Large Cap strategy and there is significant overlap in long positions between the Large Cap and Long-Short strategies. Chuck and I will continue to be Co-Portfolio Managers, and Chris Bingaman will continue to serve as Assistant Portfolio Manager. As before, both Chuck and I can make decisions on either the long or short portfolios in the absence of the other. Chris Bingaman has decision-making authority in our absence.

Looking Ahead

Importantly, we are not disheartened. We believe the results for the next five year period will be better, in part due to our adjustments but also due to the return of valuation sensitivity. We also believe that our absolute return focus, lack of benchmark sensitivity and five year time horizon add value for investors in our strategies. Finally, we have significant personal investment in our strategies. As a result, we experience the same investment returns, good or bad, as our clients.

 

The views expressed are those of the portfolio manager as of January 7, 2011, are subject to change, and may differ from the views of other portfolio managers of the firm as a whole. These opinions are not intended to be a forecast of future events, a guarantee of future results, or investment advice. All data referenced are from sources deemed to be reliable but cannot be guaranteed.

Diamond Hill Capital Management, Inc., a registered investment adviser, serves as Investment Adviser to the Diamond Hill Funds and is paid a fee for its services. The Diamond Hill Funds are distributed by BHIL Distributors, Inc. (Member FINRA), an affiliated company.

An investor should consider the Fund’s investment objectives, risks, and charges and expenses carefully before investing or sending any money. This and other important information about the Fund(s) can be found in the Fund’s(s) prospectus or summary prospectus which can be obtained at www.diamond-hill.com or by calling 888-226-5595. Please read the prospectus or summary prospectus carefully before investing.

Beta is a measure of the volatility of a portfolio relative to the overall market.

Morningstar rankings and fund performance do not reflect sales charges and are based on total return, including reinvestment of dividends and capital gains for the stated periods. The Diamond Hill Large Cap Fund (Class I) was ranked 1,211 of 1,289, 247 of 1,255, and 249 of 1,182 in the Morningstar Large Cap Value Category for the one-, three- and five- year periods ended December 31, 2010, respectively. Past performance is no guarantee of future results.

010711_grpah_04

 

010711_Sign_05

Originally published January 7, 2011

 

 

>>>>