Most managers (traditional or quant) seem to view the world in a static sense. They may look for high quality, low valuation, momentum, growth, good company, good management, or any one of a myriad of other “current states of being.” We believe that many (if not most) of these approaches indirectly touch on a particular phenomenon that occurs in the market. Our approach directly embraces this phenomenon. It differs in that our focus integrates these “tangential” ideas into a single holistic philosophy regarding the interaction between changing fundamentals and shifting sentiment. Specifically, we believe that companies/stocks are constantly in motion, seeking an equilibrium that will never be reached. We believe we can take advantage of the interaction between the cycling of company fundamentals and the associated shift in sentiment. We look to initially invest in companies as they enter or attempt to extend a fundamental cycle. We target those for which sentiment remains low but is on the rise. We will hold companies throughout the cycle (even when expensive) until the point at which fundamentals begin to turn down prior to a reversal in sentiment. This opportunity is based on the following three beliefs:
- Company fundamentals ebb and flow. Sometimes these “cycles” can be quite pronounced in both size and length of time. Accurately predicting either size or length of time is foolhardy.
- The relationship between current price and future cash flows is not like gravity. It is a flexible relationship that contracts and expands based on the market’s interpretation of the change in company fundamentals, as well as across the overall market.
- Sentiment is sticky. Once human sentiment is set in a particular direction, it tends to trend in that direction. Once set, it is difficult for human beings to emotionally “change directions” as new information is presented to them.
A static approach to investing (see No. 1) can lead to significant buy/sell discipline issues. Because value managers sell the stock once it becomes expensive, they are often tempted to push purchases earlier in the cycle, attempting to capture more excess return by timing the upward inflection point. This can lead to a phenomenon referred to as “catching a falling knife” and is, in essence, a buy discipline issue. Growth managers rarely experience this. Rather, because they typically invest only after it has established a track record of earnings growth, they can tend to push the divestiture of that investment as close as possible to the downward inflection point of the earnings cycle. Mistiming this inflection can lead to significant losses as sentiment quickly sours this is often referred to as a lack of sell discipline. Our approach allows us to hold investments as they transition from value to growth. This extends the range over which we can make buy/sell decisions so as to avoid the temptation of timing inflection points. Our goal is to simply identify the direction of the current (for both fundamentals and sentiment) and move along in that current.
This area became a primary concern to us post-2009. At that point we realized that the industry’s accepted view of risk measurement and control was not adequate to manage risk in the new risk on/risk off world that exists post the financial crisis of 2008/09.
Clients pay us to take informed active risk. In today’s investment community the dominant measure of active risk is Tracking Error (TE). Most view TE as consisting of two components: systematic and stock-specific risk. However, we believe that these two components (as commonly estimated) contain a hidden third critical component of TE – strategy risk. Systematic risk is typically defined as that component of risk that is common across a broad number of stocks in an investment universe. Typical sources of such risk include such generic dimensions as industry, country, or company capitalization. Stock specific risk is then defined as that portion of risk that is not explained by such generic common factors. The key issue here is that risk defined in this general way fails to take into account the target of the risk evaluation. Specifically, we believe that an important source of common risk among the stocks contained within any “informed” portfolio is the risk that comes from the commonality introduced through the stock selection strategy (ie philosophy) of the manager of that portfolio. Regardless of the manager’s approach (eg quant or fundamental), generic risk factors simply capture systematic risk associated with exposures to generic risk. Successful strategies, however, employ non-consensus proprietary approaches. The systematic risk associated with exposure to such unique strategies would then, by definition, not be adequately captured by consensus measures. We believe our focus on strategy risk addresses this issue. We believe the active risk of our portfolios (and all informed portfolios) consists of 1) generic uncompensated systematic risk, 2) compensated strategy risk, and 3) stock specific (ie residual) risk.
We manage strategy risk in a number of ways integrated throughout our investment process. For example, we integrate the idea into weighting the various dimensions of our strategy within our multifactor model. We apply it in matching off the shelf risk models with various CV strategies. We directly measure it within customized risk models. Most importantly, we understand qualitatively when and how it impacts our portfolio. This understanding allow us to manager through those most difficult periods where strategy risk is heightened and strategy exposure needs to be moderated in order to maintain a steady portfolio impact. They also guide us in those easier periods where strategy risk is more mitigated and we can correspondingly increase our strategy exposure to take advantage of a favorable stock selection environment. In summary, we believe that the improvements we have made in managing strategy risk has led to dramatic increases in performance consistency across our suite of products after the financial crisis of 2008/09.
A common approach to valuation among managers focuses on forming proprietary estimates of earnings three to five years into the future. They often map these estimates into price through a proprietary discounting methodology. We take a different approach. We believe in today’s uncertain world that it is difficult, if not impossible, to both accurately and proprietarily predict earnings three to five years in the future. As the forecasting horizon lengthens, things we do not yet know multiply and inevitably dominate the few things we know now. Instead we utilize the valuation framework to obtain a relative measure of market sentiment with regard to the markets’ own explicit forecast of future earnings. Specifically, sell-side estimates form the market’s explicit forecast of future earnings. Price reflects the current level of cynicism/optimism regarding these forecasts. We use a proprietary methodology to focus on those companies for which the market price implies a higher relative level of cynicism regarding future earnings.
We are unaware of any other industry that makes what we believe to be a specious and unproductive differentiation between competitors based on the tools they utilize. Note that nobody refers to Toyota as a robot user because they use robots to make cars. Whatever the source of this differentiation, it has resulted in a polarized industry wherein both quant tools and traditional insight are too often viewed suspiciously by the “opposing” sides. This leads to an underutilization of each approach by the very managers that arguably most need them to add marginal value to their investment process. We believe that quant tools and traditional work add maximum value when applied to the respective problems of breadth and specificity thus, the common phrase “a mile wide vs. a mile deep.” Accordingly, we believe in the integration of these complementary approaches to investing. Our investment begins with a bottom-up fundamentally based philosophy regarding the type of company we are looking to invest in (described above). We institutionalize this fundamentally based philosophy utilizing a proprietary quant process that focuses on narrowing the relevant universe to a theoretically optimal set of proposed investment decisions. We then address the specificity problem through traditional work that focuses on excluding those “optimal” investment decisions that we view as having been misidentified by the institutionalized process.
We believe that most managers classified as “quants” employ an anomaly-based investment philosophy. They employ significant numbers of research personnel to search the investment universe for any anomalies that might provide investment opportunities. They then model these anomalies so that the resulting multifactor models they employ are prescriptive in the sense that they define the opportunity. Our approach is different. We begin with a fundamentally based philosophy (see No. 1), distinct from any tools we might employ. We create quantitative models that are descriptive in that they are designed to reflect this philosophy, not to define it. For this reason, our use of quantitative models will always be designed around our investment philosophy rather than the philosophy around the quantitative models.
Most managers that use traditional work rely on it to do the heavy lifting of deciding which stocks from a broad universe to include in their portfolios in other words, to identify a set of best ideas. The substantial resources required for this approach necessarily lead to a reduction in breadth. However, very few, if any managers attempt to measure the amount of breadth lost to this resource hog. We believe this is an inefficient and potentially harmful use of a powerful tool.
We believe that too much traditional work, too early in the process can easily lead to a “forest for the trees” problem in stock selection. This arises from the fact that getting too close to a subject can distort that subject relative to available alternatives. What differentiates a good (or bad) company from a good investment is the interaction between 1) buy-side global investment demand for that company, and 2) the global investment supply of like companies. Successfully navigating this interaction necessitates a clear and efficacious strategy for comparing alternatives across the investment universe (i.e., a breadth problem).
In summary, we believe that traditional tools are best used where 1) they can be most efficiently used, and 2) they are least likely to distract an investor from the primary investment philosophy of the product. We believe that this is at the end of the investment process as part of a de-selection of proposed ideas. The goal of our traditional work is to 1) eliminate those stocks that are most likely to appear on our “worst contributors” list before they appear, and 2) provide a virtuous feedback loop into the institutionalized process to aid future stock selection.
The six senior members of our team have been together for more than a decade. Some of us have been working together for over 15 years. We believe that this is a people business. Consequently, team turnover is perhaps the most damaging thing that can happen to a team’s future ability to outperform. We believe that people are NOT interchangeable, nor expendable.
At this point, we have a team of experienced investment professionals who have been working together on a common, tightly defined investment philosophy across multiple styles/geographies/market environments for more than a decade. This results in a wealth of institutionalized knowledge about every aspect of the philosophy and investment process, as well as a maturation of the process that is rare in the industry.
On a related note, we believe that all steps of an investment process are highly interdependent. Bad performance often emanates not from a particular step, but from faulty connections between steps. For this reason, senior investment professionals manage every part of the investment process of the portfolio for which they are accountable. This includes all traditional and quantitative tools, as well as trading and ex-post attribution and review.
Pure quants typically view investing as a hard science. Consequently, they often prefer 1) Ph.D.s and/or 2) hard science backgrounds (e.g., physics). Traditional managers, on the other hand, tend to hire from finance, accounting, economics, and humanities. Our credentials are consistent with a quant/traditional integration approach. Our Portfolio Managers each have a passion for investing and share a fundamentally based idea for the kind of companies we are looking for. We believe that a practical education that blends a fundamental understanding of companies and the market with a comprehensive quantitative toolbox provides the best opportunity to outperform consistently through time. We believe essential credentials to investing are 1) passion, 2) character, 3) experience, 4) insight, and 5) knowledge.
We are, and have always been, the majority owners of our business. We have twelve internal owners of the business, with no one person owning more than 20%. Eagle Asset Management holds a 45% minority interest.