Berkshire Ruminations

Tuesday, February 14, 2012

Transparent Value: Moneyball for Investing

Prior to Moneyball, baseball conventional wisdom was that players could be efficiently priced utilizing the existing scouting system and accepted metrics of talent. What was troubling to the Oakland A’s was that this widely accepted approach was not consistently identifying successful professional players. Prior to an explosion in salaries, these mistakes were tolerable, but as player salaries increased, the situation became untenable for small market teams like the Oakland A’s with lower payrolls. What allowed the Oakland A’s to forever change baseball was that they were unwilling to be victimized by the status quo. They identified and implemented statistical research that had been ignored by the baseball establishment and this allowed them to value players better. Their new approach allowed them to effectively compete with large market teams in an increasingly unfair environment.

Similarly, Transparent Value rejects the conventional wisdom of investing. We recognize that traditional active management has not been able to consistently outperform benchmarks. This is not because research analysts and portfolio managers are not knowledgeable, dedicated individuals; it is because they are being asked do the impossible. The future is inherently unknowable, and yet traditional active management often relies on the ability to predict the future. Inability to predict the future is not isolated to investing. Significant research has been conducted on the accuracy of expert forecasts across many fields, including investing, and the results are dismal. The bottom line is that research analysts and portfolio managers should not be expected to be better at prediction than other experts, including baseball scouts.

This was not a problem for the traditional active equity management industry in the bull market from 1982 through 2000, when equity returns were historically high and benchmark-linked investments were not widely available. However, since 2000, equity returns have been historically low, and there has been an explosion of exchange traded funds (ETFs) that track benchmarks. In this low return environment, if active management cannot outperform, its failure to do so has a greater negative impact on investors. And now advisors have low cost ETF alternatives to underperformance. However, Transparent Value believes the choice between active equity management, predicated upon the ability to predict the future, and indexed investments, that abdicate the opportunity for outperformance, represents a failure of imagination. Consequently, Transparent Value has created an investment methodology that challenges conventional wisdom in pursuit of benchmark-beating performance. We have reversed the conventionally accepted approach to valuing companies and used statistics to gain new insights into the likelihood of which stocks may win or lose.

In Moneyball, the Oakland A’s edge was a result of Billy Beane and his statistical analysts processing knowable information more effectively than the competition. In baseball, runs win games and each team has an opportunity each inning to score runs, but only until it makes three outs. A batter’s ability to avoid making an out, whether it be through a hit or a walk, is thus crucial to preserving the opportunity to create runs. Yet batting average, the conventional metric for evaluating a batter’s skill, does not factor in walks. Since a key component of batting skill is avoiding making an out, ignoring walks is a serious flaw. The Oakland A’s analysis determined that a team’s aggregate on-base percentage, which accounts for walks, had a much higher correlation to winning games than aggregate batting average.

Recognizing this inefficiency, Billy Beane chose to sign undervalued players with higher on-base percentages. Exploiting this inefficiency allowed the Oakland A’s to remain very competitive while having a fraction of the payroll of big market teams.
In our opinion, in investing there are two ways to gain an edge. One is to obtain private information from a company’s management and the other is to more effectively process public information and capture a new insight. As a result of the passage of Regulation Fair Disclosure in 2000 and the government’s rigorous enforcement of insider trading laws, gaining an edge through private information is harder and riskier than ever. However, like the Oakland A’s, Transparent Value believes there is an exploitable opportunity to more effectively process a company’s publicly available information to seek an investment edge.

While Billy Beane and his team exploited the on-base percentage metric, Transparent Value exploits our proprietary Required Business Performance® Probability (RBP® Probability) metric. Our methodology starts with a company’s current stock price and most recent financial statements, and by reversing a generally accepted discounted cash flow valuation model, determines the Required Business Performance® needed to support the current stock price. Then we statistically determine, using management’s past business performance, the probability that they will deliver the Required Business Performance® to support the stock price. The methodology, while fundamental in nature, is fact-based and forecast-free.

Unlike baseball, in investing we keep on playing the same team, the benchmark. As discussed earlier, traditional active managers attempt to beat their benchmark by forecasting the unknowable future. Evaluated using the percentage of active managers that consistently beat their benchmark, this approach has not been optimal. In contrast, for a given benchmark, Transparent Value, utilizing RBP® Probability, ranks every company in the benchmark based upon management’s ability to deliver the Required Business Performance® to support its current stock price. By only investing in companies with the highest RBP® Probabilities, we are effectively avoiding those companies in the benchmark that we believe have the highest likelihood of disappointing and underperforming relative to other companies. Put simply, if you can avoid investing in the worst relative performers, you should outperform the benchmark. Like baseball’s on base percentage, Transparent Value’s RBP® Probability is all about using knowable information to avoid investment strikeouts.

When you consider the success of Billy Beane and his team, the key was in their ability to use an objective process to avoid the value destroying impact of behavioral biases. Scouts, the traditional arbiters of talent, brought significant subjective bias to their evaluations. They wanted to draft players that looked the part and had the tools, while often ignoring what players had actually done. For scouts, the value of the player was tied to what the scout could see the player becoming in his mind’s eye. Fortunately for the Oakland A’s, Billy Beane understood that the mind’s eye can frequently deceive. He was one of those highly valued recruits, who looked the part and had all the tools, but never lived up to his major league promise. Consequently, for Billy Beane, the value of a young player should not be based upon what he looked like, or what he might become, but what he had done.As Michael Lewis put it in the book, “He (Billy Beane) thought of himself to be fighting a war against subjective judgment.”

At Transparent Value, because we believe exposure to behavioral biases destroys performance, we are passionate about identifying potential behavioral biases and avoiding them. The field of behavioral finance has identified many investment behavioral biases. From our perspective, it is not as important to identify which bias is at work, as to be able to recognize and avoid bias in the aggregate. We identify potential behavioral bias in stock prices through our Required Business Performance® Methodology. The counterpart to RBP® Probability is our Behavioral Risk Indicator (BRI). If a company has an RBP® Probability of 75%, its BRI would be 25%. The BRI gives us an indication of whether the current stock price is misaligned with management’s ability to deliver the Required Business Performance® to support its price. Since a stock’s price is set by buyers and sellers who are subject to behavioral biases, a high BRI is a warning that there is a high likelihood that behavioral biases have impacted a stock’s price and that the stock should be avoided.

As did Billy Beane with baseball players, the value we place on a company’s stock is not derived from what the company looks like, or what it might become, but what the company’s management has done. Company managements are like baseball players in the sense that they don’t change their nature overnight. What can quickly and substantially change is a company’s stock price as determined by buyers and sellers. This is why our Required Business Performance® Methodology starts with the current stock price and then, using management ’s past performance, solves for the probability that management can deliver the Required Business Performance® to support the stock price. Unlike past investment performance, when it comes to a baseball player and a company’s management team, past performance can be a good indicator of future results. Additionally, using a baseball player’s or a company management team’s past performance as a guide is a more objective approach than subjective assumptions about what might happen in the future.

When evaluating the Oakland A’s approach to valuing players, Michael Lewis wrote, “In human behavior there was always uncertainty and risk. The goal of the Oakland front office was simply to minimize the risk. Their solution wasn’t perfect, it was just better than the hoary alternative, decisions by gut feel.”
If this same evaluation was made of Transparent Value, we would agree with it. Not every high RBP® Probability stock is going to outperform on a relative basis and not every high BRI stock is going to underperform on a relative basis. However, we manage this risk by building a rules-based portfolio of stocks with high RBP® Probabilities. We take a rules-based approach to prevent behavioral biases from creeping in at the portfolio level. While not every stock will perform as expected, in aggregate we increase the odds of outperforming a corresponding benchmark. While our solution is not perfect, we believe it is better than relying on forecasting the unknowable future or abdicating the opportunity to outperform a benchmark.


Post a Comment

<< Home