Andy Hall's castle in Germany (SCHLOSS DERNEBURG).
Source: Hall Art Foundation.
Andy Hall is best known to the investing public as the crude oil trader who made $100 million shorting oil during the 2008 financial crisis when Citibank owned his Phibro Energy firm. He is also referred to as the "god of crude oil trading." The way things are going, he may become better known as the oil trader who lost $3 to $4 billion.
When his bonus from Citibank was blocked under TARP, his firm was sold to Occidental Petroleum. Mr. Hall later established a new fund, Astenbeck Capital Management. Assets under management (AUM) were $4.8 billion in January 2013. By the end of 2015, his fund had dropped in value to about $2 billion, after suffering a 35% loss in 2015. Given published returns in the media, I estimate that he experienced net redemptions of about $1.15 billion over the past 3 years.
Assuming he has not changed his strategy, which seems like a safe bet, given the conclusion in his January 6, 2016 letter ("We therefore continue to believe that the shorter term headwinds are ultimately trumped by the longer term outlook for prices which remains firmly to the upside."), Mr. Hall's fund is down another 20% in 2016 YTD, to $1.6 billion. If prices drop to $20, his AUM will be down to about $1.06 billion, assuming no "long" strategy change or redemptions.
Note: Estimated cumulative returns based on published sources.
I think that his results reflect a combination of two problems that investors and energy traders in particular need to pay keen attention to: "confirmation bias" and failure of price risk management.
Behavioral finance describes this emotional phenomenon this way: "the tendency to search for, interpret, favor, and recall information in a way that confirms one's beliefs or hypotheses, while giving disproportionately less consideration to alternative possibilities." People also tend to interpret ambiguous evidence as supporting their existing position.
It is considered to be a systematic error of inductive reasoning. People are biased toward confirming their existing beliefs. The effect is stronger for emotionally-charged issues and for deeply entrenched beliefs. This would certainly apply to trading a massive commodity fund, where billions of dollars are involved.
Confirmation biases contribute to overconfidence in personal beliefs and can maintain or strengthen beliefs in the face of contrary evidence. According to one investor close to Mr. Hall, "He's taking this the way he's taken every other dislocation in the oil market, he absolutely thinks he's right."
Mr. Hall's great financial success over the years seems to have led him to believe that he could not be wrong, notwithstanding what the market was telling him and his mounting losses. For example, Mr. Hall had founded the Hall Art Collection in 2007, which has four locations, including his castle in Germany (see photo above).
In September 2014, Mr. Hall had come to the conclusion that oil prices were headed to $150 a barrel by 2020. His employees and advisors confirmed that he was buying long-dated oil contracts to be delivered as far out as 2019. His thinking, before oil prices had begun to collapse, was that the shale oil boom would slow by the end of 2014 and fizzle as soon as 2016. The simplest of his reasons was that producers had already drilled in many of the best areas, or "sweet spots."
I debated the rationale for his continued long position that was contained in his July 2015 letter to investors in a post here. In summary, he had four reasons: (1) oil demand is "elastic" and would rise due to lower oil prices; (2) the fundamentals "continue to improve," (3) citing "Phenomenal Demand Growth," and (4) "Price Risks Skewed to the Upside." In refuting each of these reasons, I submit that they clearly illustrate confirmation bias in his beliefs.
Missing: Price Risk Management
In The Intelligent Investor, Benjamin Graham states, "the essence of investment management is the management of risks, not the management of returns." At the heart of this approach is loss minimization, deliberately protecting oneself against serious losses. Warren Buffett described this book as "by far the best on investing ever written."
Nobel economics laureate Robert J. Shiller wrote in the 1980s about the failure of the efficient markets theory to explain stock market prices. He found that prices exhibit "excess volatility," that prices move more in each direction than the fundamentals could explain. He attributed this phenomenon to one of the oldest theories about financial markets: rising prices create enthusiasm by investors to get into the market to prosper (greed). But such bubbles eventually burst. When they do, falling prices create pessimism (fear) among investors, causing a "negative bubble" until the market reaches an unsustainable, low price.
As a result of investor behavior, rational price forecasting fails. If the market can be expected to act irrationally, price risk management is the only rational way to achieve loss control, as promoted by Graham.
There are two realities investors face if they do not control the drawdown in their accounts: if there is a large drawdown, the percentage gains must be larger than the percentage losses to get to even; and once losses exceed risk tolerances, investors "lock-in" the loss by exiting the position, and often do not reestablish the position until prices have recovered and they feel "safe" again that the market is going up.
To illustrate the first point, if an account loses 50 percent of its value, a 100 percent gain is required to return the account to its original level. To illustrate the second point, once equities lost 50% in 2008, there was mass exodus from the market. Studies subsequently showed that investors kept their remaining funds in money market account for years, after having experienced such large losses.
In Mr. Hall's case, last January 2015, he stated that $40 was an "absolute price floor." Given that's what he thought, he should have hedged his risk at that level, to prevent further losses in the event prices were to break further below his expectations. Instead, by early December, he was down 26% for the year and wrote, "Now is not the time to exit the market." In January, prices are down another 20%, which is probably reflected in his returns.
To avoid falling victim to the confirmation bias, and to control downside losses as advocated by Graham, I favor a quantitative approach to risk management. One advantage of a quantitative approach is that the decision-making process can be thought out in advance, rather than being subject to emotional-reactivity when losses are mounting. Another is that an algorithmic approach can be back-tested to see how it would have performed in the past under a variety of market conditions. While such back-testing does not provide any guarantee of future performance, I sure do not want to use a process that would have failed in the past.
Shiller's observations about market behavior were not new, as he explained. They are basically that "fear" and "greed" can drive prices lower or higher than they should go. So what I did was to try to quantify what gains might explain greed, and what losses might drive fear. I back-tested my hypotheses during an "In-sample" period, and then applied the same algorithm during an "Out-of-Sample" period. The results can be seen here.
Nothing is perfect, but it does a good job of controlling draw-downs. Sometimes it gets out when it should have stayed in, but that's better than staying in when it should have gotten out. The interesting thing is that the really big drops in the prices of oil and equity markets occur after there has been plenty of advance warning detected by the algorithm. For example, applying the same algorithm to equity markets had the position in cash prior to the October 1987 crash and avoided (simulated) much of the loss in equities from 1929-1933.
It should be recognized that taking less risk does not necessarily produce lower returns, and taking more risk does not necessarily produce higher returns. That is the nature of risk.
The dual challenges of confirmation bias and risk management are formidable for behavioral reasons. Carefully developing a trading process in advance can provide many benefits and is consistent with Benjamin Graham's very definition of investment management: the management of risks.
Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.