I Have No Idea Where Oil Prices Are Going (And Neither Do You)

|
Includes: AAL, CVX, DAL, F, GM, LUV, OIL, USO, XOM
by: Boris Marjanovic

Summary

The sharp decline in oil prices was completely unexpected - a black swan event - that caught even the experts by surprise.

Up until the day the crash began, everyone anticipated prices to remain stable as they had been in past years.

Even the most trusted price forecasts were so spectacularly wrong that they had to be downward adjusted by more than 50%.

Where prices are heading from here is anyone's guess - investors might as well flip a coin to decide.

For over three years leading up to the 2014 oil crash, oil prices around the world were fairly stable at around $105/barrel. There seemed to be little reason to believe, given signs of economic recovery in the U.S. and abroad, that prices would drop anytime soon. Unfortunately, this illusion of stability would not last long. Starting around mid-2014, prices suddenly began to fall rapidly - a black swan event that caught the entire financial world (and governments) by surprise. Even the most trusted oil price forecasts for 2014, which over-relied on past price trends, were so spectacularly wrong that many of them had to be adjusted downward by more than 50%.

However, now that the dust has settled somewhat, more and more experts (often the very same ones who were wrong before) are confidently predicting where oil prices are headed from here. Some suggest prices will continue heading lower; others think prices will rebound to their prior levels and beyond. Of course, making investment decisions based on such conflicting predictions is about as effective as flipping a coin. In fact, as this article will now show, expert oil price predictions are no more accurate than random guesses, and should be ignored altogether.

The Hindsight Bias

The most important events in our lives, and in history in general, are outliers (sometimes called "black swans") that could not have been predicted, even though they appear to have been predictable in hindsight. For example, consider World War I. In the summer of 1914, the driver of Archduke Franz Ferdinand's car took a wrong turn while driving through Sarajevo. The driver's seemingly trivial mistake gave Serbian nationalist Gavrilo Princip the perfect opportunity to approach the car and assassinate the Archduke and his wife - an act that triggered the deadliest conflict, up until then, in the history of mankind.

With the benefit of hindsight and historical records, it is relatively easy to understand the sequence of chance events that led to World War I. Our brains are wired to think in narrative structures, and most often remember these random historical facts as smaller versions of a larger story. It is like re-watching a movie or re-reading a book - once we already know how events have played out, it makes them seem more predictable than they really were. However, for those experiencing history as it happens, the future is always full of uncertainty. Even seemingly trivial and totally random events - like a driver making a wrong turn - can have disproportionately large consequences, which makes precise predictions impossible. Perhaps Kirkegaard said it best he simply stated, "Life can only be understood backwards, but it must be lived forwards."

This "hindsight bias" or "knew-it-all-along effect" is ever-present in the financial markets. Following every major financial disaster - whether it was the Mexican crisis in 1994, the East Asian crisis in 1997, the Russian default in 1998, the bursting of the dot-com bubble in 2001, or the subprime mortgage crisis in 2008 - there is always a large number of people who claim that they predicted these major events. But what is interesting is that we usually only find out about these so-called "accurate predictions" after they have already occurred, rarely before.

No One Predicted the Oil Crash

The financial news media's convincing-sounding explanation is that the oil crash was triggered by a global oversupply of oil. This oversupply, we are told, was caused by three main factors: falling demand from Europe and Asia, booming North American production, and the OPEC's refusal to cut production. As I already explained, after-the-fact explanations such as this make the past seem more predictable than it actually was. It is like watching a movie for the second time - you already know the outcome. However, for those of us experiencing history as it happens, we can never be sure about what will happen next. This is why we must always be skeptical of those who claim to make accurate predictions.

I recently got into a heated argument with a market strategist who publishes a monthly market-timing newsletter. He was boasting to me, as well as anyone else who would listen, about how he predicted the recent oil crash. The reason he knew oil prices would decline, he confidently explained, was because supply had been exceeding demand for some time (essentially the same after-the-fact explanation given by the media). But when I asked him to simply show me an old newsletter in which he alerted his subscribers about "an impending oil crash," he angrily refused to do so. The reason, as I later learned from one of his long-time subscribers, was because no such newsletter was ever published. Had he known that oil prices would be cut in half in the latter half of 2014, he would have surely told his subscribers about it!

The truth of the matter is that no one predicted the oil crash in advance. If anything, most experts anticipated oil prices to remain fairly stable, as they had been during the previous few years. In early 2014, Bloomberg's survey of the "most accurate" oil price forecasters yielded a forecast of $105/barrel for the year. In a similar survey conducted by The Wall Street Journal in January 2014, the economists surveyed expected oil to end the year at about $95/barrel. The U.S. Energy Information Administration projected oil prices to remain around $99/barrel in 2014. OPEC expected 2014 oil prices to "remain at relatively stable levels" at around $100/barrel. Goldman Sachs confidently stated "we do not expect a material collapse in oil prices," and projected 2014 prices to average $100/barrel. To say that these experts were off the mark would be an understatement. The chart below shows what actually happened.

Exhibit 1: Brent Oil Prices (US$ per Barrel)

Source: A North Investments, U.S. Energy Information Administration

Oil prices ended up falling more than 50% in just a few short months. But along the way, the experts constantly revised their useless forecasts. When the price of a barrel of oil was $105, they thought it would go up to $108; when it was $100, they thought it would go up to $104; when it was $90, they thought it would go up to $100; when it was $85, they thought it would go up to $90; and when it was $65, they thought it would go up to $80. It is $49 as of this writing, but I am sure they are hastily working on their new forecast as I type these words. This is a strategy that has been used by fortunetellers and astrologers for centuries - make enough predictions, and at least a couple of them are bound to be end up being correct. The correct predictions get all of the media attention. Thus, the public is selectively exposed to "correct" predictions and almost never hears about the thousands of failures.

Exhibit 2: Evolution of Brent Oil Price Forecasts (US$ per Barrel)

Note: Each line represents the spot price, the three-month forecast, and the 12-month forecast.

Source: A North Investments, Gavyn Davies (Financial Times)

To further show that the oil crash was completely unexpected, all we need to do is examine the historical Google search volume trend for the keyword "oil prices." As can be seen below, with the exception of a few small spikes in 2011, search volume was in a long-term downward trend. This makes sense, given that the price of oil was fairly stable during this time period. In fact, search volume was at its lowest point in mid-2014 (right before the crash began). But once oil prices were clearly falling, search volume increased more than tenfold within a matter of months. The fact is, if people really had been anticipating an oil crash, search volume would not have been in a downtrend during the first half of 2014.

Exhibit 3: Google Search Volume for the Keyword "Oil Prices"

Note: (1) Google Trends shows search volume over time for various keyword searches. (2) A well-known study has shown that people do more finance-related searches when they are worried about the state of the market.

Source: A North Investments, Google Trends

The reason why it is so difficult, and I would argue impossible, to predict future oil prices is because the price of oil - like the price of all commodities - bounces around as a result of changes in supply and demand. These changes are influenced by an infinite number of unknown variables, which makes them entirely random and unpredictable to us. To further complicate matters, even a small supply and demand imbalance can have a large impact on prices. For instance, we are told that the 2014 oil price crash is the result of an oversupply of less than 1.5 million barrels of oil a day. To put things in perspective, the whole world consumes over 90 million barrels of oil every day. Simply put, the chance of correctly predicting future oil prices is as likely as winning the lottery!

Exhibit 4: Global Oil Supply and Demand (Million Barrels per Day)

Source: A North Investments, U.S. Energy Information Administration

Profiting from Oil's Uncertain Future

Everyone is unsure about where oil prices will go from here. Some bearish experts suggest prices will be range-bound for a while before heading lower, even as low as $10/barrel. In this particular situation, large car manufacturers like Ford (NYSE:F) and General Motors (NYSE:GM) would benefit as consumers upgrade to larger, less-fuel-efficient vehicles that sell at higher margins. Low oil prices also benefit airlines like Delta Air Lines (NYSE:DAL), American Airlines (NASDAQ:AAL), and Southwest Airlines (NYSE:LUV), because it diminishes their cost to refill planes, thus making fares more lucrative and driving profits higher. On the other extreme, those who are more bullish, like billionaire investor T. Boone Pickens, believe prices will rebound to $100/barrel. In this case, it is obvious that buying oil stocks like Chevron (NYSE:CVX), Exxon Mobil (NYSE:XOM), and/or anything else oil-related is the best choice. Of course, this extremely wide spread of oil price forecasts is of little help when it comes to making investment decisions, but it reflects the huge uncertainty in the marketplace.

Given this uncertainty, perhaps the best investors can do is flip a coin: heads, oil prices will rise; tails, oil prices will fall. So, depending on which side of the coin lands up, investors will know whether they should be bullish (bet on rising oil prices) or bearish (bet on falling oil prices). Since oil prices tend to fluctuate widely over time, it is not necessary to consistently make the right directional call. As long as the bets have an asymmetric payoff structure (i.e., huge upside and minimal downside - like an option), an investor can be right less than half the time and still make substantial profits. It really could not be any simpler than that.

Summary and Conclusion

If the oil crash taught us anything at all, it is that oil price movements are unpredictable - no one knows if they will go up, down, or sideways - least of all expert forecasters. The only thing we can anticipate with any certainty is that prices will fluctuate widely over time, which makes oil an excellent trading vehicle. Assuming the trades are set up correctly, clever investors can profit greatly from these large and unpredictable price fluctuations.

Disclosure: The author has no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.

The author wrote this article themselves, and it expresses their own opinions. The author is not receiving compensation for it (other than from Seeking Alpha). The author has no business relationship with any company whose stock is mentioned in this article.