Denial Is Not Just A River In Egypt

Apr.19.12 | About: Peabody Energy (BTUUQ)

There are three important cognitive biases that every investor needs to be aware of. These are denial, the confirmation bias, and the self-serving bias. Indeed, there are many more interesting cognitive biases that greatly influence one's decision-making in the markets - together they constitute a field known as Behavioral Finance. Here's a bunch of links that cover most of them. But the three I mention are particularly important.

Why am I bringing this up? Because every day investors get confronted with these biases. Either by themselves being affected by them, or by being influenced by others that are under their spell. As a short reminder, let me define each of the biases, before I take on a particularly strong case for one of them.


This is perhaps the most amazing of the biases. It's simply the tendency of people to completely ignore or reject any evidence that's in contrast to their previously held beliefs - no matter how obvious, unavoidable and overwhelming the evidence might be.

Confirmation bias

The confirmation bias refers to the tendency people have to look selectively for information that confirms their beliefs, and to attribute greater value to such information. This is particularly important in the market, because for every thesis, there will be hundreds of data-points that can at any time provide relief, even if they are rather irrelevant. It is thus easy to keep holding onto a flawed thesis, even if in hindsight it cannot predict or explain the outcome.

This for instance happens quite often on (AMZN), with people alternatively holding themselves to approaching Xmas season, the Kindle fire sales, the upcoming tablets, how Kiva robots can increase efficiency, etc., as reasons to hold the stock, when in fact the much larger and important aggregate fundamentals, namely the earnings, are going lower and lower and should obviously override any of those positive tidbits one can come up with to hang onto his beliefs.

Self-serving bias

This bias refers to the tendency people have of attributing positive outcomes to internal factors, and negative outcomes to external factors, as in, if something goes right they attribute it to skill, and if something goes wrong, they attribute it to bad luck or manipulation.

Commenter bearfund on one of my articles on , put it succinctly:

I can't speak for anyone else, but I've never owned Amazon and I'm quite happy with that. Crying in my soup? Nothing could be further from the truth. Because for every Amazon there's a and 50 companies that go nowhere, and I'm not delusional enough to think I can pick the winner. The people who bought AMZN long ago and are crowing about it are a classic survivorship bias story and nothing more. I'm happy that you were lucky. I know you don't see it that way; part of the magic of survivorship bias is that the survivors can always find reasons to believe themselves smarter than their otherwise indistinguishable brethren who failed. But while you were rolling the dice, my money was working for me. Someone has to win the lottery, but it probably won't be me so I'm better off investing my money.

Congratulations, though. Many a dynasty has been built on nothing but dumb luck, and there's no reason you shouldn't enjoy yours.

So why am I bringing these biases to the fore?

I'm doing it because in the last few weeks, regarding the coal (KO) debate I'm having with other Seeking Alpha contributors, we're getting to see an extreme example of the cognitive biases in question, namely denial and confirmation bias.

This debate revolves around whether there can be substitution of coal-fired generation for natural gas generation that isn't the result of the generators that have the capacity to use both fuels switching, or through retrofitting or coal generators to burn natural gas (UNG). The debate thus revolves around switching that is the result of dispatching of natural gas generators ahead of coal-fired units.

In the power markets, with some constraints regarding reserve power, contracts, grid stability, etc, the power plants are dispatched according to their variable costs. Usually, coal is the cheaper fuel to burn, but given natural gas's price plunge in the recent months, in many places natural gas is now cheaper (especially taking into account the higher efficiencies natural gas generators usually enjoy). This is leading to substitution of coal for natural gas due to more units burning natural gas being dispatched to run than was otherwise normal.

Mark Anthony flatly denies this is the case. He attributes the fall in coal usage to weather; he shows that coal has not been displaced over many years (which is true - this kind of substitution is taking place mostly in the last 4 months); he shows that coal enjoys much higher utilization rates than natural gas (which is true but irrelevant, what matters is how those compare with similar periods in the past, not what they are in absolute terms).

Up to here, it's all normal - mostly different interpretations of the data. But when does it become an obvious bias? Well, that happens when not only is there no recovery in coal usage (whereas his theory that it's all the weather would imply that when the weather becomes less important, coal ought to recover), but also when I quoted numerous studies and evidence of such dispatching substitution, and yet he still holds to his beliefs. Just as a summary, I have already quoted:


Short‐Term Energy Outlook Supplement, "The Implications of Lower Natural Gas Prices for the Electric Generation Mix in the Southeast", May 2009 . This entire document is dedicated to explain precisely how dispatching would be affected by lower natural gas prices. They very thing Mark says can't exist.

Just one of the many paragraphs in the document, illustrating my point:

"Based on December 2008 average delivered coal prices of $2.58 per million Btu (MMBtu) in the ESC and $3.06 per MMBtu in the SA, a decline in the average delivered natural gas price from $4.75 to $4.25 per MMBtu in each region could boost natural gas consumption for baseload electricity generation in the electric power sector by about 2.1 billion cubic feet per day (Bcf/d) in the ESC and SA combined."


In this presentation, "Potential for Displacing Coal With Generation from Existing Natural Gas Plants", November 19 2010, slide 7, EIA considers that using 2007 data, natural gas generation could replace as much as 32% of the coal-fired generation through increased utilization.

National Petroleum Council

"POWER GENERATION NATURAL GAS DEMAND", NPC North American Resource Development Study, September 15, 2011:

"Generally power plants are dispatched based on variable generation costs with lower cost power plants being dispatched first. With low coal prices in most regions of the country, coal-fueled power plants will nearly always dispatch ahead of natural gas fuel power plants. Only where we find very efficient gas plants (NGCC) and low gas prices ($3-$5/MMBtu) does a gas-fired plant move ahead in the dispatch. A CO2 allowance price would likely disadvantage a fossil coal plant more than a gas-fired plant, and produce a similar dispatch scenario. However, there will be regional differences related to power prices and energy supplies (coal, gas, etc.) that will impact how generation plants are dispatched. With a CO2 emission allowance cost added to the variable production cost of electricity, the spread between delivered coal and gas prices is expected to be lower. However, a lower cost NGCC might not dispatch if there are transmission or other reliability constraints to displacing a more expensive coal plant."

Encana (ECA)

Encana's Matt Most had this to saw in a recent interview:

"So the competitive relationship has gotten to the point that in broad parts of the United States, natural gas is a less-expensive fuel, it's dispatching ahead of coal units, coal units are taking longer shutdowns, longer maintenance outages, staying on their minimums outer so that utilities can use natural gas more."


There is little doubt, for any clear-thinking person out there, that switching from coal to natural gas is occurring due to the dispatching mechanism. This is crucial to understand the sector, because if the drop in coal consumption was just weather-related, the effect would be temporary and soon enough coal consumption would be back to normal.

Under that scenario, investing in coal suppliers such as Peabody Energy (BTU), Alpha Natural Resources (ANR), CONSOL Energy (CNX), Arch Coal (ACI) and James River Coal (JRCC) would be a no-brainer. But, as I've shown, that isn't the case. Coal demand is dropping hard due to natural gas substitution through dispatching, and this substitution won't go away until the relative prices of natural gas and coal change. So either natural gas goes up a lot, to negate its advantage, or coal drops like a rock to regain its position. Coal prices dropping or volume continuing down are obviously incredibly negative for the coal industry and lead to the chance we might observe a panic in it.

In short, one needs to think clearly and not let himself be caught in these obvious cognitive biases, especially in light of overwhelming evidence, as is the case. At this point there is no doubt that coal is being substituted by natural gas due to natural gas generators being dispatched ahead of coal-fired plants. Over time coal might even be a good investment, but before that happens this issue must see some kind of resolution.

Disclosure: I am short AMZN.