One of the questions that plagues me constantly is, “Why is energy journalism so bad?” Most mainstream articles about energy will leave you horribly confused at best, or horribly misled at worst. Today I will try to teach you how to read reports on energy without getting lost.
The topic came up again last week when my colleague Christopher Mims pointed up a sharp discrepancy between three recent stories by Reuters, published between November 21 and 24.
The first reported Saudi Aramco’s CEO Khalid al-Falih as saying that unconventional oil (heavy oil, synthetic oil from shale and tar sands, coal-to-liquids and so on) had eliminated global supply concerns, and would rise from 2.3 million barrels per day (mbpd) today to 8.4 mbpd by 2035. This would shift the global balance of power, he said, and reduce U.S. dependence on oil imports. Further, he expected conventional oil supply from Brazil and Iraq to rise. All of this was by way of explaining why Saudi Arabia had recently halted its ongoing $100 billion program to expand production capacity beyond its claimed 12.5 mbpd current capacity, and would not seek to expand it to 15 mbpd (a fact that was already widely assumed by most who follow the energy markets, but apparently was still considered newsworthy).
The second said that oil prices should remain high because global demand had remained strong, and would reach more than 89 mbpd this year according to the IEA. (The International Energy Agency, or IEA, based in Paris and serving at the pleasure of the 28 industrialized countries in the OECD, currently shows 88.7 mbpd for Q3 2011 under a liberal definition of “oil” which includes biofuels and certain types of natural gas liquids. The Energy Information Administration, or EIA, based in Washington D.C. and serving under the U.S. Department of Energy, shows 86.7 mbpd under a more restrictive definition, which excludes biofuels, non-associated natural gas liquids, and other components.) But while demand has been strong, the article noted that supply has been "inconsistent," citing the loss of 1.6 mbpd from Libya, and various "hiccups in production in Russia, Britain, Norway and Nigeria."
The third suggested that high oil prices "could strangle economic hopes," and quoted the IEA’s chief economist Fatih Birol: "I hope that colleagues from the producing countries are also looking at the market indicators carefully, including the diminishing OECD stocks levels and the fragility of the global economic situation." In his view, too little worldwide investment in oil supply would keep prices high enough to stifle the recovery of the global economy.
So which is it? How can three stories from a single source, published over a five-day span, simultaneously claim that supply is adequate and inadequate; and that prices would remain high due to strong demand, but would be so high that they would destroy demand?
Even worse, how can the IEA simultaneously claim in its new World Energy Outlook 2011 report that by 2035, the world needs to invest $20 trillion in oil and gas supply and infrastructure to add 47 mpbd of new capacity (equivalent to about five times Saudi Arabia’s production, or twice the production of all OPEC countries in the Middle East) to compensate for the decline of mature fields, or else risk "far-reaching consequences for global energy markets". . . while at the same time asserting that the world must slash subsidies for oil and gas development and transition to renewables quickly in order to arrest climate change, starting now?
Welcome to my hell.
Let’s review a few of the common errors in energy journalism.
Uncritical acceptance of authority
Citing subject matter authorities is a necessary element of journalism, but so is casting a critical eye on what they say. Unfortunately, most journalists repeat what their selected authorities say verbatim, and rarely mention contrary views.
The main authority cited in nearly all articles about energy is Daniel Yergin, the Pulitzer-winning author of The Prize and an economist with the oil industry consultancy IHS CERA. Journalists love to quote Yergin. He always has an optimistic outlook on the future of oil, forecasting abundant supply and low prices. Editors love him too, and regularly grant him high-profile space in their publications (like this recent Wall Street Journal op-ed) to offer sunny pronouncements and spew invective on those who believe peak oil is a serious issue. The fact that his predictions have been continuously and badly wrong for the last decade straight doesn't seem to phase them in the least, nor does his cozy (and well-paid) business relationship with oil companies. Hey, he won a Pulitzer, and is regularly described as "one of the world’s foremost energy authorities." Good enough!
Other authorities often cited are active traders and portfolio managers in the oil market, but it’s a rare day indeed when the authors bother to ask whether those authorities might just be "talking their books."
The authority problem leaks downstream, too. When a reputable publication makes errors, it often leads to a series of repetitions by even less skilled journalists who cite the original as fact.
The problem isn't just authority, though. Dozens of serious petroleum geologists and analysts with real bona fides in petroleum forecasted the current oil supply plateau and the price volatility of oil beautifully over the last decade, while economist Yergin missed it. So why don't journalists consult their views? This brings us to the next problem.
Consider this recent example in the Financial Times. The author elatedly painted a picture of an impending era of energy independence as unconventional oil from gas shales caused a "U-turn in US oil supply," even suggesting that, "in the coming decade the US will leapfrog Saudi Arabia and Russia to become the world’s largest producer of liquid hydrocarbons."
The article went on to cite a jumble of relevant and irrelevant data, along with highly speculative forecasts represented as near-certainties. After sorting it all out, I found that the cited numbers didn’t add up: 10 mbpd currently produced by the U.S. and Canada (including about 1.5 mbpd from Canadian tar sands), plus a projected 2 mbpd of new "tight oil" production from U.S. shales, plus another 1.5 mbpd of projected new production from the tar sands gives me 13.5 mbpd, but the author slid directly from that data to a forecast by the National Petroleum Council (NPC), an oil industry group, projecting that the two countries would produce a whopping 22 mbpd by 2035!
Further sleuthing revealed the problem: the author had uncritically reported the best-case "2035 High Potential" scenario offered in the NPC’s latest report, in which production from tar sands rises to 6 mbpd; offshore Gulf of Mexico rises to 3 mbpd; "tight shale" (not to be confused with "tight oil") rises from zero today to 1 mbpd; enhanced oil recovery technology adds another 0.6 mbpd, and Arctic production rises to over 2 mbpd.
The author did not mention that the NPC report also included an alternative, "2035 Limited" scenario, in which North American production actually falls about 1 mbpd from 2010 levels.
Now, I have a certain amount of sympathy for the author. The NPC report, like most such documents, was a real slog: 155 pages of extremely overwrought, redundant text, with conflicting data across multiple scenarios. The sad fact is that most journalists don’t have the time to really absorb a report like that, let alone offer a critical view of it in a world where getting 1,800 words out within a day of the report's release is considered far more important than accuracy or context. It's all about grabbing those page views before someone else does, and if the errors are criticized somewhere in the blogosphere, nobody important will see it anyway. Another unsurprising fact is that journalists are, well, journalists. They're not scientists, and most of them have no background in energy, even at publications like Scientific American. The best they can do is to rewrite the summary or the press release, and they will remain blissfully unaware if a politically-minded editor writing the summary deliberately misrepresents what the scientists who wrote the report actually said.
But a simple reality check should give any author pause before suggesting that some magical array of technologies will somehow double North American oil production over the next two decades, or that within the next decade unconventional supply in the U.S. and Canada, with a production cost in the range of $80 to $90 a barrel, will exceed supply in Saudi Arabia or Russia, where their production costs are half of that or less. And it doesn't seem too much to ask that journalists bring a modicum of skepticism to their coverage of oil industry propaganda.
Let’s have another look at the chart of historical U.S. oil production, freely available on the EIA's web site for anyone who cares to look at it:
That little bump at the end is what all the fuss over unconventional oil in the U.S. is about. I suppose if one ignores the production costs and flow rates and takes plenty of antidepressants, one could imagine that bump is a U-turn that could eventually send U.S. production back up and over its 1970 peak. But I assure you that once you spend a few thousand hours gaining literacy in the subject, such that you can read and understand these scenarios, such optimistic visions begin to sound more like whistling past the graveyard than serious forecasting.
Just remember this: In the eyes of most editors, an optimistic take on future supply is just good energy journalism. And a balanced, nuanced article with indeterminate conclusions doesn't sell papers. But a pessimistic take (no matter how true, or buttressed by facts) is editorializing, which is bad.
Another particularly painful problem is the inability of writers and editors to do a little simple arithmetic to see if their own numbers add up, or if the data they’re citing is current. Far more often than not, they don’t, and it isn’t.
Without belaboring the point, I’ll just offer one recent example from the blog of an energy reporter at the New York Times. As you read it, try to do the math in your head:
He gives the example of pumping water from a pipe; if the power plant that supplies the electricity starts with 100 units of energy, it will lose two-thirds of that in making the current and another 10 percent in transmission and distribution. The motor will be only 90 percent efficient; so pumps, motors, drive trains and throttling valves along the way will lose more, leaving the plant with 93 units of energy.
I don’t know why this happens as often as it does, but I’ll be charitable and speculate that perhaps everybody is just working too fast to bother checking the math.
Missing the fine print
Anyone who has ever signed a contract knows better than to sign before reading the fine print. Yet energy journalists appear to do so regularly.
By now you've certainly heard that hydrofracking has unlocked "vast" deposits of shale gas in the U.S., and heralded a new "golden age for gas" (as the IEA recently put it). Based on the reports you've seen, you probably think that the U.S. now has a 100-year supply of gas, which will enable us to convert a significant portion of our coal-fired power generation to gas, along with a few million 18-wheelers.
The problem is: neither claim is quite certain, or true.
First, it should be obvious that if you transfer large loads to natural gas, a 100-year supply at current rates is no longer a 100-year supply. It might be a 10 or 20-year supply. Oddly, I don't think I've ever seen that mentioned in a single gushing article about shale gas. Again, the math isn't hard, and the data is easy to come by. Why doesn't anyone ever bother to do it?
What's worse are the claims about the size of shale gas resources. I have seen numerous references to the EIA's assessment of the prospects for shale gas, including their note that "from 2006 to 2010 U.S. shale gas production grew by an average of 48 percent per year." Most articles also cite the next sentence, projecting a three-fold increase in shale gas production from 2009 to 2035, but fail to examine the highly speculative details of the Reference case from which it comes. But what about the sentence after that? "However, there is a high degree of uncertainty around the projection, starting with the estimated size of the technically recoverable shale gas resource." Or the following paragraph: "the estimates embody many assumptions that might prove to be untrue in the long term?" Or the paragraph after that: "Moreover, across a single shale formation, there are significant variations. . . and as a result production rates for different wells in the same formation can vary by as much as a factor of 10?"
Then there are the really nitty gritty details. Many journalists completely miss crucial differences between terms like resources (how much of the stuff exists underground), reserves (the portion of the resources that are considered technically recoverable using today's technology), and economically recoverable reserves (what portion of reserves can be produced at a profit). This leads to reports claiming that the U.S. has 1,230 trillion cubic feet of gas yet to burn, when the EIA clearly states that that number applies to the "estimated unproved technically recoverable resource base," while in their Reference case the number drops to 827 trillion cubic feet, and in their Low case it drops to 423 trillion cubic feet — one-fourth the headline number.
By 2035, the EIA projects production under the Low case to be 22.4 trillion cubic feet per year, 7.7 trillion cubic feet lower than under their High case, and what's more, that the U.S. will have once again become a net natural gas importer. Whether that gas will be economically recoverable, or whether the projected production rates might be off by a factor of 10, are questions never addressed in the press.
How to read energy journalism
I realize that this all may seem a bit wonky and hard to understand, so I'll finish with a few words to the wise for the unwary reader.
1. Be skeptical. You will have to make up for the missing skepticism and curiosity of the journalists you're reading. If the article is all sunshine and roses, and includes no caveats or alternative views, it will be more useful to you as fishwrap than information.
2. Discount the sources. If the cited authority represents the oil and gas industry, you should view their forecasts as propaganda, not truth. Particularly when the authority is from an OPEC producer. OPEC (like the IEA) is a fundamentally political organization, and everything they say in public has a political calculus behind it. For example, I read the unconventional oil optimism expressed by the Saudi official cited at the top of this piece as their way of jawboning down peak oil fears, and throwing analysts off the scent of a trail which leads to serious questions about whether Aramco can increase spare production capacity, and whether the world's most productive oil field, Ghawar, has indeed gone into decline.
3. Do the math. If the numbers cited don't add up, then you would be wise to question the validity of what you're reading. Most of the time it's simple arithmetic you can do in your head. More ambitious readers will want to bust out a spreadsheet and have a go at the details.
4. Look for context. If the article only talks about resources or reserves, and doesn't mention production rates, you can safely ignore it. Yes, America may have 1.5 trillion barrels of oil shale (not shale oil, which again is an entirely different thing), but right now we're producing exactly zero barrels of it, and for good reason: it's a highly marginal source of hydrocarbons, and too expensive to produce with today's technology. Remember this: Only flow rates matter, not how much is in the ground.
5. Look up the references. If you really want to know how valid the forecasts are, look up the original sources. Nearly everything is available for free on the Web (except for IHS CERA's data, naturally), and it's not hard to find. The crucial caveats are usually stated somewhere in the documents, but you might have to dig for them, and the Find function (Ctrl-F) is your friend when plowing through a 350-page PDF file.
6. Compare to reality. If a claim sounds outlandish, it probably is. Go have look at the data on the EIA's Web site, and if you're a beginner, start with their Energy Explained site. See if the forecast looks remotely close to reality. Caveat emptor.
Now, I don't mean to be unduly harsh on my fellow energy journalists. I'm sure most of them are serious, nice people who are pressed to do a difficult job under tight deadlines. I don't think most of them intend to mislead or confuse their readers. And no doubt some of them labor under political editorial agendas that may contradict their findings, so they do what they must to keep their jobs. Nor do I mean to pick on these particular journalists; I could have made the same points with thousands of recent examples.
But we’ve still got a long way to go on energy literacy, and informed readers can push journalists and editors to do better work. I hope these clues have equipped a few readers to be more discerning, and less gullible.
And speaking of gullible, ever hear that rip-snorter of a story about Gull Island? Oh, nevermind. . .
Hat tip to energy economist Andrew Leach for his sharp eye on the IEA WEO.
Photo: credit (sfllaw/Flickr)