For SmartPlanet this week, I took a close look at a new report from Citigroup claiming that shale oil production had “buried” the peak oil hypothesis and that the U.S. is on its way to energy independence this decade. Instead, I found that the story of energy independence is just that: a story. The data tells us that we are losing the race against depletion, oil shortages are in our near future, and we had better plan as individuals to confront the impending oil shocks. Read it here: Energy independence, or impending oil shocks?

Gasoline is back over $4 a gallon here in California and U.S. crude is holding firm over $100 a barrel, even though domestic demand is off more than 8 percent from the 2005 peak on an annual basis, and a whopping 16 percent on a monthly peak-to-trough basis. And that can only mean one thing: We’re entering the silly season in oil punditry.

A deluge of recent articles have asserted that the U.S. is on its way to energy independence thanks to the miracle of shale oil, or “tight oil.” (Shale oil is actual crude oil produced from tight shale formations like the Bakken. Tight oil is a broader term including shale oil and natural gas liquids produced from shale gas plays. Horizontal drilling and “fracking” are used in both kinds of shale production.) None of them, however, have demonstrated how we would get there.

An amateurish report from Citigroup last week entitled “Resurging North American Oil Production and the Death of the Peak Oil Hypothesis,” garnered the most attention, perhaps because it claimed that the U.S. could achieve energy independence this decade. Here’s what they called their “back-of-the-envelope” calculation:

U.S. crude and product imports are now about 11 million barrels a day, with about 3 million barrels a day of product exports. This leaves import reliance at 8 million barrels a day. If shale oil grows by 2 million barrels a day, which we think is conservative, and California adds its 1 million barrels a day to the Gulf of Mexico’s 2 million barrels a day, we reduce import reliance to 3 million barrels a day. Canadian production is expected to rise by 1.6 million barrels a day by 2020, and much of this will effectively be stranded in North America, and there is the potential to cut demand both through conservation and a shift in transportation demand to natural gas by at least 1 million barrels a day and by some calculations by 2 million barrels a day.

Presto!

Naturally, most of the subsequent coverage of the report just repeated and amplified these extremely dubious claims under headlines blaring the dawn of our new energy independence, but a few energy journalists offered more balanced and skeptical views, notably Steve LeVine in Foreign Policy, Brad Plumer in the Washington Post, and James Herron in the Wall Street Journal.

But as ever, the task of digging up the hard data and examining these claims closely fell to yours truly.

The tight oil treadmill

The poster child for the tight oil ‘miracle’ is the Bakken play centered in North Dakota. Its production has gone from almost nothing a few years ago to over half a million barrels per day. We’re getting another 0.083* mbpd from the Eagle Ford Shale according to the Texas Railroad Commission, the state’s official oil-reporting agency. But that’s where the data trail runs dry. As I have reported previously, data from state agencies outside Texas is spotty at best, and the EIA does not offer detailed production data for tight oil at all. The EIA estimates overall U.S. tight oil production in 2011 at 0.54 mbpd, but I have seen other estimates (without the data to prove it) as high as 0.9 mbpd.

When the data runs dry, my go-to guy is French petroleum engineer Jean Laherrère, who maintains a detailed database from both public and private sources. He offered this chart of Bakken production in North Dakota:

It tells the real story of tight oil production beautifully. Each well produces a mere 150 barrels or so per day on average, and like shale gas wells, their output declines rapidly after initial production. As LeVine learned from a Bakken executive, the decline rate can be over 90 percent in the first year, then gradually tapers off. After seven or eight years, wells will have produced over 60 percent of their recoverable reserves. Therefore, you have to keep drilling like hell just to maintain production, and drill even more to increase it. Per LeVine’s source, “if the rate of drilling stays constant for a long time, the growth rate of field production will decrease, then plateau, then begin to drop.” But at around $7 million per well, these wells are not cheap.

As Laherrère’s chart shows, it takes about 1,200 wells to increase production by 150 thousand barrels a day on the Bakken tight oil treadmill. Compare that to the deepwater Gulf of Mexico, where a single gusher can produce 250 thousand barrels per day. By this metric it would take another 16,000 Bakken wells to achieve Citigroup’s projection of an additional 2 mbpd from shale oil, or five times the existing 3,200 Bakken wells.

Initial production rates from the next-biggest shale oil producer, the Eagle Ford play, appear to be substantially higher at around 350 barrels per day (including both oil and gas) over the first month, but then they decline to less than 100 barrels per day over the first year. The costs in the Eagle Ford are also substantially higher: as much as $10 million per well. Fracking the Eagle Ford is also challenged by the enormous requirements for water, an increasingly scarce resource in drought-ravaged Texas.

Elm Coulee: A cautionary tale

To see what happens to a shale oil play over time, we can look at one of the older portions of the Bakken. The Elm Coulee field in Montana, one of the Bakken “sweet spots” and the first commercially successful Bakken field in the entire Williston Basin, was discovered in 2000 and quickly ramped up to become the highest-producing onshore field in the Lower 48 by 2006. But after about five years of drilling, it was pretty well tapped out and the rigs moved on. It gave Montana a quick bump, then production fell off rapidly. Laherrère finds that Montana production is now falling by 1 percent per month.

The Saskatchewan portion of the Bakken likewise peaked in 2008. And although the North Dakota portion of the Bakken is far larger, it will eventually follow suit. A ten-fold increase in rigs since 2001 made North Dakota the envy of the country, but that can’t go on forever.

It takes an enormous leap of faith to see shale oil production rising another 2 mbpd from here, along with several leaps of logic, which the Citigroup report had in abundance. The EIA projects that all onshore tight oil production will only rise a little, to a total 1.3 mbpd by 2030, or about one-third the growth that Citigroup forecasts by 2020.

The narrow ledge of tight oil

In short, increasing our already-frenetic rate of drilling for tight oil requires sustained high oil prices. At today’s $105 a barrel for West Texas Intermediate, that’s no problem. But if prices were to fall to $70 a barrel, LeVine’s source says, drilling in the Bakken would become unprofitable and would cease, causing production to fall rapidly.

At the same time, the last few years have shown us that $100 oil translates to roughly $4 gasoline, and that’s the pain tolerance limit for most of America. Gasoline demand in the U.S. tends to fall off beyond $4, as does economic activity in general.

This is what analyst Steven Kopits of Douglas-Westwood call the “narrow ledge” of oil prices, as I detailed in 2009. Given the extremely volatile global marketplace for oil, influenced by everything from geopolitical aggression, to climate change, to the headline risk of Greece defaulting and being forced out of the Eurozone, it will be very difficult for the oil industry to cling to that ledge for a sustained decade or more as the Citi analysts breezily project. They simply wave away cost inflation, and don’t acknowledge a price ceiling at all.

The narrow ledge isn’t a problem for the surging developing economies of the world, however. China alone has replaced all of the oil demand lost in the U.S., and more, and it’s still growing fast with an 8.2 percent growth rate projected for this year by the IMF.

Want your blankey?

This point was highlighted in a fascinating debate last week, which I highly encourage you to watch, between former Shell Oil president John Hofmeister and Tad Patzek, professor and chairman of the Department of Petroleum and Geosystems Engineering at the University of Texas at Austin. China, Hofmeister noted, is “on a journey, ladies and gentlemen, to go from 9 mbpd consumption today—a year ago it was 8, now it’s 9—to 15 mbpd by 2015. India from 4 to 7 mbpd in the same time frame. There’s not enough oil out there to meet that demand.”

These nations use oil far more efficiently than we do, and derive far more economic value from it. As I like to say, picture three guys and a chicken burning one gallon of gas on a scooter to go to the market each day in India, versus a soccer mom in the American suburbs burning three gallons a day to run errands without generating any economic gain. Asia can outbid the West for the declining available net exports for a long, long time, and that demand will continue to pull prices above our pain threshold.

Hofmeister waxed apoplectic in frustration over U.S. policymakers’ failure, for four decades straight, to do something about our oil dependency. “We are in an oil shock, and we are facing impending shortages,” he warned, and went on to recall what those shortages were like here in 1973-4. His solution? To create a federal board that would assume control of the nation’s energy strategy—a proposal that seems not altogether crazy to me, but also not likely to succeed in America’s dysfunctional politics—and command greater domestic production of oil and gas in the short term, while plotting a transition to renewables.

Patzek’s prescription was far more pragmatic, and aligns closely with my own: “Don’t wait on government programs; don’t wait on people to tell you what to do; start to insulate yourself from these shocks.” As I wrote last month, the revolution will be bottom-up. “Try to think about simple things, simple steps that you can do…to insulate yourself as much as you can from the upcoming shocks. They’re coming! So take it for granted, so no matter what Daniel Yergin and others will tell you.”

Patzek acknowledges that this message is a tough sell in America. Forecasts like Citigroup’s aren’t based in reality, but in politics. As another perspicacious friend of mine, who has been looking at such rosy forecasts for decades, remarked to me this week, “the masses (and the cheerleaders) love this story because it is one of Abundance and Manifest Destiny in this Exceptional Country of ours.” We like supply-side solutions. We don’t like anybody telling us that we need to cut our consumption.

Patzek muses: “If I think about the United States, this great, magnificent country of ours, I think about a grown-up baby, who now is very large, and on a cold night wants to cover herself with a baby blankey. And no matter what she tries, there is always a part of her body that is uncovered and exposed to the cold weather. That’s where we are.”

That’s what the Citigroup report was: a blankey. It was obfuscation by oblation, the offering of a cargo cult that desperately wants to bring back the good ol’ days of US production. They discussed the decline rate of mature fields—about 3.7 mbpd of lost production each year—at some length, then failed to do the simple addition to balance that against their production outlook, instead wandering off into a long musing about capital efficiency. Several of their charts were clearly wrong, like the production levels of the Bakken and Eagle Ford shown in Figure 2, and 80,000 barrels per day per Bakken well in Figure 26 instead of 150. They plotted production separately from well or rig count, obscuring their intimate relationship. I could go on for several thousand words, but I’ll spare you.

If the Citigroup forecast were serious, it would have shown how many rigs, and where, it would take to generate another 2 mbpd from shale oil, and discussed a price scenario that would foster continued drilling for another eight years. Instead, they buried the reader in 18 pages of mostly irrelevant data about reserves and formation characteristics, and tossed off an optimistic production forecast based on little more than hand-waving, along with the obligatory jab about Thomas Malthus. It was an excellent piece of political posturing, but I wouldn’t give a plugged nickel for their forecast.

Now look at the chart I posted at the top. All the talk of incipient U.S. energy independence is based on a mere 15 percent production increase in 2011 over 2008, which required thousands of wells and great expense. Retaking our 1970 production peak of 9.6 mbpd with such expensive and low-output wells, as Citigroup suggests, is a fantasy. Production from the Canadian tar sands will probably grow a bit over the next decade, but doubling its production is unlikely. Let’s remember that tar sands production was projected to grow from 1 mbpd in 2006 to 2.8 mbpd in 2012, but actual production is currently just 1.6 mbpd; this is a direct consequence of the narrow ledge getting narrower. California’s Monterey Shale will not be exploited any time soon. Another 2 mbpd from the Gulf of Mexico would take a truly Herculean effort, at great expense, with the risk of a complete ban on drilling there if another blowout happened.

Meanwhile, the treadmill continues to speed up. As Patzek notes, incremental new oil production per rig in the U.S. was about 1,000 to 1,500 barrels per day in 2005. In 2011, it was one-tenth that, at 100 barrels. The story of energy independence is just that: a story. The data tells us that we are losing the race against depletion, oil shortages are in our near future, and we had better plan as individuals to confront the impending oil shocks.

Credit: Top chart from EIA, marked up by Chris Nelder

* Correction, March 13, 2012. I slipped a decimal point in the original version of this article, citing Eagle Ford production at 0.66 mbpd; it should have been 0.066 mbpd, based on the January-November 2011 data the TRRC had available at press time. They have now updated their data to reflect full 2011 production, which works out to 0.083 mbpd.