Saturday, March 17, 2012

22 December 2010 – Piqued oil and the 10-year rule


Having touched on some energy-related subjects over the past few months (and for that matter, over the past few years), articles discussing conventional energy reserves tend to catch my eye.  A few such popped up over the last two weeks.  Lexaria, an oil company in Kentucky, announced that it had increased its estimate of proved reserves by 81%.  North Dakota oil production has quintupled over the past five years.  Syria, Ghana and Sudan have all expanded their estimates of proved reserves.  Notwithstanding temporary recession-induced perturbations in consumption patterns, we're using more fossil fuel than ever - that's indisputable.  And yet, proved reserves never seem to decrease.

You have to wonder whether this is a perception, or if it's reality.  After all, the Earth is a finite oblate spheroid, and most of its mass consists of things other than fossil fuels.  Some day, unless we figure out how to transmute silicon into carbon, we're going to run out; that's mathematically undeniable.  The big question is when.  And there are a lot of different answers out there.

Figure 1 - US Natural Gas reserves, 1979-2009 (trillions of cubic feet)
(Source: Energy Information Administration, “Summary: U.S. Crude Oil, Natural Gas, and Natural Gas Liquids Proved Reserves 2009” (Washington, D.C.: Energy Information Administration, November 2010), figure 1, page 3.)
Well, who keeps track of this stuff?  For the US, of course, it's the Energy Information Administration (EIA).  Last month, the EIA published its annual report on the state of proved crude oil, natural gas and natural gas liquids reserves in the US at the end of 2009.  The results, taken from the summary, are fascinating.  According to the report, proved reserves of natural gas increased by more than 11% from 2008 figures, from 255 to 284 trillion cubic feet; while proved reserves of crude oil increased by 8.6%, from  20.6 to 22.3 billion barrels.  You have to think about the dynamic to appreciate it; every year, the US consumes more and more of this stuff - and yet, every year, the total amount left continues to climb.

Let's try to put this in some sort of consumption context, shall we?  According to the EIA energy production reports for 2009, the history of US natural gas production looks something like this:

Figure 2 - US Natural Gas Production, 1949-2009 (billions of BTUs)
(Source: Chart by author, data from US EIA)

That’s in billions of British Thermal Units (GBTU).  As a rule of thumb, depending on its composition the energy content of natural gas is about 1000 BTU per cubic foot.  So, for example, the 22,000,000 GBTU of gas that the US produced in 2009 translates to 20 trillion cubic feet of gas (or 566 billion cubic metres).

Looking at figure 1 above, in 1979 the US had 210 trillion cubic feet of proven gas reserves.  At 1979 production rates, in other words, the US had 10 years of natural gas left.  Let's try this in decennial snapshots, shall we?

1979: Production rate: 20 TCuFt/year.  Reserves: 210 TCuFt 
- Amount of natural gas left in the US?  10 years.

1989: Production rate: 17.5 TCuFt/year.  Reserves: 175 TCuFt
- Amount of natural gas left in the US?  10 years.

1999: Production rate: 19 TCuFt/Year.  Reserves: 190 TCuFt
- Amount of natural gas left in the US?  10 years.

2009: Production rate: 22 TCuFt/Year.  Reserves: 285 TCuFt
- Amount of natural gas left in the US? 13 years.

Hmm.  That's interesting, non?  For the past 40 years, the US, at contemporary production rates, has always had about 10 years' worth of natural gas left.

That leads to some interesting aggregate statistics.  For example, in 1979, as noted above, the total US proved reserves of natural gas were about 210 TCuFt.  Since then, the US has produced 589 TCuFt of natural gas - roughly three times as much as the US had in proved reserves in 1979.

Companies are getting better at finding it - the biggest increase in US natural gas reserves in recent years has come from shale gas deposits in Alaska, Arkansas, Louisiana, Oklahoma, Pennsylvania and Texas - and they're getting better at extracting it.  Shale gas reserves in the US have gone from 23.3 trillion cubic feet in 2007, to 34.4 TCuFt in 2008, to 60.6 TCuFt in 2009.  Extraction technology is improving so much that no matter how much natural gas America produces from domestic sources, there always seems to be 10 years' worth of it left.

Well, does the same peculiar situation apply only to natural gas?  Here's the EIA chart on US oil reserves:

Figure 3 - US proved crude oil reserves, 1979-2009 (billions of barrels)
(Source: Energy Information Administration, “Summary: U.S. Crude Oil, Natural Gas, and Natural Gas Liquids Proved Reserves 2009” (Washington, D.C.: Energy Information Administration, November 2010), figure 7, page 8.)

And here are the EIA figures for US crude oil production over the past 60 years:

Figure 4 - US Crude Oil Production, 1949-2009 (billions of BTUs)
(Source: Chart by author, data from US EIA)

First of all, did you notice how the gas reserves and gas production curves (from figures 1 and 2) for 1979-2009 both trend upwards?  While the oil reserves and oil production curves (from figures 3 and 4) for the same period both trend downwards?  Is there a causal correlation between reserves and production?  And more importantly, if there is a causal correlation, which way does the arrow of causation point - does change in production cause change in reserves?  Or vice-versa?

Back to the oil charts.  A barrel of oil has an energy content of about 5,600,000 BTUs.  The 18 quintillion BTUs worth of crude oil that the US produced in 1979 therefore translates to 3.21 billion barrels of oil.

Taking a look at the same ratio relationship as before, here's what we find:

1979: Production rate: 3.21 Gbbls/year.  Reserves: 32 Gbbls
- Amount of oil left in the US?  10 years.

1989: Production rate: 2.85 Gbbls/year.  Reserves: 27 Gbbls
- Amount of oil left in the US?  9.5 years.

1999: Production rate: 2.4 Gbbls/year.  Reserves: 24 Gbbls
- Amount of oil left in the US?  10 years.

2009: Production rate: 2.05 Gbbls/year.  Reserves: 22.5 Gbbls
- Amount of oil left in the US?  11 years.

In 1979, the proved reserves of crude oil in the US amounted to 32 billion barrels.  Since 1979, the US has produced 81.26 billion barrels of oil from domestic sources - roughly three times as much as the US had in proved reserves in 1979.

You have to admit, that's a little peculiar. about coal?  The statistical data is a little hard to - you'll forgive the expression - "mine" from the EIA site, principally because it's archived and reported differently.  The data on US domestic coal reserves only go back 10 years, and the "world statistics" that the EIA site offers are only for 2005.  The coal supply, demand and pricing reports likewise go back only 10 years.  The question, though is this: where coal is concerned, does it really matter?  For the past decade, the US has been consuming about a billion tons of coal a year, nearly all of it for electrical generation.  The EIA estimates US recoverable reserves at 261 billion tons, from a reserve base of 488 billion tons.  In other words, using current technology only, at present consumption rates, the US has more than 250 years' worth of recoverable coal, and if technology improves only slightly, nearly 500 years' worth of recoverable coal.

Kind of makes you wonder why coal is treated so differently from gas and oil.  There are a lot of reasons, but from the perspective of figuring out why there is so much less uncertainty in the coal production-to-reserve ratio than there is in the case of gas and oil, the most important characteristic is production cost.

Figure 5 - Cot of producing energy in the US, 1949-2009 (dollars/MBTUs)
(Source: Chart by author, data from US EIA)

You see, over the past 60 years the cost of producing coal has been remarkably constant.  The coal is there, there's a lot it, and while it's not precisely easy to exploit, extracting coal from the ground is technologically very straight-forward.  The cost of production experienced a little bump in the early 1970s, coinciding with the oil shock, but since then it's been relatively constant.  There are two interesting features in this chart.  One is the incredible "bumpiness" of the production costs figures for oil and gas, but NOT for coal, after 1973; the other is the incredible stability and predictability of the cost of producing energy from ALL sources in the US up until the oil shock.  Monopoly is part of the answer; the formation of the OPEC cartel allowed producers rather than consumers to dictate oil prices, and this fact, coupled with stagflation during the Carter Administration and the recession of the early 1980s, is commonly thought to have been the key contributor to the the massive spike in production costs.  But that can't be the whole story, can it?  After all, stagflation and the recession affected the whole of the US economy.  Why would the costs of natural gas and oil production have been so heavily impacted, without any significant or enduring impact on the cost of coal production?

The answer is probably made up of a lot of complicated and interrelated factors, but one of the major influences is probably speculation.  Oil and natural gas, being highly fungible and allegedly scarce commodities, make good instruments for futures trading.  Commodity prices are especially responsive to perceptions of scarcity; remember how we all spent several weeks paying $1.25 for gas after Hurricane Katrina?  According to EIA statistics there was no noticeable decline in the production of motor gasoline in 2005 [Note B].  The fear that the hurricane would adversely impact US refinery capacity was unfounded - and yet it formed the basis for frantic speculation that led (again, according to the EIA) to much higher pump-price increases in 2005 (a 22% increase over 2004 prices) compared to 2004 (17% increase over 2003) and 2006 (13% increase over 2005).  Futures trading and speculation driven by perceptions of scarcity in a monopolistic environment are one of the principle reasons that the price of a barrel of oil fluctuates so much.  They're also the reason, incidentally, that diamonds are so expensive, as well as the reason that tulip bulbs were so pricey for a very brief period in 1638, and real estate so costly in New York in the early 1980s.

So whenever somebody talks about oil shortages, it's worth remembering that over the past 40 years the US has produced nearly three times as much oil and gas as it thought it had 40 years ago.  When folks talk about there only being "10 years of oil or gas left", remember that they've been saying that since Nixon was in the White House.  Tomorrow - per the dulcet warblings of Little Orphan Annie - is always a day away.

And when people prognosticate on the topic of "energy shortages", ask to see their figures; and remember that the US could quadruple its electrical generating capacity using coal alone, and still have enough coal to last for the next century.

As for uranium (i.e., uranium oxide ore)...well, global proved reserves are 5,404,000 tonnes against consumption of 68,000 tonnes/year, or an 80 years' supply at current consumption rates.  And reserves continue to increase - not surprisingly - proportional to the cost of uranium per kilogram and the annual aggregate expenditure on exploration.

In fact, this chart suggests an answer to the question I posed earlier on in this message - whether production influences reserves, or reserves influence production.  This chart suggests that more exploratory funds lead to more proved reserves - hardly a surprising conclusion, that - while sharp increases in funding for exploration lead to equally sharp increases (see the post-2007 period) in the amount of proven reserves represented by very high-quality uranium ore.  So the arrow of causation, in short, appears to point both ways: the discovery of more reserves provides more incentive for exploration and R&D into improved exploitation technologies, and more exploration and exploitation leads in turn to more reserves.

As for the matter of the 80 years' supply, it would seem that uranium - like coal - isn't yet scarce enough, either in fact or in perception, to prompt the kind of speculation and futures trading that leads to the profits, and the consequent wild price swings, that we see in the oil and natural gas markets.  Maybe Canada should team up with Australia, Russia and Kazakhstan to form a cartel - the Organization of Uranium-Exporting Countries (OUEC), perhaps - that would control 60%+ of the world's supply of uranium, and thus would be able to control production, fix prices, generate perceived or actual shortages, cause panics and bubble behaviour, and prompt floods of nervous articles about "peak uranium".

Maybe that will happen anyway.  If there were eight times as many reactors operating, industry analysts would be able to claim with a straight face that "the world only has 10 years' worth of uranium left."  It would be interesting to see whether the same dynamic that has characterized the oil and gas industries for the past 40 years would take over the nuclear industry as well, resulting in a dynamic where ever-increasing consumption results in ever-increasing reserves, while mankind spends the next fifty years paralyzed with worry over the fact that the uranium is always ten years from running out...despite the fact that, for some inexplicable reason, it never does.

Meanwhile, with the holidays upon us, I'll be doing my part to provoke an up-tick in consumption of three endlessly renewable classical carbohydrate energy sources: ethanol, sugar, and wood.

Merry Christmas.



The 2009 oil and gas report from the EIA is available here.