2010 Oil Story: Drawing Down the Inventories

Hat tip to the Economist Magazine for catching a key, energy data point from week’s BP Statistical Review: in 2010, the world consumed about 5 mbpd (million barrels per day) more oil than it produced. Anticipating the discrepancy between the two figures, the BP Statistical Review authors write on page 9 of their PDF: Differences between these world consumption figures and world production statistics are accounted for by stock changes, consumption of non-petroleum additives and substitute fuels, and unavoidable disparities in the definition, measurement or conversion of oil supply and demand data.

I agree with the Economist Magazine’s conclusion that most of the difference between 82.095 mbpd produced and 87.382 mbpd consumed came from stock changes. In other words: the drawing down of oil from inventories. Indeed, this was a key driver for oil’s advance last year from the high $70’s to over $90 a barrel. In the chart below, from the latest IEA Paris Oil Market Report, you can see that starting in mid-year, total OECD inventories started a new decline. Moreover, the histograms in the below chart also show the difference to the five year average, which also illustrates the global stocks drawdown. This coincided by the way with a resurgent, mid-year advance in the price of oil from a low of $69 to $92 by year end.

A data point of interest that the Economist did not mention, however, is that BP’s oil consumption data include global biofuels, while BP’s oil production data do not. Does the inclusion of biofuels in global consumption skew this figure meaningfully higher? My calculations show the following: No, it does not. Switching to BP’s preferred Mtoe energy unit (million tonnes oil equivalent), I find that total biofuels production was only 59.26 Mtoe in 2010. Compared to 4,028.10 Mtoe of oil consumed, there is at best 1.47% biofuel consumption embedded in BP’s total consumption figures for oil. Thus, the story of 2010–and it’s a very important story indeed–is intact. Unable to meaningfully increase global oil production to meet demand, the world ate through inventories. You have been warned.


Trading Note: There was some market chatter today that a concerted release of oil from OECD inventories could be considered to drop prices from current levels. But you can see the problem: if markets were already so imbalanced in 2010 that consumption outran production by 6.4%, then a discretionary release of oil from OECD stocks will only force the global futures market to price in less of a safety cushion in the event of war, or some other serious disruption to a major supplier like Russia or Saudi Arabia. Inventories are meant to manage disruption in flows. Not the inconvenience of “high prices.”

Further Research: Nota Bene that BP’s Oil category includes natural gas liquids (NGL’s). NGL’s are not oil. Not in any sense. Furthermore, NGL’s only contain 65% of the BTU of oil. In recent years, as crude oil production has oscillated below a ceiling, there has been growth in NGL supply. Aggregating NGL supply into “oil” categories is misleading to say the least. For a broader discussion of this accounting issue, see my essay at The Oil Drum: Secrecy By Complexity: Obfuscation in Energy Data, and The Primacy of Crude Oil.

Update, Saturday 11 June 2011: There has been considerable commentary and reaction to the BP data, as interpreted both here and at the Economist Magazine. In short, it is perhaps a more complex calculation to determine the components which cause the difference between Consumption and Production as reported by BP. One of the reasons that Gregor.us tries to deal almost exclusively with Crude Oil only–and not Liquids such as biofuels, NGL’s–is that the latter create enormous complexity on many levels, not least of which are the myriad energy inputs to creating Liquids. Moreover, both EIA Washington and IEA Paris will often include, and then disclude, various Liquids from various accountings of “oil.” This is why in the original post here, I pointed readers to some of these issues, and my Oil Drum post from earlier this year. The bottom line however is that BP is clearly trying to capture the difference between global crude oil production–and–the global demand for liquids. That still strikes me as a worthy effort–even if in my own work I try to work exclusively with crude oil only. In addition, there is absolutely a correlation between the price of oil and the ROC (rate of change) and the direction of changes to global inventories. Observations of OECD inventory changes starting in mid 2009 showed a directional drawdown which coincided with the movement in the price of oil. For context, US oil stocks form only a small portion of OECD stocks, and OECD stocks are of course themselves only a part of global stocks. In conclusion, while it remains difficult to quantify the entire inventory drawdown in 2010 on a global basis, I think most data available points to that phenomenon as having taken place, and, as having influenced the price of oil rather predictably.

Further Reading:

OECD oil stocks data offer limited insight, March 2011, Financial Times, Javier Blas.

Global Deficit Between Oil Consumption and Production Remains the Norm, June 2011, Rigzone, Trey Cowan

  • The inclusion of biofuels — don’t forget refinery gains — is a good narrative. It suggests that more innovations are right around the corner, all we need to do is click those ruby slippers — and trust the big energy companies such as ‘The New’ BP..

    Reality can and will take care of itself. Those lacking the means will simply not consume. This is the solution to any and all energy ‘problems’

    Anyone can tell a self-deprecatory joke: it takes real talent to put the joke on someone else! Here is the reason to love our moderne, liberal society, its predatory design is now revealed where all can see it.

  • Brian Ledell

    Hello Gregor,

    I am having trouble reconciling the graph you posted, “Closing Industry Stocks of Oil,” to the 5 mbpd figure.  A 5 mbpd shortage in 2010 equates to 1825 mbp-year, but the graph only shown a measly peak-to-trough difference of 100 bp.   Shouldn’t it be a lot larger if most of the 5 mbpd shortfall was through inventory depletion?

  • gregor.us

    Yes, I understand. There are deeper layers of complexity here. But the short answer is that I used the best graphic I can obtain freely to illustrate the direction of global inventory drawdowns last year.

    Here are some of the complications:

    1. BP is reporting on global consumption, production, and the implied global drawdown.
    2. IEA Paris is just reporting on OECD inventory drawdowns.
    3. BP is reporting all consumption, including NGL’s–and including NGLs in their production.
    4. IEA Paris is only reporting crude oil products as part of oil products that form the total oil figure.
    5. While IEA Paris also reports all liquids in production figures, it’s not clear that they report biofuels of NGLs as part of their Total Oil Stocks.

    If I can ever obtain non-OECD stock drawdown data, I will do a second post. I would also mention that in general, BP is attempting very difficult work in reporting non-OECD consumption, and the implied stock drawdown. However, as I followed the story all last year of OECD stock drawdowns, and also the no-data reported stock drawdowns in non-OECD, I still generally conclude the implied stock drawdown reported by BP is “true.”

    In general, most of this work involves cobbling together different data from different sources. Revisions year to year can be significant. But as I said, I am taking the implied stock drawdown from BP seriously and I find that the OECD data confirms the direction in 2010.



  • Gregor, 

    We have had 17 year highs in stockpiles – supplies spilling over the brim — why would drawing down of those supplies (now still above the 5-year avg.) cause you to ring warning alarms?  Inventories have NOTHING to do with supply or spare capacity – and even less to do with price.   

    Storage of oil should always be measured by WHAT”S IN THE GROUND and ACCESSIBLE.

    Not rigorous work and incendiary.  

  • gregor.us

    Hi. Are you referring to US inventories? If you are, could you give me your view about the proportion of US inventories to both OECD inventories, and global inventories.



  • I was referring to US inventories in quoting vs. the 17 year highs, but your chart refers to global inventories in a similar – and definitely misdirecting way….by only going back to 3/09, also (coincidentally?) the bottom of the equity and oil markets —   Any historical context to either OECD and non-OECD will show stockpiles vs. the five year avg. to rotate both at positive and negative levels (that’s what average implies, after all), without much if ANY connection to price.  Between 1980 and 2000, price basically held between $16 and $24 a barrel, with zero correlation to stockpiles. (I should know, I’ve traded it daily both on the NYMEX floor and off of it since 1982).   WHO CARES about stockpiles?  Storage is inflexible, leased at long-term almost exclusively and slow to be built.  Real storage of Crude is IN THE GROUND.  Bringing it up to store it is expensive and profit-dissolving — and is always avoided  by commercial physical producers when they can.  Storage (inventories) is just a red herring, inapplicable, unrelated.


    regards – 

  • gregor.us

    Hi. Just a couple of factual notes first, then I’ll give you a broader response.

    First, the chart shown here is from IEA Paris. The timeframe is not selected by me, rather it’s simply what you can get for free from IEA.

    Second, US inventories as I think you will allow only form roughly more than 10% of total OECD inventories.

    Now, I think the more legitimate criticism of this post is that I was too quick to assign 2010 differences between consumption and production to stock draws. Also, the post may give the impression to the first time reader of my work that I think stock draws are a primary driver of price. Actually, I favor a longer cycle structural change to the cost of marginal extraction as being the key driver to oil’s repricing, starting in 2004. This has been amplified by reduced spare capacity globally. And in particular, both the much much higher costs and reduced spare capacity in Non-OPEC  supply, which as you know accounts for nearly 60% of global supply.

    However, your focus if I understand it is that there is absolutely no correlation between stock changes and price. So for you, price can rise and fall completely free of stock changes. My response: I would not dismiss such a view outright. I too have seen the poor correlation between stocks and price throughout the last decade. Moreover, I have also witnessed the increasingly irrelevancy of stocks at Cushing, OK as a determinant of price.

    If I understand you correctly, you have no problem whatsoever with the currently high level of stocks at Cushing, OK (and in total US) with the current high prices of oil. Because, your thesis is that the two have nothing to do with each other. Well, I might tend to agree with that but for a different reason: US inventories are no longer a determinant of price.

    It would be helpful if you would use terms like Spare Capacity and Recoverable Reserves, in your other remarks. I have to guess this is what you mean, but I’m not sure. As far as I can tell, you think Recoverable Reserves are what matters to price. Or perhaps you means Spare Capacity. Again, I’m not sure. However, just on this point alone, I have written several times about Hotelling’s views on reserves, cost of extraction, and the reinvestment prospects of pulling Oil out of the ground–as a better determinant of price.

    Back to the themes in this post: many predicted in 2009 that stock draws globally would appear quickly should demand return, as production would be slower to respond. If you look at the detail of OECD stock draws you will find that is exactly what happened. Also, while it’s true that OECD stock changes can’t account for global stock changes completely, the ferocity of stock draws in the OECD the past 9-12 months I think are clearly one factor driving price. (I know, you disagree).

    There is an irony here which I mentioned in my update: I frankly detest the accounting of Liquids, and in nearly all my work I try to work only with Crude Oil. However, given that 2010 saw the largest ever disparity between BP’s accounting for Consumption vs Production of Liquids I am probing further here and I don’t see that the difference can merely be accounted for via Biofuels, for example. I’ve concluded this will merit an updated, second post but again–based on my observations of global stock draws for the past two years, the primary direction implied in BP’s numbers, imo, stands.