Tracking the EIA Short Term Forecasts

This is a guest post by Gilles.

Hi all. Thanks to Khebab, I have the opportunity to repost a study that I made about Short Term Energy Outlook (STEO) EIA's predictions. I posted it originally as a comment to Khebab's last post , but it was buried under more than 100 comments and he thought it was a good idea to repost it in a guest post.


Animation of the EIA Short Term Oultooks from January 2005 to October 2006. Fine lines: past forecasts; Line with circle markers: forecast for the month displayed. Click to enlarge.

Please let me introduce myself in a few words. I'm a French professor in astrophysics, and I discovered the problem of PO less than 2 years ago (before that, I was rather confident that nuclear energy and hydrogen engines would solve the problem of the depletion of fossil fuels. I changed my mind a lot since this time of course, but it is worth reminding that this opinion is still mostly shared among the scientific community !) I participate regularly to the French forum oleocene , and I follow of course very closely the very interesting discussions on TOD and other similar sites. My work here is minor: I just made a compilation of recent EIA predictions. I'm not of course an expert in energy sources, but I was curious to test how the official agencies could manage the possible apparition of a plateau, in comparison with their cornucopian, ever growing production forecasts.

I drew the following graph in May 2006, by superimposing all monthly published outlooks since the beginning of 2005 (I took off some months for sake of clarity). The spectacular feature for me is that the prediction curve dropped abruptly in August 2005, i.e. just before Katrina hits. After the unusual 2005 hurricane season, the following months have accentuated the decline of the curves, but the EIA regularly predicted that the production would rise again after a few months.

However, this hope has been largely disappointed, since the inflexion point has been continuously slipping, giving the impression of an everlasting postponement. The real curve is following a more or less flat line, exhibiting the famous plateau that Stuart commented a lot of times.


Figure 1: Compilation of EIA monthly STEO predictions from January 2005 to May 2006. For each month, the solid line refers to past quarters and the dotted line to predictions for the present and future quarters.

The same kind of figure was drawn this month, showing a similar pattern: after a few stable predictions in summer, the curves started to be revised downwards again.


Figure 2: Same as Figure 1, but from July 2005 to December 2006 (beware, colors have changed).


Basically, EIA has been telling us: "don't worry, the production will start again rising within 3 months...." for more than one year now ! The cumulated gap since the beginning of 2005 is now reaching around 3Mbd. They did not offer of course a clear explanation why THIS has happened !!!

For the fun, an Oleocene moderator (Sylvain) constructed an animated gif to show the EIA "snake" crawling on the ground in real time that you can watch on top of that post.

One last comment: in the last STO issue, the EIA predicts a gap between the world demand and supply of 0.3 Mbd in 2006 (84.7 vs 85.0) and 0.4 Mbd in 2007 (86.1 vs 86.5). While these figures are not yet really catastrophic (there was an extra production of 0.5 Mbd in 2005), they are nevertheless puzzling, since they indicate that the suuply drop cannot be attributed to a demand destruction. (The cumulated crude oil stock decrease would nevertheless reach 200 Mbl, two thirds of the total US stocks). In any case, I think we are founded to ask EIA naively : could you please explain us, with simple words, why do you predict that the 2007 production will not exceed 86.1 Mbd, whereas the demand is planned to reach 86.5, and that only 15 months ago, the world was supposed to be able to produce 88 Mbd in this last quarter of 2006?


Gilles