<DrumBeat: November 26, 2008 The Oil Drum Drum(stick)Beat: November 27, 2008>

78 comments on Predicting Future Supply from Undiscovered Oil

Comments can no longer be added to this story.

| Show without comments | PDF version

aangel on November 26, 2008 - 11:13am Permalink | Subthread | Comments top

Thanks for doing this analysis, Khebab.

Bringing this one step further, given that Pickering calculates his a in the formula below:

P= aR + b

(where P is production, R is reserves)
to be:
OPEC: a= 0.0096 mbpd/Gb, b= 0.2323 mbpd
Fringe: a= 0.0466 mbpd/Gb, b= 0.093 mbpd
Small Fringe : a= 0.0435 mbpd/Gb, b= 0.0418 mbpd

we should expect your #6 (increasing difficulty in extracting oil) to show up as a (much) lower a on the other side of the peak. At this point by how much is just a guess, but it could be argued that Pickering's 0.0466 line in Figure 5 should be lowered by 15% to 25% "soon" since his data set was from 1980 through 2002.

Comments can no longer be added to this story.
WebHubbleTelescope on November 26, 2008 - 2:51pm Permalink | Subthread | Parent | Comments top

Interesting that the shock model typically discards the "b" term in Pickering's formulation. The problem with this term is that doesn't apply any kind of proportionality, in that big reserves produce more and smaller reserves produce proportionately less. When the reserves go to zero and you still have this term, you are left puzzled and shaking your head. Yet the strangeness of this number for OPEC/SA implies that they must be really cooking the books for reserve accounting, otherwise the "a" term would be much stronger for proportional draw. Another way to think about it is that "a" is the capitalist/greed factor and "b" is a socialist/control term. Yet at some point "b" has to disappear as reservoirs deplete.

Comments can no longer be added to this story.
aangel on November 26, 2008 - 4:10pm Permalink | Subthread | Parent | Comments top

Perhaps that's a valid way to look at the terms. I would be able to comment on that more easily if I had the paper.

From the abstract (emphasis added as it pertains to b):

This paper considers the relationship between the extraction rates and remaining reserves of oil. A simple exhaustible resources model suggests a linear extraction rule, with slope terms common to all extractors when discount rates are homogeneous. Differences in pricing behavior and costs determine the intercept. Panel data from the world oil industry exhibit a robust and stable extraction–reserves relationship across countries and through time. Common slopes characterize extraction over large ranges of reserves although there is a concave relationship across non-OPEC members, and a significantly lower estimated slope within OPEC countries. These findings may be explained by risk aversion, differences in discount rates, and measurement error in the reserves data.

There is not much to go on from that to understand how he derives b but a seems to me self-evident.

Comments can no longer be added to this story.
WebHubbleTelescope on November 26, 2008 - 4:32pm Permalink | Subthread | Parent | Comments top

So the 'b' term looks like a residual of the fitting process, as it essentially determines the y-intercept of a best fit when they plot P against R.

Comments can no longer be added to this story.
aangel on November 27, 2008 - 12:45pm Permalink | Subthread | Parent | Comments top

Yes it could be just a "residual" but in the abstract he suggests that in this case b has meaning — something to do with cost and pricing behavior.

Comments can no longer be added to this story.

78 comments on Predicting Future Supply from Undiscovered Oil

Comments can no longer be added to this story.

| Show without comments | PDF version