Shaun recently posted material noting that the IEA had revised its world oil reserve estimates downward and suggesting that the US had put pressure on the agency to keep them artificially high.
That post took me back to my long lost past, when I wrote my PhD dissertation on social factors affecting estimates of how much oil was left in the ground. The diagram at left shows some of the data from my dissertation, illustrating that the pattern of estimates fell into distinct historical periods. Each dot represents a particular estimate of how much oil is left in the ground in the US plotted against the year that the estimate was published. As you can see, revisions are a fairly common occurrence :+) And, as I argued, significantly political.
Unfortunately, I can't embed the pdf of the article on the blog. But anyone that is interested in can take a look at the original here. If you aren't at a location that will allow access to the journal, the citation information is: Gary Bowden, "Estimating U.S. Crude Oil Resources: Organizational Interests, Political Economy, and Historical Change" The Pacific Sociological Review (the journal changed its name, now it is Sociological Perspectives), 25(4): 419-448. October 1982.