Monday, February 18, 2008

EIA Revisions

For this first chart, I used about 12 data sets I had saved from the EIA. The first one was from November 2005. Since it does not use moving averages, I have included 2004 data here.

In this second chart the numbers are 3-month averages. I used this approach to eliminate some of the noise. I was pleased with the result. Anyone see any patterns?

This third chart shows how over time(about two years) the final revisions tend to settle on the higher level.

And of course, the one that includes price. I threw in some extra verticle lines around where I thought the downward trends in price are. I'm not seeing much of any correlation with short-term increased production levels.

Over this period the highest value for the difference between the high and low revisions was about 400,000 bpd, with an average of about 250,000 bpd (or 0.33%).


Anonymous said...

The amount supplied alledgedly went up by >2% in just two months, an amount that might be expected for a WHOLE YEAR.

Normal economics says this should have brought the price down - but, actually and undeniably, during that same period the price of Brent crude went UP ~20% from ~$75 to ~$90.

IMO somebody, somewhere, for some reason, is deliberately trying to manipulate/falsify the oil supply data - I wonder who ... and why?


Anonymous said...

I understand your point about the supply and price movements, I woonder about these relationships constantly, but there is something you need to keep in mind.

These are monthly numbers. The 2% is from one month to the next(or in this case over two months). Statistically, production rises OR falls by about this amount every month.

I'll print another chart tomorrow after I combine some data I've got to give a better picture of this.

There is a huge amount of oil (maybe six months worth, I've calculated it before) that resides between the ground and the gas tank worldwide. Price fluctuations are more likely based on the markets perception of how that amount is changing or will change rather than the monthly supply numbers.

You will see when I publish a chart on revisions how much actual uncertainty there is in the actual amount being produced.

Anonymous said...

Interesting graphs. I think the marginal barrels which determine the price of WTI are 'net export' and not 'all liquids' - how do 'net exports' and 'all liquids' vary relative to each other?


Anonymous said...

Yeah, I'll do that next. I might not get to it today, but probably by tomorrow. I still need to update my exports spreadsheet for this month, too.