Previously, I was comparing data from the annual BP Statistical Review of World Energy with the annual averages of JODI Oil data, and when I cast my eye over a table of differences, it was easy to spot that something happened in 2009 – the data from the two sources jumped to more closely correlate. For some countries and product types, if it didn’t happen in 2009, it happened in 2010; but since then some data lines have begun to diverge again. Either somebody was lying prior to 2009 (and by “lying”, I mean, making errors in reporting on hydrocarbon refinery), or something changed in the definitions of the sub-categories of hydrocarbon products from petrorefineries. At this stage, I cannot tell if the corrections were done by BP or by JODI Oil, but the corrections show a step change. This intrigued me, so, here follow a few diagrams and some summary notes.
The example of North America is dominated by a correction in the data for the United States of America (whether the correction was in the JODI Oil data, or in the BP data) for the “Others” category. Since 2009, the data lines have been coming progressively closer, until it seems they are reporting from either the same sources, or using the same industry data to base their calculations on.
Data from South and Central America as a whole is rather random when compared between BP and JODI – however there is a clear correction in the category “Others” in 2009, and perhaps a further correction to both “Light distillates” and “Others” in 2011. Since then, the trend is for BP and JODI data to diverge.
The 2009 correction for the “Europe and Eurasia” region (an artefact) is mainly due to the big correction for the European Union in 2009 for “Light distillates” and “Others”. The data for CIS undergoes a smaller correction, and this is in 2010, for “Fuel oil” and “Others”.
The “Others” category is also adjusted for the Middle East in 2009.
There are minor corrections in the data for Africa in both 2009 and 2010, and recently a large divergence for “Middle distillates”.
Asia Pacific data is corrected for “Light distillates”, “Middle distillates” and “Others” in 2009, reflecting corrections in both China and Japan data.
Corrections in 2009 for OECD data are the main reason for the differences between BP and JODI to snap shut; whilst Non-OECD data still remains divergent.