This sounds familiar .<\/p>\n
Felix Salmon<\/a>: What are the consequences of allowing multi-billion-dollar systemically important multinational corporations to report their assets using proprietary mark-to-model tools involving discredited Monte Carlo simulations? I think we all know the answer to that one. But unbelievably, after such shenanigans contributed enormously to the greatest financial meltdown in living memory, the SEC is now set to allow more or less exactly the same thing in the oil industry.<\/p>\n Otto<\/a> points to a stunning report<\/a> by oil consultant Alan von Altendorf which spells it all out. Up until now, oil companies needed to actually prove they had reserves before they reported proven oil reserves. Now, however, the SEC is allowing them to use internal, proprietary computer models to essentially pull their “proven reserve” numbers out of thin air (or the nearest friendly Monte Carlo simulation).<\/p>\n We’re not quite as appalled as Felix is — for one thing, oil companies don’t pose systemic risk the same way as financial companies, so just because you can draw a line between overuse of models and the financial blowup, it doesn’t mean we’ll suffer similar repercussions if these models turn out to be total BS.<\/p>\n And, of course, models are useful, and frequently better than actual counting (see: the census controversy)<\/p>\n \n