Register

Peak Oil is You


Donate Bitcoins ;-) or Paypal :-)


Page added on December 24, 2012

Bookmark and Share

EIA’s crappy forecasting

Public Policy

For instance, EIA, by its own admission, states that they had
overestimated crude oil production 62% of the time; they had
overestimated natural gas production 70.8% of the time; and they had
overestimated natural gas consumption 69.6% of the time. Not the best
track record by anyone’s estimation except perhaps EIA’s.

It is also noteworthy that EIA had overestimated the energy intensity
ratio a whopping 96.5% of the time. This is a ratio of total energy
consumption and GDP. They tended to overestimate energy consumption and
underestimate growth in GDP.

In short, EIA is not very good at forecasting.*But what is even more
interesting is that Dr. Montgomery, the lead author of this new study,
was once in charge of models and forecasts at EIA*. Both NERA and
Deloitte used EIA forecasts as the basis of their report models.
Deloitte even stated that they considered EIA’s forecasts to be too
conservative in spite of the fact that EIA has not projected natural gas
demand accurately 70% of the time.

As any student of economics soon learns, economic models are only as
good as their inputs. In fact, it is neither difficult nor unusual for
models to be designed to favor one outcome over another. In other words,
models can be essentially reverse engineered. This is especially true
when the models have been commissioned by industries that stand to gain
significantly in monetary terms. Or government agencies which are
perhaps pushing a political agenda.

Energy Policy Forum



7 Comments on "EIA’s crappy forecasting"

  1. BillT on Mon, 24th Dec 2012 6:27 am 

    This is only news to the uninformed. They are owned by Big Petro. All you get are lies from these outfits as someone has to sign their paychecks and no one is truly independent these days.

  2. ken nohe on Mon, 24th Dec 2012 10:37 pm 

    As it goes for economics, so it does for CO2 and the rest: Models are models and their limitations are blatant. Our strength is in understanding their weaknesses and improving them step by step. The EIA should be lauded for admitting it and doing the first necessary steps, not criticized.

  3. BillT on Tue, 25th Dec 2012 2:57 am 

    More propaganda from the Empire. The EIA is doing nothing positive. Just spouting the words that they are told to by their masters. Models today are much more accurate than in the days of slow computers and minimal input.

  4. DC on Tue, 25th Dec 2012 9:45 am 

    While one could make a separate argument that modelling like that is difficult on a good day, that IS what the EIA is supposed to good at. But then again, once you understand the EIA purpose is not to provide useful or accurate information about anything, the notion that EIA make crappy forecasts is no longer really ‘new’s. The EIAs job was never intended provide accurate information. If they did that, there would solar panels on every roof, windmills in every field, and electric trams on the streets, instead of gas-powered garbage bins. Nope, the EIA real job is to blow sunshine up everyone;s backside. After all, what is penalty for that outfit being wrong all the time?

    Nothing whatsoever.

  5. ken nohe on Tue, 25th Dec 2012 1:58 pm 

    I did take the time to read the full EIA report and sure enough I was not very impressed. It looks like the Government would save some money by just getting a copy of the equivalent Exxon report.

    It is actually hard to pinpoint something specifically wrong as the report is one long list of linear thinking, just projecting current trends for as long as possible.

    By and large the future in 10 years looks very similar to the present. Well, nothing much to worry about then? On the one hand such a report should not be a doomsday document exploring every hopeless scenario. On the other hand, some potential hurdles might deserve a footnote, here and there…

    So fracking will save the US, maybe but not cheaply. The economic analysis is nowhere to be seen whereas it should be the most important part of such a report. Frankly, I now agree: What’s the use of such a report?

  6. econ101 on Wed, 26th Dec 2012 5:48 pm 

    Ken, the report is political. It is meant to support current policy directions. It’s the linear thinking that bothers me too.

  7. econ101 on Wed, 26th Dec 2012 5:55 pm 

    The climate models have been proven to be reverse engineered. Those things have been massaged to produce the desired result no matter what the inputs. Our recent snowfall of 15 inches was a record and guess what? It is predicted by the models as further proof because global warming causes extreme weather events. Pretty handy conclusion wouldn’t you say? Especially when you consider our averages are nothing more than the blending of a hundred years of extreme events.

Leave a Reply

Your email address will not be published. Required fields are marked *