Heres what I did with 10 minutes to kill:
3.7 mil barrels per day Iranian oil production * 365 = 1.3578 billion barrels/year. Add Irans 90 billion known reserves and 38 billion newly discovered and divide it by current production (2002) and you get about 94 years. This of course assumes current production, 100% recovery and no further discoveries.
My guess is the author needed a motive to support his thesis so he either lifted the 20 year field life figure from some environmental publication or he did something like dividing the highest daily production number ever by the lowest reserve estimate. Probably took him at least 20 minutes to pull that one off.
Our collective effort is showing different answers for good reason. Your equation is flat in that it does not consider increase in global demand. For my 40 year guess, I estimated that Iran will average a 3.75% contribution to global consumption over the next twenty years. This is not far off from where they are today. If you look at the increase in global consumption from 1980 to 2000, it's easy to determine a linear equation, y=M*x+b. Use calculus to speed things up Take the integral of the linear global consumption equation from 2006 to 2026 and it will reveal total global consumption over the next 20 years. 3.75% of that is about half of Irans proven reserves. Therefore, if Iran is going to pump half its reserves in 20years, itll pump them all out in 40years.
There are flaws in both our models as Im sure there are with Spenglers. There is no proven method to predict the future for systems as complex as the one were trying to model. When petroleum is as expensive as it is now, the energy economy provides opportunity to other technologies which are currently cost effective. Unfortunately, if the price of crude settles below the alternative energys cost effective threshold, investors will lose their investments as markets transfer away from alternatives energy back to crude. So thats one aspect that is difficult if not impossible to account for when modeling consumption. Another variable I didnt bother with is this We know population growth and industrial growth are modeled with exponential equations and my quick estimate does not consider either. I bet Spengler's did and that's where he came up with 20years.
But I digress... What we've done is provide three estimates for the duration of viability for Iran's proven reserves. Spengler says ~20, I say ~40 and you say ~90. If Iran pulls a unilateral embargo or gets hit with sanctions each of our estimates will hiccup. Nevertheless, we should still try to predict the future! What the hell else would we do with all of these brains? lol BTW: Good Commentary BTTT.