We'll build the most simple model possible, using two stocks: oil (in the ground, in tanks, anywhere), and a capacity for burning that oil (rigs, refineries, pipelines, cars, power stations, etc). The stock of oil is non-renewable, and can only decline. The stock of burning capacity can change as a fixed yearly percentage, which makes sense, as it takes a certain amount of processing capacity to build new capacity.
According to IndexMundi, the world used about 31.7 billion barrels of oil in 2010. We'll assume that this number represents our capacity for burning oil as it existed in 2010. For the initial volume of World Oil, we'll use 'proven oil reserves' - known oil that is economically feasible for extraction. BP estimated the total volume of proven reserves in 2010 as 2383 billion barrels. If we assume that usage stays constant (ie, no growth in processing capacity) then our model says we'll run out of oil in about 75 years.
But of course, oil processing capacity (blue, right axis) has been growing over time. From 1981 to 2010, consumption grew at an average of 1.26%. When we use this growth rate in our model, we're out of oil after about 53 years.
If growth rates were to (hypothetically) average 3%, we would be out of oil in 39 years:
![]() |
(I would love to be able to set the scales so that the charts didn't look identical, but alas, it's not in insightmaker's functionality yet.) |
For now let's set an upper bound on how much time we have left. We'll go crazy and assume that the top 5 km of the entire planet is made purely of oil. This comes to about 21,418,957,600 billion barrels, or about 8 million times the current proven reserves. The weird thing about exponential growth (set at 1.26% annual) is that even this huge excess capacity doesn't buy that much more time. We run out after about 640 years, only about eight and a half times as long as with currently proven reserves.