Technological advances and innovation change every industry, and energy is no exception.

In 2008, the Energy Information Administration (EIA) was estimating that crude oil production in 2020 would be 13.4 quadrillion British thermal units (quads) and would fall to 12.0 quads in 2030.

In the most recent estimates (from EIA’s 2017 Annual Energy Outlook), 2020 production is expected to be 20.7 quads, with 22.0 quads in 2030 under baseline assumptions. The EIA also looks at other scenarios including one involving more growth from technology, which indicates that production could be even higher.

The EIA’s models and personnel are top quality, and believe me no one understands the challenges of forecasting more than I do. The fact that the EIA’s models failed to capture the dramatic rise in production enabled by fracking and other technological advances illustrates just how fundamentally those advances shifted the market.

Remember the “peak oil” argument? The idea was put forth in the 1950s by geologist M. King Hubbert, who had begun to talk about the idea even earlier. It was embraced by many market watchers, academicians, and people in the industry. The peak oil theory essentially said that oil production would follow a bell-shaped curve with a peak when half of the world’s oil had been produced and a decline after that time as it became increasingly difficult (and expensive) to extract. Hubbert predicted the peak to be about 1969 in the United States. The year of highest U.S. production occurred in 1970, seemingly confirming the theory. However, while peak oil theory would have had production trending down in a smooth decline, new technologies have changed everything.

Some peak oil fans continue to hold to the theory. They point to a technicality: the fact that US annual crude oil production did peak in 1970 and has never reached that point again despite price and technology changes. Don’t be fooled. Oil production in the United States did reach its all-time high in 1970 of 9.6 million barrels per day (on average through the year) and was generally declining for decades after that (with some bouncing around in response to market conditions and a blip when the Alaskan north slope came on line), but that’s not the end of the story. With fracking and other technology developments and prices high enough to make using them economically feasible, production jumped from its low in 2008 of less than 5.0 million barrels per day to 9.4 million barrels per day in 2015. Production declined when prices subsequently fell, but is starting back upward now.

I am not suggesting that there is an infinite supply of oil, because we all know that’s not true. The point is that with the right incentives, research, and smart people putting their minds to it, we’ve been able to drastically increase what we can extract. Even now, I’ve seen estimates that only about 60 percent of oil in most reservoirs can be extracted with today’s technology regardless of oil prices, economics, and how much is spent. I feel confident that we will see further advances which improve efficiency, effectiveness, and economics of production (not to mention our ability to locate new deposits).

Another recent development is that the U.S. is now estimated to hold more oil reserves than any other nation. Estimates in 2016 by Rystad Energy indicate the U.S. tops Saudi Arabia and Russia. More than half of U.S. oil is in unconventional shales, which rely on fracking to produce. Although the cost to produce such oil can be higher than conventional wells, when the economics are right it will be tapped. Rystad Energy also estimated that total global recoverable oil reserves were about 2.1 trillion barrels, about 70 times the current production rate and far in excess of cumulative production up to 2015 of 1.3 trillion barrels.

Future oil production will depend on several things. One is the market and price levels, because using advanced recovery techniques costs money and prices must make the outlays worthwhile. Of course, costs can also be reduced as technology advances, which we have seen dramatically just in the past few years. Production will also in part depend on new advances and how they continue to evolve. The geology is obviously an additional factor, but what we know about in the ground is enough to last for a long time.

Only a few years ago, there was credible talk of the ongoing, inevitable decline in oil production. The models were projecting that would happen and intelligent people were on board. Even so, those projections were based on “ceteris paribus” (or “all else being equal”) which essentially means that the models were assuming things basically stayed the same as they had been. The models weren’t factoring in the shale oil plays, because producing them wasn’t feasible at the time. Now it is, and while oil supplies are finite, the decline has been pushed far out into the future. We can now talk in terms of centuries rather than decades.

History is replete with such phenomena. The best minds of the early 19th century were convinced that the population would outstrip the food supply and bring the end of civilization. Then, as now, human ingenuity and powerful market incentives to unleash it came to the rescue.