Probably a key point of the energy decline movement has been the idea of “peak oil”. The theory has been that oil and gas would imminently decline, leading to catastrophic consequences for civilization, such as collapse. Although this group has a number of different theories (declining net energy, declining EROI, and so on), the primary and central theory of the group has been peak oil. Indeed the energy decline movement (the subject of this blog) was often referred to as the “peak oilers” by the press back in 2008 when the group had publicity. As a result I will discuss the peak oil theory specifically in this post.
As always, I think the theories of this group are quite wrong, including the peak oil theory. To be sure, oil will peak and decline someday, but that will be caused by high oil prices and a switch to alternatives. We will never “run out” of oil, and its decline will have only modest and barely-noticed effects on society. There are clear substitutes for all usages of oil, and the economy will transition when the time is right. Peak oil will be caused by not needing it as much anymore, not by depletion.
Not only do I think the group is wrong about the cause and consequences of peak oil, but I also think the group is wrong in their estimates of how much oil remains and the timing of peak oil. The group has often portrayed peak oil as imminent because of depletion, but that is simply not correct, in my opnion. Even if there is only a very gradual switch to alternatives (EVs), oil supplies are adequate for decades. We do not face any kind of energy shortage during the next 50 years, even if the transition to EVs is very gradual.
In this post I will develop a very simple mathematical model of oil supplies and depletion. I will show that oil supplies are adequate for the next 50 years or so. I should mention that I have no training whatsoever in petroleum engineering or geology. I’m a math/econ sort of guy. However, I’ve noticed that most members of this group (even the thought leaders) have no training in the field either. So it seems fair (to me) to construct an extremely simple model as an amateur, based on very basic math. I will use only a few key facts and very simple math to develop the outline of a basic model.
The Model
The first fact to understand is that oil extraction is a process that currently takes many years and has many stages. First, there are petroleum geologists who scour the earth looking for new oil deposits which haven’t been found yet. After which, there is a planning and engineering effort which can last for many years before any oil comes out of the ground in a particular project. At any time, there is a PIPELINE of oil development projects, at various stages of completion. Old oil wells are continuously depleting and being replaced with newer ones. The oil companies manage of pipeline of projects years in advance so new oil wells will come online as old ones deplete.
At any time, there is a useful figure called the “reserves to production ratio” (henceforth R/P ratio) whcih measures how much oil we have in reserves compared to present usage. Right now, the R/P ratio is about 55, which means that we would have 55 years of oil remaining even if the pipeline discussed above stopped totally and immediately, and no new discoveries were made. Of course, problems would emerge long before the 55 years were up. At some point before the 55 years that we ran out, drilling of new wells could not keep pace with depletion of old ones, and oil production would plummet, probably near the end of the 55 year period.
It is important to understand that oil production for the world will not follow a Hubbert curve. Individual oil fields may well follow a Hubbert curve, but the world will not. This is because any decline in global oil production increases prices, which causes more furious drilling or other technologies which drive the production of oil back up. For example, an individual oil field may undergo depletion, but operators can then use technologies such as steam injection (injecting steam into the well) which drives production back up. As a result, any tiny decline in global oil production will cause increased prices and increased production, preventing any further decline. The eventual decline in conventional global oil production would happen drastically, at the end of the oil era, when no amount of furious drilling or steam injection could overcome the rapid depletion.
We mentioned before, that we would have 55 years of oil remaining if we stopped all discovery right now, and didn’t develop any unconventional oil reserves. Probably, oil production would continue as normal for 35 years or so then would suddenly plumment to zero during the last 20 years. Furious drilling of existing reserves and techniques such as steam injection would initially prevent a decline in production. However, depletion of existing wells would grow worse and worse over time, until any amount of drilling or technology could not arrest the decline, leading to a sudden plummet at the end.
Thus, at present, we face 35 years or so of steadily increasing prices and then sudden crisis. This assumes all discovery ceases right now and no unconventional oil is ever used.
Of course, we have more time than that. Discovery and development of unconventional resources are ongoing. It is not possible that discovery would suddenly and totally cease, that we would fail to develop any unconventional oil, and so on. Discoveries and development of unconventional resources have been keeping pace with depletion for years. The R/P ratio has remained basically unchanged at higher than 50 since 2010 (https://www.statista.com/statistics/682098/oil-reserves-to-production-ratio-worldwide/). The R/P ratio is the result of exploration and technological advancement by tens of thousands of people in different areas, many years in advance. It’s a statistical phenomenon which follows a smooth curve. It will start declining some day, but it will follow a smooth curve and will not have any kind of sudden discontinuity. As a result, the 55 year figure is a big underestimate of how much oil could ultimately be extracted.
Here we can begin sketching out the basic parts of a model. Oil is being extracted and thereby depleted. This reduces the R/P ratio over time. On the other hand, there is a pipeline of upstream projects, new discoveries, and development of unconventional oil reserves, all of which increases the R/P ratio over time. These two factors are in competition with each other. Production decreases the R/P ratio and discovery and unconventional development increase the R/P ratio.
There is a third factor which needs to be taken into account. At present there is a a transition to electric vehicles underway. At present, 18% of new car sales worldwide are EVs. Electric trucks (the tesla semi and others) are starting to come online also. Any construction of EV factories or deployment of EVs pushes to date of oil depletion back into the future and increases the R/P ratio. It represents oil not burned now. If discoveries kept up the pace they have been doing for the last 13 years, but EVs became more widespread, then discoveries would stay constant but depletion would slow down, leading to an increase in the R/P ratio.
Thus, we have three factors tugging at the R/P ratio: 1) depletion, 2) discovery and unconventional development, and 3) EV adoption. The question is: which factor is winning? What will happen over the next 50 years?
Let’s make some pessimistic assumptions and see the results. Let’s assume that EV adoption is very gradual, much slower than it has been in the last few years, and it takes a full 50 years for half of vehicles on the road to become EVs (here the word “vehicle” is an abstraction which includes cars and trucks in proportion to their fuel usage. For example, a class 8 truck might count as 8 “vehicles” because it burns so much more fuel). At the same time, all oil discovery slows down drastically, starting now, and only little unconventional oil could ever be extracted, so the additions to oil reserves decline to 0 over the next 50 years. These are highly pessimistic assumptions. What would be the result?
The result is that the R/P ratio 50 years from now would be 85, which is much higher than today. We start with 34*55 years of oil, which is 1.87 trillion barrels. However, discoveries and development of unconventional oil declines linearly to zero over 50 years, which implies we gain an additional 0.85 trillion barrels (34*50*.5) during that time, so we have a total of 2.72 trillion barrels to use over the next 50 years. At the same time, EV penetration increases linearly to 50%, gradually, so the total amount of oil burned over the next 50 years is only 75% of what was initially expected (34*.75*50). As a result, we burn 1.275 trillion barrels out of 2.72, so are left with 1.445 trillion at the end. Oil consumption is half at that point (EVs constitute 50% of all vehicles then). The end result is an R/P ratio of 85 (1445/(34*0.5)).
Thus, using very pessimistic assumptions, the R/P ratio INCREASES over time, and the date of exhaustion gets further away. Of course, the R/P ratio won’t really increase to 85 because oil companies will curtail discoveries and curtail development of unconventional resources. Oil companies keep the R/P ratio at 55 as a kind of inventory management, so they will curtail upstream development if it gets too high. Incidentally, the curtailment of discoveries and upstream production has already happened at oil companies, which implies that the date of exhaustion is getting further away.
I pointed out earlier than we cannot just keep production constant until we suddenly run out. Near the end of the oil era, oil production would suddenly plummet despite furious drilling. Here I will define an R/P of 15 as an oil crisis where oil is imminently plummeting and we must rush to convert vehicles and car factories to EVs.
Using any kind of assumptions, it is very difficult to reach an oil crisis (R/P of 15) over the next 50 years.
Let’s try making drastically pessimistic assumptions. Assume it takes 100 years to reach an EV penetration of 50%. Furthermore, all new oil discoveries and all unconventional development drops linearly to zero over only 40 years (not 50). The result is an R/P ratio of 43.4, fully 50 years from now. Such an R/P ratio is still higher than it was in the 1990s. Even using drastically pessimistic assumptions, oil production is quite adequate 50 years from now, and reserves are quite adequate.
Of course, this has not taken into account the billions of people in China, India, and southeast Asia who wish to start driving and join the middle class, as well as population growth in the developing world. However, those areas are far more densely populated than the USA (which consumes almost 25% of all oil by itself). People there are much more likely to follow the model of Japan in terms of oil consumption, and even that will take many decades. As a result, global oil production has been growing fairly gradually for decades and is only 33% higher than 30 years ago. This factor is obviously greatly outweighed by EV adoption and would not change the above analysis by much.
Conclusion
It does not appear that an oil crisis is imminent. No plausible mathematical analysis yields an imminent oil crisis. At least for the next 50 years, and probably much longer, oil supplies are adequate to meet demand.
I saw a video from Nate Hagens recently showing an imminent drastic drop-off of oil production in the near future. That video made two mistakes, in my opinion. First, it assumed that we are already halfway through our total oil endowment, which is incorrect. Second, it did not mention or take into account EV adoption which is already underway, is growing exponentially, and which pushes out of the date of depletion into the future. When these two factors are included in a basic mathematical model, we see that there is no imminent oil crisis.
So far, I’ve offered some scenarios which stretch 50 years into the future. I don’t like to predict further than 50 years into the future, or even that far, because technological developments occur which render such predictions useless. For example, almost nobody foresaw, 8 years ago, that the biggest car company by far would be an EV company, that EV sales would be increasing at a rate of about 20% per year, and so on. There have repeatedly been technological developments which start out quite small, but which totally change the long-term trajectory. Thus any long-term projection I would have made would have been FAR too pessimistic because of technological developments. The peak oil movement in particular has frequently been very wrong even 5 years out. As a result, the 50 year estimates indicated above are almost certainly far too pessimistic because they ignore technological developments which are disruptive and which could change the picture completely (for the better) over the next 50 years. This makes it even less likely that we face any kind of oil crisis.
Finally, all of this is assuming relatively stable geopolitical conditions. It is always possible there will be a nuclear war because of Ukraine or something similar. I have no way of predicting that, or even calculating its probability. However, we do not face any kind of imminent disruption because of oil depletion, no matter what assumptions are made. Fifty years from now, R/P ratios are likely to be similar to what they are now.
One final thing. Oil is by far the scarcest substance in the world that we use, relative to demand. Nothing else is so concentrated in a single geographical area, so widespread in its usage, and so difficult to substitute. Yet even oil shortages pose no serious problem for the next 50 years. Shortages of other things (like coal and minerals) are MUCH FURTHER AWAY than shortages of oil. Since we don’t face any crisis of oil, we don’t face any crisis of anything else we dig out of the ground either.
No comments:
Post a Comment