If you prefer to listen rather than read, this blog is available as a podcast here. Or if you want to listen to just this post:
The theme of this blog comes from the book of Jeremiah, verse 20 of chapter 8, and though that verse is included conspicuously in numerous places, it doesn’t hurt for it to appear again.
The harvest is past, the summer is ended, and we are not saved.
For this post I’d like to focus on that first phrase. What do I mean when I claim that the harvest is past?
We all basically understand harvesting, it’s when you take the food that has been grown and remove it from the plants where it grew, presumably storing it for later use. If we look a little deeper, the harvest is really the final step of turning energy from the Sun into something usable.
It’s not just food that is ultimately made by the Sun, with a few exceptions (nuclear, geothermal and tidal) all energy comes from the Sun. Fossil fuels included. I’m sure this won’t surprise any readers of this blog, but I think that many people are unaware of the fact that fossil fuels are just stored solar energy. Thus when we burn oil, and coal, and natural gas for power, we are withdrawing solar energy that was saved up millions of years ago.
Consequently, when we talk about the total energy available, if we exclude nuclear (geothermal and tidal power being not significant enough to matter) we’re limited to the amount of sunlight which reaches the Earth. We’ve been able to reach back and withdraw the energy of past sunlight, but at some point, as every family knows, you have to stop spending more than you bring in. When we say that all energy comes from the Sun we’re more or less saying that everything comes from the Sun. (I know what you’re thinking. I promise we’ll get back to nuclear power before the end.) Our economy, the conveniences of modern life, economic growth, employment, etc. are all intimately connected with energy usage. And all of these things have been supercharged by the extraction of additional energy. That is the harvest. That is the summer, and as I will show they can’t last forever, if in fact they haven’t ended already.
As we get into things, I’m hugely indebted to a couple of posts by Tom Murphy, a physics professor at the University of California, San Diego. If you want to really get into the math of things and if you’d like to see more charts and graphs, I would check out both articles:
Murphy begins by pointing out that since about 1650, a century before the Industrial Revolution, the United States (or what would become the US) has grown at a fairly steady rate of 2.9% per year, on average. This has continued down to the present day, though recently there are signs that it’s been slowing. (Average growth since 2001 has only been 1.8%.) One might usefully ask, what was the long term average growth rate before 1650, or in any case before the industrial revolution? As it turns out it was next to zero, perhaps a long term average of 0.1%. So hurrah for the industrial revolution, but how did we go from nearly zero to 2.9%? What changed?
Put simply, we did it by spending a million years worth of accumulated solar energy, in the space of a few centuries.
Now, when I say something like that, there’s a danger that you’re going to tune out, thinking, that this is going to be some kind of environmental rant. You may even be thinking that the next step is for me to start talking about Peak Oil. Certainly a discussion of environmental issues, or whether we’re about to run out of oil is interesting, but for the purposes of this illustration it’s beside the point, because it doesn’t matter if we’re about to run out of oil, or if oil is naturally produced deep in the earth and effectively unlimited (a theory with a surprising amount of traction in Russia) or if we’re going to destroy the earth with global warming, or if fusion will save the day. When you actually look at the numbers, in the long term none of those things matters.
Yes, I do think that for a variety of reasons that the enormous growth rate spike we’ve experienced over the last several centuries is nearing an end. That we have essentially had the biggest harvest ever, as we’ve extracted the accumulated solar energy of all the previous epochs. But even if we leave fossil fuels out of the equation we are still reaching the limits of growth. And that is what I found so interesting about Murphy’s posts, and why I decided to write about them.
As I said, let’s keep fossil fuels out of it and just focus on the solar energy we’re receiving at this moment. Also let’s follow Murphy’s lead and reduce the annual growth rate from 2.9% to 2.3%. This translates into an easy to remember 10x increase every 100 years. So if we’re currently using 12 terawatts of power every year then in 100 years we’ll be using 120 terawatts.
First the good news. We get more energy from the Sun every hour (174,000 terawatts) than we use all year. Of course only 70% of that energy reaches the surface (the rest is reflected back into space.) And only 28% of the earth is land, and thus currently eligible for solar collection, and of that land we’re using 50% for agriculture (which sounds high, but who am I to argue with National Geographic?) And finally our current solar panels are only 15% effective. Taking all of that together we end up with 2558 terawatts of currently usable solar power which still seems pretty good. It’s over 200 times what we’re using now.
The problem is when you have exponential growth, things that aren’t problems can quickly become problems. And they have a tendency to sneak up on you. In 100 years, at a growth rate of 2.3% as I said above, we’re still only using 120 terawatts. Still lots of room, but then 200 years from now we’re using 1200 terawatts (10x every 100 years remember). That’s still less than half, how bad could it be right? Well it only takes another 35 years and we’re out. So in 235 years, at the current rate, we’re using all of the sun’s available energy. Now I understand that 235 years seems like a long time. But it’s less time than the US has existed.
In other words if we expect the US to be around for at least as long as it’s already has been (if we don’t want to believe we’re past the mid-point) and further if we expect it to continue in roughly the same trajectory. Then we have to increase the efficiency of solar panels, start putting them onto the ocean and/or into space, or grow less food. The problem is, that because of exponential growth, none of that buys us very much time.
Let’s assume that we’re able to increase the efficiency of the solar panels to 100%. That buys us 319 years (an additional 84 years.) Or if we’re looking in reverse back to around 1700. Once again that seems like a long time, but we’ve had longer than that to become accustomed to 2.9% growth as a law of nature. Once again we’re looking at being past the midpoint of our growth, in a best case scenario that involves every square inch of land being covered by solar panels or used for farming, and which also involves a premise (100% efficiency) which is physically impossible.
Let’s then assume that we get rid of all the land being used for agriculture. I’m not sure how, perhaps we grow things underground. Doing that adds another 31 years and gets us to 350 years from now. Or looking back it gets us to around 1666, very close to the start of the big harvest. But now of course every square inch of land from Antarctica to the remotest part of Siberia is being used for solar collection.
As you can see once you start running up against the limits, then even doing something as radical as doubling the amount of land being used doesn’t buy you much time. And recall that we are talking about theoretical limits. This is as good as it’s possible to be. Practical limits are likely to be 10x more restrictive. In other words we might be bumping up against those limits a lot sooner than we think. But let’s continue with our thought experiment.
We can add in the oceans, which buys us another 55 years. 405 years into the future or back to about 1600 or when Shakespeare was alive. Historically we’re still talking about a time that was fairly recent. In fact as Murphy points out in order to continue on the same growth rate for 1200 years (so the Dark Ages if you’re looking back) we’d have to use all of the Sun’s energy. And I mean all of it, the Sun would have to be completely encircled by solar panels. And if we wanted our energy use to continue to grow at the same rate for 2500 years we’d need to use all of the solar energy produced by our entire galaxy. Which to put it bluntly, seems unlikely.
I know that some of you have been screaming, “But what about nuclear power? What about fusion?” Here we run into a different problem. Heat. I said above in 1200 years our energy requirements are going to be equal to the entire output of the Sun. What do you think happens if you’re generating as much energy as the Sun, through fusion, in a space much, much smaller than the Sun? The implacable laws of thermodynamics dictate that the Earth, because it’s much, much smaller, would end up, much, much hotter. As I said if you want to look at the math and see some graphs I refer you to Murphy’s paper, but according to his calculations, if terrestrial energy use continues at it’s current rate, the Earth gets hotter than the surface of the Sun in less than 1000 years.
I think you can see now that things are unsustainable. To steal Murphy’s key point:
Continued growth in energy use becomes physically impossible within conceivable timeframes.
Thus far I’ve been using something of a sleight of hand. I’ve mostly used growth and energy growth interchangeably. And in fact for most of human history they have basically been the same thing, though in the last 50 years or so we have started to see a divergence between economic growth and energy growth. This is good news, if we don’t need growth in energy use to get economic growth, then perhaps everything’s okay, and we won’t need to cover the entire Earth in solar panels, or heat the surface to the point where steel melts (730 years). It would be nice if it were true, but there are also insuperable problems with this line of thinking as well.
Basically if at some point we have to keep energy use flat, while the economy keeps growing then the percentage of the economy made up by things that don’t use any energy get’s bigger and bigger. What part of the economy doesn’t use energy? Stuff like fashion, certain innovations, education, and most of all financial transactions. Perhaps this sounds familiar? Perhaps it sounds like the world we already live in? And maybe you’re thinking hey this isn’t so bad. Is it that big of a deal that the UK’s GDP is 10% financial services, and that the economy of New York City is 35% financial services?
First I’m not sure that financial services are as divorced from energy use as even Murphy thinks. Secondly, yes it may be fine that 35% of New York City’s economy is financial services. (I actually think the housing crash proves the exact opposite, that it was not fine, but let’s set that aside for the moment.) Will it be fine when that number is 90%? What about when it’s 99%? Or 99.9%? Because at some point that’s what has to happen if we divorce energy growth from economic growth, and economic growth continues at the same rate.
And what happens to the people who are forced to work in the 0.01% of economy which still uses energy? For example we are presumably still going to have to eat, and presumably even if all of the food is planted and harvested by robots, that someone is going to have to be involved with food production on some level. How do these people get paid? In order for all this to happen, food, manufactured goods, new houses, etc. would all have to be virtually free.
Accordingly, it’s not just energy growth which has to stop at some point. In fact, at some point, all growth has to stop. The problem is, that’s not the way the modern world is set up. All of our assumptions, all of our institutions, all of our systems are built around the idea that growth will continue. Therefore transitioning to a system where growth is flat is not going to be pretty.
Perhaps you’re comforted by the fact that 235 years is a long way away. Long enough away that even your grandkids won’t have to deal with it. But I would argue that we’re already starting to see this world. I already mentioned the increasing percentage of the economy that’s made up by financial services. And of course there’s the giant share of the economy devoted to services (which is sort of halfway between in terms of energy use.) This is not to mention the persistent negative interest rates, not only with government bonds but now extending even to corporate bonds. And of course I explicitly avoided talking in any great detail about global warming or any of the other potential catastrophes which may befall us before we even get to the point where we’re worried about covering the world with solar panels.
For a long time we have looked to progress and technology to save us. But if they can (and I would offer the opinion they can’t) then they’re running out of time.
The harvest is past, the summer is ended, and we are not saved.