Moore’s Law and The Future of Renewable Energy, or Why We Can’t Get to 80 Percent Clean Energy in 2035: Part 3

Jason Harrow

3. Two Other Ways That Energy Doesn’t Work Like Information

This is the third post in a multi-part series. In Part 1, I explained the idea of Moore’s Law, which states that computer technology gets about twice as good, for the same price, every two years.  In Part 2, I argued that the goals President Obama set for clean energy in his State of the Union are overly optimistic, because energy technologies do not obey Moore’s Law. In this Part, I give two other reasons why energy technology cannot be adopted as quickly as computer technology. Feedback is welcome to

I do not mean to be overly pessimistic, but when you look at the energy industry, you see that things are even worse than I’ve previously explained when it comes to the speed that we will be able to implement innovative, renewable technologies. Even if the rate of our technological development happens faster than most people are predicting, it still will be exceedingly difficult to transform our energy use in 25 years to conform with the President’s goal. That’s because there are at least two other ways that the Green Age does not work like the Information Age. I’ll call the first the “old stuff” problem and the second the “permissions” problem.

The “old stuff” problem arises because we already have an entire power grid working well, and the equipment is really, really expensive. So what do we do with this “old stuff” when we have a bunch of “new stuff” ready to go? There’s no easy answer for this. Vaclav Smil, a Professor at the University of Manitoba and perhaps the leading historian of energy use, has described the scope of this problem in a 2008 article here. He notes that the current fossil-fuel and nuclear generation infrastructure has a replacement value of $1.5 trillion dollars. Who will pay to take much of that offline when, as the President hopes, we are generating 80% of our power from “clean” sources in 2035?

As Smil notes, though, there is still more “old stuff” we have to worry about than just the power plants themselves. For instance, we don’t have the transmission infrastructure to move electricity from where the wind and the sun is—the plains, and the Southwest—to where most people are in the East. So we’ll need to build, in Smil’s estimation, “at least 40,000 additional miles of new high-capacity lines” at a cost of around $100 billion. Presumably, that will render some of the existing lines virtually useless. Where does all the old stuff go in 2035? The fact that we have lots of old stuff online means that the cost and difficulty of transitioning to renewables is higher than many people realize. You have to not only consider the cost of the new, which is what my prior post was about, but also the cost of getting rid of the old.

This raises an interesting question: why isn’t there the same kind of “old stuff” problem in the tech industry? The problem is there to some degree. We don’t all rush out and get new laptops every month, even though the cost of components does keep falling. That’s because the improvements in the new stuff isn’t worth us chucking our old stuff in only a month. But the reason that it’s not nearly as big of a problem in the tech space has, once again, to do with Moore’s Law. Computer technology gets so much better and so much cheaper so much more quickly than other sorts of technology. That means that it very quickly becomes far less efficient—and therefore more expensive—to keep operating the old, slow stuff than to buy the new, fast stuff.

An example illustrates the point. Lots of people paid $1,000 fifteen years ago ($1,300 in today’s dollars) to get a Motorola StarTac, the coolest, smallest, best cell phone on the market (see the picture below—you’ll remember it). But who is clinging to that investment today? In just a few years, the new stuff got too good and too cheap not to make the switch. Apparently, Motorola sold about 60 million StarTacs in the 90’s. I would guess that almost every single one of them has been thrown away.

Two lines of LED text for $1,000!

But, of course, you can’t justify throwing away a coal-fired power plant until solar or wind are so much cheaper and better than coal that it’s easier just to throw the old power plant away and build a new one based on solar. We are not even close to that point, and because of the way energy technology changes, we will not be there soon. We need solar power to work like cell phones, where we can just chuck our StarTacs in the trash. But it doesn’t work that way.

Smil calls pretending that we can overcome this infrastructure problem easily “Moore’s Curse.” He notes that such green-energy optimism not only ignores the pace of technological change, but it “it completely ignores the massive infrastructural needs of new modes of electricity generation.” I have not seen anyone propose any kind of solution to this problem that responds to its massive scale.

That brings me to a final problem that demands mention. I’ll call it the “permissions” problem, in light of an excellent article that Professor Larry Lessig wrote this fall about the movie The Social Network. In it, Lessig lamented that the film did not engage at all with the most incredible feature of the Internet: that “Zuckerberg’s genius could be embraced by half-a-billion people within six years of its first being launched, without (and here is the critical bit) asking permission of anyone. The real story is not the invention. It is the platform that makes the invention sing.” The information technology sector thus combines rapid technological change with a kind of radical democracy that lets anyone with a few good lines of code take over the world in a heartbeat. YouTube went from not existing to being valued at a $1 billion in 12 months. We get used to hearing craziness like that, but the blinding speed of growth on the Internet should make our heads spin every time we hear that. That speed is not normal. But a key driver of this incredible speed is that online innovators don’t need to ask permission. They just need to start running code.

The opposite is true, of course, in the energy sector. No one has ever said that a feature of our energy grid or regulatory framework is that it makes “invention sing.” It grinds invention to a halt.

One of the wonderful things about being in law school is that you get to see how the sausage is made in lots of industries, and seeing how things work in the energy industry is not encouraging on this front. Reading a decision by the Georgia Power Commission that allowed a new nuclear generation plant to be built will make your head hurt. Though there apparently have been some small steps to allow new generation facilities to come online faster, companies still need permission from a whole lot of people if they want to enter the energy space. It’s not just regulators—there are also vendors, distributors, retailers, manufacturers, and partners that all need to be on board.

It’s tempting to think that the Mark Zuckerberg or the Bill Gates of the solar panel industry can just get hundreds of millions of people to use her “insanely great” new product, because it’s better and cheaper than everything else. That’s what happens when Apple introduces a new product and tens of millions of people have it in their hands months later. But the Internet has skewed our thinking. In the energy sector, that innovator will need to ask a lot of people’s permission first, and some people won’t want to give her permission, perhaps because they don’t grasp the technology, or perhaps because granting her permission is against their vested interests, or perhaps because they don’t really care either way and just want to go home at five o’clock. Change can happen, but it will necessarily be slower and more expensive when innovators have to keep asking permission from other people. That, too, cuts against the President’s goal.

So what’s left? What should policymakers do once they stop deluding themselves into believing that the dizzying speed of innovation in the information technology sector will be replicated in the clean technology sector? That is what the next post in this series will explore.

Old Paper by