This blog started out a year ago as an explicitly techno-optimist blog, and I feel like I’ve gotten away from that a little bit. This isn’t because I’m less optimistic about technology, but because it’s very easy to get distracted by stuff like economic policy, social unrest, China’s economy, Covid, and so on. But it’s time to go back to my roots a little here.
It’s still too early to tell whether we’ll get the Roaring Twenties that some have predicted. There are certainly reasons for doubt. After a great first quarter, labor productivity is now actually falling, due mostly to supply chain snarls disrupting production, but also possibly due to Covid variants and the uncertain and chaotic process of workers partially returning to the office. Goldman Sachs, one of the early proponents of the Roaring Twenties thesis, is now forecasting anemic growth in the year ahead.
Also, one of the most optimistic tech trends — the shift to remote work — might not be panning out as hoped. Some still expect remote work to reshape our economy in positive ways, but some new research suggests that physically separating workers can reduce information sharing, and many companies are now demanding that workers return to their desks. So the jury is still out on whether that’s going to offer a productivity bonanza.
But despite these setbacks, stumbles, and headwinds, I’m still massively optimistic for the decade ahead.
The energy revolution
Energy is fundamental to all physical technology — to move atoms around in real space, you need energy. And in the early 1970s, what had been a trend toward greater and greater energy use suddenly stopped, making it much harder to innovate in the physical realm:
We compensated for this by innovating in digital technologies (“bits”), but these are somewhat less conducive to rapid productivity growth, so productivity stagnated (this is the answer to the famous question of “WTF happened in 1971?”; it was actually 1973, and it was the Oil Shock).
The oil shocks were political, but fundamentally the energy stagnation was technological. The Industrial Revolution had been powered by a move from wood and animal power to coal and then oil, but the next transition — to nuclear — was stymied because people were too afraid of the risks. We can argue about whether that was a mistake all day (I think it probably was), but it’s in the rearview mirror now.
Fortunately, after the stymying of nuclear, our scientists and engineers worked very very hard for a long time, and came up with safe, environmentally friendly technologies that are even better in most ways than nuclear: Wind and solar. Decades of staggering cost declines — made possible at first through government-funded research, then by government-subsidized scale-up in the private sector — mean that renewables are now ready for prime time. To translate those cost declines into cheaper energy, we actually have to built the generation capacity. And we are now building it.
Bloomberg New Energy Finance has the numbers. In 2020, most electric power generation sources shrank — especially coal and natural gas — but solar and wind grew.
The International Energy Agency, which has traditionally been too conservative in predicting the rise of renewables, now forecasts that the trend will continue strongly throughout the decade:
By 2026, global renewable electricity capacity is forecast to rise more than 60% from 2020 levels to over 4 800 GW – equivalent to the current total global power capacity of fossil fuels and nuclear combined. Renewables are set to account for almost 95% of the increase in global power capacity through 2026, with solar PV alone providing more than half. The amount of renewable capacity added over the period of 2021 to 2026 is expected to be 50% higher than from 2015 to 2020.
Solar, the real miracle technology in this story, will be more than half of that.
Now, as everyone knows, renewables are intermittent; the sun is not always shining and the wind is not always blowing. Batteries can store electricity overnight, and solar can be overbuilt for the winter, but that doesn’t solve the whole problem. Intermittency means that on the margin, we can get cheaper electricity by replacing fossil fuels with solar, but as a larger percent of the grid becomes solar and wind, the cost advantages will go away. So to really keep that energy revolution going, we’ll need a lot of storage, which companies and researchers are now working on. The most promising technologies for rapid cheap large-scale storage include iron flow batteries, thermal energy storage, and hydrogen. Countries are also building more nuclear reactors — these are expensive, but will help firm up power generation.
Cheap solar, cheap wind, and cheap storage mean that we could see the first large sustained decrease in electricity costs in over half a century. People are finally starting to realize this, and are speculating about what could be done with cheap abundant electricity. I offered my thoughts here and here:
Many more good ideas can be found in the replies to Jason Crawford’s thread. Here are some of Tyler Cowen’s thoughts:
How about heating and cooling? It might be possible to manipulate temperatures outdoors, so Denmark in January and Dubai in August would no longer be so unbearable…Eventually, more flying would be automated. Robots would become far more plentiful…Cheap energy would also make supercomputing more available, crypto more convenient, and nanotechnology more likely…[T]echnologies…to remove methane (and carbon) from the air…are also likely to be more feasible and affordable.
The reality of cheap electricity — not 30 or 50 years in the future, but in the coming decade — is thus starting to sink in. This is really happening.
And amazingly, solar/wind/storage might not be the only cheap electricity-producing technology on the horizon. There has been a massive explosion of funding and startups in the field of fusion power:
Source: https://www.fusionindustryassociation.org/about-fusion-industry
Though it isn’t commercially viable yet, the field is advancing by leaps and bounds, with notable breakthroughs in multiple types of fusion technology. The old joke that fusion is always 30 years in the future is just about played out.
And as if that weren’t enough, geothermal electricity is poised for a big expansion. Check out this new policy paper by Daniel Oberhaus and Caleb Watney on geothermal’s potential, especially in the Western U.S.
So cheap electricity is on the way. The fossil fuel age that began with James Watt is about to finally yield to something — perhaps several somethings — cheaper and more abundant.
But here’s the thing. Cheap electricity is only one of the two energy revolutions that’s happening right now.
The other is a revolution in how we store and transport energy. Battery technology has advanced amazingly quickly, dropping in cost by 90% over the past decade:
Further declines are forecast in the next few years, despite the supply crunch. As always, Bloomberg New Energy Finance is the best data source here, so check out this thread by the excellent Nat Bullard:
This will obviously allow electric cars to replace internal combustion vehicles, which will be good for the environment, as well as higher-performing in many ways (electric cars accelerate much faster, are naturally much quieter, have far fewer moving parts that need servicing, and generate less particulate pollution). But the battery revolution goes far, far beyond passenger vehicles. It changes the entire way humans are able to carry energy around with them.
Some people scoff at batteries because their energy density is less than the energy density of fossil fuels. This scoffing reflects a fundamental misunderstanding of how energy portability works. The weight and volume of a system to transport fossil fuel energy also involves all the machinery required to extract the energy — the combustion chamber, the piston, the crankshaft, the valves, and all of that stuff. A battery may not be as energy-dense as gasoline, but all you need in order to extract the energy from it is a few tiny little strips of metal. This is a big part of why you have a battery in your phone instead of a gas tank.
It also means that we’re entering an age when small, light, portable machines can carry quite a lot of energy around with them. Drones are the most obvious application; small battery-powered quadcopters may change the face of warfare and offer all kinds of delivery services. High-capacity batteries will also power the robot revolution, allowing bots to move around factory floors, hospitals, and many other environments freely and giving them the power to do all sorts of tasks. And better batteries will also enable the creation of all kinds of cheap, light, portable appliances — something I first realized when I saw protesters defeating tear gas attacks with electric leaf blowers in Portland in 2020.
Finally, cheap batteries are already enabling new modes of transportation, which will change the entire way that cities function. E-bikes are the most promising technology here, and the government is preparing to spend big on them in the coming years. Electric scooters and personal air taxis are two other important examples that could revolutionize how we get around within the decade.
In other words, batteries aren’t just useful for storing energy, they’re useful for moving small bits of energy around very lightly and efficiently. And that is going to transform our physical world. It’s highly complementary to cheap electricity — whether our electricity comes from solar, fusion, or geothermal, batteries will be a big part of how we put that energy to use.
These two energy revolutions aren’t on the far horizon; they are happening right now, before our eyes. And they will define this decade.
Biotech (or, the Age of Miracle and Wonder)
At the same time that the energy revolution is poised to reshape our relationship to the physical world, an explosion of biotechnology is poised to reshape the very nature of human life. This is a decade that began with mRNA vaccines arriving just in the nick of time to save millions of people from a global plague. And mRNA promises vaccines for everything from malaria to various kinds of cancer to any number of other diseases. But amazingly, it’s only one of many bio breakthroughs that are just now coming to fruition! Here’s a short list of miracles that have recently floated across my screen, just within the last couple of months:
1) Brain-computer interface:
2) Gene therapy:
4) Neurotechnology:
Brain Implant Gives Blind Woman Artificial Vision in Scientific First
5) Animal-to-human organ transplants:
In a First, Surgeons Attached a Pig Kidney to a Human, and It Worked
6) Gene synthesis:
These advances are being driven by several breakthroughs in fundamental science: advances in synthetic biology, the discovery of Crispr and other gene editing techniques, the mRNA breakthrough, stem cells, and a couple of others. Those in turn were made possible in part by advances in computing — a case of one technological revolution laying the foundations for another.
As for how this will affect the economy, it’s not yet clear. A vast array of startups is already emerging to take advantage of all of these advances, but it’s not yet clear whether the most successful ventures will be capital-intensive projects like pharma or low-overhead synbio stuff that can be done in a basement. Moderna and BioNTech certainly showed one route to riches, but that may actually be the exception — drug discovery has been getting more expensive over time, and even mRNA and other new magical technologies may not reverse the trend. Anything that goes through the medical system has to pass through a forbidding gauntlet of regulation and testing. It’s not yet clear how many technologies will be able to circumvent that system completely and make an end run to consumer markets. It’s also unclear whether there will be a big political push for regulatory reform in order to allow American business to exploit the new technologies; the pandemic certainly made it painfully clear that the FDA needs deep reform, but that doesn’t mean we’ll get it.
Also unclear is how biotech will affect aggregate productivity. Cheap energy is likely to show up in the productivity statistics, since it always has in the past, but bio might end up being more like social media — something that transforms the way humans live our lives in deep and meaningful ways, but struggles to move the needle on productivity. (I aim to write more about the economics of biotech in the upcoming months.)
But it’s clear that the shape of human life will change, and soon. The nature of disability will change. Mortality risks will change. The nature of aging will probably change. And the natural limits of human capability will start to flex and bend in ways that will make us question the distinction between removing disability and adding superpowers.
In other words, we are finally starting to access the source code of humanity — to gain the ability to hack and rewrite ourselves. The 2020s will be a decade where we wrestle with the implications of that awesome power.
A.I., nanotech, space…
I’ve focused here on the two things I’m personally most optimistic about for the coming decade: Energy and biotech. But those are far from the only big things going on in tech!
There’s also the continuing A.I. revolution, which may finally start to bear fruit in terms of aggregate productivity, replacing all sorts of service industry tasks, vastly improving the abilities of robots and machine tools, and facilitating advances in many areas of research.
There’s nanotechnology, which has benefitted from decades of government investment, and is now starting to transform chemical engineering and materials science:
Tools that make use of massively parallel arrays of nanoscale tips are now being used to build combinatorial “megalibraries” consisting of millions of nanostructures, each with slightly different sizes, compositions and shapes…In fact, a single megalibrary contains more new inorganic materials than scientists have collectively synthesized and characterized to date.Consequently, when screening new materials, scientists have identified catalysts that fuel processes in the clean energy, automotive and chemical industries…Scientists are also using megalibraries to identify structures with important physical properties like magnetism, luminescence and high-temperature superconductivity.
And then there’s space. A new space race with China, combined with the Herculean efforts and startling advances of SpaceX, are suddenly bringing the space age back from the realm of shattered dreams and science fiction. Casey Handmer has a great blog post about how revolutionary SpaceX’s Starship rocket platform really is:
Starship matters. It’s not just a really big rocket, like any other rocket on steroids. It’s a continuing and dedicated attempt to achieve the “Holy Grail” of rocketry, a fully and rapidly reusable orbital class rocket that can be mass manufactured…Historically, mission/system design has been grievously afflicted by absurdly harsh mass constraints…As a result, spacecraft built before Starship are a bit like steel weapons made before the industrial revolution…Starship obliterates the mass constraint and every last vestige of cultural baggage that constraint has gouged into the minds of spacecraft designers…
Starship could be used for the entire Artemis program, and probably will if the program continues. Indeed, for the same annual cost Starship could deliver perhaps 100x as much cargo to and from the Moon, meaning that instead of two or three dinky 10 T crew habs over the next decade, we could actually build and launch a base that could house 1000 people in a year or two. We probably won’t, but we could.
This has the potential not just to accomplish amazing feats (build a Mars base!), but to make space development a viable commercial activity in terms of tourism, research, and possibly even mining. It also obviously has transformative military applications. And Starship is only one of the two space revolutions SpaceX is carrying out, the other being Starlink, a satellite network that could soon make the global internet far more efficient and robust, as well as circumventing local internet shutdowns by authoritarian governments.
(And notice that I haven’t even mentioned crypto; that’s a subject for future posts.)
These are the big transformative tech trends I’m most aware of, but there are probably others whose importance or potential I’ve overlooked. Feel free to let me know in the comments.
The upshot, however, is this: We are potentially at the dawn of not just one new technological revolution, but several at once. These revolutions form a sort of tree —just as industrial technology unlocked the information age, computers, the internet, and A.I. are allowing rapid progress in a number of other fields. And as these fields mature, they will open new gateways.
The 2020s are going to be a very cool decade.