Energy Bell: The sketch of an idea

How can we get America back to the forefront of energy research?

MBisanz talkBell_Labs_Holmdel,_The_Oval2.png, CC BY-SA 2.0

There are many great blogs out there these days, but my favorite is probably Construction Physics, written by Brian Potter. Again and again, Potter churns out insightful deep dives on different industries, with a focus on innovation and productivity. This week, he wrote a characteristically excellent post about the construction of the Bell telephone network:

Building the Bell System

This post got me thinking about an idea I came up with a couple of years ago, which I wanted to flesh out a bit more. I call it Energy Bell.

Basically, the U.S. needs more and better energy technology — better batteries, better solar panels, better electrical grid systems, better electric motors, better nuclear reactors, fusion power, green hydrogen and ammonia, fuel synthesis, and so on. This is a priority right now for three basic reasons.

First, this is a very special time in the history of technology. The rise of solar and batteries is promising to finally give humanity an energy source that’s cheaper and more portable than any fossil fuel. This will be the first true energy revolution in a hundred years. Countries that master these new technologies will be able to harness cheap energy in untold ways.

Second, the AI revolution is going to be very energy-hungry. Already, data centers are straining local power supplies around the world, and AI companies are planning a massive scale-up for the next generations of algorithms. That means the demand for energy is set to skyrocket.

And third, the U.S. is in danger of losing the lead in energy technology to its main rival, the People’s Republic of China. With its mastery of battery, solar, and electric vehicle tech, as well as nuclear, China is dominating the global markets for products that depend on those inputs — EVs, drones, electronics, and so on. Cheaper energy will also give Chinese manufacturing companies an additional cost advantage in all sorts of industries.

And yet despite the crucial importance of energy technology, the U.S. government devotes precious little research funding to the topic. Federal energy research has grown a bit, but is still dwarfed by other priorities such as health:

Source: AAAS

In the coming age of austerity, I don’t see a lot of scope for a huge increase here.

So if not the government, who will do the energy research in the U.S.? The private sector. In fact, over the past 35 years or so, the private sector has been taking over a greater and greater share of the country’s R&D spending:

Source: NSF

In fact, this is sort of how China does it too. A 2021 report from the Center for Strategic and International Studies shows that most of China’s vaunted R&D spending is done inside corporate labs. Of course the line between government and the private sector is very blurry in China, but the point is that most of China’s research is not government-funded academic science; it’s practical stuff being done within corporations, with the aim of making those companies into market leaders.

The economists Ashish Arora and Sharon Belenzon have been arguing for years that the U.S. needs its big corporations to do more scientific research. In a 2015 paper with Andrea Patacconi, they wrote:

Our results indicate that the willingness of large firms to invest in scientific capability has declined…A pessimistic interpretation of these results is that private research is in decline. Established companies can no longer emulate firms such as DuPont, AT&T, or Merck, whose investments in research in the past have significantly advanced the frontiers of human knowledge. Unless public funding can make up the deficit, technical progress will slacken and eventually reduce productivity growth.

In a 2023 paper with several co-authors, they argue that federally funded science has difficulty trickling down to commercializable innovation — in other words, they believe there’s just no substitute for corporate labs.

Their most intriguing idea is in a paper published in 2021 with Lia Sheer. Arora, Belenzon, and Sheer argue that big companies invest more in science when they’re less worried about rivals appropriating and using their discoveries:

We find that private returns to corporate research depend on the balance between two opposing forces: the benefits from the use of science in [their] own downstream inventions, and the costs of spillovers to rivals. Consistent with this, firms produce more research when it is used internally, but less research when it is used by rivals. As firms become more sensitive to rivals using their science, they are likely to reduce the share of research in R&D.

This would certainly fit the story of Bell. In addition to building out the country’s telephone network as Potter describes, Bell famously did much of the basic science for the modern electronics industry, and invented many of the core products we depend on today. Just a few examples include:

• the transistor (the foundation of all computing)

• the solar cell

• the laser

• radio astronomy

• information theory

• the communication satellite

• the UNIX operating system

• the fax machine

…and plenty more. But when antitrust authorities broke up the Bell telephone system and Bell Labs was sold, the flood of world-changing innovations stopped.

Contrary to Arora et al.’s pessimistic conclusion, I do think that corporate labs in the mold of Bell Labs are alive and well today, in one sector of the economy. Google’s internal labs — funded by the company’s search ad monopoly — basically invented modern AI, as well as the transformer technology that powers the new wave of generative AI and the revolutionary new methods for predicting protein folding. In fact, Google publishes more AI research than any university. Microsoft’s AI research division is no slouch either.

And AI is only one part of U.S. tech companies’ massive software research apparatus. Of the eight companies in the world that spend the most on R&D, six are American. All of those six — Amazon, Microsoft, Google, Meta, Apple, and Intel — are in the software or electronics sectors.

But where are the big U.S. companies pouring billions into energy research? Tesla has done wonders to push the frontiers of battery and electric power train technology in the past, but it’s going to have trouble continuing in this role — it’s now pivoting to AI after an increase in competitive pressure from China’s state-backed auto brands. Ford and GM spend a lot on research, but much of this is for their gasoline-powered auto businesses.

Meanwhile, Chinese battery and EV giants BYD and CATL, who do a significant amount of China’s innovation in the energy space, are powering ahead. They’re pumping up their R&D spending, with BYD overtaking Tesla. CATL claims to have flown a 4-ton airplane on batteries alone, and says that in three or four years it’ll have a battery-powered aircraft twice that size that can fly over a thousand miles. BYD just unveiled a plug-in hybrid car with a range of 1,300 miles. If true, these are absolutely incredible advances, and there’s no U.S. company doing anything to keep up in this space.

So I’m not sure if the U.S. has any big corporations who have the cash and the sense of security to spend truly meaningful amounts of money on pushing the frontiers of energy research, the way Bell Labs pushed the envelope in electronics and Google did in AI. And the problem is especially acute in the face of Chinese competition.

That means too much of the responsibility for America’s private R&D in the energy field falls on startups and the venture capital ecosystem. VC firms have been trying their best here, funding startups in fusion, solid-state batteries, and many other cool energy technologies. But the VC business model, which depends on relatively small amounts of capital being deployed for quick returns, just isn’t very suited to the task of funding energy research, which tends to take a long time and be very expensive.

It would be great if the U.S. had a giant company that could do for energy tech what Bell did for electronics and Google did for AI. But how could such a company be willed into existence? Perhaps the U.S. could create a national private electrical company.

Unlike the long-distance telephone network that Bell built, the utility industry is incredibly fragmented — electricity is generated locally, and electric power lines are local networks. (For you history buffs out there, this represents the triumph of Westinghouse’s model of electricity over Thomas Edison’s.) But with the rise of renewables, the equation is changing. Long-distance power lines are becoming more important for moving power from place to place depending on where the sun is shining and the wind is blowing. China is building out a long-distance power network; the U.S. is falling behind in this regard.

One idea for catching up would be to create a nationally sanctioned long-distance power grid monopoly. In case you’re horrified at the idea of a government-sanctioned monopoly, there’s actually a precedent here. Local power utilities are commonly regarded as natural monopolies, because it’s simply not economical to run multiple power networks through a city. Thus, utilities in the U.S. tend to be either regulated government-sanctioned local monopolies, or state-owned enterprises.

A national electricity company — an Energy Bell — would be analogous, only larger. It could also own and operate solar farms, wind farms, and nuclear and geothermal plants that are located far away from most customers.

A key feature of Energy Bell would be price controls on the electricity the company provided. For those of you who shudder at that term, remember that when a market has a monopoly, a price cap increases economic efficiency. In plain English, what that means is that if Energy Bell was limited in terms of the prices it could charge customers, it could only make more profits by one of two ways:

1. Selling a larger volume of electricity to customers

2. Doing innovation to drive down the cost of selling electricity

Both expanding service and lowering costs via innovation would be good for the country, and price caps would encourage both activities. Again, there’s a precedent here, in the regulations that the U.S. already imposes on the prices utilities can charge and the rates of return they can earn on their investments. The key here is that in exchange for a government-sanctioned monopoly, Energy Bell would have to focus on serving customers better instead of squeezing them for profit.

A second requirement for Energy Bell would be that it would have to create its own equivalent of Bell Labs, and spend a certain amount of money on research. It would already have incentives to do this — its status as a national monopoly would make it less worried that competitors would copy its innovations, and price caps would give it an incentive to find new ways to provide electricity more cheaply. But a hard mandate to spend money on R&D would be useful in addition to those incentives.

This requirement would need some other important features. Limiting the markets Energy Bell could enter — much as the original Bell was limited to telecom-related markets by a 1956 agreement with the U.S. government — would make sure Energy Bell’s lab would focus its efforts on energy tech. And like Bell Labs, Energy Bell would be required to license all of its innovations cheaply, to encourage rapid diffusion of ideas throughout the U.S. economy.

Basically, the idea here is to reverse-engineer Bell Labs, in order to propel the U.S. to the forefront of energy technology. But Energy Bell might have other significant advantages on top of this. A national monopoly might have the scale to more effectively deal with two of the main regulatory barriers to building long-distance power lines and solar and wind power in America: NEPA and the interconnection queue.

NEPA is a law that allows people to sue companies to block development by forcing them to complete long and onerous environmental reviews. Energy Bell, due to its sheer scale, might have in-house legal expertise to be able to complete these reviews more quickly, or even to get courts to reduce the requirements more often. It would also serve as an effective lobby for permitting reform.

The interconnection queue is a system that requires companies to line up to connect power to the grid. Currently, the system is set up so that each company bears a significant risk of being forced to pay for a grid expansion all by itself, if its place in line is unlucky. This would be much less of a risk for a very big company like Energy Bell than for a small local power company.

Anyway, this is only the sketch of an idea. There are many, many features of the electric power market that I glossed over, and many which I don’t even understand well enough to fit them into the Energy Bell concept. The full development of this idea would be extremely technical and complex.

In addition, there are some important regulatory questions I’ve completely glossed over here. First, there’s the question of how, exactly, Energy Bell would first come into being. Simply yelling “OK, we now allow a national grid monopoly!!” from the roof of Congress would not be effective in bringing one about. More realistically, the government could establish some system by which long-distance power providers are allowed to buy each other without triggering antitrust law. Another (perhaps complementary) approach would be to create a government-sponsored enterprise like Fannie Mae to get the ball rolling by buying up a bunch of companies that provide long-distance transmission; eventually this would be privatized.

There’s also the question of how Energy Bell would relate to local utilities, grid operators, and power providers. The original Bell was allowed to buy local telephone companies, but only with restrictions. An analogous system might work well here, but the electricity market and the telephone market are different enough that the old system couldn’t just be copy-pasted.

Anyway, the Energy Bell idea is an industrial policy idea — not a subsidized “national champion” to compete directly with Chinese companies like BYD and CATL, but a domestic service monopoly aimed at boosting U.S. innovation in the energy technology space. Unlike Tesla or other manufacturing companies, it would never be under threat from being outcompeted by China, because it would be in a nontradable industry. And as a vehicle for funding research, it has at least two key advantages over simply pouring more taxpayer dollars into the Department of Energy.

First, Energy Bell’s research budgets would be insulated from the political process, and thus not subject to catastrophic reversal every time partisan control of Congress switches hands. And second, Energy Bell’s innovations, like those of the original Bell, would be focused on tangible products and commercializable technologies — a good complement to the basic science funded by government organizations and performed at universities.

As I said, this is just a sketch of the idea, to get people thinking. If something like Energy Bell ever comes about, it will probably be in a considerably altered form from what I’ve described here. But the U.S. has a great big gaping hole in energy research, at a crucial time in the evolution of this technology. Something along the lines of Energy Bell could be just the thing we need to plug that hole.

---