Are data centers and electrification going to break the US power grid, or are they the secret to making it cheaper for everyone? In this episode, I talk with Pier LaFarge of Sparkfund about Minnesota’s landmark decision to let Xcel Energy deploy batteries directly into local distribution networks. We look past the politics and map out how a battery-saturated system can socialize the benefits of load growth, ushering in an era of boringly reliable, low-cost energy by 2030.
📌 Instructions to add paid episodes to your preferred podcast app via mobile / desktop
(PDF transcript)
(Active transcript)
Text transcript:
David Roberts
Greetings, salutations everyone. This is Volts for May 20, 2026: “Sooner than you think, electricity is going to be cheap, abundant, and boring.” I’m your host, David Roberts.
Late last year, up in Minnesota, the power utility Xcel Energy proposed a new program called Capacity*Connect that would have it deploy 200 megawatts of battery storage — not in the form of large utility-scale installations, but in smaller chunks distributed throughout the grid, placed atop commercial and industrial buildings like big box stores. It didn’t draw much attention outside the energy world, but my guest today believes that it augurs a fundamentally new chapter in the history of the US electricity grid, and for the US economy more broadly.
Back in 2024, I talked with Pier LaFarge, the founder and CEO of Sparkfund, about his vision of a utility-led distributed energy expansion. Now he’s back to make an even stronger version of the case: what Minnesota foretells is that utilities are going to start deploying distributed batteries under conventional utility cost-of-service rules and those batteries are going to solve all our grid problems.
They will help utilities better utilize their currently wildly underutilized grids, which will in turn enable the addition of data center load that will reduce, rather than raise, electricity prices. Over time, with the addition of these and other batteries to the system, US electricity will become cheap, abundant, and boringly reliable.
It is, to say the least, a provocative case. Lots of people object strenuously to the notion of utilities owning distributed energy. Lots of people are skeptical that storage can really, in practice, substitute for more long-distance transmission and more generation. Lots of folks think raw materials or geopolitics will impose limitations on the number of batteries we can access. Lots of people think all kinds of things, and I’m going to talk with Pier about all of them today. I’ve been looking forward to this.
With no further ado, Pier LaFarge, welcome back to Volts.
Pier LaFarge
David Roberts, thanks for having me. Great to be here.
David Roberts
I’ve been thinking about this episode for a while, thinking about how to structure it, and I think in my mildly neurodivergent adjacent way, I have come up with a structure for this episode that makes sense to me, may or may not make sense to anyone else, but I’m going to lay it out for you and then we are going to do it.
Pier LaFarge
I can’t wait.
David Roberts
We’re going to start with just the facts of Minnesota, what’s happening in Minnesota, just the structure of it, and then from there we’re going to vault all the way up into the stratosphere and talk about batteries on the grid. What are the benefits of having a crap ton of batteries on the grid during a time of load growth and electrification?
Then I want to back out from that. What are the benefits specifically of distributed batteries? Why might we want those batteries to be distributed throughout distribution networks rather than clumped up in large utility-scale transmission-based facilities?
Thirdly, insofar as we do have some preference for distributed batteries, why is the way Minnesota is doing it the way to do it as opposed to other alternatives? For all of those out there who have objections and complaints about this Minnesota program, we are going to get to them at the end, but I want to work my way backward to it so we understand what we are aiming at. Let us start with Minnesota. Tell us, what is the Capacity*Connect program?
Pier LaFarge
In Minnesota, the largest utility there is Xcel Energy, as many of your listeners know, and what Capacity*Connect was, was a program brought forth by Xcel and filed as part of their Integrated Resource Plan, which is their plan for how they’re going to grow the grid in response to load growth and grid needs. They put an item in there called Capacity*Connect. It’s an idea that Sparkfund helped develop and bring to market broadly as a distributed capacity procurement, which is pretty straightforward. It’s just what it sounds like.
It’s a procurement of a bunch of distributed capacity because in Xcel and the commission’s view, the grid can benefit from several hundred megawatts of batteries that can charge up when there is plenty of energy on the grid, dispatch when there is peak and constraint, and in doing so help operate the grid more efficiently, sell more electrons over that grid, and make the grid stronger, more reliable, more capable of handling intermittent renewables, and also make power cheaper if there is someone to buy it.
Minnesota Capacity*Connect was a 200 megawatt regulatory filing that was approved in a historic 5-0 vote several weeks ago at a commission hearing.
David Roberts
That’s significant. We should pause on that. That’s not taken for granted at all. This was a long discussion and contentious.
Pier LaFarge
No question, and received quite a bit of pushback from, in particular, the distributed solar industry. But at the end of the day, it was exciting to see a commission ultimately evaluate a proposal to make batteries — again, to use your word — a relatively boring part as the next generation of distribution grid infrastructure. They said, “Look, batteries have these properties that the grid really wants. We can use them to benefit every ratepayer. Let Xcel invest in them, operate them to the benefit of the grid.”
Ultimately, that’s no different than transformers or substations or capacity banks. It’s just the next type of distribution grid infrastructure.
David Roberts
This, I think, is a key feature to understand here: these are not Genco. Xcel is a restructured utility. It is only a distribution utility. It does not own generation assets. This filing treats batteries as distribution infrastructure, the same as if they were building transformers or power lines. Xcel is going to own the batteries. This is crucial. Talk about the financial arrangement.
Pier LaFarge
One correction. Xcel is vertically integrated and they source their power through competitive RFPs from the market. They’re part of MISO, which is the regional operator and power market. They are vertically integrated. But it doesn’t change the importance. The reason that this decision is historic is regardless of the regulatory structure or market structure of Minnesota, this is a commission saying, quite simply, that batteries fix many of the problems on the grid. It helps us utilize it better. It makes it more reliable. It makes it better at handling and delivering intermittent renewable power. It makes it cleaner.
Most importantly, this is a commission saying, “Look, putting batteries all over the grid, hosting them at customer locations, paying those customers to rent their land.” There’s another interesting element of Capacity*Connect, which is that instead of a complicated bill savings or a customer taking out debt, this is a model that lets customers and community groups all over Minnesota benefit from the wealth of infrastructure build that’s used to support a growing grid.
David Roberts
To clarify, the utility owns these batteries, so it goes to a facility, a big box store, and says, “Can we rent your roof?” Just treating the roof as a piece of land, just as if it went and put a field of batteries in some greenfield outside of town somewhere. The financial structure is basically the same. You’re just renting a piece of land.
Pier LaFarge
That’s right.
David Roberts
The only financial deal that the big box store owner is involved in is he just gets a check every month, a rent check. He is not involved with the energy, does not own it, does not operate it, does not mess with the batteries. The batteries are owned, operated, and maintained by the utility.
Pier LaFarge
That’s right. We’re renting sides of parking lots, back fields, unused land. It’s a few thousand square feet. Think about a church. One of our first hosts in Minnesota is going to be a local church. That church is going to get real money every month for 20 years to help them run their community programs and operations. Participating in infrastructure wealth is best when it’s unburdened by asking that customer who — look, this is a church. Yes, they do a lot of things well in their community, but they don’t have to think about it.
David Roberts
They probably don’t have an energy division.
Pier LaFarge
They’re not meant to be taking complicated bets on long-term energy arbitrage or regulatory structures or making sure that their price of debt is covered by their revenue. These are people who wake up every day and participate in their community and go to their jobs and focus on what they do best. The utility in this case shows up, rents land, gives them money that improves their core business, and then the grid gets stronger and more capable of supporting growth.
David Roberts
When we talk about chunks of batteries, give the audience some sense of scale here. How big is a typical utility-scale battery field? Where is the line as you go down where you cross into “distributed”? Is there a line? Is there an accepted threshold?
Pier LaFarge
Most utility-scale batteries which are connected to the transmission system and help smooth out the timing of power are, call it, 150 megawatts to 300 megawatts. But even those big utility installations are built out of a lot of the same sized one- to two-megawatt batteries. They’re in shipping containers.
David Roberts
It’s all Legos.
Pier LaFarge
Stacking little Legos in a field and then they inject all at once and they look like a power plant from the perspective of the grid. Utility-scale batteries are fantastic. They are absolutely part of why batteries are going to transform the grid. Utility-scale batteries are the first to scale and a critical part of what I think you’re framing in that. But these distributed batteries, they are one to three megawatts per location as filed. They are shipping container batteries. They are prepackaged, they are engineering templatized, they’re standardized. You get a lot of the critical cost efficiencies that you get from a utility-scale battery. You’ve just broken it into little pieces.
Two hundred megawatts is a substantial utility-scale contribution in aggregate. But it’s put exactly where the grid needs it most at the local level. The utility and Sparkfund are working to determine where on the grid these batteries will provide the best value to ratepayers. When you incorporate them into the grid and operate them 100% for the benefit of the grid, that is just the same as a utility investing in a new substation or transformer that ultimately delivers that value to society. That’s why this commission decision is so historic.
This is the first moment that hundreds of megawatts of distributed batteries are going to be incorporated on the distribution grid as normal grid infrastructure. We believe it’s the largest such deployment and largest such approval. It really is a big step in the entire United States. Last year in 2025, the country installed just shy of 200 megawatts of these middle-sized commercial industrial battery installations, call it 191 megawatts in the entire country. To see one state put a 200 megawatt deployment into play is a huge victory for what batteries can bring to the grid.
David Roberts
That’s Minnesota. There are lots of questions about the model and about different models, but for now we’re going to leave that there and we’re going to rocket up into the stratosphere and then we’ll slowly parachute back down here to Minnesota.
Pier LaFarge
I’m ready.
David Roberts
Let’s jump to the big question, because this is why this all matters. Let’s talk about our shared vision of a grid with crap tons of batteries on it. Let’s put aside for now the question of where the batteries are located, how big the batteries are. Let’s just talk about a grid with cumulatively a crap ton of batteries on it. Why do we want to get there?
Pier LaFarge
The simplest way to put it is batteries are electron time machines. Why does that matter? Why do you need an electron time machine? We built our grid to peak, which means we built the grid to be big enough to handle the tightest, highest demand, 50 to 100 hours of the year. That wasn’t a mistake. A lot of people hear that and they say, “Oh, you overbuilt the grid.” No, you built the grid to be the size that gives us the economy and the society. We have the reliable power, the cheap power. We built the grid to peak.
It’s worth noting that America has the cheapest, most reliable power in the world. It’s easy to forget in these discussions that this thing works on a delivered basis and a reliability basis and a cost basis. We built the grid to peak and by necessity of the technology available. Big centralized generation, transmission, distribution. That meant that for most of the hours of the year you were almost by definition not using much of the grid you just built.
David Roberts
A familiar problem. Same with cars. Same with a lot of big systems. It’s something that you don’t really notice until you start noticing. With cars, you just look around, you’re like, “Oh, most cars are sitting most of the time. Most roads are unoccupied most of the time.”
Pier LaFarge
Exactly. People buy cars so that they can go to the grocery store, but they don’t go to the grocery store every hour. It is very important that when you need to get somewhere, you have effective transportation.
David Roberts
I’ve been thinking about it. It’s interesting. We built electricity to the peak, meaning we have made a social contract saying, “You can use as much as you want and we will provide it, period, full stop.” We don’t do that with highways or healthcare. With highways, we do restrict capacity. There are going to be times when there’s congestion and you can’t get to your destination in your preferred amount of time. We do restrict supply in almost every other system in our society.
This is a unique and interesting and different kind of system where we’ve just pledged, “Use as much as you want whenever you want, and we’ll build a system big enough to make sure it’s there when you want it.” That’s a remarkable thing.
Pier LaFarge
It’s a historic triumph, and I think it’s core to this discussion.
David Roberts
And a populist thing. I think this is what you’re getting at, but an incredibly egalitarian and populist thing.
Pier LaFarge
Absolutely. I think it’s one of the great triumphs of American infrastructure. There are several things in our society that we deem to be, to put it formally, inelastic, which means people deserve to use as much of them as they need and not be limited in that quantity. Water is one of those things. You wouldn’t have a water system that said, “Oh, sorry, it’s hard for you to have drinking water now or you just can’t take showers this time of day, sorry, let’s do pricing to make sure only rich people get the water.” That would be terrible. We would see that as a real political —
David Roberts
Although that is the way it works in many places —
Pier LaFarge
In many places —
David Roberts
Egalitarian, that we don’t do it that way.
Pier LaFarge
Yes, sir. This is again why I count it as a uniquely American infrastructure triumph. We live on a big abundant continent with a lot of resources, and we have, as a society, made some dramatic choices. You can code them however you want, but they are egalitarian choices to say we are going to have enough power, enough water, clean air. We are going to make investments as a society with a cost to deliver things that help human flourishing and keep people safe, happy, and productive so that we can achieve that national goal.
Power was one of the early — it was before we invested nationally in our water system — we did this in the 30s. Even more than the commitment to have enough power for society’s needs and growth to build a great nation, we also made an incredible decision of rural electrification that said you have an obligation to serve. If you build a house or a cabin anywhere in the territory of a US electric utility, that utility is obligated by law, chartered in public purpose, to deliver power to that cabin, string a line, cut down trees, get up the driveway. That is just a staggering and now taken-for-granted part of this conversation that shapes every aspect.
David Roberts
One of the things I always say to people is try to imagine if we had a similar social compact around driving. If you want to drive, you should be able to drive straight to your destination without slowing down whenever you want to go, wherever you want to go. Imagine the scale of highways and roads that we would have to build to accommodate that. If we just told people, “You can do it whenever you want,” and then if you did that, people would immediately start thinking, “Maybe we should figure out ways of persuading people to drive at different times so that we don’t have this enormous peak and we don’t have to build this elaborate roadway structure.”
Segueing back to electricity, we’ve promised everyone electricity whenever they want it. Consequently, we have built a system that will provide the peak — as you say — the most anyone could want at any moment. Within a year, we’ve built that much infrastructure, and now we have a bunch of infrastructure that’s not being utilized. This gets back to the grid utilization discussion. We’re utilizing our grid at a rate of about 50%. Fifty percent of all this money and investment we’ve plowed into this infrastructure is sitting idle because of this social compact. This is where batteries come in.
Pier LaFarge
That’s right. We’ve deemed electricity to be an inelastic good, a natural monopoly, something that we want as much as we need for the purposes of our society. We’ve regulated it to make sure that the way in which that’s done is balancing all of the real tensions between cost and availability as you described.
The job of a utility regulator is to determine, in your metaphor, how many highways are enough to let society have the quantity of this good it wants. One further digression you can remove if you want is driving is in the category of elastic goods. We think it’s okay politically and societally to limit how much people drive. It is not okay to limit how much water they can get out of their faucet or how much electricity they can put into their house.
David Roberts
One final note, this social compact about electricity is incredibly significant and important, but is only going to get more significant and more important as more and more of the services we use in our day-to-day life are run on electricity. Eventually, as our entire economy is run on electricity, then all of a sudden you can have as much as you want, whenever you want. It really starts to seem like a much bigger and more consequential pledge. Then you really have to start worrying about overbuilding. If everything is electric, the amount of overbuilding that you would have to do if you didn’t care about peaks would just be insane. We have to start caring about peaks again. This is where batteries come in.
Pier LaFarge
That’s right. From our daily lives to national security to critical infrastructure, the grid is going to be one of the great foundational inputs of this next chapter of our economy, as it was in the last.
Batteries have this unique superpower which, as I said before, they’re electron time machines, which has one simple function: charge up when there’s plenty. In those periods of underutilization, when the grid is not being stressed, you can charge a battery, you fill it with electrons, and then you can wait and move them in time until the grid cries out for relief and you discharge that battery.
When that battery is in the right place, it, at a physics level, relieves the constraint on that grid. It meets the peak in a more dynamic and locationally valuable way. There you go, you can sell more electrons over the same wires.
David Roberts
Let’s talk about that, because this is something that people break their brains on. It’s really important to understand. With — this is an important caveat — with sufficient batteries in place, added load can reduce rather than raise prices. Furthermore, and I’ll make it more specific and more provocative, we could accommodate all these data centers and they could lower rates. That is a possible world. Explain how and why.
Pier LaFarge
This is why the whole debate, which is now so politically hot around electricity, is going to end up being pretty boring. Cheap batteries and data center demand happening at the same time, plus some normal building out of new transmission and generation — we’re going to need some of that too — but those two things happening at the same time pretty much solve the grid. The cost of electricity is downstream of a pretty simple formula in most places, which is, how much did you spend on your wires to get the power to people? In most cases, that’s 60, 70, or even 80-plus percent of all the cost of power. Generation is just a smaller piece.
It’s the wires. Simply put, you spend money on the wires and then the price of electricity is how many units of your product can you sell across those wires? If you can invest in a technology like batteries that fundamentally at a physics level lets you sell more over the grid, then the only other thing you need, David, is someone to buy those electrons. Which is why data center demand and demand from electric cars and home electrification are such good news. The more we buy over the same wires, the cheaper power gets for everyone.
David Roberts
The premise here is, and I think it’s important to surface this: additional load can bring down average prices, as long as adding the additional load does not require you to add a ton of new infrastructure. If you’re adding more load but using the same fixed infrastructure, you’re spreading the cost out over a wider customer base and thus per-unit costs are lower.
Pier LaFarge
Exactly.
David Roberts
If you add load, and in order to meet that load, you have to build a bunch of new wires and transmission and generators, then you don’t get that effect. This effect only is in places if you’re adding load, but without having to add new infrastructure to carry the load.
Pier LaFarge
Or at least while you are having to, you have to do both.
David Roberts
You’re adding batteries, which are technically the infrastructure, but they’re much cheaper than wires and generators that they’re almost different in kind.
Pier LaFarge
They have the effect of letting you sell more over the grid you have. I think that ultimately we are going to need new transmission, new generation.
David Roberts
I wanted to ask about that. This effect, adding load reduces prices, presumes slack capacity on your system, right? You’re adding load to soak up that slack capacity. That’s what the batteries allow you to do. Whenever you have generation capacity, you just put it in the batteries and then absorb it whenever you need it. New load absorbing slack capacity means that you can add new load and bring down prices, right?
Pier LaFarge
That’s correct.
David Roberts
The question then becomes how much of that slack capacity. I want to give people a sense of scale. How long can you do that before you eventually have squeezed out the slack capacity and are forced to build more generators and more transmission lines? Because you need fundamentally more capacity. How much runway do we have before we need to build new stuff anyway?
Pier LaFarge
It’s an important question to ask for two reasons. To level set, we have a grid that’s about 1,200 gigawatts of nameplate. There are a lot of different ways to approach this and we’ve done some studies on it. Utilize Coalition has done some studies on it, which Sparkfund’s a part of. Horseshoes and hand grenades, it’s 200 to 400 gigawatts that you could get from increased utilization and you’d end up putting a couple hundred billion dollars of downward pressure on rates. It’s significant. It’s more than the very near-term demand from data center growth, but it’s not as much as the long-term growth of electrification, which is now becoming fundamental to every aspect of our daily lives, our economy, our data infrastructure.
David Roberts
This gives us, call it 10 years, call it a decade of runway where we can accommodate all this data center rush and lower prices without building a gigantic amount of brand new infrastructure, which we can’t do in the short term anyway, as we know. But the larger electrification of the economy — never mind data centers, as you say, everything’s coming onto the electrical grid — we are eventually going to squeeze out that slack capacity and we are eventually going to have to build more, more, more. We are going to have to expand.
The reason I’m making this so explicit is that, as you probably are following, there’s a weird very dumb fight happening in the green community between the utilization people and the permitting reform people. The people who say, “Hey, we need to squeeze out this slack capacity,” and the people who say, “Hey, no, we need to build more, more, more.” If you get anyone in a room and sit them down and talk to them reasonably, everyone’s going to agree. You need both. I just want to frame this.
You acknowledge we need both and you’re mostly pushing this as a short- to midterm strategy, right?
Pier LaFarge
That’s right. It’s not just eventually we’ll need to build more. It’s starting right now. We need to build as much transmission and as much new generation from gas to wind to solar to industrial — geothermal. We need as much as we can. The reason we should do that is because, exactly as you said, it is hard and slow to build large centralized infrastructure projects. It takes five, seven, sometimes 10 years or longer for transmission. Utilization is a way to maximize the total energy abundance of the US electric infrastructure today to deliver on near-term growth and to do that in a way that specifically puts downward pressure on rates and makes power cheaper.
Oh, and by the way, invest in a bunch of machines — like batteries and grid flexibility, demand response, grid-enhancing technologies that help the grid be stronger, more reliable, better able to resist weather, and better able to deliver intermittent resources. It’s a good thing to do now, it’s a good thing to do as much of it as we can. But we need to be in parallel building as much new American power infrastructure as we can to win the future.
I’d also like to point out that the standards by which utilities determine the engineering, safety, and requirements of the grid do matter. We do not want the grid to melt. I often say to people, before you cast shade on the safety requirements of our public infrastructure, imagine you’re a parent and you have a car parked under a transformer or you have a house in a neighborhood. If that transformer melts because those standards were not robust enough and that thing turns into liquid metal and drops on a car or on your neighborhood and you hurt people, you start fires. What if your kid was in that car?
Think like a parent, not an energy nerd, and take a breath on this endless optimization of infrastructure. Remember that some of the reasons we build these systems the way we do is because we want to keep people safe and we want a big economy that can grow and invent the future. Energy is for society. It is not just an optimization challenge of a bunch of very bright nerds.
David Roberts
I want to emphasize this point one more time. I know there are people who are concerned that the current focus on affordability among Dems is pushing them toward —
Pier LaFarge
Degrowth and scarcity?
David Roberts
Degrowth and scarcity-coded solutions.
Pier LaFarge
Yeah.
David Roberts
It’s important to surface what you’re saying here: you can shovel all the data centers on the grid you want and bring down rates and serve affordability. These are not in tension as long as you have a crap-ton of batteries.
Pier LaFarge
That’s right, that’s it. Batteries and data center demand solve it. The tragedy, David, is not only that politics of degrowth and scarcity are toxic and tragic, it’s also just wrong on the merits. You have this opportunity to have more and make things cheaper, and we need to take it. It’s responsibly managed growth.
David Roberts
The thing is, the Democratic Party, Pier, it causes me such — they are on the right side of history here, if they would just argue for their side. But you look out at the landscape and everything they are doing is everything they are doing in the name of affordability begins by conceding the premise of Republican attacks. They are rolling back energy efficiency programs now. We have been arguing for decades that those things lower rates and now you are going to concede to Republicans that they are raising consumer prices? That is your savvy affordability play? It just drives me mad.
Pier LaFarge
I can’t get pulled down that road.
David Roberts
To summarize the point you’re making here, when you have a crap ton of batteries on your grid, it solves the affordability problem, allows you to bring in new load in a way that uses slack capacity, doesn’t require a bunch more building, and reduces per unit costs, reduces rates for everyone. This is the future that you’re looking toward: a future where the grid is filled with batteries and consequently it’s boring.
It just works. It’s cheap. You get electricity for pretty cheap. You get as much as you want, it works, it’s reliable. Then, hallelujah, you can stop thinking about it and start thinking about other things you want to do with your life.
Pier LaFarge
Welcome to the age of boring. Cheap batteries, data center demand, and a few other good investments in the US electric grid will bring us back to where we have been for much of the last hundred years, which is the grid is public infrastructure we don’t have to think that much about. We can go about our lives, build our economies, have our infrastructure, and win the future in a way that I think is really compelling.
David Roberts
This is the vision of a grid with a lot of batteries on it, a grid that goes from no storage to a lot of storage. Fundamental change allows accommodation of a new load without raising prices, allows you to bring prices down, solves these reliability issues. With smart inverters, you get your ancillary services, you get your black start, you get your frequency regulation, you have a very safe, electronically software-controlled, stable grid when you have a crapload of batteries.
The second question I want to address then is, are batteries batteries? Is any battery on the grid as good as any other battery on the grid? Or is there any reason to prefer distributed batteries over utility-scale batteries?
Pier LaFarge
That’s a great question and I’ll answer it in two ways. One, it’s important to make the point that my view is that we are going to put batteries everywhere and of all sizes. Big ones in fields, medium-sized ones — which is what Sparkfund does for a living — in parking lots and in fields near churches and Walmarts and stuff like that. Then you’re also going to have residential batteries in the home and garages. You’re going to have embedded batteries in heating and air conditioning, like what Carrier’s working on and others.
David Roberts
Let me put an exclamation point on that because it’s huge and I think a lot of people don’t see it coming, which is that ACs are going to start embedding batteries and there are a lot of AC units out there in the world. Once your HVAC has a battery in it, your stove has a battery in it, your water heater will probably have a battery, and you’re going to have batteries in every device, in every appliance, in every house, in every garage, in every car, in every field, in every parking lot, whatever.
Pier LaFarge
I would point out, David, we already have them in every pocket and on every desk. It’s not like the battery revolution we’re pointing to is some science fiction future. We’re carrying them around.
David Roberts
Let me ask you, why do that? Why not just put them all in utility-scale installations in a big field somewhere and not mess with the rest of it? What performance or social advantage do you get from the distribution of them?
Pier LaFarge
There are two big groups of batteries. One are batteries that are big enough and have specialized telemetry that utilities can see and operate in their control room today as true grid assets. Utilities have an incredible amount of data about power plants, what a power plant is doing every second. Same with utility-scale batteries. Batteries that are, call it, one megawatt and bigger, you can afford to build that telemetry onto them and you can show it and connect it into a utility — they can operate it in their normal course today.
That is a big part of Capacity*Connect in Minnesota. These are utility-controlled assets because they have the data quality and telemetry that allows a utility to integrate them into their control room. When a utility can see a battery that clearly, it can dispatch it and it can accredit it, give it credit for the capacity it is creating on both the distribution and transmission system. It knows that it is not going to trip any thermal limits or voltage.
David Roberts
You can plan around it, you can rely on it.
Pier LaFarge
Yeah.
David Roberts
Because it’s yours.
Pier LaFarge
If you can’t see something, you can’t use it in high critical environments like the grid. Stuff melts, fires get created. That accreditation allows utility-scale batteries and the distributed capacity components — the missing middle — that Sparkfund is helping fill with these distributed capacity procurements. Those batteries will go first because they’re the easiest to bring into the core of grid operations as they stand today.
But the batteries we’re putting near buildings have a particular characteristic. They can be sited on the distribution grid closer to the infrastructure that needs it most. The closer the battery is to the load that it’s serving and the infrastructure it’s using, the more it can directly benefit that customer and that infrastructure. If you have a residential neighborhood filled with houses that have EVs, electric cars, and a bunch of air conditioners that are now Toyota Priuses, they’re hybrids. As they spin up and it gets cold in your house, it’s coming from a battery that charged up while there was plenty.
It’s the same metaphor. They charged up when there was plenty, when no one was using their electric appliances in that neighborhood. When everyone turns them on at once, they all self-serve. The proximity of a battery to load is, in some ways, a direct relationship to the value it has to alleviate that load on the grid. That’s why air conditioners are a great way to think about this because it’s like, put the battery inside the air conditioner and it can directly alleviate the peak stress that that air conditioner puts on the grid.
Same with the Capacity*Connect batteries. You put them in a commercial industrial feeder or a part of the grid that has a bunch of coincident load and needs that stress relief. You put a battery there and when the grid gets crowded and congested, you turn the battery on and alleviate that stress at the level of that distribution node or feeder.
David Roberts
To rephrase this, people talk about the battery stack, the value stack of batteries, the different values they bring, the different benefits they bring.
You could say that utility-scale battery installation in a field somewhere has certain benefits, but if you break them apart and put them on the distribution grid, you get the same benefits and then more benefits stacked on top. You get performance benefits, resilience benefits on top of the basic battery benefits.
Pier LaFarge
That’s right. Something I think our industry needs to get more disciplined about doing, and something we’re working on broadly at Sparkfund, is mapping out the distinct revenue streams or value streams — not all of them are revenue — but value streams that batteries and things like demand response and grid flexibility can provide. There are, depending on how you splice it, seven or eight of them. We need to make a list and just stack up which ones each company or each type of battery thinks they provide.
David Roberts
You’d think that would have been somewhat formalized already.
Pier LaFarge
Even worse, David, I think we’re really suffering from a bunch of companies whose equipment does one, two, three, or four of those things, coming in and just talking about all of them. Then utilities and regulators and legislators get a little confused and sometimes feel a little annoyed when they realize, “Wait, we’ve been spending a year talking about this and it doesn’t do the thing that I need it to do.” They’re like, “Well, no, of course it doesn’t. But if we changed a bunch of variables and had a ton of new software, then it could in the future do it.” People are just disappointed because they’re like, “I have a very urgent problem.”
David Roberts
I did a pod a couple of weeks ago about an attempt to do something like that for VPPs, a rating system — like how close to an actual power plant is it? I think something like that for batteries too. Of the stackable value streams that batteries could potentially bring, how many are manifest in this installation?
Let me ask you then, because Southern Company — listeners will be familiar — big utility down in Georgia, notoriously vertically integrated, jealous of its territory, unenthused by climate activism, etc. They have gotten religion on batteries, but they are not distributing them. They are clumping them up in giant utility-scale installations. But it sounds like they’re getting a lot of the benefits out of them.
As you mentioned, you now can’t sell one of your distributed capacity procurements to Southern because the amount of batteries they’ve already installed has reduced that arbitrage opportunity to the point that you can’t squeeze a lot more out with distributed batteries. Talk a little bit about Southern Company, their experience, and then just tell me, if I’m an IOU executive, a Duke executive or something, and I’m looking at Southern Company and I was like, “Well, they just are building and owning large-scale battery facilities and seem to be getting all the benefits they want out of that. Why should I futz about with distribution when I could just do that?”
Pier LaFarge
To put some numbers to it, in one of their integrated resource plans, they had something like 50, 60 megawatts of batteries over the next bunch of years coming onto their grid.
David Roberts
That was just a few years ago.
Pier LaFarge
In this latest update, they signed a bunch of contracts with large load customers and data centers which will put billions of dollars of downward pressure on rates for everyone who lives in Georgia. They are making power cheaper every time they sign a new data center. That’s the thing. If you just take one thing away from this podcast, every utility, every state who signs a big data center will end up with cheaper power if that load is managed correctly.
David Roberts
If they’ve got a bunch of batteries and they manage the load correctly, which is a substantial if.
Pier LaFarge
To your point, that’s why this example with Southern is so compelling, which is this is a utility who is chartered in the public interest to spend money to invest in the grid to deliver the lowest cost, most reliable way to meet growth. Growth shows up, data center asks for power, utility planners go to work and it turns out right now, based on cost and value, the answer is batteries. The reason they’re all utility-scale is that the grid in the Southeast is most constrained by transmission. It is not very constrained by distribution.
They put the batteries closest to the problem they have and the customer they want to serve. Batteries are just as good at charging up when there is underutilization and charging from a gas plant or a coal plant as they are from wind and solar. Batteries just smooth out electricity and time. That is why I call them time machines. You end up getting more out of the grid you have, dividing the cost of the wires, and you sell it to a data center and everything gets cheaper and better per our age of boring framing. Southern Company is a great example of a utility, a large IOU that basically did this overnight.
David Roberts
Wait, you forgot to give us the numbers. They had a few. I think it was like 2020. They had like 50 megawatts of batteries in their IRP. What is the latest?
Pier LaFarge
It’s 6,500. It’s 6,800 — thousands. 6.8 gigawatts of batteries or so. I forget the exact number.
David Roberts
Not a small — that bespeaks someone inside that utility having a revelation of sorts. You don’t go from 50 to 6,000 with somebody inside the —
Pier LaFarge
Yeah, it’s 100x. We spend so much time in this industry, David, talking about how utility incentives are broken. The thing that was missing is load growth. Utilities have been sitting there the whole time saying, “Thank you very much, I’m incentivized to spend money to build infrastructure for America. My incentive is build things society needs that the regulator says is the lowest cost, most reliable way to do that.” When these machines get cheap enough and when load growth comes back and creates the need, a utility simply does the math and delivers the outcome.
David Roberts
A utility, not all utilities, I guess. Southern Company is the iconic — if people come up with examples of who’s going to be the last to get on something clean or something progressive or whatever, they cite Southern Company. The fact that this revelation has happened inside Southern Company is quite significant. Should we assume that other utilities are going to be like, “Well, if Southern...” This is not Xcel. This is not California. If Southern is doing this, surely it must be the obvious thing to do. Is this going to unlock a similar 10x in other utilities?
Pier LaFarge
Age of boring, David. There are other examples — Indiana Michigan AEP just announced that with data centers and a bunch of flexibility and batteries, they’re going to lower rates for everyone. PG&E said every time we get a new gigawatt of data center load, we’re going to drop power prices by 1 to 2%. The ones we’ve already signed are making power 11% cheaper than it would have been. Now, California, you’ve got a lot of wildfires to pay for, so it may not get actually cheaper, but California’s got its own problems.
David Roberts
Yes, but the data centers are pushing down prices because of the batteries. We can’t say that often enough because it bounces against people’s instincts.
Pier LaFarge
That’s right. We’re getting the politics and the grid discussion of data centers’ impact 100% wrong. We’re getting it backwards. Every state that does not get data center growth is going to have an aging grid they have to pay to replace. The cost of the wires is going to rise. They’re going to be divided over too little new growth and power will get more expensive.
Power prices have been going up mostly because of distribution costs for the last bunch of years. People say that all the time, but they don’t then subset what are the inputs into that distribution cost. A big chunk of it is that some of this stuff was installed in the 60s and 70s and entropy always wins the argument. You got to replace it. It’s old, it’s breaking.
The second piece is that we’ve had a bunch of renewables come onto the grid that are cheap to generate but require the grid to take the shock of intermittent power when it gets very windy or very sunny. We’ve made some investments in things — like capacity banks — to handle that.
David Roberts
Affordable batteries at scale are a new thing.
Pier LaFarge
That’s right.
David Roberts
That is a new category of distribution infrastructure that is now available to utilities.
Pier LaFarge
Batteries, because they are products, because you can put them in shipping containers and standardize the design and build millions of inverters and cathodes and anodes and power electronics and factories, you can drive the cost way, way down. Just like solar panels, the cost of a battery since 1999 has gone down by 99.5%. In the last five years, they have come down another 85%.
David Roberts
Here we are looking at this future of massive load growth. Forget data centers. Beyond that, load growth to the horizon. Transportation, heating and cooling, industry load growth. Load growth, load growth, load growth. Here we have in our hands a tool that renders load growth a mechanism of reducing rates. Here we have load growth and batteries in hand — abundant energy for cheaper.
Pier LaFarge
Amen.
David Roberts
Just going to emphasize that one more time. We’ve discussed the virtues of having a crap ton of batteries on the grid and discussed when and where there are merits in distributing them versus clumping them. Is your congestion problem on your transmission grid or is it in your distribution grids? You want to clump the batteries near the problem, and in many places the problem is distributed and you’ll clump the batteries in a distributed way. Assuming then we want some distributed batteries, let’s talk about the merits of the way it’s happening in Minnesota specifically.
The first thing I’d ask is how committed is Xcel to this approach? I know the PUC told Xcel, “Hey, you should also study the VPP model that they’re doing in Colorado.” This is one approach still, among others. A lot of questions about this, maybe the first one is just compare this to a VPP. Why might this be preferable to a VPP? How do you view this in relation to VPPs?
Just the distinction for anybody who is not following it: these batteries that we are talking about in Minnesota are owned and operated purely by the utility, purely for grid benefit. They are not operated for the benefit of the customer who is hosting them, except insofar as everybody on the grid benefits from grid stability, but they get no specific benefit.
Whereas a VPP is taking behind-the-meter, not in front of the meter, but behind-the-meter resources, and doing this dance where they’re primarily operated for the benefit of the building owner and then secondarily operated for the benefit of the grid. That’s what a VPP is. I’m curious your thoughts about how these two relate and the relative merits in the world we’re talking about.
Pier LaFarge
The best distinction is there are true grid assets that are 100% operated for system benefit. Every value dollar that’s created for the grid gets socialized to everyone. Then there are assets that are paid for by customers, and therefore the customers have a right to get benefit from them. Those assets primarily serve the customer value. But then also if they can be used for the grid, the grid should pay them. That’s the difference between the VPP program in Colorado and the Capacity*Connect program in Minnesota. In both cases, this is Xcel bringing an innovative proposal to their regulator and saying, “Here’s a thing we could try to make the grid more flexible, better utilized.”
David Roberts
It’s Xcel in Colorado too? Somehow I hadn’t put together that it was the same.
Pier LaFarge
Same utility. They’re trying different models to learn which one delivers the best benefit. When utilities — this industry is very quick to point out when utilities don’t do things that they like, but when they do things that are innovative and forward-leaning, they end up attacking them anyway for it. Xcel was attacked in both states for both of these proposals. Then they were attacked, comparing the two to each other, and it just, it’s so bad and everything is broken and people say monopoly and forget to say public purpose, regulated monopoly.
Are utilities perfect? No. But the regulated compact in which they are allowed to invest private capital for public purpose to grow a grid that supports our economy is a pretty good system that really does matter. This is a system operator coming to their regulator in Minnesota’s case and saying, “We now believe that batteries have the price dynamics and technical capabilities at this scale, put in these places to be the lowest cost, most reliable way for us to handle part of our grid’s growth.” Two hundred megawatts is big for distributed batteries, David, but it’s tiny compared to the gigawatts and gigawatts.
David Roberts
That’s a fifth of one giant data center.
Pier LaFarge
Just tiny. We have to remember the scale of the grid. This is a landscape-scale machine. The advantage of the Capacity*Connect program is that it’s asking the question — and hopefully if we in Xcel do our job, it’ll prove the point — that batteries are just the next boring generation of distribution grid infrastructure. They’re no different than substations, transformers, capacity banks. They just have this new superpower which is that they can move electrons through space and time and you can put them next to the problem and deliver the value.
David Roberts
The central objection that people have had to the Capacity*Connect program is that it puts utilities in charge. Utilities own and have a monopoly over these battery installations. That excludes the private model that currently rules where private providers go house to house, sell these things to homeowners, etc. There is a real question about democracy here, about energy democracy, which I would like you to address. I know because we have known each other for a bit I know you to be committed to local community and local democracy specifically.
I know that you value local democracy. A lot of people are coming to this with energy democracy, with local energy democracy in their minds, saying, “Why shouldn’t people own these things? Why shouldn’t communities own these things? Why should we let utilities have a monopoly over these things?” How do you think about the energy democracy question?
Pier LaFarge
Absolutely. It’s a funny part of my life. I spend much of my personal and political time advocating for locally led decentralization solutions in the abundance movement. In energy, I’m advocating for decentralized batteries and people are like, “This guy likes monopolies.” There’s a difference between civic infrastructure for public purpose and just a generic monopoly. At the end of the day, the reality of a grid is that it is public infrastructure in Minnesota. This was very specific to the decision that the commission came down with, David. In this 5-0 vote, they put this intentionally and with great legal specificity. I encourage those among your listeners who are truly the most wonderful nerds to go and look at the transcript, look at the hearing. This is public record. It’s a civic process.
The commission said very particularly, legally speaking, the grid is owned by the ratepayers. Grid operators just take private money, Wall Street’s money, they build the grid and then are given a guaranteed or partially guaranteed return so that everyone can use it. They use it by paying for power.
The legal status of the grid is it is owned by the ratepayers. When a battery is installed on it, the utility compact says if it is used entirely for the benefit of the grid, if it is owned and operated only for that purpose, then the utility should own it. It is also hard to own things that you cannot control. Ultimately, it is just a normal part of the public infrastructure that drives the grid forward.
David Roberts
The word infrastructure or the concept of infrastructure is key here because we have been — the conventional model to date is to treat distributed energy resources as a consumer item, as a consumer market item. We sell to the consumers who want them. That is not how we treat infrastructure financially.
Pier LaFarge
And to your point, I very much support and I think Xcel supports and has a range of programs that are specific to allow third-party ownership. There is a program called Battery Connect where third parties can bring and own batteries and get paid for the value they give to the grid.
But remember, all those batteries are splitting their value between the customer and the grid. This program says these batteries are just 100% dedicated to the grid. Batteries are now so valuable to the grid in the way that we have been talking about on this episode.
The transition you’re seeing is when batteries were more expensive and there was no load growth, customers had to pay into them to have batteries exist at all and the grid could pay for a portion of the value based on its need. That should never stop. This is America. People should be allowed to own things, they should be allowed to invest in things. They should be allowed to participate with their money and time.
David Roberts
To pause here in Minnesota, private distributed energy resource developers still exist, can still sell their wares to customers. The PUC, the utility is not preventing them from doing anything in particular.
Pier LaFarge
No, there are many existing programs that explicitly allow, and not only allow, but are only for third-party ownership and participation. There is a huge community solar industry and law in Minnesota, with almost a gigawatt of third-party owned community solar.
Frankly, one of the opposition points that I found surprising and somewhat disingenuous was that this one program which is designed to test and validate the idea that batteries can become a core part of normal grid operating infrastructure — the boring stuff on the grid — was attacked on that ground by a group of advocates who I think have gotten a little bit lost in whether or not the grid exists to let them invest in and privately profit from the ownership of their assets or whether it’s public infrastructure that under certain conditions of mutual value and coincident need, those private owners can and should be allowed to apply to own assets and take some value if they can provide some value.
I’m all for third-party ownership and I think the trade of value is often good for the grid and good for the private owner. But it is not the point of the grid.
David Roberts
Another objection is if we allow private actors to make these investments, private investors take on the risk. If the demand forecast or the economics business model, whatever it is, doesn’t work out, it’s private investors who lose the money. In this model, the utility takes the risks, but it’s ratepayers who pay if things go poorly. This is another common criticism. How do you think about that?
Pier LaFarge
It comes down to value and risk. When private investors build assets and apply to participate in the grid, they take risk and they also take their opportunity for profit and upside, which is appropriate. The upside they create, they keep. The grid is operated by a regulated system — a civic compact that finds other private owners. The other weird thing about this is utilities are also private owners. It’s always strange when one side says to the other, “You have a capital incentive to deploy more money or have a shareholder incentive to give returns to your shareholders.” Do all companies not have shareholders and an incentive to grow their business? I’ve never quite understood why that’s a ground for debate.
One private owner, the utility, is the regulated grid operator who gets a fixed return. Every dollar of value created above that return is socialized to the benefit of all ratepayers, because, again, legally the ratepayers own the grid that the operator is chartered to deliver. That distinction, I think, is neither one is better or worse. Both should be allowed, both should be celebrated. Think about the beginning of our conversation when we were talking about how one of the foundational inputs into the future of America’s success as a nation is, from our daily lives to our national defense, to have more electric power for more purposes.
We want all of it. This is the grid growth version of all of the above. Let’s go, let’s get this stuff out there.
David Roberts
What about the idea if you’re a private actor, you’re somewhat disciplined by this risk of loss. Whereas if you have a guaranteed high return on investment, as the utilities do, the criticism goes, you’re not going to get spending discipline, you’re going to spend ratepayers’ money profligately and without proper analysis because you’re not disciplined the way the private market is. Do you think that high guaranteed return on investment removes discipline? Do you buy that critique?
Pier LaFarge
The thing it misses is that the regulator exists. Something that bums me out about this is there is this whole group of civic servants and a bunch of staff who spend their whole day with legal power and charter from the state making sure that the way in which these procurements are built and delivered meets a reasonable standard of prudency.
David Roberts
But we know that PUCs are of varying quality, varying levels of knowledge, varying levels of friendliness to industry. These are long-standing critiques of the regulators, too.
Pier LaFarge
They are. One of my other political beliefs is the next chapter of American politics needs to fall back in love with civic engagement and civic compacts. We need to pay regulators more, we need to have more staff.
David Roberts
Administrative capacity! Wish I had a little sound effect machine so I could play some sort of... [plays horn sound]
Pier LaFarge
I’m not just pro-utility, I’m pro-utility regulated compact. That system, you can debate the history and its individual decisions. The point I wanted to make, which is more specific to your challenge of does a regulated system incorporate the dynamics of competitiveness and innovation? The answer is yes. Not because the regulator is perfect or all-knowing or all-powerful, but because the regulator can simply require, as it did in this procurement, that the way in which these assets are procured includes competition.
Sparkfund’s job in this Capacity*Connect program is to build and manage on behalf of Xcel a local competitive ecosystem of vendors, civil, mechanical contractors, including union labor and local contractors. We bid out. This is right in the filing, this is public information. Every time, 80% of the value of this procurement of distributed batteries will be bid competitively in Minnesota. Eighty percent of the value of this procurement will be bid competitively to local businesses in Minnesota.
The reason for that is that regulators and utilities build in market-based competition into these procurements. They do it for utility-scale power with all-source RFPs when they procure wind and solar and gas. You end up with constant rolling market-based competitive price discovery when a utility is incorporating distributed batteries onto its grid for public purpose.
David Roberts
If you’re in the DER business, they’re probably going to yell at me for saying this, but you still get to bid on the job, but what you don’t get to do is go out and find and sign up customers and structure financial deals with them. That just sounds to me like the worst part of the job. The worst and hardest and least rewarding part of the job. It seems like if you’re a DER person and you just get to skip straight to the part where you’re doing the job, it seems better.
Pier LaFarge
A lot of the opposition in Minnesota was downstream of that. Protecting a finance business model that requires that part of your business model. This is again alongside other programs in Minnesota that allow for exactly that.
The exciting part of Capacity*Connect, it is a major utility doing a forward-looking, innovative filing that says we’re going to make these machines part of the normal course of the distribution operations plan of our utility and we’re going to do that because we think everything we said in the first half of this episode is true. They’ll charge up when there’s plenty, they’ll dispatch. They agree that batteries are going to be valuable.
David Roberts
If this model, utility-owned distributed batteries, takes off, works, they like it, they spread it, they expand it, other utilities grab onto it, it will have the effect of squeezing out business models that rely on third parties in between the utility and the customer. That includes VPP aggregators, it includes solar finance people such as Sunrun, and I think it includes Sparkfund.
Do you not think that to the extent this model catches on and becomes standardized and becomes boring, all these third-party business models, all these middleman business models, are going to get squeezed out over time, maybe including your own?
Pier LaFarge
Maybe when things go to scale and become normal. Of course, in big public systems they tend to commodify and the value chain that delivers them changes. I think that’s just true in economic history. Part of what we’re debating here is not really about utilities or this filing or that filing. It’s just about batteries and distributed energy having a moment where they’re going to become a fundamental pillar of how we grow and manage our grid, because our nation needs it.
Will there be change? Absolutely. In particular, consumer finance-based models that initially were about helping customers buy these machines versus the public grid are probably more at risk.
David Roberts
Let’s not rush past that. Sunrun is very large and that is its core business model. I know they were among the entities that opposed this in Minnesota and I will assume they are going to fight you all the way down the line on this, are they not?
Pier LaFarge
They will. They spent more money than anyone fighting this in Minnesota. We were a bit surprised in some ways, but when a grid has more batteries, it is better at handling and delivering solar energy. People defend their business.
David Roberts
They’re pivoting to batteries. They’ve said it a couple of times now. They’re pivoting to batteries and they’re doing this finance model, this finance model of selling batteries is now their core thing. This model that Xcel has adopted, if it became the default, would shrink or eliminate that business model.
Pier LaFarge
That’s right and I think that’s why they oppose it. It is a competition of business models, one in private financing and one in public infrastructure. I also like to come back to VPP as you made an important point. I also don’t want to dodge the question about whether Sparkfund will be necessary to the future of this.
David Roberts
We’ll talk about VPPs first. I’m curious what role you see them playing in this battery-filled future.
Pier LaFarge
Much more durable, because I think most VPPs, as you defined it, which is largely behind-the-meter resources that in one moment serve customer value, in another serve the grid. That is really broadly speaking demand response. I do not mean that dismissively. Demand response assets are critical parts of relieving pressure on the grid in aggregate. Demand response is absolutely one of — we have been talking about batteries today with a lot of focus, but batteries are on a short list of grid flexibility-creating assets, including demand response and grid-enhancing technologies that really drive the future that we are talking about.
David Roberts
I have a pod running. I think it’ll be out by the time people hear this, wherein someone argues that what we ought to be doing is making price-responsive appliances, everything electric. All these embedded batteries, all these electric devices you’re talking about ought to just be price responsive. There ought to be a competitive retail price that is updated every five minutes and is time and location dependent and broadcast such that every device can receive and know the retail price at every moment. Then the devices respond to the price on their own.
The benefit of this is that then all the value of the flexibility is captured automatically by the homeowner, by the owner of the devices, rather than split 50-50 with a VPP, which is just a middleman doing the same function in a slightly cruder way. It’s interesting that there are other ways that you might eliminate the VPP model. If then all the devices are immediately price responsive, then you don’t need a VPP in the middle and they become much more directly grid assets, I think.
Pier LaFarge
What you’re talking about there is really how much does an aggregator get paid as a share of revenue? That sounds like one plausible way that an aggregator could get paid less as the middleman. The more important question, though, is will demand response be a service the grid wants and will it reach into customer devices to do that? Will there be benefits for the customer and the owner and the grid? The answer is yes. Nothing about deploying a bunch of big batteries changes that. They’re just both filling two different parts.
David Roberts
But why not though? If you’ve got enough of these 1, 2, 5 megawatt chunks scattered around your distribution grid, you get your stability, you get your boringness. Why then do you need to go on and tackle the complexity of trying to coordinate consumer-side devices? Is there not some possibility that this could prove so cheap and easy for utilities and so effective that it really just slightly moots the need for consumer-side coordination?
Pier LaFarge
I think there is. The question is how much does it moot it and is it 100% gone? Remember we talked about the value of batteries at every scale and every size being how close you can put them to the problem, and batteries in the home and flexibility devices in the home are closer to and easier to put in and pay for at homes.
David Roberts
But I guess the question is, is the distribution level close enough?
Pier LaFarge
It may be. Someone very wise in this industry once said to me, the reason the centralized planned grid has lasted for 100 years is that many of these questions have been posed. This happened with postwar industrialization, it happened with HVAC and air conditioning. It turns out that often the lowest cost, most reliable way to deliver these solutions is through more centralized assets than not. What we’re talking about is batteries that are slightly less centralized than 200 megawatts in a field, but are still shipping containers filled with batteries that are far beyond the scale of the home.
We have to remember that the grid is not built for individual value to come off of it. In the world in which every homeowner can perfectly participate in these dynamic real-time pricing markets, think about who the citizens are who are likely to buy those smart-enabled devices and get that value. They are going to be richer, they are going to have more time after work to think about that.
One of the reasons we built the grid, going back to our very first history point on this conversation, David, is that there is an egalitarian sort of American infrastructure compact vibe to the US electric grid, which is that you want to plan things to be lowest cost, most reliable, so that you can share the benefits with everyone, whether or not they have the time or money to think about it or not.
I’m not against wealthy, educated people participating in the grid. It’s how we’ve invented some of these devices and scaled them — that is a critical part of an innovation economy. It should never be disallowed, banned, or limited. I don’t want to ban third-party ownership at all, but it is not the point of the grid. We did not build public infrastructure so that enthusiasts can take rents from it.
David Roberts
This brings me back to another point I wanted to make about the energy democracy thing. I understand and am compelled by the merits of local ownership of energy assets, and I’m compelled by them in terms of energy democracy and in terms of local empowerment. But I think if you are measuring that against the benefits of the same community having reliable, cheap, ubiquitous electricity, even on the metric of local empowerment, the latter wins.
Putting cheap, abundant electricity in people’s hands is the most effective economic development tool on the planet. Every bit that you make a community’s electricity cheaper and more ubiquitous and more boring and more cheap, that is economic development, that is local empowerment. That’s not an alternative.
Pier LaFarge
It’s human flourishing. I love that, David, I really do. That’s a beautiful way to say that and I think it needs to be said more. It’s in this important and under-discussed category in our tribe of nerds, which is what is energy for, what is infrastructure for? I think you just articulated exactly what it’s for.
David Roberts
It is for flourishing. Let me ask you about this. One of the things — I had Lorenzo Kristov on the pod, love him. He’s a big fan of local democracy, local action, local community, local resilience, self-sufficiency, all that kind of stuff in electricity too. One of the cases he makes, which I’ve always found compelling, is that there are benefits to local communities being able to plan their power alongside planning local transportation policy and local building policy and local economic development policy. Power should be of a piece with that and should be integrated in that.
There’s something artificial about all that planning being up at the utility level, while all the other planning is happening at the local level. Are you not at all moved by the idea that local community ownership of local energy assets carries some benefits?
Pier LaFarge
First of all, Lorenzo is brilliant and he and I have had some wonderful conversations about this and I think we share a real passion for localism, broadly as a basic idea of subsidiarity, which is — push the planning and the decision making down to the lowest competent level that also can efficiently deliver the outcome.
His point, I think, I’d break between two things: where it’s planned and where it’s owned, which are two different but related questions. Ownership is important, but really about who receives the rents versus the risk. The regulated compact was designed to smooth or socialize both the risk and the rents to all participants at all scales of planning.
The next question is, when you go smaller with infrastructure, do you have the equivalent vehicles of debt financing and capital that allow you to do that at that level of aggregation efficiently?
David Roberts
Let me put a finer point on that. One of the things Lorenzo has envisioned, and I think they have them in the UK now, is what’s called DSOs — the distribution grid equivalent of the transmission system operators, the TSOs at the transmission level. A smaller entity devoted particularly to that community. You get closer touch with the community, you get better knowledge of the community, you get perhaps greater investment in the community. As you, I think, are getting at here, you might lose some of the financing benefits of size and scale.
Pier LaFarge
If you don’t, then I think there’s a lot to like about that model. Lorenzo has an incredible — again, I really like and admire how he talks about this stuff. The question really is that if you zoom back, the utility regulatory compact was designed to pull off a magic trick, which is take Wall Street’s money, private money, give them a 3.5% net dividend every year, every 100 bucks they put in, they get three or four bucks. Let a private operator who has some profit incentive build the grid, which is the foundation for our economy and human flourishing, as we’ve just discussed, and then submit to a civically accountable regulator to control every one of its decisions, actions, and profits.
It’s Wall Street’s money building public infrastructure. That’s easy to say. The reason that’s important, David, is that you don’t have to choose in a public budget, a general fund or a town’s budget between building the next rural hospital or paying teachers or repaving roads and building a substation or a new transmission line.
David Roberts
Because you got the Wall Street money.
Pier LaFarge
We did it. We got private money for public purpose, regulated as best as can be done. Is that system perfect? No. Does that system have flaws and imbalanced incentives? Yes. Can you name me a landscape-scale, public regulated compact and institution that doesn’t? What’s amazing about the utility compact is that it delivers its outcome pretty much 100% of the time. There are grids where the grid is only off 80 or 100 minutes a year.
In most of our state’s economies, at the level of aggregation of debt and planning, the power is pretty cheap, the capital is super efficient. What comes out the other side of that is that in a lot of states, the percentage of the economy that goes to pay for the infrastructure that lets us have as much power as we want, anywhere we want it, all the time, is 3 or 4% of the total state’s economy.
David Roberts
To return to a point I made earlier, you could compare that 3 or 4% to the market value of the assets it has purchased and is deploying. I think that will look pretty favorable.
Pier LaFarge
Incredibly favorable.
David Roberts
But again, the market value of the assets is a small sliver of the value you get from having ubiquitous, reliable, unlimited, cheap power. The second- and third-order benefits of that are vast. They are not captured by the value of the transformers.
Pier LaFarge
Not just vast, they’re the rest of the economy. I think it’s easier to just assume. I’m sure you could do some dumb research paper which is, “What percent of GDP is not enabled by electricity?” Who cares?
David Roberts
That’s growing and growing and growing, as we said.
Pier LaFarge
All of it and more, uncaptured GDP. That’s exactly right. That is what energy is for. The compact we have delivers it with private capital and it powers a whole economy with most people not having to think about it.
David Roberts
Could a CCA do this in California or wherever there are CCAs? I know there are several other states where they —
Pier LaFarge
In terms of distributed batteries?
David Roberts
Yeah, and just implement an Xcel-style program.
Pier LaFarge
Absolutely. We’re having conversations with grid operators across municipal utilities, co-ops.
David Roberts
Munis could do it. Co-ops could do it. If a muni was doing it, you moot some of the energy democracy complaints, at least because then it really is still locally owned. It’s still local.
Pier LaFarge
Batteries are going to win the future and we’re going to put them everywhere because they are time machines for electrons and it’s what the grid needs. What we saw in Minnesota is a simple case of a threatened business model paying money to advocate to protect its business model in a moment of real transition when these assets that used to be in their protected market camp start to become a core part of grid infrastructure, and it freaked them out.
David Roberts
You could also flip that around. People have objections to this that are not just generic, not just business model exceptions, but Xcel specifically. People have reasons specifically not to trust Xcel. There are people who say Xcel has had these distribution resources for a long time. They don’t use them for public benefit. There have been a lot of price events where they could have used them for public benefit and they didn’t. They’ve been dragging their feet on distributed energy resources. They got fined by the PUC for dragging their feet on distributed energy resources.
There are reasons to distrust the utility and why should we believe that this, when they have betrayed the public trust in the past, why should we believe that this program will genuinely be run for public benefit?
Pier LaFarge
Definitely true. We’re in an age of distrust of institutions and the public institution of utilities and the regulators is no exception. The only solution to that is to deliver the value and hope that the future we’re describing, the age of boring, comes true. Cheap batteries and data center demand happening at the same time make power cheaper, the grid more reliable, and we can go on our merry way and work on other things.
David Roberts
That’s not an answer to the question. “I hope they do this right,” I guess, but you can imagine why people who have been burned by Xcel or have been fighting Xcel for years do not just say, “Well, let’s see what happens.” They do not find that particularly adequate. They also say that this first tranche of these distributed batteries are not going to churches and hospitals and civic facilities. They are not being assessed and chosen based on community resilience benefits, at least yet. Right now they are just being placed purely for grid benefit.
Why shouldn’t locals, why shouldn’t they be suspicious, and why shouldn’t they look at what’s happening and say it looks like Xcel is just making the macro grid better and is not doing what we would do with distributed energy — which is specifically local benefits?
Pier LaFarge
Something I think is important to know about this one filing is it did and does have a robust energy justice portion. There is targeting of those resources, categorized by EJ zones. That’s why we got a lot of support from that community and from a bunch of local groups advocating, including advocates like Fresh Energy and CEE and others that really engaged constructively in making sure we got that portion of the filing right. They were part of the hearing, they spoke at length, they offered decision options that were accepted by the commission. They got to directly modify so that the — literally, just to put a point on this — social justice and the energy justice community was at the table helping to write those requirements to their standard.
That’s a commitment that Xcel has, that Sparkfund has, and that the commission required. Maybe we’re going around a little bit in circles here, but the fact that that has been such a common talking point about this and it’s right in the order and the whole section is a bummer.
I don’t know how else to say it. Politics is a dirty game. There’s some real misinformation that has come out around this and it has been a surprising journey. I think it landed in the right place and I think, frankly, the commission and its members — it was genuinely impressive to watch Chair Sieben operate a civic compact under that type of duress and really ask questions and get stuff on the record and challenge Xcel and Sparkfund and advocates to explain and defend their positions and to provide evidence. No civic process is perfect. Minnesota is, Minneapolis is like Norway on the prairie. That civic compact is as strong and robust and chock full of civic nerds as I’ve ever seen.
David Roberts
We’ve all had occasion to celebrate Minnesota recently. You have said, I think maybe being slightly deliberately provocative, but you have said to me, and I think maybe in writing, that you think it is possible that the old-fashioned, vertically integrated utilities might move faster on this and reach that exalted, cheap, abundant, and boring plateau before restructured areas. Why do you think that?
Pier LaFarge
There is a U curve of utility operator and market success. At one end, there are things that are closer to pure markets — like ERCOT in Texas. They work really well. ERCOT added a whole New York worth of electric load in just the last four years.
David Roberts
Right now Basepower is down there in ERCOT operating on the retail energy model. They’re a retail energy provider, but they’re going now to vertically integrated utilities knocking on the door saying, “Hey look, we figured out how to get batteries on the grid faster than any other model. Wouldn’t you like some of this juju?”
Pier LaFarge
That’s right.
David Roberts
Very interesting to see that happen.
Pier LaFarge
What makes Base such an interesting company is that they’re not wedded to one business model. There are put-batteries-everywhere companies —
David Roberts
Put batteries everywhere!
Pier LaFarge
That makes them a durable part of this story and they should bring it to other markets. They’re like a residential version of what we’re doing in the larger batteries. I think it’s great. The answer to your question though is that IOUs, if you’re not going to have a full connect and manage market that also — and this is really critical — Texas’ superpower is very, very permissive local permitting. You can build things really fast.
There are so many details about this, David, that you know better than I do. When you talk about PJM design, all this stuff, if you have bad local permitting that makes it hard to build stuff, it almost doesn’t matter what your ISO rules are. You just can’t build big power plants or even small ones or even batteries. If you are at the bottom of the U curve, a restructured market that has a weird mix of esoteric rules that make it hard to participate in your market and allow rent seeking, auctions that guarantee capacity because not enough people can participate and no one can build things anyway and are also still highly regulated and have lots of rules and lots of laws from states that want a bunch of things — that is the thicket.
David Roberts
Kind of a worst of both worlds.
Pier LaFarge
That’s the worst of both worlds. The heart of the U curve. The thicket of PJM’s discontent, which I should make T-shirts that say, but you’d really know who the nerds in the airport are if you walked through a terminal. IOUs are not perfect. They have this basic advantage in moments of great infrastructure need and build: they can organize capital and conduct their planning at the level of generation, transmission, and distribution as one integrated system.
A lot of the advocacy that you and I and many people in our tribe have been on about for years is ultimately that thinking of the grid as a system, that seeing distributed assets as part in that system will ultimately be a key part of growing the grid.
What we’re seeing in Capacity*Connect and others is that large vertically integrated utilities are doing the math and they’re agreeing. That is not because this was the absolute first moment they could have done that. But it is pretty recent that these batteries got cheaper. I’m not going to pretend that big utilities aren’t slow and for some good reasons in terms of safety, reliability, critical function, also some reasons of bureaucratic inertia and scale and slow regulated proceedings. Good and bad. But it is not very many years ago that batteries were not the cheapest way to do this. I would argue that they’re just barely becoming the cheapest now.
I think we’ve crossed that line permanently, irrevocably. That’s sort of the thesis of this pod. Welcome to the age of boring, because that has happened. But it only — as the Dude said, new shit has come to light.
David Roberts
Are you — this is orthogonal to our whole point here, but I just want to get you on record — are you among those people who predict a return to vertical integration in the US in part or in whole?
Pier LaFarge
Yes.
David Roberts
What time frame? Let’s get a date here — ish. A date range.
Pier LaFarge
It’s important to clarify what vertical integration is today. Most vertically integrated utilities procure generation competitively through big all-source RFPs from private owners and IPPs. This is not your grandfather’s vertical integration where a utility hires a crew and builds a power plant directly and operates it with its own union and its own pension. This is still, like I said, tucking market price discovery, competitive delivered advantage, innovative designs and scaled private capital into a relationship with a public purpose infrastructure compact. There’s a little bit of a distinction without a difference.
Do I think that vertical integration is important to allow for the proper planning of society’s needs over time to deliver infrastructure that is, as we’ve discussed, the critical input into a nation’s success into human flourishing — from our daily lives to our national defense? Yes, I think there are advantages that that type of civic compact has in a planning level that allow us to go faster, go further, and make the sort of leap you saw a Southern Company make with those batteries when those conditions merit.
The rapid scaled adoption of gigawatts and gigawatts of effective utility-scale batteries is a good example of how to meet the moment when the nation needs to build data center infrastructure to fundamentally power the next chapter of the American economy. I don’t think that vertical integration has to be universal.
I think ISOs and interoperable markets that expand the pools. I think IPPs will be a durable and critical part of competitively procuring supply, supplying, and operating power. That private capital is the whole point of the utility compact. It’s Wall Street’s money for public purpose. There is no market in the United States today where a political and societal judgment does not set the rules for how electric power infrastructure forms the foundation of that state’s economy — not Texas — ERCOT has a critical regulatory role. PJM has a critical regulatory and rulemaking role. Every state in PJM has commissions.
We have not thrown societal caution to the wind and just said let markets eat the world boom or bust on electric power. No one would suggest —
David Roberts
No one has ever said that. No one would suggest in any region, in any market, anywhere in the world. I love talking about this for a bunch of reasons. One of them is there are so many things, new possibilities, new opportunities around the world of clean energy where I find myself on the edge of my seat white-knuckling it, thinking, “God, we’ve got to do this right. It would be so easy to do this wrong and screw it up generationally.”
But when it comes to this — batteries — if you put a battery on the grid, it helps. It almost doesn’t matter who puts it or where or why or how they pay for it. It’s almost impossible to do this wrong. Just put the batteries on the effing grid and they’re going to improve the grid. We don’t have to do this exactly right. We just need to do it at scale. We just need to do it big. We don’t have to be precise about it. Every bit of battery helps. You almost can’t screw it up.
Pier LaFarge
Welcome to the age of boring.
David Roberts
The age of boring. Final question then is: right now, I think probably conventional wisdom in our world is that we are currently facing a bit of a gauntlet. A bunch of new demand is showing up on the grid. We’ve got these old sleepy institutions, these old sleepy, not fit for purpose social and financial and political and regulatory arrangements. We’ve got 20 years of no growth that has lulled everybody into hypnosis. We face this gauntlet where prices are going to go up and everything’s going to be shaky and not really work. Then, 20, 30, 40, 50 years from now, once we’ve made it through the gauntlet and we’re able to start building the big transmission lines, we’re able to build whatever big geothermal plants, then we’re going to recover and bring prices back down.
I think you disagree. I think you think — and tell me if I’m right, I want to get a date, another date range from you — I think you think that batteries are going to solve these problems sooner than people think they will and that the grid will get boring and reliable and abundant sooner than people think it will. How soon?
Pier LaFarge
The best way to answer that question is you are already seeing data center announcements, particularly in IOU territory, but not exclusively, directly result in rate cases where the divided cost of energy goes down because you have divided the fixed cost of the wires by more sales and power gets cheaper.
David Roberts
For example, give us a concrete example.
Pier LaFarge
PG&E, in California, one of the hardest places to operate a grid anywhere with wildfires and stuff, said that for every gigawatt of data center load, you get 1 or 2% cheaper power for everyone in their territory. To date, 11% cheaper rates because of the data centers that have come.
David Roberts
Maybe not visible to the people in California because wildfire costs are simultaneously pushing rates up.
Pier LaFarge
Indiana Michigan, AEP, has actually made power cheaper. Southern Company, I believe, has actually made power cheaper. I believe about $100 a year per ratepayer was their public filing number, or maybe it was 80 or something. That’s billions of dollars of cheaper power. All of that was related to exactly this effect of managing the growth of large load customers to public benefit.
It’s also important to say that data centers are the hero of the story. It doesn’t matter how much you invest in batteries or anything else if you can’t sell the extra power. If you don’t have electrification and load growth and data centers, you’re just spending money and you can utilize the grid better, but no one is going to buy the power, so it doesn’t matter.
David Roberts
What about people who would say, “AI is terrible and we should be electrifying transport faster, we should be getting that boost of demand from somewhere else. In some sense, by data centers eating up all the available new capacity, we’re delaying further electrification.” You don’t buy that at all?
Pier LaFarge
I’ll go once more to the Dude and say, “That’s your opinion, man.” That’s way outside the scope. That’s way outside the normative scope and frankly, something I don’t love. There’s a lot of things I think that are wonderful about the community of climate energy nerds that we’re both lucky enough to be a part of. These are system-oriented, normatively engaged, philosophically interesting people. We care and we’ve organized our careers around that. That’s fantastic and it’s a great thing to be a part of.
One of our faults is sometimes we breach the walls of the energy discussion and we imagine the grid is this portal to unlimited social control and power. We’re saying, “Let’s use energy policy to determine, should we have more electric cars or more data centers.” That’s not a job for this. Take a breath. This is me back to my genuine, localist, decentralized, slightly grumpy — make fewer rules and trust that people can figure stuff out. There’s a real chance for an abundant America.
David Roberts
You think this phenomenon whereby data centers are going to create the new load that takes the batteries’ new capacity and thereby reduces rates, all of that is going to happen automatically if they get on the grid. In other words, you don’t need the extra step that a lot of people are talking about of making them pay extra, some extra fee to get on the grid, making them pay for some specific capacity. Maybe you still think that is a good idea, but the effect you are talking about doesn’t depend on that.
Pier LaFarge
That’s right. It’s very important. Brian Janous cleverly beautifully said this in a LinkedIn post around the announcement of, “All data centers will pay their fair way,” which is great. It’s good politics. It’s good to make sure that we’re saying that out loud. It is proper regulation and political governance to say out loud in one of the largest capital events in human history, by the way, that the participants and beneficiaries of that investment must pay into a public system appropriately. That is good and true.
The basic Bonbright principles — took us this long to say Bonbright, by the way — is inherent to utility rate making: all participants should pay their fair share. Is that always perfectly arbitrated? No, it’s also complicated to get right and how you judge which cost was caused by what type of stuff, system growth and then there’s a storm or a fire and then there’s —
David Roberts
What about the idea that a lot of people have, that we need a bunch of spending on the grid, we have a bunch of needs, the public’s grumpy, power prices are high, they don’t want to cough up for it. Along come these wildly wealthy companies that desperately need to hook up quickly. Why not extract more than their fair share? They’ve got money coming out their ears. We need money. Why not extract a little out of them in the process of letting them hook up?
Pier LaFarge
Because it’s no more fair to them than if they didn’t pay their fair share. Fair share means cover the costs. If you manage growth well with batteries and grid flexibility and other stuff, you’ll spread the benefit of their participation to everyone. Just the fact that they’re willing to buy gigawatts and gigawatts of new power over public infrastructure and participate in the grid as good grid citizens is the contribution. You do not need to penalize them, you do not need to tax them.
This is a broader philosophical point, but we’re in this moment politically between moments. We don’t have a political order to guide our shared agreement. Neoliberalism is gone. Before we figure out what the next political order is, there’s this real chaotic risk of villain politics. People are searching for villains. What we’re getting wrong with data centers and utilities is we’re casting them as these satisfying villain characters for the technocratic left. We’re missing the forest for the trees in it. We’re getting the story wrong.
David Roberts
At the least, I think we should separate out. The problem with these villain politics is that once someone gets the scarlet V, people just displace all their complaints onto the villain. We need to distinguish the effects on the grid — the physical, energetic, and economic effects on the grid. We need to be able to separate that in our heads from whatever you think the social impact of AI may or may not be. Those are, at the very least, separable questions.
Pier LaFarge
The T-shirt I’d like to make or the sign I’d like to bring to a town hall where the community is determining whether a data center is right for them — and not all communities want or need a data center, it’s not right, there should be local control and local voices — but the sign I would bring is, “This building is not a witch.” Let’s not create a parasocial category that’s hard to disprove and then imagine all of your worst fears flow through this one thing. Villain politics sucks.
As Americans, we should challenge ourselves to build a politics of abundance. Be curious and kind and empathetic about how — look, society is hard and participation isn’t easy, but we can get this right. We can build American infrastructure, we can grow the grid, we can make it more reliable, more clean, and we can make it cheaper. Maybe it’s the opposite of boring. Maybe we’ve gotten it all wrong. Maybe that’s a really exciting place to end.
David Roberts
Give me my date then. When is US electricity cheap, abundant, and boring? When am I going to look up and say, “I need to figure out something else to pod about?”
Pier LaFarge
2030. Just a few years. I think the narrative will be clear enough by 2030.
David Roberts
Wow!
Pier LaFarge
That this trend is permanent and structural and inside the existing systems and incentives that data centers are driving most of the near-term electrification that makes it possible, that electric vehicles and heat pumps and broader electrification are coming behind it. New manufacturing.
David Roberts
Will average electricity rates in the US be visibly falling by 2030?
Pier LaFarge
Yes, I think so in the states that have substantial data center growth. I will make that my hottest of takes. I think by 2030 in the states where there is substantial data center growth and other electrification, power will become cheaper on a nominal basis. If you have a lot of wildfires, if you have terrible storms, there are other things that can make a landscape-scale machine expensive for unrelated reasons. I’ll give myself one type of contextual hedge. But I do think that in many markets, in many states that can get this growth moment right, power will be cheaper, the grid will be stronger, and we’ll be on our way to a new chapter.
David Roberts
That’s a perfect place to leave it then. That’s a prediction that’s going to come up soon enough that we’ll probably be able to talk about it on my pod.
Pier LaFarge
Ten-year predictions are so boring too, right? It’s like ten years.
David Roberts
I know. Everything’s possible.
Pier LaFarge
The death of accountability bores me in predictions. I want to be clear. I have no particular information. I’m almost certainly wrong in countless ways. But it’s much more fun to say. At least by 2030 we’ll be —
David Roberts
2030 it is. Thank you. As always, Pier. This has been fascinating. Fascinating to see, with all the dysfunction around us in so many other areas, I really think it is important for listeners to absorb and take heart in the fact that electricity does really seem to be on a good path heading to a good place, so at least not 100% of things are awful.
Pier LaFarge
Let it be abundance then. Thanks, David.
David Roberts
Thank you for listening to Volts. It takes a village to make this podcast work. Shout out especially to my super producer Kyle McDonald who makes me and my guests sound smart every week. It is all supported entirely by listeners like you. If you value conversations like this, please consider joining our community of paid subscribers at volts.wtf, leaving a nice review, telling a friend about Volts, or all three.
Thanks so much and I’ll see you next time.
