Episode description

In this episode, Saul Griffith (co-founder of Rewiring America) and Arch Rao (founder and CEO of Span, which makes smart electrical panels) discuss the need to electrify US homes, the challenges standing in the way, the kinds of solutions that will ease the process, and much more.

Full transcript of Volts podcast featuring Saul Griffith and Arch Rao, June 28, 2021 

(PDF version)

David Roberts:

Those of you who have been reading or listening to Volts for a while know that I am fairly obsessed with clean electrification, which involves shifting all the things we do now with fossil fuels over to electric equivalents (while cleaning up electricity supply).

One important nexus of electrification is the residential sector. US homeowners are in a position to electrify their power supply (with solar panels), their heating and cooling (with heat pumps), and their transportation (with electric vehicles). How can we induce millions of them to make the decision to electrify, starting today? How can we make it cheaper and easier for them? 

To discuss that and related issues, I was excited to connect with two of the smartest people working in this space. The first is analyst, inventor, tinkerer, and entrepreneur Saul Griffith, who will be familiar to longtime readers — I've cited his work numerous times, especially his most recent work with Rewiring America, which advocates for rapid electrification. There is probably no one on earth with a better understanding of the US energy system. (He’s got a book on electrification coming out in October.)

Griffith is a backer of and investor in a startup called Span, which makes smart electrical panels that offer homeowners fine-grained control over all their individual appliances, lights, and devices (via an app on their phones, of course). The founder and CEO of Span, my other guest, is Arch Rao. Rao was the project lead for Tesla's Powerwall home battery before leaving to start Span, so it goes without saying that he is intimately familiar with the technical and economic challenges of home electrification.

Welcome to Volts, Saul and Arch! Saul, I want to start with you. We're going to talk about home electrification today, and just by way of setting context — it's pretty easy to make the case that home electrification is fun, it's cool. But what is the case that it is necessary, and not only necessary, but necessary quickly? Set the bigger picture for us.

Saul Griffith:

There's a few components to that. Let's start with the climate component: the urgency.

There's a concept called committed emissions — that is the emissions that a machine that exists today will emit while it lives out its lifetime. So if you bought a petrol or gasoline car last year, it'll keep burning gasoline for another 20 years; if you bought a natural gas furnace last year, it'll keep burning natural gas for 25 years; a hot water heater, 15 years; an oven burning natural gas, another 12 years. So those are committed emissions, the same as a new coal plant opening last year would go on operating for another 50 years. We now know that if all of the machines that exist on the planet today live out their natural life, the committed emissions of those machines take us to about 1.8 degrees Celsius, over three degrees Fahrenheit of warming. So the practical reality is every time any of our machines fails or needs to be replaced, we need to upgrade it with a zero-carbon option. And the only real zero-carbon option that has emerged is electrification, and that's electrification of our heat with heat pumps, of our vehicles with electric vehicles, and then tying that all together and balancing the grid.

David Roberts:

Right. And what chunk of emissions comes from residential? 

Saul Griffith:

Historically, we put emissions into sectors: residential, commercial, industrial, and transportation. The residential sector is responsible for 10 or 15 percent of total emissions, by that measure, but it's actually much higher than that because in reality, you make the decision about your car and your home. And when we electrify our cars, they're going to be charged at home. And then today, as it stands, a huge amount of our economy in the U.S. — close to 10% — is used to find, mine and refine fossil fuels, so that's the pipelines and the trains moving coal. And that's all filed under industrial emissions. So if you wrap up your pro rata share of that in your household, you wrap up the electrification of your vehicles, the decisions you make around your kitchen table are actually about 40 or 42 percent of our total emissions. In our small businesses and offices, what's traditionally known as the commercial sector, it's about another 20 percent. 

As I like to say now, there's two types of emissions. There's a small number of big machines, and there's a large number of small machines. The small number of big machines is a few hundred coal plants and a few hundred LNG terminals and a few hundred oil tankers, but the real game in town is the 200 million vehicles, the 128 million households, the 70 million natural gas furnaces, etc. That's the large number of small machines, which is what we need to electrify when we electrify the household.

David Roberts:

And it’s individuals in charge of those decisions.

One objection I hear to home electrification is that, from an household point of view, the boring stuff like insulation and weather-sealing is a better value. What's your general response to that? 

Saul Griffith:

I don't think that's true. It’s empirically wrong. Retrofits like the envelope and ceiling can be very, very expensive, because you have to remove walls, you have to stuff new insulation in those walls.

I just got a quote on doing that for my house, which is in a mild climate but is nevertheless freezing. It was about $28,000. That would give me a small efficiency win in this house, of maybe 20 percent less energy use, whereas buying the heat pumps to heat the house is about a $2,500 project. They will lower my energy for heating the house by two-thirds compared to how it's currently heated.

So the big efficiency win in this house is the heat pump.  

Arch Rao:

Just to expand on that, we have to be looking at technologies that change the outlook for us, not just looking back, but looking ahead. Things like fundamentally changing how we heat our homes or how we think about cooking and heating our water supply are the big game changer here.

David Roberts:

Following up on that: Span makes these smart electrical panels. What's the pitch that intelligence is important, or a necessary component of this, versus just dumb-appliance replacement?

Arch Rao:

That's a very good question. Often people have to be walked through the second-order effects that a solution like the Span panel can have for your home and your carbon footprint.

A very useful analog, perhaps, is: in order to enable faster adoption of electric vehicles, we have to think about building charging networks, where we are working to deploy a large number of charges around the country and around the world. It's kind of similar.

The electrical panel has a critical place in the electrical grid infrastructure, especially for your home. Without thinking about data controls and intelligence flowing in and out of it, it is very hard to envision a future state where switching over to electric appliances is practical or inexpensive.

Anecdotally, when you think about replacing your water heater or your home heating system, it doesn't happen outside of an event. You have equipment failing, then you think about upgrading it, and when that decision point arrives, the cost and the timeline for upgrading your electrical system is often not compelling. It’s too much for any homeowner to take on, so they end up adopting the easiest solution, which is often calling somebody on Yelp and saying, “Hey, please replace the water heater with another natural gas water heater.” They're then locked into those committed emissions, like Saul said, for the next decade or more. So that's one part of the problem. 

The other part of the problem is, when you think about the continued adoption of electric appliances — be it electric vehicles, electric induction cooktops, or self-generation storage solutions like solar and batteries — the electrical panel, which is what everything connects into, is insufficient from a capacity standpoint.

What we're allowing for is a smarter electrical panel that gives you controls and visibility down to every appliance and makes it easier to lock in these electric appliances and also potentially avoid the cost of upgrading your incoming service.

David Roberts:

The idea is that Span will juggle the loads and the timing, to smooth out the demand curve so you don't have these spikes where you might need bigger hardware.

Saul Griffith:

So the average US household today has two cars in the garage that burn petrol or gasoline or diesel, and it has natural gas heating, and it uses about 25 kilowatt-hours per day of electrical energy. If you electrify both of the vehicles in that household, you'll add about another 25 kilowatt-hours per day to the load of that house. And if you electrify the heat, you'll add about another 20 kilowatt-hours.

For the majority of U.S. homes — and this is true around the world, when we electrify for purposes of decarbonizing, or for purposes of having a quieter, cleaner car, or because you're trying to improve the respiratory health of your children, because you don't want to burn fossil fuels inside your house — you're going to double or triple the loads in that house. 

The other phenomenon that is happening — and where I'm dialing in from today, in Australia, is an incredible example of this — rooftop solar is now providing 5 cent per kilowatt-hour electricity in Australia. Australia got the right mix of regulatory environment, politics, financing, that there is no way the grid will ever provide electricity to you as cheaply as solar.

That will be true in the US — it is starting to be true, but it's going to be very, very true by 2022 or 2023. And then you're going to want to run your hot tub when the sun is shining, and you're going to want to charge your car battery when the sun is shining. That needs coordination, and it needs a computer, and a brain.

What we're trying to avoid is having both cars on a type-two charger running at the same time as your oven, at the same time as your stove, at the same time as your hot tub. And with a small amount of intelligence coordinating those loads, there's actually quite a big economic win — it means that the total retrofit that we need to do is much smaller.

Arch Rao:

Thinking about it from the bottom up as well, the home electrical grid is built very much like traditional electrical infrastructure. You think about the worst case scenario, you try to build out capacity to support a very low likelihood scenario of these EV chargers, the heat pumps, your induction cooktop, and all of the different appliances in your home being powered on at the same time.

That rarely ever happens, or in fact never happens, right? To be able to manage these, you need a solution that sits at the nerve center of your electrical system. That's really what Span does.

David Roberts:

It’s quite analogous on the household level to the larger grid: you're just peak shaving. You're installing some intelligence to avoid big peaks and valleys, right?

Arch Rao:

There’s an elegant comparison to be made here with fractals. We're moving into systems where your edge of grid looks more and more like the grid — you have generation, storage, you have different types of loads.

So all these things we talked about, in terms of energy efficiency — reducing your consumption, which is a concept that's existed for decades now — is just not going to buck the curve quickly enough. If people want to continue to have the conveniences they have today, the solution is to electrify, to power it with known, available, increasingly lower-cost renewables.

David Roberts:

A lot of the promise of intelligence or software in buildings is that grid managers will be able to interface with the grid edge and coordinate and co-optimize it with central generation, so that everything works well together.

A lot of what Span offers is: here's what you the owner can do; you can set this, or if this happens, you can turn this down, and turn this up. Over the years, I have picked up a real skepticism about how much we can expect people to do. Even if it's directly in their self-interest and super easy, people generally, as a behavioral matter, will not do things.

In thinking about Span, what's the balance of user control versus what’s automated? Do you think about that a lot? 

Arch Rao:

We do think about that a lot, actually. At the homeowner level, what we've zeroed in on is, monitoring without controls is not valuable. We're not just offering customers information and hoping that they'll change.

Alongside that, controls without intelligence is not scalable. Ao we have to work toward solutions that take the human out of the loop, which means doing nuanced but subtle things within a home that don't impact your everyday life, but at the same time, are the right thing to do in terms of power flow management.

So the solution is at the home level, but you want to take the homeowner out of the everyday decision-making process. 

The second part of it is, as we deploy more and more of our systems, we are seeing the benefit of the fleet level. In fact, we've already signed on a few utility partners that are looking to give this away to their customers or significantly rebate it, because the panel, given where it sits, is the natural intersection point between the grid operator and the home. Everything that you source power from, sink power into and store power in, all naturally connects, which means you can have a single gateway, so to speak, that can monitor and control everything at a fleet level as well.

Whether it's an EV charger, from any brand, make, model, or company, or whether it's a smart thermostat, all of these can be monitored and controlled to a single gateway.

David Roberts:

To what extent are grid managers controlling things through Span now? Or does that need some sort of extra signoff from the homeowner?

Arch Rao:

Te're thinking about this as an opt-out model. We recently announced a program with Green Mountain Power, and another one with Silicon Valley Clean Energy here in California, where the customers are by design enrolled into some form of load monitoring and demand management. This doesn't preclude them from overriding those requests, but now the utility is given access to a piece of technology that allows them to control water heaters and EV chargers and space heaters, etc.

David Roberts:

Is Green Mountain giving them away? 

Arch Rao:

That's right, the first part of their program is, they're offering 100 systems to customers in their territory — some that have batteries, some that have EV chargers, some that are just getting a panel up because they're getting an electric appliance — to understand what the benefit will be. The plan is, as we demonstrate successfully that this has value to both the homeowner and the grid operator, to scale this up. They have around 200,000 homeowners.

There's another interesting benefit to this, which is that there are still a number of utilities in the US that are filing for rate cases to adopt smart meters — but frankly, smart meters are not really that smart. They’re communication radios. Our panel gives the energy company or the utility a lot more visibility at the whole home level and at the circuit level.

So imagine by design being able to monitor the consumption of EVs, the production of solar, the major appliances, etc. and the controls. There's a very strong case for how this can be offered nationally through existing utility companies and potentially even being rebated by the government to enable electrification. 

David Roberts:

So just slotting these in where smart meters are now? It sounds like a better technology, but for the same general purposes.

Arch Rao:

That's right, this takes the place of an existing smart meter. Our product is revenue-grade metering at the home level and at the circuit level.

David Roberts:

There’s another complaint I often hear about home electrification, and I'm curious what both of you have to say about this. Whenever I bring it up online, one thing I hear is that electrification would raise homeowners’ costs quite a bit. And it's true, probably, in places where electricity is more expensive than natural gas per therm — which is, I think, a lot of places.

So I'm curious what percentage of homeowners can save money by doing this now, by virtue of existing economics, versus those who need some sort of incentive?

Saul Griffith:

I’ve spent my whole year answering this question, with Rewiring America, which is an organization founded not only to answer that question, but to make the economics better and better and better.

One thing we looked at was, at what point in the future the household economics will be positive for everyone. That point is when the US achieves Australia's cost of solar installation, at $1 per watt on the roof, when we've electrified our heat, when we get batteries installed for under $200 per kilowatt-hour at the household, and when we get EVs to cost parity with gasoline.

That moment is about 2024. At that point, every American household saves more than $2,000 per year on their energy bills.

We're actually getting closer and closer on EVs — a lot of the savings are driven by EVs. Just for perspective: 10 cent per kilowatt-hour electric vehicle versus $3 gallon gasoline; three or four cents a mile for the electric versus 20-plus cents per mile for the gasoline. Carrier, an American heating and cooling company, just created a variable-speed heat pump that's got a COP of 4. 

So the technology is there. If you can install these things in the household without too much headache, then households will realize savings.

I admit we're not quite there today. One of the reasons is that we haven't trained enough HVAC technicians, we don't have enough electricians, there's a lot of excess permitting that needs to be done, there's just challenges to the practical reality on the ground. But they're falling day by day, and there's more and more contractors that will do a heat pump and not recommend a natural gas heater.

The economics are flipping right now. But I do agree that the challenge is, we've got to design and build for the future that's going to happen in 2024, and have to sell it in 2021.

That means there are ZIP codes where it works. We just did a study trying to help the White House on how many low- and middle-income households would already save by flipping off natural gas, and it's tens of millions of homes. You can pick the ZIP codes in the country where it's most favorable, where it's a reality already.

David Roberts:

What's the differentiating feature of those areas?

Saul Griffith:

For heating specifically, it's either because there's high natural gas prices or because there's low retail electricity prices in that location, or because the climate is mild. So across the South and Southeast, there's a lot of homes in the money, across the Southwest and the West Coast there's a lot of homes in the money. It gets a little bit harder in the middle and up at the top of the country, but honestly, every year we're gaining 5 degrees of latitude in heat pump performance. 

David Roberts:

Arch, you must have to deal with the question of upfront costs.

Arch Rao:

Saul provides a really useful macro perspective, through the longer-term horizon, for us to enable this for every home. Expanding on one of the points he made: how do you get to that dollar-per-watt solar? How do you get to the sub-$200-a-kilowatt-hour storage?

I think we're coming at it from an inside-out perspective of affordability. That's what's driving our product design and innovation. When you think about the cost of delivered, behind-the-meter solar today, to homeowners here in the US, a lot of the cost is operational costs, installation cost that stem from design complexity and customization on site.

That's part of what we're trying to solve here: there is too much design complexity, and there's too much design customization, for this to get to a low enough cost to become affordable for every household. 

If you don't have to worry about relocating nodes, if you don't have to worry about changing the ideal size of the solar that goes on your roof, because you're not as constrained by how much capacity the service has, it makes it that much easier for us to reach more customers at a lower dollar per watt. That's what we're solving.

When we think about storage — and this is actually a very debated topic for us internally — resiliency is arguably more of a need for low- and medium-income households than it is for larger, more affluent households. But in order to get your whole home backed up, or your essential loads backed up, most folks cannot afford to have two or three batteries. Part of what we can offer is to combine a single small battery with the Span panel, and effectively you can manage everything in your home.

Again, it goes back to the same point of affordability: lower capex, lower opex. That's how we're going to move closer to a smaller carbon footprint, but also closer to enabling these customers to take that next step towards electrifying their appliances. 

David Roberts:

Spell that out a little bit, why a Span panel plus a small battery can do the same work as two or three big batteries. What exactly is the magic there?

Arch Rao:

There's two pieces to it. One is, when you think about backing up an entire home, you're having to place a large bank of batteries to meet what your maximum load could be. And you have to wire it up such that you're placing this bank of batteries upstream of all of your circuits, or all of your loads.

In some places, like California, that requires quite a bit of installation labor, because you're having to disconnect the meter, move all your circuits and loads into a separate critical-loads panel, etc.

With the Span panel, we are a one-for-one replacement for your panel — or it can also serve as a sub-panel or critical-loads panel. And it takes in all the circuits that you already have in your home, with all of the existing breakers, so there is no on-site customization required.

Once installed, we are able to see and manage your load such that you can get more outage protection with a single battery than you can with two batteries, because you're aware of what's consuming power during an outage. One surprisingly common piece of feedback we got from customers when installing home batteries for Tesla was that they inadvertently discharged their home battery into their car during an outage, because they didn't know any better. They didn't know they were experiencing an outage.

We can solve the problem for some people by giving them more batteries, but I think we can solve the problem for all people by giving them a more controllable and dynamically managed load-control panel. 

David Roberts:

So in the event of a brownout or a blackout, your Span panel can turn off or turn down inessential loads. It can prioritize.

Arch Rao:

Yeah. Today, the interface is very simple. It is designed for anybody to use, where it says, here are my must-have loads, here are my nice-to-have loads, and here my non-essential loads. It all goes to the Span app.

Depending on the state of charge of your battery, or more importantly, how many hours and minutes of backup you have left, based on your load profile, we will shed those non-essential loads first. When you approach, let's say, half of the capacity of the battery, it will shed those low-priority or nice-to-have loads and send you a notification saying, “Hey, David, so you know you have another five hours of battery remaining, would you like to share anything else?”

That's what we can do today. Where we are headed, because of the massive amount of compute and communication we have built into our panel, is the ability to talk to your appliances. We will change the set-point of your thermostat; we will change the frequency of your compressor turning on or off inside your refrigerator. We can get a lot more out of the same number of kilowatt-hours than you would if you were just blindly discharging it into your home.

David Roberts:

Explain that a little bit. Today, it can basically turn up or down the amount of power going to the appliance. You're saying it'll have more fine-grained control over appliances in the future?

Arch Rao:

That's right. Today we turn off circuits — which is a harsh sort of customer experience. Very soon, we will have the ability to interfere, to change, let's say, the rate of charge of your EV, or change the set-point of your thermostat. And in the future, we're working towards capabilities where we can control most if not all the devices in your home. 

David Roberts:

Is that going to require any change in the devices, or is it more intelligence on the Span side?

Arch Rao:

It’s just more over-the-air software updates that we can release. And as long as your device is “smart,” meaning it has WiFi, we should be able to talk to it. 

David Roberts:

Is this also true of fossil fuel appliances, like if my Span is connected to my natural gas furnace? Or is this an all-electric type of thing?

Arch Rao:

It works with any appliance that has a digital interface. So if you had a fossil-based heating system for your home, but if it was controlled through a Nest thermostat, then we will soon be able to talk to the Nest thermostat and ask it to turn off or go to a higher setpoint. So that's possible today, but directionally, where we of course want to get to is a place where the heating device itself is electric, and we're talking directly to it. 

David Roberts:

Let's turn to another big topic, which I know Saul has burned a lot of brain cycles on. We need to do all of this home electrification quickly, and the main thing we need to do to make that happen is to fund it. That's always the choke point — funding and financing by the customer. There are the appliances, there's the smart management software, there's the rewiring of the home electrical system to be ready for 220-volt appliances, and all of that costs money.

With customers, we know it's legendarily true that even if the lifetime cost of ownership of something is much lower, the upfront costs scare people off. Basically, people don't like upfront costs. So what are the right new mechanisms or tools to fund these things such that we can get them going, quickly, at scale?

Saul Griffith:

It's a great question. It matters a lot which part of the market you're in. The good news is, we have historically low global interest rates. So the set of goodies that we're describing are things that people buy every decade, and there's a lot of evidence to show that most likely, you're doing any one of these things either because you just moved house, or you're refinancing your house, or you've just bought a car, or, the new entrant in this game is, you've just put solar on your roof.

At any one of these interventions in your life, most households typically will do a couple of things at the same time, and at those points we need mechanisms that tie the financing of these things to the refinancing of your mortgage, and do it at the lowest possible interest rate. Certainly if you could apply mortgage-level interest rates to all of these items, that helps enormously with the financing of it.

There are about eight things we're talking about — the heat-pump water heater, the heat-pump space heater, the two electric vehicles in the garage, the two vehicle chargers in your garage, the Span load center that connects all of them and has the brain, and the induction stove and induction oven in your kitchen.

There is a good argument that we should declare this national infrastructure, because there will be a date in the future where my solar panel is running your cooktop, because I'm out of the house, and there'll be a time where your home battery is supplying my car with some electrons, because that's the nature of electricity — how this is all going to get connected together and balanced.

There's actually precedent for financing the suburbs and homes as infrastructure. FDR created Fannie Mae in 1936 under the Federal Housing Authority, and the government stepped in and provided guaranteed infrastructure-quality financing to build out the American suburbs. We will make this project collectively cheaper for the nation if we do something similar for this bag of goodies.

And if we recognize that if you install any two of these things at the same time, it's cheaper than installing two of them individually. So there's a discount if we're smart about how we do it.

That's a pretty good story for people who are homeowners, in the top 50 percent of homes.

The other thing we really need to recognize is that half of American homes are struggling to find $800 and have credit rating problems. This was before COVID, in the economic disaster that was 2020. So I think we've got to be more honest with ourselves, providing tax write-offs for building these things is not enough — we need point-of-purchase rebates. We know that when people buy hot water heaters or air conditioners, more than half of the time it's under financial duress — your wife is pregnant, your partner is ill, if the water heater fails, you need a hot shower tomorrow, right? You go to the store and they say, well, the natural gas one will cost you $100 less, and I've got a contractor ready to go. That's why people make that decision. 

So we need to think about this, not as tax incentives for the top households, but point-of-purchase rebates for all households to do it? How do we do clever on-bill financing? How does the utility play a role in building out this infrastructure? Quite frankly, given the time frame required, every single possible mechanism that lowers the cost and lowers the brain damage of doing it at that point of purchase is what we have to do.

Arch Rao:

Mechanisms like the investment tax credit already exist and products like Span already qualify for it, because when you install a solar system or a battery system, the new measure allows for standalone storage to also take incentives with the 26 percent tax credit. They are good and they are in place now. 

But I agree with Saul 100 percent that moving toward a model where we think about products like Span, home-electrification products, EV charging, as part of your home infrastructure and potentially part of the grid infrastructure, allows us to go a lot farther in terms of what types of financing we offer.

Even before we brought Saul on board as an advisor to Span, he and I were independently thinking about the idea of mortgage-based financing for improvements to your home. We see a path towards offering solutions like this to every home through mortgage lenders and insurance providers.

David Roberts:

And is there anything interesting to say about renters and landlords? This is another thing I hear whenever I talk about this online: “I'm a renter, I have such limited control.” What kind of problems does that raise?

Saul Griffith:

I don't think any of us, if we're honest, have a satisfying answer to that question yet. Somebody will figure out the business model innovation or the piece of policy that will help. I think what we can do is describe the world we'd like to get to.

That $2,000 or $3,000 per household in savings is $300 billion a year in the US, and there's a lot of options of where that money could go. Those savings could go to the bank that wants to finance you and make it in the interest rate, it could go to a company like Tesla or Sunrun who'd like to somehow benefit from selling the grid services, it could go to the utility.

I think we should go with a guiding principle that we should always choose regulations where we return as much of that money as possible to the household, then try to figure out how to make that happen. In this case, you'd like to figure out the mechanisms and the regulatory environment that would motivate the renter or the landlord to both do these things and to pass some portion of those savings along to the household.

Arch Rao:

There's a combination of thinking about this problem from a technology standpoint, which is primarily what we're doing, and thinking about the regulatory policy standpoint, which is a big part of what Saul and his team are doing. The alignment with the economic incentives, be it the landlord, the homeowner, the bank, the solar installer, etc. — I think there isn't one form that fits all here. 

Going back to your original question, David, about funding, there's one piece I think deserves just as much attention as the rest, which is building out the workforce that can scale this up, by offering vocational training to a larger number of Americans so that we can actually get these deployed at a scale and pace that will have an impact. There's about 60,000 licensed electricians in the US now, and that's woefully inadequate if you want to electrify every home in the next decade.

David Roberts:

Right. I was gonna return to that in just a sec. 

One more thing about the barriers for homeowners: it's pretty well known at this point that, the closer you get to home and hearth, the less you're talking about economics, the less rational interest maximizing you have, and the more time constraints and psychological constraints come in — which is to say that electrifying a whole home today is, aside from the cost, a huge pain in the ass.

So even in places like California, which you’d think would be cutting edge on this, I hear, “I had to talk to six different contractors and their permitting boards, and there's different vendors, trying to coordinate, trying to figure out which piece to do first, and how to finance one piece with another piece…” So the transaction costs seem enormous right now. And I can say from my own point of view, I would pay a lot of money to have to think about this s**t less.

So how do you reduce those transaction costs? Is there a future where I can just call somebody and say, here's my check, please do all my stuff, and then that's the last I have to think about it? 

Saul Griffith:

I think you've led yourself to the correct answer here. That's already emerging, and it is one business model innovation away from being able to do it.

There's a company in Australia called Brighte that has similarities to a company in the US called Mosaic. They offer the financing; they have contractor networks. There's a whole lot of companies starting to do that. They’ll be able to afford to pay full-time lobbyists to help change the laws and the regulations.

Rewiring America is focused on this, because we've got to fix the rooftop-solar regulations in America. In Australia, there’s suburbs with greater than 50 percent solar on their roof, all enjoying electricity that's a third the price that the grid can offer. Yet in America, we still have less than 2 percent of homes with solar on their roof, because it's still too expensive. And we have a prevailing wage of $35 an hour in Australia, so it's not because of minimum wage.

This is red tape. We've red-taped the future out of existence. There's been 120 years for fossil fuels to build regulations and nurture workforces that work for it. 

So I think it will need some combination of regulatory change and organizations lobbying for those changes, with some business-model innovation. This is going to be true not just in one affluent ZIP code, but every ZIP code.

We're really at the inflection point. There's a famous photograph of New York City with 200 horse-drawn carriages in the foreground and one car, in like 1908, and then they show you a photograph taken from the same point 10 years later and there's 199 cars and one horse. We’re at that 1908 moment, with one electric house and 200 non-electric houses, and it's all going to change this decade. It will take a lot of contributions, like what Span is doing, to enable that. No one exactly knows the answer, we're all feeling our way forward here, but we're doing that with a fair amount of intelligence about what needs to happen. 

Arch Rao:

I recently wrote a blog about this being our decade to decarbonize. If you think about the growth in the adoption of solar over the last decade and a half, if you look at just the rate at which EV adoption has grown in the last half a decade, I'm strongly a believer and a proponent of the fact that this decade is going to be about electrifying many things in your home. 

To your question about home electrification companies, I don't think there's going to be one — there's going to be many companies offering a combination of products that allow you to choose electric over gas, to the point where it truly does become as easy to adopt these as it is to get a home appliance.

If you compare the shopping experience between getting a washer/dryer installed in your home versus getting an EV charger or a battery system installed, the latter is an order of magnitude more complex, because of all the red tape we just talked about. The path to simplifying that is from the top down — improving regulation, simplifying standards, moving toward policies like we have in California, where we have solar mandates and EV mandates and soon hopefully an electrification mandate — but also from the bottom up, which is how do you build pieces of technology that become the new standard for every home, that truly make it as easy to adopt a home battery or an electric appliance as it is to get a refrigerator delivered to your home from Home Depot. 

Saul Griffith:

Just think about the events of the last month: Joe Biden driving an electric F-150, which in and of itself is amazing. The F-150 is the most produced car in the history of mankind — 42 million F-150s have been built so far, double second place, which was the ubiquitous Volkswagen Beetle. So this really matters.

What was not nearly as much in the headline, but I thought was fascinating and really interesting, is that Sunrun is partnering with Ford. So they might say, “You can make the installation easier, while you're financing your car, let us just suggest that it might be a great time to finance the solar installation and this battery and this upgrade as well.” These things are happening and they are about to happen at lightning speed and enormous scale.

David Roberts:

I did a podcast with Lynn Jurich, the CEO of Sunrun, a couple of weeks ago in which she discussed that, among other things.

Arch Rao:

Sunrun is a good example of a company that grows up to become a home electrification company — much like I think many other companies will. Tthey have to, because just being a solar installer or a solar financier is not going to cut it.

David Roberts:

Are you thinking a lot about, not just the technology of the Span box and how to make it better, but business models and partnerships? How to package things and ease the transaction costs?

Arch Rao:

Absolutely. We're thinking about three pieces.

We're thinking about how we generate demand through channels that haven't been the primary focus. So today, I think most green energy solutions start with a conversation around the coffee table about solar. That's a missed opportunity, because you can be thinking about a panel upgrade along with adopting an electric heater or an EV charger, or even just a battery without a solar system for your home. That's what we're seeing happen more and more often.

These are all technologies that can be installed with any electrical contractor, so we're seeing electrical contractors as a key channel. We're working with mortgage lenders and financiers to make it as easy as giving away a Span solution as part of your refi, because that now makes your home that much closer to being a clean energy home, and an entry point to offering better products or more products in the future.

The second pillar for us is the fulfillment side, how do we build a network of qualified service providers like electricians that can install the system for us, but at the same time, provide ongoing service to the customers.

The third is technology partners — thinking about any product that we're going to build ourselves or we're going to partner with OEMs. That's where we're headed.

David Roberts:

On the regulatory front, Lynn said this as well, about Sunrun, that her mantra now is soft costs, dumb regulations, red tape. It's close to being conventional wisdom now in the US, at least in energy circles, that this is the big barrier at this point. So is Span getting involved at all in lobbying for specific regulatory changes or laws? What's your top priority there?

Arch Rao:

Well, there are a couple of things. There are big-picture opportunities, like the work that we're doing around Rewiring [America], asking for or setting up programs that can rebate the adoption of upgraded electrical panels. Then there are more nuanced but specific things like helping define standards with NFC or UL, that really help us rethink our arcane model of distributed electricity grids. We think about building electric conductors and circuit panels with oversized capacity for that larger load to come on.

So it's across that whole gamut, but I think the biggest levers for us are going to be better financing and, like Saul talked about, upfront rebates. 

David Roberts:

Saul, do you have any particular regulatory targets that are top of your list?

Saul Griffith:

This is how extraordinary the soft-cost problem is for solar: your cost of solar modules now is 25 cents per watt, yet the cost of installed solar on an American rooftop is $3 a watt. So let's just dispense with the belief that this has anything to do with the cost of the solar [panels].

Of that $3 in the US, about 75 cents is the cost of the sale. Because American rooftop solar is slightly more expensive than the grid, you gotta spend a lot of money to sell it — it's as much for the cost of the sale as the cost of the hardware. It’s crazy.

There's another 50 cents to $1 in the costs of permitting inspection and those regulatory soft costs. So we’ve just got to get to the binary flip where you get over that, getting rid of that sale cost because you lower the cost of electricity. That's huge.

Pretty much in every ZIP code, when it gets to $2 a watt installed, it's cheaper than the grid. That's the rule of thumb. Just making bigger systems, so that the people that are putting it on your roof can install eight kilowatts instead of four kilowatts, is enough to pretty much get you to $2.

We're really on the cusp of a very exciting moment. A wonderful guy called Andrew Birch has done a lot of work to do a solar app, which is an online portal that's going to make the application for the permits much more streamlined and easy. That's important. Every utility in every state has a slightly dysfunctional relationship with time-of-use billing and reverse metering.

We need a future of what I would call grid neutrality, meaning any generation resource is treated equally, any storage resource is treated equally, any opportunity for demand response is treated equally — so that homeowners are motivated to put the maximum amount of solar and the maximum amount of [electric] vehicles in their garage, and benefit from participating in all of these transactions to balance the grid.

The utilities traditionally resist this because they also sell natural gas, but the reality is, we're going to double or triple the load in every household and rooftops aren't big enough to meet all of that so the electric utilities are going to see load growth. They win no matter how they play here, and we just have to kill off the protection of the natural gas part of the business component and move towards grid neutrality, which will be hard-fought regulatory changes. 

The Federal Energy Regulatory Commission is debating a bunch of issues right now that are super important, that will determine the future. We know where the North Star is, it is grid neutrality, where everyone gets to play and benefit from the resources that they install, and that is also going to bring down the cost and maximize the savings.

David Roberts:

What country is farthest along the pathway toward grid neutrality?

Arch Rao:

The countries that I've seen as the most forward-leaning when it comes to understanding the need for and supporting the adoption of distributed energy and grid neutrality are Germany, the Netherlands, and Australia. Now, this is a blanket statement, they each have their own deficiencies in terms of how these programs are implemented, but I think they're certainly farther along than the U.S. right now, where red tape is really delivering a crushing blow to the rate at which we can and should be adopting rooftop solar, batteries, electric vehicles, etc.

Saul Griffith:

I’m going to give you two different answers, both of which might be funny. I think the unregulated markets in Africa and Malaysia and smaller countries in Southeast Asia are being very innovative precisely because they have so little infrastructure, they can write new rules as they go. I wouldn't look to any Western nation to be perfect here.

Australia is having to face up to this reality first, and that is because they got to very cheap rooftop solar first, they go to very high penetrations — 50-plus percent — first, but Australia is dysfunctional in its own unique way.

What you really want is a country that has Australian rooftop solar policy, Californian or Norwegian electric vehicle policy, and South Korean or German heat-pump adoption. That's the country where the economics are very positive for the household. So we know how to do this, we just don't know how to do it in one place.

David Roberts:

Are there, in the US or in the world, places that are too hot or too cold to fully electrify a home?

Saul Griffith:

Very cold places are difficult, although heat pumps are getting better and better. But places that are cold enough generally are also quite wooded so there is a good opportunity for wood pellets or just traditional wood stoves. So there are places in the world where a wood stove is still going to be the cheapest way to heat your house, and it can be renewable.

In Australia, there are states like Tasmania, where that wood-fired heat supplements the electrification of the rest of the house. They're already a 100 percent renewable grid with hydro. There are other places in the world, such as Arizona, or Queensland in Australia, where the biggest mismatch between electricity generation and load is the peak summertime air-conditioning hours. So the solar peak is at two o'clock in the afternoon and the afternoon peak is at 6 pm, and we've got to match that.

The ones where that is the case for heat, you know, small amounts of storage, whether it's thermal storage or electrical storage, can bridge that four-hour gap and you're in good stead. There are technologies that we're working on, thermal storage technology that can bridge the 16-hour gap between the solid generation periods and when you want heat at 4 am. So that's the heating problem.

There will always be a few small ZIP codes where there are exceptions, but I think now everyone can squint and see that we have answers for nearly all of these cases.

David Roberts:

Arch, you're selling all over the country?

Arch Rao:

We are selling nationally,, across the US. We've now deployed, I want to say, in over 18 states across the country, and the efficiency of heating or cooling hasn't been a barrier to adoption of Span panels or adjacent technologies for an electric home.

David Roberts:

And are Span customers saving money now? Are you tracking usage or trying to get data on usage and finding out what actual results on the ground are?

Arch Rao:

We certainly are. The metrics we are most closely tracking are, are they getting empirically this benefit of going with fewer batteries? We're seeing that with a Span panel you get roughly 1.65 to 1.7x value per kilowatt-hour available, because of how we can manage the loads. That's quantifiable value.

The other metric that we often look for is, what is the reduced time to install and cost of installation labor? That then directly translates into a lower dollar per watt or dollar per kilowatt-hour to customer, and we're seeing that play out more and more significantly, especially right now, for a couple of reasons.

The electric panel as we have it designed is a Ul 67 box. What that means is, it's an over-the-counter permit, so the permitting for a product like Span is, by design, a lot shorter than the permitting for a rooftop solar system or battery system. We're able to help our installation partners get projects going very quickly.

Right now, because of global supply-chain constraints, we're seeing that there's a significant supply shortage of lithium-ion batteries. especially in those cases, as we approach summer here in the Northern Hemisphere, we have installers that are able to go back to their customers and offer them a Span panel plus, say, one battery, as opposed to the previously designed two- or three-battery solution, and help them move to a more self-sustainable lifestyle.

David Roberts:

Air-source heat pumps versus ground-source — how far do you think ground-source is going to get? Or do you think air-source heat pumps are going to get good enough to do everything?  

Saul Griffith:

So ground-source heat pumps, for everyone's benefit, are where you use the temperature of the ground as the input to your heat pump. The ground about four feet down, around the country, is about 55 degrees Fahrenheit all year round. An air-source heat pump just takes the outside air. 

Ground-source heat pumps require something like an enema for your front yard or your back yard. It's a lot of drilling equipment; it's a very high capital item. If you're doing a new-build home in New England, or you have large acreage in New England, it's a no-brainer to go with ground-source. For any milder climate, the ease of installation of air-source looks like it will be the answer.

As with nearly everything on climate, electrification, and energy, these are giant markets where there's room for both players, and the details will be around local regulations, local geography, local climate, so the answer is yes and/or both.

If I was picking the market share in 2036, I'd say 80 percent air-source, 20 percent ground-source is ambitious for ground-source. Unless we get robots that can drill holes really well and take all the headache out. But you still need the physical space to drill the hole, and most people don't have enough land under their house to sink the heat.

David Roberts:

Especially if you're building new developments — if you're doing a common ground-source heat pump for multiple buildings, that's probably the most efficient way.

Saul Griffith:

This is the decade for change, so don't rule out unexpected solutions. We've got to do something with the 4.4 million miles of natural gas pipelines in the country. Why don't we use that as a collective ground source? It's not impossible to think there might be a solution in lane three that helps solve this, but I think ease of installation and low cost of capital certainly favors air.

David Roberts:

I recently found out that the battery in the new Ford pickup is bigger than a Powerwall, bigger than a home battery. So in terms of home power backup, what percentage, in 10 years, will come from home batteries versus connected EV batteries?

Saul Griffith:

Overwhelmingly electric vehicles. Any electric vehicle is going to do 250-plus miles, which is basically where real range anxiety just goes away. The battery in a sedan that does that is going to be 60-plus kilowatt-hours and for a truck it's going to be 100-plus kilowatt-hours. So that's four days of your current electrical load in your house. There's 1.88 cars in every driveway in America, so that's like a week's backup.

The Powerwalls and LG batteries that people bring in the side of the house, they're like five to ten kilowatt hours. That's a very, very small fraction. I'm working on thermal storage technology, as are other people — being able to turn your space heating, your water heat, and your vehicles into storage is a big opportunity.

Those two batteries dwarf any other battery that will be on the grid 10 years from now.  

David Roberts:

So what's the case for home batteries then? 

Saul Griffith:

Well occasionally, both cars are out of the house and you still want to play your Xbox.

David Roberts:

So, backup for the backup.

Arch Rao:

Actually, I quite disagree. I think that the vehicle-to-home concept is a powerful one. The size of the battery in your car is roughly 10x or at least 5x the size of a home stationary storage battery, but the emotional choice for a homeowner to say, I'm going to take the energy available in my car's battery to power my home, I think it's a difficult one. It's kind of analogous to saying I have a large power generator in my car with my petrol engine, and I'm going to use that as a genset for my home, right? 

David Roberts:

But if you're in the middle of a blackout?

Saul Griffith:

Honestly, we're not even disagreeing. It's just that this is a huge battery. There are so few time periods in the year where you're gonna want these things. This is not even an argument. There's a huge number of reasons to have a small battery in the home. But you really just need to understand that our car batteries are enormous. Even if we put those batteries in 200 million cars, the whole grid for the whole U.S. can run off our cars for days at a time. But that never happens, because you still have solar, you still have wind, you still have all of these things happening. But the anxiety we have about storage goes away as we deploy batteries in vehicles, batteries in homes and batteries in our thermal systems.

Arch Rao:

And the cost of stationary batteries comes down as we continue to adopt more electric vehicles and batteries in your home as well, which makes room for not just thinking about storage as being a backup to a backup type of solution, but instead a device that can provide you with energy management — being able to charge when you have excess solar, etc., benefits to it that aren't provided by vehicles, where the primary use is actually to get around.

The second part of it that I think has spun up a lot of conversation since the F-150 announcement is, how does it actually power your loads, because it's a giant battery, but you still need a whole host of electronics and grid-connect and grid-disconnect devices.

David Roberts:

That's what Sunrun is trying to sell alongside the pickup truck; that’s the piece they're trying to provide.

Arch Rao:

Yes, they're going to be installing those solutions. But to generalize the problem statement, there are going to be millions of electric vehicles, not just F-150s, so what is the ideal solution for being able to “island” your home and power it off of your vehicle battery? I think those are the kinds of products and solutions that we're thinking about here at Span — within our panel, we already have a grid-disconnect built in. 

David Roberts:

So you can island if you have a Span panel?

Arch Rao:

That's right. Our main breaker has its own control system built into it, where it's monitoring the health of the grid, and if it determines the grid is sagging, or it's out of spec from a frequency or voltage standpoint, we automatically, safely disconnect and bring up your battery as the main power source for the home.

It can then orchestrate your solar and your loads as needed, to then manage that “off-grid” home. And then when it's safe, it rejoins the grid, making sure it doesn't cause issues on the grid — you want to be careful about frequency when you rejoin the grid, you want to be careful about how much power you push back into the grid, etc.

That's all built into our current generation product. But as we think about future products, the types of things we spend a lot of time thinking about are exactly the question that you're asking, David, which is: can cars provide power, not just when off-grid, but even when your on power from the grid? Does it make sense for us to use vehicles as a day-to-day thing?

And what are the types of technologies you need built into the nerve center of your home, i.e., a load panel or an EV charger, that allow you to do this in a seamless way, without having to go through a complex design process and having a smorgasbord of stuff on your garage wall just to be able to use that expensive battery in your car?  

David Roberts:

If you could create this theoretical country with all the best distributed energy policies together, do you guys seriously think that we can decarbonize 100 percent of US homes by 2035, which is the 100 percent clean grid target year? 

I think people get daunted because it's so distributed, the work itself is so distributed, the use cases are so distributed and so various, and there are so many different kinds of homes and owners. It just seems so daunting. 

Arch Rao:

Yes. The alternative scenario is a bleak one. We have to align products and policies and people to move toward at least decarbonized homes (decarbonizing industry is a whole other can of worms). I would say, yes, it's possible, and we need to be working toward doing that. 

Saul Griffith:

We need to make it possible. It is physically and technically possible, and the more things you electrify, the easier you make it to electrify everything. We will electrify steel making, we will electrify the manufacturing of aluminum, we will electrify a huge amount of industry. Then you dial industry up and down, there’s huge inertia in those machines, and so they will be a great asset, just like your car batteries and just like our home heating systems, they will be critical to balancing the grid. People really don't appreciate that.

It'll be about the same amount of wire that's distributing electricity to the grid, but three times as much electricity will go over it. More than half of your utility bill cost is the cost of that distribution network. So if we're putting two or three times as much electricity over it, the price of that distribution cost is going to come down. The more we electrify, the cheaper it gets to electrify everything, the easier it gets to electrify everything.

So this is possible. Yes, it is heroic. Yes, the timeline is now. A good climate outcome is, we do this in 15 to 20 years, a terrible climate outcome is, we drag our feet and we take 40 or 50. 

David Roberts:

Well, there's an appropriate call to arms to end with. Thanks, guys, for taking all this time. Appreciate it.