Why is rooftop solar so expensive in the US?
The cost of solar hardware has fallen 80+% over the past decade, driven largely by China’s massive production scaling. Rooftop solar in many countries fell nearly in proportion to the hardware costs. Total installed cost of solar in Australia, Germany & UK are now very low, and adoption rates are correspondingly high in these countries. In Australia, for example, average rooftop solar installed cost was $0.99/W in 2023, while in the US it was $3.20/W. So why is the cost of rooftop solar so high in the US? There are a few reasons for this:
Permitting and inspection processes in the US are manual, confusing, slow, and hyper-local. Every jurisdiction has unique permitting forms and rules and inspection practices. Mid to large residential solar companies in the US pay large teams of lawyers, inspectors, officials, etc to navigate these messy processes across all jurisdictions the company operates in. The US has an additional step absent in most other countries: after permitting and inspection, an application must be submitted to the utility company for review (which requires payment) before the system is allowed to be turned on. All these processes are costly, and due to their complexity increase the barrier to entry and reduce competition (vast majority of rooftop solar in the US is deployed by 3 companies).
Electricity prices are very high in the US. High electricity prices means the US solar industry has little or no downward pressure on their pricing. As rates continue to rise, these companies can keep raising the price of rooftop solar while still presenting their customers with a good long term cost saving opportunity. This partially explains why average profit margin in the US solar industry is far higher than Australia and other solar leaders.
Customer acquisition costs in the US are extremely high. In fact, the cost of acquiring a customer in the US are higher than the cost of all the hardware combined. There is a saying that “rooftop solar is sold in the US, and bought in Australia”. This is alluding to the fact that US the solar companies work very hard to gain customers (though advertising, sales commissions, etc), while in Australia rooftop solar sells itself. In Australia, majority of customers come asking for solar since its an economic no-brainer, and the whole process takes one week rather than multiple months. This delay is responsible for more than doubling customer acquisition costs, as over 50% of potential customers fall out of the process during this period between initial contact and installation.
You’ll notice the reasons above are all interrelated / reinforcing. Through this reinforcing effect, the impacts are amplified, leading to the cost of rooftop solar being over 3x the price it would be if we followed Australia and Germany’s lead. We have designed our solar program to minimize these “artificial costs”, which is necessary to make the economics of our program viable in the long run.
Haven’t recent CA solar policy changes negatively affected rooftop solar? How does Potrero Solar deal with these changes?
The California Public Utilities Commission (CPUC) recently enacted a policy known as Net Billing Tariff (NBT) or Net Energy Metering (NEM) 3.0. NBT fundamentally changes the economics of residential solar by changing the way homeowners are credited for solar energy they export to the grid. Rather than credit exported energy at the same rate (in $/kWh) as consumption, the credits are time varying (hour by hour), and roughly equal to the utility’s best estimate of costs avoided. Thus grid users are incentivized to maximize avoided cost, which is equivalent to minimizing total cost. A lower cost grid means lower electricity rates and more tailwind for electrification (thus decarbonization). California is generally the national, and even international, leader when it comes to solar policy, so it is widely expected that many states and countries will adopt something similar to NBT in the coming years.
The US solar industry has been strongly opposed to NBT since the credits for export are extremely low during the hours solar panels are generating electricity, and high only in the late evenings (and extremely high during summer evenings when air conditioner load is high). Thus the economics for rooftop solar have taken a big hit, unless you can store the energy for a few hours and export it during the evening when the utility really needs it. Without storage, the typical payback period rises from 8 to 15 years, making solar-only a very hard sell. With storage, the opposite is true. Typical payback period for solar + storage was 15+ years prior to NBT, with the high cost of energy storage systems limiting adoption to the small set of customers who care more about having power to their home during a utility power outage than they do about the return on investment. With NBT in place, the high cost of storage is partially offset by these lucrative evening export credits, bringing typical solar + storage payback period down to a reasonable ~10 years.
Payback period is only part of the story, however. The size of the initial investment in a solar + storage system is much greater than solar-only. As an example, a homeowner prior to NBT would make a$20k solar-only investment and expect to recover the investment in 8 years. The same homeowner is now being asked (under NBT) to put roughly twice as much capital at risk with a $40k solar + storage investment, and expect to recover the investment in 10 years. Thus you can see why NBT has caused a massive slowdown in the CA rooftop solar industry, leading to the layoff of ~40% of the workforce.
The CA rooftop solar industry has responded to the new NBT policy with lobbying, lawsuits and layoffs. We saw the industry failing to step up, so we started Potrero Solar to re-ignite the CA rooftop solar industry, starting with our local community. We designed a new business model with NBT front of mind.
Aren’t batteries effectively required to make solar a reasonable investment (due to new Net Billing Tariff)?
See question above for more context on why battery storage is often seen as ‘required’ to make the economics of rooftop solar work under the new CA residential solar billing system referred to as Net Billing Tariff (NBT). However, we believe solar-only is still economically viable in the new billing environment, which is key to making our incremental deployment program work. A true incremental approach must start with solar-only, else the first incremental investment is too large / too risky. We wouldn’t be able to cover 100% of the capital risk without a 5+ year commitment. The addition of storage simply carries too many costs:
Energy storage unit(s), including battery modules and additional power electronics (to convert to/from panels to battery, battery to home/grid, grid to battery). These units currently (2024) cost ~$10k each (roughly $1,000 / kWh).
Electrical wiring and panel upgrades, often needed to allow battery to export high power during the evening timeslots when export credits are most lucrative (thus helping offset battery costs)
An additional inspection carried out by the local fire department
We are prepared to cover all these storage related costs for you starting in phase 2. By that point we’ll have a better sense of the likelihood you choose to remain in our program longer term, and of your evolving usage patterns. Another reason we wait until phase 2 is our belief that energy storage systems are still quickly evolving in terms of technology, performance, cost, and safety standards (electrical and fire codes). By the time our customers begin entering phase 2, we expect storage systems will be closer in maturity to solar panels, making it a better time to invest.
Now that we’ve provided motivation for solar-only in phase 1 of our program, we share the two methods we use to make the economics of solar-only work under NBT:
We design a standard, solar-only system with a 10 year performance horizon, rather than the typical 25 year horizon. The short term focus allows us to avoid the scope creep of features, flexibility and reliability one would want in a system designed to stay in place for 25 years. Focusing on a shorter time horizon enables us, for example, to source used solar panels. As utility scale solar farms around the country upgrade panels, they sell their used panels at 50%+ discount (despite having plenty of useful operating life)
Maintain minimum qualifying electricity usage patterns, mitigating the risk that deployed hardware is under-utilized for years. High utilization is required to make the economics work.
In summary, batteries are not strictly required to make a reasonable return on investment under NBT, and this fact is key to enabling our incremental deployment program!
Aren’t large solar farms a more economical decarbonization solution than thousands of small rooftop installations?
We need both. Rooftop solar is more expensive per unit capacity, but it has several unique advantages over larger solar farms:
No transmission and distribution costs. Grid costs are dominated by transmission and distribution, not by generation. Thus, even if PG&E effectively had an infinite well of free power to pull from whenever you need it, your PGE bill would only fall ~20%. The distribution infrastructure to move the power to wherever you need it is 80% of the overall cost (check your bill to confirm!). By generating (and storing) power at home, these costs are avoided. In addition, grid costs are actually dominated by peak power demand, not by total energy delivered. Distribution networks typically have a peak power capacity 5x greater than the average power. By generating power at the home, loads on the grid can be more evenly distributed in time, leading to higher utilization and lower costs in the long run.
No grid interconnection required. Large utility scale solar farms must connect to the grid via an ultra-high power, ultra-high voltage subsystem. This circuitry is extremely expensive and long lead (there are multi-year lead times for these unique transformers). Major grid interconnections also require extensive permitting and analysis. The waitlist for connecting to the grid transmission network is multiple years in most of the US, a major problem that is currently the main bottleneck to decarbonizing our electricity grids (via connecting new renewable sources). The Federal Government is finally working to improve these issues, but it will take time.
No impact on land use. Utility (and community) scale solar farms, by contrast, impact land which is not otherwise affected by human use.
No transmission and distribution losses. Power losses from transmission and distribution lines are minimal on average (~5%), but losses increase when the path to your home is heavily loaded with demand (high current). Losses on parts of the grid reach 20% during peak times, and those peaks will be more and more common as the world electrifies. Since grid costs are dominated by peak power demand, these losses raise grid costs by over 20%.
Doesn’t the California grid already have enough solar? Isn’t it saturated with solar at times?
While its true our regional electricity grid has enough solar capacity to serve demand for a few hours on sunny days, we need to keep expanding solar (paired with storage) until it can serve demand around the clock in the way that a fossil fuel power plant does. Only then will we fully displace the fossil fuel plants. Solar has an effective capacity factor (utilization rate) of ~20%, meaning that roughly five times as much solar generation is needed to serve demand around the clock. In addition, our regional (and national) electricity demand is projected to at least double over the coming 10-20 years. This expected increase is due to the electrification of transportation, heating, industrial processes, and more. For both of these reasons, we need far more solar capacity in our region, we are nowhere near true saturation.
What types of behavior adjustments will help me qualify for the next phase?
The most useful adjustments to make are (in no particular order):
Program smart thermostat to work hard during 10am-4pm, and rarely have to work at other hours. For example, over-shoot your desired temperature by a few degrees during 10-4 or 9-5, so its remains in a comfortable range during the evening/night without needing to keep running the heat pump / AC. Think of your home as a giant thermal battery, storing extra heat/cool during sunny hours and gradually releasing in the many hours after.
Program water heater similar to smart thermostat (see 1. above). Think of your water heater as a dense thermal battery, storing extra heat during sunny hours and gradually releasing it as you use it.
Program your EV similar to smart thermostat and water heater. Often this is already front of mind for EV owners, who are generally used to utility “time of use” rate changes impacting their daily charging patterns
Wait to run dishwasher and laundry the following morning when its not too inconvenient
Software exists to help automate 1,2, &3. If you’d like help, please reach out and we’ll provide more detailed advice.
Do I need to switch my PG&E billing plan to participate?
Not during phase 1. When entering phase 2, you’ll be switched to E-ELEC when we inform PG&E that you have a solar/storage system configured to export energy back to the grid. See how E-ELEC plan rates compare to other plans here.
E-ELEC is meant for homes that have electrified some or all of their large energy loads. Such homes have a high electricity consumption, so the E-ELEC plan was designed to have a fixed cost ($15/month) in order to allow lower average marginal cost ($/kWh) for high consumption households. It’s debatable whether E-ELEC is a better overall deal for most electrified households compared to other billing plans, but it is the only plan offered to homes exporting energy back to the grid.
E-ELEC is not quite as favorable for those EV owners on the EV2 plan who do most of their charging during off-peak (midnight-3pm). Average annual off peak rates do rise from 35c to 40c per kWh (+14%) when switching from EV2 to E-ELEC. Fortunately these cost increases are usually offset by lower peak and partial peak rates, and fortunately California incentivizes participation in the plan with ~$110/year in bill credits.
Do I need to continue to meet the qualification criteria for the current phase?
Yes. Homeowners must continue to meet the usage pattern qualification criteria associated with their current phase. If the criteria, assessed over a trailing 12 month period, are not met, we may choose to return the homeowner to an earlier phase of the program. Returning to an earlier phase means we will collect some of the solar generation and/or storage infrastructure at your home, and PG&E bill savings will fall.
Returning to an earlier phase is rare, and rest assured you’ll be given 6 months of monthly notices to allow you time adjust usage patterns in order to re-qualify.
What if the system stops working? Who is responsible for maintenance? Is there a warranty?
If the system has an issue, we’ll know immediately via our continuous data monitoring. Potrero Solar is responsible for maintenance. We will come to fix or replace it as needed. Our response time will vary in proportion to the severity of the issue. Our incentives are highly aligned with yours, we both save/earn money only when the system is operating well. There is no warranty provided to you as the homeowner since the hardware is technically owned by Potrero Solar. We have warranties with our vendors, but that doesn’t directly affect you.
What if I want to exit the program? or sell my home?
Call us, we’ll come collect the equipment within a month. No cost to you. We are confident this won’t happen often, so we’ll absorb the cost in the case it does.
Some materials may be left behind (electrical conduit, small ‘feet’ where solar panels mount penetrate the roof) in order to keep the labor cost of un-installation reasonably low, and to avoid any risk of introducing a roof leak. We' take care to ensure any remnants are not an eye sore.
If you sell your home, we will continue to bill the new owner by default. They, of course, can exit the program at any time. Its likely you will even fetch a premium on the sale price by having our system in place, actively lowering monthly electricity bills.
What if I get my roof replaced / re-shingled?
Give us a call, we will uninstall the solar panels and mounts, and reinstall them after your roof work is complete. We understand that it’s hard to know how many years you really have left in your roof, but we can’t let that delay deployment and decarbonization!
What if I sell my home?
We will uninstall the system whenever you want. We are confident that uninstallations will be rare enough that we can absorb this (labor cost) risk.
What is the timeline for getting started?
The expected timeline at the beginning of the program is as follows:
Day 1: Sign Up
Day 3: Receive phase 1 qualification assessment
Day 5: Receive install plan, provide any feedback / requests (over next 2 days)
Day 7: submit permit application
Day 21+: phase 1 installation (single day)
I don’t think I have enough area on my roof with good sun exposure. Will I qualify?
We’ve looked at all roofs in Potrero Hill via satellite imagery, and over 90% have sufficient area and sun to make it economically viable for us to deploy. We have a low cost design, and razor thin “soft costs”, allowing us to make relatively small (and shaded) installations economically viable.
South facing roof is not strictly needed, east and west facing are acceptable. Our design is resilient to any partial shading coming from your neighbor’s roof, tree branches, etc.
We may decide to include you in the program even if your home is marginally economically “non-viable”. We place a dollar value on avoided carbon, and if the expected value of carbon reduction is great enough, we’ll include you in phase 1! This will all be communicated through the qualification assessment we provide.
Will the system be able to maintain power to my home in a grid outage?
Not in phase 1, partially in phase 2, fully in phase 3. We’ll elaborate on the definitions of ‘partially’ and ‘fully’ in the near future.
How much capacity will the system have?
Generation capacity will be roughly 4kW, 8 kW, 12kW in phases 1,2,3 respectively. Storage capacity will be roughly 0, 10, 20+ kWh in phase 1,2&3 respectively.
Will you assess whether my roof can support the weight?
The rooftop equipment is light and well distributed by design, imparting a force on your roof of only 4 psf, below the limit for which the city requires a structural assessment of the roof as part of the permit approval process. This helps us minimize cost and maximize the number of eligible homes.
Will the system export excess power to grid if we’re not using it?
No in phase 1. Yes in phase 2+. Starting in phase 2, the credits we can earn by exporting to the grid from the energy storage system will be significant. In phase 1, the potential credits are minimal, thus not worth the added fees PG&E charges for participation in a “net billing” plan.
What if I already have solar?
We’re focused on homeowners without any existing solar for the time being, but if you have an undersized / underperforming system in place, we’re open to replacing it with our system.
Which roof types are supported?
We currently install on anything except clay roofs. We believe this serves ~100% of our community and allows us to standardize our mounting hardware.
What geographic area do you serve?
Please fill out the signup form if you are anywhere in the SF bay area. We are focusing on Potrero Hill and Dogpatch neighborhoods of SF initially, but we are open to expanding our service area if there is sufficient demand. At first, a very small service area has helped us minimize costs and operational complexity.
How do you track our electricity usage patterns? Is my usage data safe?
Prior to phase 1 installation, we pull data from PG&E. PG&E has a usage database which homeowners can access, and can give access to 3rd parties when appropriate. After you signup, we’ll reach out to initiate this process as part of qualifying you for phase 1.
During phase 1 installation, we place a small cellular connectivity module near one of the panels, and periodically pull the usage data over the cell network. This allows us higher resolution data which allows us to make more optimal decisions around preventative maintenance, software configuration, future scaling and/or replacement.
We follow modern security practices to ensure usage data for all customers is safe.
Do you install EV chargers, electric water heaters, electric space heaters?
No but we’ll connect you with the best companies in the area for each of these. We’re also happy to walk through the math with you, and advise you on all the incentives you may be eligible for!
Can I claim federal and state solar tax credits come tax time?
Unfortunately you cannot. The tax credit (30% of cost) is currently only available if you buy and own a system. In our program, we own the system and rent it to you. Fortunately Potrero Solar captures the tax credit, which helps make the economics of our program work.
My building is two units, is there any incentive to have both units join the program at the same time?
Yes, we’ll credit both accounts $200 at the end of the first year of phase 1 if both units are still in the program. Only requirement is that installation for both units are on the same day. In addition, you’ll have the satisfaction of having roughly twice the decarbonization impact!