How a Commercial Solar Deal Actually Pays Back, Line by Line
Last issue covered why you can't buy solar projects in REIT form. This issue opens the machine and shows what those projects actually earn, using the structure I know best because it's the one my company has been executing since 2012: the commercial and industrial (C&I) solar PPA.
The setup
A business, school, or nonprofit wants solar on its roof or parking lot but doesn't want to spend capital. A developer like us builds, owns, and operates the system at no upfront cost to them, and they sign a power purchase agreement: a long-term contract, anywhere from 15 to 30 years but typically 20, to buy the electricity the system produces at a set rate. (Our EV charging deals run shorter, usually 10 to 15 year revenue-share terms, but that's a future issue.)
The customer's math looks simple from the outside: pay the PPA rate instead of the utility rate, save money from day one with zero capital outlay. In California under NEM 3, our C&I PPA rates typically land between 13 and 18 cents per kilowatt-hour, and smaller C&I customers and HOAs can run 20 to 23 cents depending on how much of their bill is demand charges.
But here's a detail that separates real underwriting from a sales flyer: the savings calculation is not PPA rate versus the gross utility rate. We calculate against the customer's avoided cost: what they actually stop paying the utility once solar is on site, netting out demand charges (which solar may partially reduce, generally through peak shaving and time-of-use avoidance), the impact of switching rate tariffs, and how much of their usage the system genuinely offsets. Two customers with identical utility rates can have very different avoided costs, and quoting savings off the gross rate is how customers end up disappointed and developers end up with terminated contracts.
The owner's math is the other side of that contract. Here is where every dollar comes from and where it goes.
Revenue: three streams, one contract
1. Energy payments. The core of the deal. System production (kWh) times the PPA rate. Two details matter more than the headline rate:
The escalator. Many PPAs include an annual price increase, anywhere from 0 to 2.9 percent, typically 1 to 2. Over 20-plus years, the difference between a flat rate and a 2 percent escalator is enormous. The bet embedded in the escalator is that utility rates rise faster, so the customer's savings grow even as our price does.
Degradation. Panels lose roughly 0.5 percent of output per year. Year 20 production is meaningfully below year 1. Every serious model bakes this in; every amateur model forgets it.
2. Tax benefits, for the projects that still have them. For most of solar's history, the federal investment tax credit and accelerated depreciation represented a large share of a project's total value, flowing to tax equity partners as covered last issue. That era has ended: the credits have expired for new projects. The exception is projects that were safe harbored, meaning the owner locked in credit eligibility under the old rules by starting construction or incurring qualifying costs before the deadlines. Pipelines with safe-harbored projects are carrying something scarce, and the economics of new-build projects without credits now rest entirely on the energy fundamentals: the avoided-cost math above has to work on its own. In my view that's a healthier discipline, but it's a real repricing of the industry.
3. Incentives and environmental credits. Depending on the state: renewable energy certificates, state incentive programs, and for our EV charging projects, clean fuel credits. These vary wildly by market and vintage, and they're a big reason two identical systems in different states can have completely different economics.
Costs: what the brochure doesn't show
Operating a solar project is cheap compared to almost any other power plant, but it is not free, and the cost lines are where inexperienced owners get hurt:
Operations and maintenance. Monitoring, preventative maintenance, routine cleaning, timely repairs. The vendor you choose here matters more than the line item suggests; more on that below.
Insurance. Property and liability coverage. This line has moved against owners hard: premiums are up nearly 40 percent in some geographies. Models built on old insurance assumptions are quietly wrong.
Asset management. Billing the customer, managing the contract, compliance, reporting to lenders and investors. The unglamorous work that determines whether projected cash flow becomes actual cash flow.
Inverter replacement. Panels last 25+ years; inverters generally don't. You handle this one of two ways: build a replacement reserve into the model, or pay upfront for an extended manufacturer warranty. Either is defensible. Doing neither is how year-12 cash flow surprises happen.
Rent, sometimes. On third-party sites, a roof or ground lease payment to the host.
The order of payments
Project cash doesn't go to the owner first. A typical priority stack, simplified:
Operating costs (the system has to keep running)
Debt service, if the project carries a loan
Preferred returns to any and all preferred investors
The operator/owner: what's left
That last line is the "REIT-like" income stream retail investors can't easily buy, and it's why the earlier issues of this newsletter exist. Contracted revenue from a creditworthy customer, low operating costs, 20-plus years of visibility.
What can go wrong
An honest list, because every investment pitch in this industry skips it:
Customer credit. A 20-year contract is only as good as the counterparty, and underwriting the customer matters as much as underwriting the sun. There are two schools of thought on managing it. One: forecast the probability of default directly into the cash flows, solving for a higher risk-adjusted return across the portfolio. Prudent, and it works portfolio-wide, but it can't fully protect you from an industry-wide or economy-wide shock that hits many customers at once. Two: buy credit insurance. More expensive, but closer to bulletproof. Which one a portfolio uses tells you a lot about how its owner thinks about tail risk.
Production shortfalls. Weather variance, equipment downtime, shading you didn't account for. A seasoned developer models these upfront; just as important is working with a trustworthy O&M vendor offering a gold-plated maintenance solution, because good monitoring catches a 2 percent issue before it becomes a 10 percent one.
Development costs that bury the deal. Before a project ever produces a kilowatt-hour, it accumulates transaction costs: legal, real estate diligence (geotechnical evaluations, surveys, title), accounting. On a bespoke C&I deal these can quietly consume the economics. The defense is discipline: keeping processes, deal documents, and underwriting consistent across every project. C&I will never be as cookie-cutter as residential, since each project is bespoke, but treating it like residential from a process and documentation standpoint is what sets you up for success during development.
The host relationship. Roofs get re-done, buildings get sold, tenants change. PPA contracts handle these scenarios on paper; managing them in practice is a real skill.
Development costs. Deals can die before they're born. Legal, real estate diligence (geotechnical evaluations, surveys, title), and accounting costs pile up during development, and on a mid-size C&I project they can bury the economics entirely. The defense is consistency: keeping your processes, deal documents, and underwriting identical across every project. C&I is bespoke by nature, with every site and customer different, but the developers who survive are the ones who run it like residential from a process and document standpoint: standardized paper, standardized diligence, standardized underwriting, repeated across the portfolio.
Policy change. Incentive programs get revised. Projects underwritten on aggressive incentive assumptions are the ones that disappoint.
The public market mirror
Here's the comparison this newsletter exists to make. A public investor can buy exposure to cash flows like these today: utility stocks own regulated generation, yieldcos own portfolios of exactly these kinds of contracted projects. So what's actually different?
When you buy the stock, you own the waterfall above through several additional layers: the sponsor's management fees and corporate overhead, the payout ratio the board chooses, the dilution and dropdown decisions of the parent, and a share price that moves with interest rates and market sentiment regardless of how the underlying projects perform. In exchange, you get the two things direct ownership can never give you: liquidity (you can exit tomorrow) and diversification across hundreds of projects for a few hundred dollars.
Direct ownership hands you line four of the waterfall with no layers on top, and charges you for it in illiquidity, concentration, and the work of underwriting everything this article just described yourself. Neither side is the right answer. But most investors evaluate the stock without ever seeing what the underlying machine looks like. Now you've seen it.
The takeaway
A well-built C&I solar PPA is a machine that converts sunshine and a signature into two decades of contracted cash flow. The economics are genuinely attractive, which is exactly why the structures around them (project debt, preferred equity, funds) are so competitive and so institutional. Next issue: what happens when you add a battery, and why storage revenue is both more lucrative and less predictable than solar's.
If someone forwarded this to you, subscribe below for the rest of the series.
Figures reflect our experience in the markets we operate in; your market will differ. Not investment advice.