Not too many years ago, homeowners that wanted to go solar faced a daunting set of obstacles: $30,000 or more to buy the equipment, countless hours negotiating a labyrinth of state and local bureaucracies, $2,500 for permits and unknown maintenance costs. Enter solar-as-a-service, sold to homeowners by a new generation of providers that will lease solar equipment to homeowners with little or no upfront cost and sell them power at a predictable, low rate. One of the first, and arguably one of the most successful, companies to offer solar-as-a-service, is Sunrun, a five-year-old venture started by Ed Fenster and Lynn Jurich, Stanford MBA 2007 classmates. Fenster started his career in venture capital; Jurich also worked in venture capital before helping to found Sunrun and was named one of the "10 most powerful women entrepreneurs" by Fortune magazine in 2009.

Since 2007, Sunrun has placed solar-voltaic panels and generating equipment on the roofs of approximately 28,000 homes in 10 states, with installations on both coasts, the Southwest and Hawaii. Fenster, who shares the title of CEO with Jurich, spoke to Bill Snyder about Sunrun's business model, and the challenges the San Francisco-based company faces as natural gas prices drop and government subsidies for solar power installation are gradually reduced.

Five years ago, no one had heard of solar-as-a-service. How did you develop Sunrun's business model?

Ed Fenster: When we started looking at solar power in 2007, the first two things that jumped off the page were "Holy crap, this stuff is expensive." And as compared to other forms of energy it has a really small minimum-efficient scale, that is, the smallest unit that you can make something and have it work. And that's unique in energy generation. At the other end of the scale is a nuclear power plant. It will make enormous amounts of money, generate enormous amounts of power, and cost enormous amounts of money to build.

At the time everybody was thinking bigger is better: "Where can I put the most solar panels to achieve the largest scale?" We came at it differently. The home is the best use case for solar in the long run.

Why was the market slow to take off?

By 2007, two million people or so should have had solar power in their homes, but only about 100,000 actually did. So something was broken. What was broken was the way it was delivered. It was a business problem, not a technology problem. And that was something Lyn and I knew how to fix.

If you imagine yourself in the shoes of a person considering getting solar power, the demographic is early 50s, dual wage earners with kids. They are exceptionally busy people. They have never bought a solar system before. Out of nowhere someone shows up at their door and says this thing is going to last for 30 years, it will require no maintenance, and this is how much power it is going to make, and I'll give you a 10-year warranty, and it is going to be great.

And you're saying I've never bought anything that lasted 30 years, I don't believe that it is maintenance free, I have no idea how to maintain it myself, or even to know if it is working or not. You're telling me it is going to produce this much power, but I have no idea if that's the case.

So how did you modify that approach?

We thought if we could really simplify the sales pitch so that when you walked into a home you'd say we'll put solar on your roof, we'll charge you less money per kilowatt hour for the power we deliver you, and you don't have to pay us anything upfront. It's pay for performance; it will save you money, why wouldn't you want that for your family? You've now transferred the risk and the upfront cost from the homeowner to Sunrun.

Customers don't pay upfront for the cost of the installation. I'd assume that's the main selling point.

It's not. If you actually survey our customers it turns out that our providing the upfront costs is only the second most valuable thing to them. More valuable is that we maintain it for them. In fact, 30% of our customers prepay the entire 20-year agreement (for the cost of power) upfront. People just don't want to own the power plant that goes on their roof. Everybody is busy; who wants to come home and deal with another broken thing? Electricity is something people don't want to think about.

Where do you get the money to buy the equipment?

We partner with large financial institutions, notably U.S. Bancorp, and have financing for over $1 billion in equipment.

Sunrun isn't vertically integrated; you don't do the installation, you don't manufacture the equipment. Why not?

Three-fifths of the cost of residential solar is soft costs. Customer acquisition, municipal permitting, and things like that. The real winners in the marketplace will be those who can minimize those costs.

We want a scalable business, which involves spreading fixed capital investments in marketing, software systems, and so on across millions of customers. But construction on a residential basis has never been successfully scaled nationally. It works as a local business because you win by having density; you win by knowing the building department; and by not having too much overhead. In addition to that, homeowners like to work with local businesses for construction.

What about manufacturing?

It's not a good business. At the time we got going, there was a global silicon shortage but what you could see was that the silicon manufacturers had taken enormous prepayments from their customers to build new production capacity. You could very clearly see a crash in panel price coming about 2009.

Is there hope for the U.S. panel industry?

There's no hope. Solar manufacturing is not a good match for our capabilities and costs. The U.S.'s comparative advantage is in very advanced manufacturing. It is easier to manufacture solar panels than even basic semiconductors, and the U.S. has long been uncompetitive in basic semiconductors. End-to-end solar panel manufacturing has high fixed-to-variable costs, resulting in unprofitable pricing at average cost in times of excess supply.

Some energy experts believe that fracking will boost natural gas production and lower prices, which in turn will lead to declining electricity prices. If that's the case, won't that make solar-generated electricity less competitive?

Electricity is actually a relatively small use of natural gas compared to other uses. There are various schools of thought as to where natural gas pricing will go as a result of fracking. Even if natural gas prices fall, retail electricity prices will continue to increase.

Why?

The retail price of power is a function of two things — the wholesale power cost and the cost to transmit and distribute that energy. The amount of required investment in transmission and distribution is so significant that it would more than offset savings on the wholesale side. Over $100 billion needs to be invested. If the utilities have to invest in maintaining their grid, the amortized cost of doing that will exceed any savings from the low cost of natural gas.

But since we put the plant on top of the user's house, we don't have to deal with transmission costs.

What about utility-scale solar power; will it suffer if natural gas prices drop sharply?

It is a major problem for utility-scale wind and solar developers. If you build a power plant in the desert, which is where large scale solar installations need to be built, you need transmission and distribution to get that power to the consumption point. And that's too expensive. In fact, the pace of RFPs (requests for proposals) for utility-scale renewable projects has already slowed to basically zero. The only reason someone would build utility-scale solar now is if a regulator forced them to.

There are significant tax breaks for home owners wishing to go solar, but those will be shrinking. How will Sunrun adapt to that change?

We have a 30% tax credit today nationally. On Jan. 1 2017, that becomes 10%. Even if you assume the cost of power doesn't increase over that period — and we think it will — we would need to eliminate 20% of our costs to stay even. We are eliminating 10% to 15% annually.

That's a big reduction. How are you doing it?

Equipment prices are going down, but most significantly the soft costs are cratering. Companies like ours are figuring ways to originate customers for half the cost they used to. We've automated system design, and have been able to reduce expensive permitting requirements in many states.

You mentioned that home-based solar got off to a slow start. That was five years ago. What do you think the market will look like five years in the future?

In another five years we'll see solar on more than a million homes, compared to about 250,000 today. Americans will realize home solar is easy and necessary for our planet, not just a nice to have.