By 2020, the US will have installed 44GW of solar PV capacity at a cost of US$100bn, according to Bloomberg New Energy Finance.

But as the Investment Tax Credit is due to decrease in 2016 to 10%, the issue of who is going to pick up the tab is becoming more pressing.

US$1.73 billion poured into project funds for third party lease companies such as SolarCity, Sunrun and SolarReserve in the last quarter, putting pressure on GTM Research to revise its forecasted US$5.7 billion in finance raised in 2016.

But for most developers, project finance is still hard work. Much has been made of the various public capital vehicles that could be transferred to the solar industry. Real Estate Investment Trusts (REITs) usually raise tax efficient capital to develop commercial properties such as shopping centres, offices and apartment buildings; Master Limited Partnerships (MLPs) allow retail investors to put their money into energy projects in exchange for tax benefits; securitisation could potentially unlock billions of dollars by aggregating solar projects and allowing them to be traded on a secondary market, much like asset-backed mortgage securities.

But at Intersolar NA yesterday, only one of those options was worth betting on, according to Dirk Michels, partner at global law firm K&L Gates.

"MLP and REITs? It's a long shot, it would be nice but I don't think it's very realistic," he said. "I don't think there is political will to get [MLPs]. We have an MLP Parity Act introduced in July and it didn't go anywhere and it's not going anywhere. If you want to participate in MLPs you have to give up your Investment Tax Credit.

REITs could be complex because of the question of project ownership, he said.

"Solar, depending on how you develop it, it may be real estate or not. If the Special Purpose Vehicle owns the ground, it…becomes an improvement to real estate.

"However, if you're thinking about rooftop solar, or distributed solar, developers…have to keep it as separate personal property separate from the real estate. Here's the problem: you have a Power Purchase Agreement (PPA) that says it's not real estate, but then you tell the IRS it’s real estate. It's just not going to happen."

But on securitisation of solar-backed securities (SBS), Michels took a different view.

"The benefits of SBS are apparent," he said. "You create an environment that allows access to public capital that allows the proverbial teacher's union or pension plan to invest and it holds the rating that it has been given by underwriters and doesn't create the problems that we had in the mortgage-backed security (MBS) debacle."

But ratings agencies such as Moody's and Standard & Poor's were struggling to assess the credit risk, he said.

"We need to see how we are going to receive the high ratings for solar asset-backed securities that are necessary to sell them and re-sell them in the market to create the secondary market.

"For the rating agencies it's an esoteric asset class. It's new for them and they model by analogy to MBS to car loans - but the analogies are not perfect.

Risks include module degradation compounded by manufacturer consolidation. "Everyone you talk with personally is comfortable with the technology," he said. "But when you get to putting a credit rating underneath a certain portfolio they come up with various issues that are not necessarily connected to performance of the technology, but with respect to certain manufacturers. For example what happens if all of a sudden a Chinese modules manufacturer goes bust in 2-3 years?

"The disclosure and liability that they take as underwriters…it's an issue and you have to get them comfortable. I don't think we're quite there but we will get there.

"It's like that first kiss – you always have to work hard for it."

But there were also risks associated with the consumer – if the price of utility-based energy declines they may be tempted to default.

"Today, when solar for a resident seems to be a good proposition, they pay 16c/kWh to the utility but only 12c/kWh to 13c/kWh to the solar provider. But what if energy and gas prices go down and solar prices are higher than utility prices? How are offtakers going to react and how are rating agencies going to model that in addition to the creditworthiness of the offtaker and manufacturer, etc?

One option would be to get the utilities onside, he said.

"Right now utilities are outside," he said. "They are the enemy of distributed generation (DG) because it takes away the customer base and increases the distribution cost for the remaining customers. Utilities could be the servicer, they could do the billing, etc. The remedy for a customer not paying could be to lose access to solar energy and utility energy - all of a sudden you have a big hammer."

The Department of Energy through the National Renewable Energy Laboratory established the Solar Access to Public Capital (SAPC) working group in March to work through these thorny issues in securitisation. We should know more soon about its feasibility as the SAPC is meeting this week in San Francisco.

Beth Dart, a PV engineer for consultants CH2M Hill, said that financing concerns were often more fundamental and pointed out that all too often developers took a risk by partnering with solar module manufacturers in the design process.

"We need to have the design aspect of a project and the financing aspect step back together. Often, you'll have someone getting ready to put together the design long after they are partnered with a module manufacturer, so the steps are not in the right order - the design and financial modelling needs to happen together."

Dart and her colleague Emily Martin had carried out detailed work into the financial implications of technology selection. Based on a client's 20MW project they found some astonishing results. Less efficient but cheaper modules had more added costs, sometimes as much as US$1.20/W to the Balance of Plant costs.

"We found that module costs didn't correlate with financial return module cost – but the cheapest module didn't give us the worst return," she said.

Sheldon Kimber, chief operating officer at Recurrent Energy, argued that the industry needed a "reality check" on costs and should aim to compete in conventional power markets.

The US is forecast to add another 12GW in utility capacity over the next four years, he said.

"How are we going to get to this 12GW? I don't see it in the current business model," he said. "There's a huge disconnect between this reality and the rosy forecasts.

"The current capital market situation is also a bit of a wreck. You've got a broken and highly illiquid market for tax equity. You've got a cartel of banks that have pricing power in the market, and the government is giving us their form of support which is instead of giving us cash, they're giving us a coupon which they can redeem at our local bank, of which there are three who might do business with us."

The 20-year PPA model was causing developers to keep their prices detrimentally low, he said.

"The price of these contracts is being eroded," he said. "You've got an enormous number of fairly desperate developers with their projects on their line, this is how they view it, their only way to get to market is to hook this PPA and flip it to an equity buyer. This business model is creating a bottleneck on the contract side and the prices are being bid down to a totally irrational level.

"That's a terrible deal for the seller, it's a great deal for the buyer. It means you can take this project into the market and realise a lot more value than locking in this contract, but you have to lock in this contract to do the business model that solar has gotten addicted to."

Shorter-term contracts as seen in other markets such as natural gas generation in Texas would generate more appealing returns, he said.

"The real opportunity for solar and the way we're going to get through to the scale we really want to be at is by taking cues from the existing conventional energy market and joining the conventional energy market."