Previously, all we could see were the smoke signals. San Diego Gas & Electric lost its attempt to clamp down on net energy metering (NEM) and the California Public Utilities Commission (CPUC) then approved a new formula to calculate the state's 5% cap, raising small-scale residential and commercial solar potential to 5,000MW.

This year, the flames are already starting to spread as the utilities push against the solar advocates on NEM, the "silent subsidy".

The CPUC will temporarily suspend NEM for new customers at the end of 2014, once it has worked out new rules based on a review of a new cost-effectiveness study of NEM to be completed this year. The CPUC's last cost-effectiveness review could have been better news for the solar industry.

Although a report by consultants E3 published in 2009 found that the impacts of NEM were small, they said that once the 2,300MW capacity of the California Solar Initiative was fully subscribed, the cost to California's three big utility companies would be $137 million. Residential NEM customers would in effect impose a net cost of $0.19 per kWh of power exported to the grid, a cost that is shifted to non-solar customers.

Pacific Gas & Electric (PG&E), San Diego Gas & Electric (SDG&E) and Southern California Edison (SCE) have told the CPUC that the current 5% NEM cap will impose billions of dollars of additional costs on ratepayers who do not participate in the NEM programme. But many in the solar industry claim that the utilities are crying foul for no good reason.

Non-profit advocacy group, Vote Solar, was quick off the mark this year by commissioning a cost-benefit analysis from Tom Beach of Crossborder Energy, formerly also of the CPUC.

Annie Lappé, solar policy director, said: "The reason why Vote Solar is adamant about doing these cost benefit studies properly, is because the stakes are so high for current and future [distributed generation] solar users.

"Utilities that are claiming that NEM is a subsidy have proposed several options to 'remedy' the situation. To varying degrees, we believe that all the proposals on the table compared to NEM as it exists today would serve to discourage the adoption of on-site solar energy."

Lappé said that utility proposals ranged from an increase in monthly customer service charges "as much as $50 a month could be reasonable", imposing standby service charges, limiting NEM capacity and replacing net metering with Value of Solar Tariff, as put forward by Austin Energy, a variant of a feed-in tariff that would be recalculated annually based on the value of PV generation to the utility.

"We first want to see a proper cost-benefit analysis and if it shows that NEM provides financial benefits to all ratepayers in a utility service area or a state," she said. "At Vote Solar, we say why change a policy that is working so well to increase use of solar."

"The question we're really trying to answer in this study is: does the retail rate credit accurately capture the value of the power that's exported to the grid," said Beach.

"Most NEM systems only export about one third of their power. As a matter of physics, those exports usually serve the NEM customer's neighbours, it serves nearby loads. It displaces power that the utility would have delivered to those neighbours from more remote sources. The principal cost is that utilities lose revenues when the meter runs backwards and the utility may incur billing and administration costs."

While the E3 report found there was a net cost of about $0.14 per kWh of power exported, Beach's new study found that the net cost in the residential market has dropped by a factor of 10 to $0.013 per kWh. In SCE and SDGE territories, NEM has a small net benefit for other residential customers, the report found.

In the commercial and industrial sector, which is 60% of the market in California, the benefits were much greater, said Beach.

"Commercial and industrial rates often include significant fixed charges or demand charges that are difficult for solar customers to avoid. They have much lower volumetric energy rates and as a result the costs of net metering are significantly lower than the benefits.

"Commercial and industrial customers tend to install solar systems that are smaller compared to their load than residential customers and their exports tend to happen in the middle of the afternoon when the power is more valuable. NEM is definitely more cost effective in the commercial and industrial market."

In total, Beach found that there were $2 million annual NEM benefits in residential and $90 million a year in the commercial and industrial sector, a sum easily absorbed by the $25 billion revenue of California's big three utilities, he said.

Susannah Churchill, policy advocate at Vote Solar, said that utilities routinely overstated the average credit on consumers' bills.

"The average bill credit is more like 22.5 cents per kilowatt hour rather than 31.7c/kwh," she said. "Utilities often account for only one of the benefits or avoided costs that NEM solar provides to the grid. They account for the avoided cost of the conventional generation that they would have had to buy but they leave out all of the cost savings associated with new generation, including transmission and distribution investments.

"Utilities also count the impacts of all NEM system output instead of only generation that gets exported to the grid. But like energy efficiency, solar power that's used on site places no burden on the utility.

"We don't see utilities claiming that customers who turn their lights off or install energy efficient appliances should have to pay a charge for the power that they didn't buy. We think it is fundamentally unfair to look at the grid impacts of solar that's used on site differently from the grid impacts of energy efficiency.

Expect more reports, and counter reports throughout the year, as the CPUC edges towards a decision.

Last week, researchers at the Lawrence Berkeley National Laboratory published a separate report, Electricity Bill Savings from Residential Photovoltaic Systems: Sensitivities to Changes in Future Electricity Market Conditions.

Although it doesn't advocate a position, it does lay out some of the pros and cons of what the utilities might have in mind to replace NEM.

"Under all other rate options and compensation mechanisms evaluated in this report, however, the bill savings from residential PV decline with increasing solar penetration levels on the grid," said the authors.

"Across all of the electricity market scenarios considered, the bill savings are significantly eroded if, instead of net metering, generation from behind-the-meter PV exported to the grid is compensated at hourly wholesale prices."

But the only thing that is certain post-2014 is that the CPUC decision will not only determine the size of California's rooftop market, but also set the pace of demand and shape consumer expectations of solar for many years to come.