Changes Ahead For California Residential Solar
By Laura Norin, Heather Mehta and Julia Getchel
Two looming regulatory developments in California will have a significant effect on the residential solar market.
The state is rewriting the rules for net metering, where homeowners who use solar to generate their own electricity can sell any excess electricity to the grid. The rewrite is expected to scale back the benefits of net metering for solar customers.
The current multi-tiered residential rate structure used by investor-owned utilities in the state, which has been an important driver of the economics of solar for high-usage customers, is being re-evaluated and is likely to undergo substantial change or to be superseded by a new structure entirely.
These developments are likely to make California a tougher market for rooftop solar companies. However, the market should still remain viable, and there could even be new opportunities.
California has long supported behind-the-meter residential solar electric generation through net metering, which allows customers to sell surplus solar power back to the utility at the full retail value of the electricity. Net metering was instituted in California in 1996 and has been expanded several times over the years to allow for wider participation and greater benefits.
Under the existing net metering program, the amount of net-metered capacity that can be added for each investor-owned utility is capped at 5% of the utility’s aggregate customer peak demand, which is defined as the sum of the maximum peak demands for each customer rather than the maximum demand for the utility as a whole.
When the cap is reached, there will be approximately 5,570 megawatts of installed solar capacity across the systems of the three California investor-owned utilities. At the end of September 2013, Pacific Gas and Electric reported that it had 902 megawatts of net-metered capacity connected to its system, which is the equivalent of 1.87% of the utility’s aggregate customer peak demand. San Diego Gas and Electric and Southern California Edison had net-metered capacity of 1.67% and 1.46% of aggregate customer peak demand, respectively.
California’s net metering program allows a customer who installs a solar photovoltaic system of one megawatt or less to receive a financial credit for power generated by his or her system and delivered to the utility grid. A typical solar customer will generate more solar power than needed during some parts of the day and less than needed, or none at all, during other parts of the day and throughout the night. The same pattern can arise seasonally, often with the generation of more power than needed during the summer and less than needed during the winter.
Under net metering, customers can send excess power to the grid and use this power as a credit to offset purchases of power from the grid that are made in the same 12-month period.
For a customer whose rates are wholly volumetric, meaning that the customer is charged for the kilowatt hours of usage without any fixed charges or demand charges, as is the case for customers of PG&E and SDG&E, this allows the customer to sell power back to the utility at the full retail rate for the power, including all generation, transmission and distribution cost components. Aside from small minimum charges, which are binding only for customers with extremely low net usage, these customers can avoid having to pay electric bills by selling back enough solar power to offset all grid purchases. Customers who generate enough solar power to more than offset all grid purchases can also receive net surplus compensation payments, but at the wholesale rate ― rather than retail rate ― for the surplus power generation.
Impact of Rates
Net metering has worked hand in hand with residential rate design structures to make solar PV economical for many customers, particularly high-usage customers. Most of California’s residential customers have inclining block rates, with prices increasing over two, three or four tiers of rates as usage increases. Net metering is particularly valuable for high-usage customers because it allows them to avoid being pushed into higher tiers and rates.
For example, the January 2014 rates under PG&E’s default residential rate schedule were 13¢ per kWh for “baseline” usage, 15¢ per kWh for 100% to 130% of the baseline amount, and more than 32¢ per kWh for usage greater than 130% of baseline amount. In other words, rates for highest levels of energy usage are about two and a half times the rate for the baseline level of usage. (The amount of energy in each tier is linked to the customer’s “baseline” usage amount, which is set at 50% to 70% of the average residential electricity usage in the customer’s climate zone.)
The steeply inclining block structure greatly increases the value of net metering for high-usage customers, and these customers have represented a significant portion of the market for residential solar in California. Under the January 2014 rates, a high-usage PG&E customer whose solar system reduces electricity consumption from 200% of baseline to 130% of baseline has effectively sold power to the utility at a rate of 32¢ per kWh. By contrast, a low-usage customer whose energy usage without solar PV is at 130% of baseline would sell power to the utility at just 13¢ to 15¢ per kWh.
Customers also have the option to select a time-of-use rate schedule, under which rates are higher during peak periods of the day and during the summer months and lower at night and during other low-usage periods. These rate schedules are usually tiered, meaning that rates vary both by time of consumption and by level of consumption. For customers on time-of-use rate schedules, net metering credits are assigned a value based on the retail cost of power in place at the time of the power generation.
As a result, solar power generated during a summer late afternoon may offset two to three times that amount of winter or nighttime power consumption. For example, under January 2014 rates, one kWh of solar power sent to the PG&E grid between 1 p.m. and 7 p.m. on a summer weekday would earn a credit of 28.7¢, which is the summer peak-period residential time-of-use charge for one kWh of baseline usage. During the summer months between 9 p.m. and 10 a.m., this credit would offset 2.85 kWh of power, since the baseline cost of power is just 10.1¢ per kWh during this interval.
Concerns that the net metering program was shifting costs to customers who do not have solar on their roofs led to legislation requiring a study of net metering’s costs and benefits for all ratepayers. AB 2514, enacted in 2012, directed the California Public Utilities Commission to undertake a study “to determine who benefits from, and who bears the economic burden, if any, of, the net energy metering program.”
The study was completed in October 2013 by an outside consultant, Energy and Environmental Economics, Inc.
The study found that the level of the net metering subsidy is highly linked to the rate structure and that the utilities’ current residential rate structures, with steeply inclining block rates and little or no fixed charges, yield a subsidy of 20¢ per kWh of solar generation.
Critics of the study claim the study does not account for the full benefits that solar PV provides to the grid and overstates the cost-shifting impacts. Despite the criticism, the study is being used to support proposals for changes in the residential rate structures of the three main California utilities.
In the fall 2013, the state legislature enacted another bill, AB 327, that will end the current net metering structure in mid-2017 or, if earlier, when net-metered systems reach 5% of a utility’s aggregate customer peak demand. AB 327 requires the California Public Utilities Commission to develop a new “standard contract or tariff, which may include net metering” for solar customers, to replace each utility’s current net metering structure when the current program expires.
Unlike the current structure, the new net metering structure would not have any cap on participation. However, the new net metering program must ensure that no costs are shifted to non-participating customers. Current net metering customers would be grandfathered under the current system for a period of time that has not yet been determined.
The practical effect of the law is that there will soon be two or possibly three distinct sets of net metering customers: one set that will remain under the current program for an indeterminate period of time, another set that will be put under the new program, and a third set that did not have solar when AB 327 was enacted but installs it before the current net metering program ends and that may be under a different set of grandfathering regulations than pre-AB 327 net metering customers.
To implement AB 327, the California Public Utilities Commission must first determine a schedule to transition from the current net metering program to the future uncapped program as well as the rules for grandfathering existing net-metering customers. The CPUC received a range of proposals for how to structure a transition period and the grandfathering rules.
PG&E and SDG&E both proposed that existing net metering customers with solar systems installed before April 2014 be allowed to remain on the current program through the end of 2023. Net metering customers with systems installed between April 2014 and December 2015 would be allowed to remain on the current program through the end of 2020. Net metering customers with systems installed between January 2016 and June 2017 would be transitioned to a new net metering program that would take effect on July 1, 2017.
SCE proposed that customers who participate in the existing net metering program before July 2017 would be grandfathered in the program through the end of 2023.
The California Solar Energy Industries Association recommended a more extensive grandfathering program that would allow customers who participate in the net metering program before July 2017 to remain under the current program for a minimum of 30 years.
The CPUC is expected to issue a decision on grandfathering rules by March 2014.
The commission has until the end of 2015 to develop the new structure for net metering. AB 327 gives the CPUC wide latitude to determine what should replace the current program. Possibilities include a feed-in tariff that allows customers to sell solar power to the utility at a fixed price or a new net metering program that reduces the amount of power that can be sold back to the utility or reduces the financial credit associated with that power.
In all likelihood, the structure of the new program will depend on the default residential rate structure in place when the program rules are adopted. The commission has a rulemaking underway to re-evaluate the current rate structure.
AB 327 gave the CPUC authority to make radical changes to residential rate design, including reducing the number of rate tiers to two through 2018, eliminating the inclining block rate structure entirely thereafter, and imposing fixed monthly charges of up to $10 per month for non-low-income customers beginning in 2015, with inflation adjustments thereafter.
A CPUC staff proposal released in January 2014 recommends that default residential rates shift from inclining block rates to non-tiered time-of-use rates beginning in 2018, and that until then the CPUC reassess the appropriate time-of-use period definitions ― for example, what hours and months should be included in the summer peak rate period ― and the rate differentials between time-of-use periods ― for example, how much higher the summer peak-period rate should be than the summer off-peak-period rate. The proposal also recommends gradually reducing the number of tiers to two between 2014 and 2018, and greatly reducing the rate differentials between the tiers to just 20% by 2018, at which time the tiered rates would be an optional alternative to the default time-of-use rates. Finally, the staff recommends phasing in a fixed charge that would start at $5 a month and increase to $10 a month by 2018 with future inflation adjustments. The staff proposal does not recommend using minimum bills (instead of fixed charges), but if the CPUC were to adopt minimum bills, it recommends that they should start at $10 per month in 2015 and increase with inflation.
These potentially substantial changes to the residential rate design structure create many unknowns for residential solar developers, as the implications of the changes for the market will depend on the details of the new rate structures.
There are three areas in which changes in residential rate design could alter the market for residential solar.
One is time-of-use rates or inclining block rates. A shift to non-tiered time-of-use rates (from the current default inclining block rates and optional tiered time-of-use rates) may improve the economics of solar PV for moderate usage residential customers with relatively high shares of energy consumption during peak periods. The extent to which this will be the case will depend on the rate differentials between the peak and non-peak periods and on how much overlap the new time-of-use period definitions retain between the peak time-of-use period and the period of maximum solar generation. These structural time-of-use definitions are likely to differ from those in place in the optional time-of-use rates that are currently available, and how they are structured could significantly support or debilitate the market for residential solar.
For example, the current on-peak period for customers on SDG&E’s optional residential time-of-use rate is weekdays from noon to 6 p.m., a period that captures about 35% of the output from solar PV systems in the San Diego area. In January 2014, SDG&E proposed to shift the on-peak period in the winter to weekdays from 5 p.m. to 9 p.m., a period when little solar generation can be expected and, in the summer, to weekdays from 2 p.m. to 9 p.m. These new time-of-use periods would lead to a significant reduction in solar PV generation during the peak periods: less than 10% of the output from solar PV systems would occur in the peak period under SDG&E’s proposal instead of about 35% under the current definitions, with the remaining output occurring in the semi-peak period.
If large differentials are maintained between on-peak and semi-peak rates, then this shift could undermine solar PV economics for customers on time-of-use rates. However, changes to time-of-use period definitions may not be quite so detrimental to solar PV economics. For example, the CPUC staff report raised an idea of a split on-peak period that would include both morning hours and late afternoon or evening hours. This structure would offset a portion of the loss of mid-afternoon on-peak hours with the addition of morning on-peak hours, which could include hours of high solar generation.
Another area where change could have an effect is a reduction in the rate differentials for inclining block rates. A reduction in the rate differences between tiers would eliminate the very high upper-tier rates that have been a cornerstone of residential solar economics. However, reductions to the upper-tier rates would be done concomitantly with increases in the lower-tier rates, potentially making residential solar economic for lower-usage customers whose rates were too low earlier for the investments to pencil out.
Finally, another area where change would have an effect is the introduction of fixed charges or higher minimum bills. Fixed monthly charges make residential solar less economic because these charges cannot be avoided through net metering. However, the extent of the impact depends on the level of the charge. Furthermore, if a higher minimum monthly bill is used in place of a fixed monthly charge, then the impact is likely to be less significant, particularly for those customers who offset most, but not all, of their electric bills with solar generation and continue to purchase a small amount of power from the utility. Under a fixed charge, these customers would pay the fixed charge in addition to their volumetric charges. Under a minimum bill, the volumetric charges would be credited against the minimum bill amount.
How the net metering program is restructured will also have a significant effect on the long-run viability of residential solar in California. The restructured program is likely to be less generous than the current program. These very important residential rate design and net metering program details have yet to be worked out.
That said, the California Public Utilities Commission has a goal of maintaining a viable and growing residential solar market in California. It is a daunting task, given the competing interests.
Residential rate structures must be designed to balance a number of equity and efficiency concerns that have implications both for solar and non-solar customers. The grandfathering provisions for customers who installed solar before AB 327 became law must be sufficient to provide confidence to customers who are evaluating new solar installations that their investments will not be undermined through future rate design or net metering changes while also satisfying concerns about cost-shifting. The design of time-of-use periods must be done carefully to provide the proper price signals to consumers while at the same time being sensitive to the potentially significant implications for commercial and industrial customers that have structured operations and entered into investments based on the current time-of-use periods. The restructured net metering program must support the residential solar market without increasing rates for non-solar customers.
Given these challenges and complexities, some amount of market disruption is inevitable.
However, there is also opportunity in that new segments of the residential population may be open to solar for the first time with the shift to non-tiered time-of-use pricing.