Wednesday, September 7, 2011
The bigger picture
I'm reminded there is a "Transmission Ranking Cost Report" that California utilities file with the CPUC. The report shows transmission costs which they add to renewable bids before the bids are ranked for selection. The purpose is to arrive at the total cost, from ratepayers' perspective, for the power from each project bid -- including the transmission to get the project's power to load. This total cost theoretically can be compared to a bid from a DG project -- comparing a rural project's transmission cost to an urban DG's more expensive land (but little to no transmission cost). This would be a valid comparison if time were not a factor. The DG can be permitted and installed in about 24 months, whereas a typical transmission project today requires a minimum of eight years to permit and install, if ever. So what is the time value benefit of faster DG? Is there a time value benefit? If not, then the more expensive land the DG project can justify is only about 2 cents/kwh. This is the average cost of transmission for delivering rural power to load in the Transmission Ranking Cost Reports. This leaves a lot of LA Basin land out of the DG picture.
Friday, August 19, 2011
CPUC approves 2011 RAM Programs
"The California Public Utilities Commission (CPUC) yesterday approved the program details to implement the Renewable Auction Mechanism (RAM), a new procurement method that will help lower transaction costs and promote the development of system-side renewable distributed generation projects of up to 20 megawatts (MW) in size." This is for PG&E, SCE and SDG&E.
Unfortunately, what was approved is a reverse auction process (seller with lowest price wins the PPA from the utility) that covers each utility's service area -- the entire service area, with no further breakdowns. My prediction is the most competitive bids will be for projects located in the rural parts of each utility's service area -- generally not where one would want a lot of "distributed generation". How can a new 10-acre project (~1MW) on $100,000 per acre land compete on price with a project on $10,000 per acre land? The more expensive land would represent approximately 25% of project cost, whereas the less expensive land would represent only 2.5% of project cost. In fact I would venture to predict there will be very few, if any, winning projects located in the LA Basin (to pick a sub-area).
This issue will have to be addressed soon if California is to have large amounts of renewable distributed generation in the right places.
Unfortunately, what was approved is a reverse auction process (seller with lowest price wins the PPA from the utility) that covers each utility's service area -- the entire service area, with no further breakdowns. My prediction is the most competitive bids will be for projects located in the rural parts of each utility's service area -- generally not where one would want a lot of "distributed generation". How can a new 10-acre project (~1MW) on $100,000 per acre land compete on price with a project on $10,000 per acre land? The more expensive land would represent approximately 25% of project cost, whereas the less expensive land would represent only 2.5% of project cost. In fact I would venture to predict there will be very few, if any, winning projects located in the LA Basin (to pick a sub-area).
This issue will have to be addressed soon if California is to have large amounts of renewable distributed generation in the right places.
Tuesday, July 26, 2011
Should there be one competitive market price for Renewable Generation (utility side of meter)?
The following relates to one of the subjects of discussion in the California Public Utilities Commission's current Rulemaking 11-05-005 "Order Instituting Rulemaking to Continue Implementation and Administration of California Renewables Portfolio Standard Program."
Should there be one competitive market price for Renewable Generation? (Answer: No, not if utilities want Distributed Generation.)
One of the significant characteristics of Renewable Generation is the need for large amounts of land (or rooftop space). Generally speaking the land required is approximately 10 acres for every 1 megawatt of Renewable Generation. Whether it is wind or solar, if one wants to collect the energy nature provides, one needs to collect that energy over a large area in order to have a significant amount relative to what we consume. So a 10 megawatt wind facility, whether it is twelve 0.8 megawatt turbines or four 2.5 megawatt turbines, will require roughly the same amount land as a 10 megawatt solar facility -- i.e. roughly 100 acres. Of course there are second order differences for example, solar thermal versus photovoltaic, that can mean a difference of + or - 20% in this number. The point is a lot of land is required for Renewable Generation projects. That's just the nature of renewable energy resources.
So how much does the land cost? Anywhere from $1,000 an acre (or lower) in sparsely populated, rural desert areas, to $1,000,000 an acre (or higher) in high-density, high-value urban areas. (These are 2011 prices.) As so happens there is a natural correlation between land value and utility load. Loads tend to be high in high-density areas and low in sparsely populated areas. Distributed Generation (DG) is the concept of generating in proximity to utility load to avoid utility transmission and distribution costs. The closer the renewable generator is to the load being offset the better. Wanting renewable DG is wanting Renewable Generation projects, greenfield and rooftop, located in the higher land cost areas.
A competitive bid situation limits the areas where a renewable project can be located.
Whether land is leased or owned, land represents anywhere from less than 1% to no more than 20% of the cost of producing the renewable energy (PV of revenues). Paying more than 20% of revenues for land is prohibitive and generally will render the Renewable Generation project uneconomic. For example, paying 20% of revenues for land in a competitive bid situation where other bidders might be paying 1% for land, is tantamount to using all of the project's equity returns to pay for land. All other things being equal, at the lowest energy sale price the high land cost developer is willing to accept (i.e. no profit), the 1% land cost bidders would be making 19% of revenues as additional profit. There simply isn't that much profit in these competitive bid situations. Competitive bid situations compel developers to site in low land cost areas. (Fixed price PPAs set at MPR have rural land costs implicit in their rates.)
This table demonstrates that solar PV developers cannot pay much more than $50,000 per acre for land and still have their greenfield projects remain economic. (For rooftop projects see discussion below.)
Unfortunately DG areas (e.g. LA Basin) are generally where land costs exceed $50,000 per acre. This is one of the significant reasons why there has not been many renewable DG projects on the utility's side of the meter. Just look at PG&E's FIT program and SCE's CREST program -- a total of only 31.3 MWs in signed PPAs.
For DG, the utility is buying electrons at a preferred location. There should be a price for each.
Having one competitive market price for Renewable Generation is fine, if that's all what utilities want. However if utilities also want to direct where Renewable Generation is located, e.g. renewable DG, then one competitive market price does not work. The utility is no longer buying just electrons, it's buying electrons at a preferred location. Location needs to be priced separately. One way to get projects sited close to load, where land costs are high, is to separate the land price bid from the energy price bid. The utility should be willing to pay up to the avoided cost of not building or delaying transmission, substation and distribution projects (including avoided right of way costs), plus the value of the renewable energy received.
As an alternative to separating land from the renewable facilities, perhaps the utility could request bids by similar land value areas (i.e. divide up the metropolitan area) -- (perhaps even LAP/LMP nodes?), asking developers to bid whole projects for each area, and then integrating the bid results with the utility's planning for transmission and distribution facilities to serve future load. This will provide a way for the utility to plan for net load (load minus DG), not to mention the ability to be more specific about what transmission and distribution projects have been avoided.
-----------------------
Rooftops
One might argue if land costs are too high, then go get rooftops! However rooftop issues are very similar. If a roof exists, it is already owned by someone. Paying pennies per square foot for rooftops is also problematic. Roof owners are reluctant to put their rooftops at risk for only pennies per square foot. To overcome this concern of owners and to make it worthwhile for them, developers are looking at costs very similar to $50,000 per acre, or roughly $5,000 per acre per year. This is only 11 cents per square foot per year, which may or may not be sufficient to interest roof owners.
Should there be one competitive market price for Renewable Generation? (Answer: No, not if utilities want Distributed Generation.)
One of the significant characteristics of Renewable Generation is the need for large amounts of land (or rooftop space). Generally speaking the land required is approximately 10 acres for every 1 megawatt of Renewable Generation. Whether it is wind or solar, if one wants to collect the energy nature provides, one needs to collect that energy over a large area in order to have a significant amount relative to what we consume. So a 10 megawatt wind facility, whether it is twelve 0.8 megawatt turbines or four 2.5 megawatt turbines, will require roughly the same amount land as a 10 megawatt solar facility -- i.e. roughly 100 acres. Of course there are second order differences for example, solar thermal versus photovoltaic, that can mean a difference of + or - 20% in this number. The point is a lot of land is required for Renewable Generation projects. That's just the nature of renewable energy resources.
So how much does the land cost? Anywhere from $1,000 an acre (or lower) in sparsely populated, rural desert areas, to $1,000,000 an acre (or higher) in high-density, high-value urban areas. (These are 2011 prices.) As so happens there is a natural correlation between land value and utility load. Loads tend to be high in high-density areas and low in sparsely populated areas. Distributed Generation (DG) is the concept of generating in proximity to utility load to avoid utility transmission and distribution costs. The closer the renewable generator is to the load being offset the better. Wanting renewable DG is wanting Renewable Generation projects, greenfield and rooftop, located in the higher land cost areas.
A competitive bid situation limits the areas where a renewable project can be located.
Whether land is leased or owned, land represents anywhere from less than 1% to no more than 20% of the cost of producing the renewable energy (PV of revenues). Paying more than 20% of revenues for land is prohibitive and generally will render the Renewable Generation project uneconomic. For example, paying 20% of revenues for land in a competitive bid situation where other bidders might be paying 1% for land, is tantamount to using all of the project's equity returns to pay for land. All other things being equal, at the lowest energy sale price the high land cost developer is willing to accept (i.e. no profit), the 1% land cost bidders would be making 19% of revenues as additional profit. There simply isn't that much profit in these competitive bid situations. Competitive bid situations compel developers to site in low land cost areas. (Fixed price PPAs set at MPR have rural land costs implicit in their rates.)
This table demonstrates that solar PV developers cannot pay much more than $50,000 per acre for land and still have their greenfield projects remain economic. (For rooftop projects see discussion below.)
Unfortunately DG areas (e.g. LA Basin) are generally where land costs exceed $50,000 per acre. This is one of the significant reasons why there has not been many renewable DG projects on the utility's side of the meter. Just look at PG&E's FIT program and SCE's CREST program -- a total of only 31.3 MWs in signed PPAs.
For DG, the utility is buying electrons at a preferred location. There should be a price for each.
Having one competitive market price for Renewable Generation is fine, if that's all what utilities want. However if utilities also want to direct where Renewable Generation is located, e.g. renewable DG, then one competitive market price does not work. The utility is no longer buying just electrons, it's buying electrons at a preferred location. Location needs to be priced separately. One way to get projects sited close to load, where land costs are high, is to separate the land price bid from the energy price bid. The utility should be willing to pay up to the avoided cost of not building or delaying transmission, substation and distribution projects (including avoided right of way costs), plus the value of the renewable energy received.
As an alternative to separating land from the renewable facilities, perhaps the utility could request bids by similar land value areas (i.e. divide up the metropolitan area) -- (perhaps even LAP/LMP nodes?), asking developers to bid whole projects for each area, and then integrating the bid results with the utility's planning for transmission and distribution facilities to serve future load. This will provide a way for the utility to plan for net load (load minus DG), not to mention the ability to be more specific about what transmission and distribution projects have been avoided.
-----------------------
Rooftops
One might argue if land costs are too high, then go get rooftops! However rooftop issues are very similar. If a roof exists, it is already owned by someone. Paying pennies per square foot for rooftops is also problematic. Roof owners are reluctant to put their rooftops at risk for only pennies per square foot. To overcome this concern of owners and to make it worthwhile for them, developers are looking at costs very similar to $50,000 per acre, or roughly $5,000 per acre per year. This is only 11 cents per square foot per year, which may or may not be sufficient to interest roof owners.
Subscribe to:
Posts (Atom)