California Looks Harder at the ‘Smart Grid’ (CPUC’s Division of Ratepayer Advocates new analysis)
“Regulators who don’t approve smart stuff are by elimination reducing themselves to certificators of dumb stuff. When nuclear optimism peaked, backers said that its power would be “too cheap to meter.” The bill for the smart grid is turning out to be too confusing to meter, but like in the nuclear heyday, the momentum is irresistible.”
I have some kind words for the California Public Utilities Commission’s Division of Ratepayer Advocates (DRA), its in-house department charged with representing small consumers in rate proceedings.
DRA has long been agnostic about the benefits of smart meters. But with the release of “Case Study of Smart Meter System Deployment: Recommendations for Ensuring Ratepayer Benefits,” the issue of high costs relative to benefits is on the table.
Better late than never.
DRA’s lightly redacted public version analyzes the gap between anticipation and reality in Southern California Edison’s “Advanced Metering Infrastructure” (AMI or SmartConnect) rollout program.
The numbers are interesting, but DRA’s big point is that the CPUC has hardly any idea about what it does or doesn’t know amid the hyper complexity of smart meter costs and benefits.
AMI is now the subject matter of lots and lots of dockets, many of which appear to have little to do with it. By actual count, the commission must track more than 130 different costs and 50 projected benefits, probably not all well-defined.
The basic value of AMI has long been far from certain. In a 2005 response to the commission’s request for a “business case,” SCE concluded that on balance it would not be cost-effective, save for some programs for customers who already had smart meters. The company came back in 2007, and this time the regulators authorized $1.63 billion for implementation on the basis of a remarkable estimate.
Specifically, over the 24-year duration of SCE’s program, the present value of net benefits would be the princely sum of $9.2 million. (Some other claimed benefits like reduced power theft were thrown out by agreement.) The commission authorized the $1.63 billion, but the business case includes another $1.58 billion of post-deployment costs not authorized but in the business case calculation.
DRA wrote page after page on the funding complexities, some already authorized in non-AMI dockets and some possibly duplicative. It concluded that it would be “practically impossible to track most post-deployment costs given the cost recovery processes adopted for SCE.” (at 25) The same, of course, holds for benefits (the meters are largely installed) that should be built into lower rates but don’t seem to be there yet. “(T)o the extent deployment period capital benefits are reflected in rates, these benefits appear to be much lower than forecasted in the business case.” (at 21)
To further complicate things, SCE’s claimed benefits include reduced generation investment, but its ongoing general rate case does not appear to include those reductions. After pointing out numerous funding requests in other areas that are really for AMI, DRA summarized that “(t)he full cost of SmartConnect will be more than double the $1.6 billion approved for deployment costs.” They hardly needed to note what that does to the $9 million.
What’s Up Now?
What’s being bought for the money is impossible to tell now, because the commission never asked for an explicit list of deliverables associated with those funds. DRA claims that SCE is loading deployment costs into higher rates but not yet booking benefits into reductions.
The company estimates savings during deployment at $35 million that hasn’t turned up in rates (actually, balancing accounts. Don’t ask.), while those rates already contain over $345 million in meter-related capital expenditures, 75 percent of the budgeted total.
Other benefits have also been slow to materialize. The business case projected over 386,000 customers enrolled in one or more demand-response programs by the end of 2010, but the company reported no participants while recording between $15 and $41 million in DR-specific costs.
Participation in the company’s Peak-Time Rebate program is 63 percent lower than estimated, customers taking time-of-use rates are less than 1 percent of the estimate and no customers have enrolled in the Critical-Peak Pricing program. (at 38).
Nuclear Power Redux
Has anyone noticed the similarities between smart metering and nuclear power?
The latter started with “turnkey” reactors built by outsiders and sold to utilities ready for operation, and the cost trends looked favorable. A variety of economic and political events ultimately put most nuclear projects in the hands of utilities that were unqualified to manage the complexities of construction, were regulated by agencies of questionable competence and knew that they could recover the bulk of their costs no matter what, save for a few imprudence proceedings. Booking smart meter costs before benefits materialize is just CWIP again.
Smart meters have become big for a reason: If utilities are going out of the generating business and into “electric services,” their managements need to rationalize some very big investment that will replace generation in rate base. Smart grid stuff is even better than nuclear because it is a technology whose possible outputs are limited only by the imagination, often hard to value and accompanied by an almost infinite list of potential cost addenda.
Regulators who don’t approve smart stuff are by elimination reducing themselves to certificators of dumb stuff. When nuclear optimism peaked, backers said that its power would be “too cheap to meter.” The bill for the smart grid is turning out to be too confusing to meter, but like nuclear in its heyday, the momentum is irresistible.