Possibly the most fascinating aspect of the Smart Grid is the absense of an economic rationale. But industry incentives being what they are (concentrated benefits, diffused costs), many have bet on much of it being built. Boondoggles must pass political tests, not economic ones.
But guess what? People are finally starting to wonder if this smart grid is worth the trouble. Intervenors, at last, are turning up at state proceedings. For a good sample of the issues and alternatives, look at Synapse Energy Economics’ July 8 filing at the New Jersey Board of Public Utilities on behalf of the state Department of Public Advocate. Synapse is possibly the best firm in the business to represent efficiency or environmental interests, but they stand with the skeptics on smart grids.
The utilities have yet to find consultants who can make an easy case for the grids. Advanced Metering Infrastructure (AMI) by itself recovers only 50 to 80 percent of its costs if all it gets used for is automated reading, data transmission, and service initiations and terminations. (See Brattle Group’s The Power of Five Percent, at p. 6.) Getting a positive cost-benefit figure requires time-varying rates for small customers and ways they can react to them, or giving their utility power to do that for them.
California is in the midst of distributing smart meters to everyone over the next few years, but it has already made certain that the necessary rate reforms and controls rate and controls won’t be there. First, the state just got a law that prohibits any mandatory form of time-varying pricing, with or without bill protection, prior to 2013. Mandatory real-time pricing without bill protection has to wait until 2020. Utility-controllable thermostats (originally deemed necessary for a positive cost-benefit figure) were removed from the state’s regulatory options a year ago by public protests.
But let’s say that redesigned rates somehow come to be. Then the increase in consumer bills to pay for the meters will be counteracted by reductions due to peak shifting. The actual levels and persistence of these adjustments are far from clear. The purported evidence for responsiveness comes from controlled experiments in which self-selected consumers got all the hardware for free and were rewarded for their participation. Brattle calculates that a reasonable shaving of the peak under smart metering will save about $3 billion a year nationwide in avoided generator investment ($2.4 billion) and operation ($0.6 billion). Relative to almost any dimension of the industry, this is a trifling figure. Edison Electric Institute (at p. 17) says that total investment by corporate utilities amounted to $84.2 billion in 2008 (these figures do not include IPPs).
Raise that to $112 billion to account for the 25-odd percent of power that comes through municipals and co-ops, and assume that about $40 billion of it goes to build transmission and distribution. Discounting at 9 percent on a 20-year horizon, Brattle calculates the present value of the savings at $35 billion, $28 billion if we look only at generation investment. After the smart glacier comes and goes, the detritus is equivalent to a four-month moratorium on generator construction, and that happens only once.
Things get better still once we realize that these are only figures from the utilities’ side. The smart grid’s promise only materializes after small consumers buy a bunch of their own equipment – flashy thermostats, premium-price appliances that can talk to the grid, controllers, assorted communications gear, etc. Add a few hundred bucks of costs like these, multiply them over, say, 80 million homes and small commercial units, and then use the same logic as environmental intervenors to calculate the “societal” cost, also known as the wealth loss to ratepayers. The utility? Unlike consumers, it gets a regulated return on its investments, and possibly even stranded cost recovery on unamortized equipment that the smart stuff replaces.
But as long as we are talking societal, exactly where is the smart grid’s environmental contribution? Just about everyone agrees that its main effect will be to time-shift peak consumption, with little if any effect on total power use, i.e. no carbon consequences. Synapse notes that there may be benefits from scaled-down versions of smartness – automated meter reading without advanced communications may be effective, and there surely are circuits that could benefit from smarter technologies. For cutting carbon they recommend mandated efficiency measures. The rollout of everything to everyone doesn’t help much, even assuming that it works and is secure. The smart grid does have one virtue – it’s something big and durable for utilities to spend on and manufacturers to produce, which is looking more and more like the entire point of the program.
Note to critics: If any of the figures here are wrong, tell me and I’ll cheerfully acknowledge it. Again: check the aforementioned reports from the Brattle Group and from Synapse.
Robert,
You haven’t mentioned an important aspect of peak load shifting. Utilization factors within the industry are notoriously low due to building generation capacity to meet peak loads. This is even more pronounced when trying to incorporate large solar and wind farms that seem to be in fashion at the moment. Shifting peak loads will enable utilization factors to increase and increasing the value in the current assets.
Incorporating smart grid technologies with energy storage (batteries, thermal) and localized solar generation would make this much more viable. Imagine, using storage as well as solar as an aggregated utility component to manage the load throughout the grid. This could be used for offsetting load during peak demand as well as shedding power when power sharply increases (think wind turbines) or when load suddenly drops.
Once installed, this could be utilized without the ‘customer’ even knowing what is going on within their system. As far as they know, electricity is flowing. The source of this electricity, whether by power generation or by drawing from the storage component, is managed by the utility via the smart grid. This would eliminate the need to include a price level to incentivize the costumer into adaptation.
I’ve been arguing that the smart grid is a bad, wrong-headed idea for months at POWER magazine’s blog (www.powermag.com). Michaels is absolutely correct that the smart grid simply shifts demand, with no benefit to society (but plenty to utilities).
My case has been for a strong grid — more robust interconnections between and among the existing grids, without regard to how smart those interconnections are.
There is no doubt in my mind that a smart grid would not have protected the nation in the most recent regional blackouts. Indeed, the cascading grid collapse might also have collapsed the information-based grid, and exacerbated the outage.
As best I can tell, it’s a con. The utilities recover the costs of the smart grid in rates, benefit from load smoothing, and consumers have to pay for the infrastructure and buy intelligent toasters and the like on their own accounts.
Sounds a lot like plug-in hybrid cars (which I regard as another enormous scam benefiting utilities and not consumers or society at large).
shaun
Nuclear gives all of the benefits you mention without the need for costly (and power wasting) storage. Further, nuclear can be located much closer to demand, further reducing transmission loss, and requires only a tiny fraction of the fragile and unspoiled lands than solar, wind, and even hydro require.
In a few years dramitic advances in nuclear generation technology will make it abundantly clear how foolish we are to not concentrate resources and efforts on nuclear and forget about the foolish and wasteful alternative power production schemes.
“Smart grid” is just a political label for grid investment. In the parts of the North American grid that I know something about (mostly Ontario), I can point to many shortcomings in historic grid investment, and I remain hopeful that the label can be defined in a way that produces overall benefits. Looking only at smart meters, they are costly, in part due to short depreciation periods for both hardware and software, but will improve outage recovery, help optimize grid design and maintenance, and provide better monitoring of power quality. Over time, every new major appliance will come with simple logic controls at approximately zero incremental cost. Of course some utilities will abuse the opportunity to invest, particularly if consumers are not vigilant about keeping them accountable. The biggest incremental grid screw-ups I see are not driven by “smart grid” but by the drive to provide costly grid connection for new remote generators, whose output would be uneconomic even if it was located on top of the load.
Robert–Very informative. The system may not be as beneficial as it is being touted, but it does seem that even if total consumption remains the same, shifting usage from peak times can have carbon reduction effects. As a resident of NY, NY, I am thinking specifically of summer months when cooling load pushes the grid of the city to its limit and the peaker plants on the other side of the rivers turn on. They are all coal plants and the oldest in the area so unsurprisingly the dirtiest–the evidence is clear in the sunsets in August. It seems to me that negating the need for those plants to ever come online by using some of that power at night would mean more power could come from the more efficiency natural gas plants instead. I suppose you could replace the old plants, but would that really be “smarter”? Cheaper? Easier?
Smart Grid expenditures are as ridiculous as spending money on facilities to makes synthetic fuel additives from corn.
The government doesn’t make money and keep people employed when the answer is simple. That answer for both auto fuel and electricity is conservation, effciency (appliances, homes & manufacturing).
It reminds me of somebody determining we need to meter our food intake and waste disposal so that we can optimize our bodies. Give me a break!
Bob, reading your article I got the idea that the Brattle article was relatively dismissive of the idea that smart grid and demand response would produce benefits. Far from it.
The first line of their conclusion is, “The potential impact of demand response is large and significant.” And Brattle is only counting the relatively direct, obvious effects of smart metering with improved rate structures.
When power consumers have bills based upon month-long interval usage, it creates a kind of common pool resource problem and associated resource waste (since the various ways of efficiently managing common pool resources are unavailable to consumers in the electric power consumption case). Metering at intervals sufficient to differentiate between use of scarce peak-load capacity and less-scarce offpeak capacity effectively “privatizes” the commons by making each consumer responsible for their own use of scarce resources rather than forcing consumers to share in each others use of resources.
a simple case that they haven’t probably considered in their study:
line lenght substation-load = 15km
load exceeding voltage/current limit per year = 250hours
necessity to take action = yes
cost line refurbishment = $100.000/km = $1.5M total
cost fancy smart grid something solving the same problem = $100.000
…
if I must be more explicit, the cost of the work on the network is passed to the final user through connection charges, and the network is a natural monopoly that has no compoetition by definition, not by regulation
Yo. Pretty sweet dude!
Yo. Sweet blog, dude!