A free-market energy blog
Random header image... Refresh for more!

Smart Grid, Dumb Economics

Yesterday, the National Clean Energy Product Summit was held in Washington, DC to discuss the Center for American Progress’ s February 2009 white paper titled “Wired for Progress: Building a National Clean-Energy Smart Grid.”  Participants included Steven Chu, Al Gore, Robert F. Kennedy Jr., T. Boone Pickens, Bill Clinton, Nancy Pelosi, Harry Reid, and pretty much everyone else who thinks they know a priori how to most efficiently organize and manage the electricity sector.   As one might expect, no good came of it.

There is no doubt that, if we’re going to build more central station wind, solar, or nuclear power plants, we’re going to have to build more wire to get that electricity to consumers.  But the price signals we’re seeing in electricity markets suggests that building a small number of remote power stations with a lot of wire to get that power to market is less efficient than building a relatively large number of smaller, more accessible power stations with fewer long distance transmission lines.  The reason we don’t have more long distance transmission, after all, is because it has been more than 30 years since large central station power plants made more economic sense than smaller, more local power generation. 

What almost everyone on both the Left and Right miss is that transmission and generation are economic substitutes for one another.  The optimal (most efficient) mix of the two is an economic puzzle, not an engineering puzzle, and the best means of solving that puzzle is to ensure that prices for both are correct and to then let market actors to sort this out.

Ironically, it may very well be that, to the extent that there is suboptimal investment in transmission, the culprit is primarily vertical de-integration, not local “not-in-my-back-yard” sentiment.  That is, the wave of state-induced restructuring (popularly but inaccurately known in some quarters as “deregulation”) might be the real problem here.     

As an aside, it’s interesting to me that the Left is all a-gog over a massive federal program to build this transmission super-highway and in so doing preempt state siting authority.  Since when did the Left support dropping the hammer on local communities or property owners who object to development because of concentrated costs and diffuse benefits?  And haven’t many of these same parties at one time or another echoed Amory Lovins and evangelized for “distributed energy” - that is, an electric power system characterized by neighborhood or even on-site power generation and a correspoding abandonment of the “big power, big-wire” model? 

I don’t have any brief one way or the other, but I can’t believe that central planning works surprisingly well when it comes to electricity markets but, for some mysterious reason, not very well anywhere else.  The fact that the grid at present is essentially a public commons certainly requires some central planning from somebody just to stay in operation.  This implies the need for policies to privatize that commons and return the electricity planning back to market actors.

Update: Lynne Kiesling comments on the conference and nicely hits another issue – the “smart” part of the grid conversation that is frequently overlooked.  My take - when residential consumers get – and pay! – accurate price signals for the electricity they are buying, pigs will fly.


1 rbradley { 02.24.09 at 11:13 am }

I remember when environmentalists routinely were opposed to high-voltage, long-distance transmission projects.

It would be good to pull out some quotations on this to add to the “When is an environmentalist not an environmentalist?” joke.

(Oh, the answer: “When it comes to windpower.”)

2 Donald Hertzmark { 02.24.09 at 11:23 am }

Just to add to this post on smart grids, what the central planners are talking about is not really a smart grid, just a lot more wires. A smart grid also provides information about prices and quantities – in both directions – so that electricity users understand the consequences of using electricity at 4 PM on a weekday in August versus 10 PM in the evening in October, and generators understand whether and when to add additional generating resources to the grid.

True demand response probably would have bought California enough time earlier in this decade to forestall the worst impacts of its ill-conceived market plan by allowing large users to scale back demand during peak periods. However, electricity users were never bought into the solution, except as the victims of load shedding.

Spending a lot more money on wire without letting consumers make informed choices about when to use electricity (and that includes everything form electric dryers to electroplating) will not add smarts to the electricity system.

In the Mid-Atlantic region the PJM transmission operator tries to substitute for the absence of retail pricing signals with local generation and transmission pricing. This works pretty well, but still does not engage retail customers. PJM also does the major share of regional transmission system planning. However, the balancing of the system occurs almost exclusively on the supply side, with customers unaware of the myriad adjustments in the locational marginal prices of electricity throughout the PJM system.

At present there are no states that allow the requisite degree of price variation and informed electricity demand to make a smart grid operational. Realistically, it will take many years to figure out workable models (trials of smart grids in, e.g. Boulder, CO, have shown mixed results and considerable consumer confusion) for two way communication and price adjustments.

Just because something is difficult does not mean it should not be tried, but we should not kid ourselves about the likelihood of the 48 contiguous states, or even the 15 member states of PJM, adopting a brand new pricing model, especially when even simple time of use pricing has proven difficult to implement.

3 Larry Dickman { 02.24.09 at 11:54 am }

You cite a 2003 paper on why remote generation is not the way to go in 2009 and beyond: “Today, however, new generation tends to be natural-gas fired.” Gas-fired capacity is used to meet peak demand, and it rarely operates. And it is not economic to run for baseload capacity.

The debate on long-distance transmission, and more to the point, on building an extra-high voltage overlay (765 kV), centers more on baseload energy sources–coal, nuclear, wind. And wind resources are located far far away from load.

A recent study by the Midwest ISO looking at the difference of tapping lower value wind near load vs. tapping high value wind resources in the Great Plains and building transmission to move it to load showed that: “All other cost impacts being equal, installed generation cost difference could allow for more than $30 billion incremental transmission costs to access higher capacity wind.”

“I don’t have any brief one way or the other, but I can’t believe that central planning works surprisingly well when it comes to electricity markets but, for some mysterious reason, not very well anywhere else.”
It’s not that central planning doesn’t work surprisely well, but that privatization doesn’t work at all. Where would our highway system be today if a central planner had not built the interstate highway system?

There are two efforts going on: more transmission to access wind (storage is coming on to solve the intermittancy problem); and the smart grid effort. Smart grid could enable people to better respond to real-time prices. Do they want to? I beleive smart grid is terribly oversold.

On the wholesale side, though, there is, and will be increasing demand response from larger customers and aggregators that will partipate directly in RTO programs.

“…and when to add additional generating resources to the grid.”
We already have real-time wholesale prices. One of the problems with these prices is that they do not provide a robust enough signal to incent new generation resources to be built. And, if you wait for price spikes to incent generation, what kind of generation will be constructed? Gas, which is cheap to construct and fast to site and build (it’s just jet engines, after all).

A problem with restructuring, or privatiziation, is that the load serving entity has no long-term obligation to serve customers, and cannot count on even having customers long-term, so they are reluctant to invest in any long-term generation plants.

Energy policy, electricity policy, is pretty important–the debate is what do we want to build, and where? Just saying set the rules and let market participants work it out is like being for motherhood and apple pie. Do you have anything that points to a history of all of the subsidies that fossil fuels have received from the government in the last 150 years?

4 Ed Reid { 02.24.09 at 5:13 pm }

The key to the “smart grid” is actually informed and motivated customers. The customer communication aspects of the “smart grid” could inform customers; the imposition of real time prices could motivate customers.

Customers are not altruistic. They are not likely to inconvenience themselves because doing so would advantage the electricity supply chain; or, because there might possibly be some ill-defined environmental or efficiency benefit. Many would do so if the costs of not doing so were high enough; and, they were aware of that fact timely and were in a position to act upon it.

Many of the experiments with demand management have failed because the experiments were of too short duration to justify investment in advanced appliances or thermostats. Others have failed because the rate differential offered was insufficient to justify changed behaviors. Still others have failed because the customers had no control over the demand management decisions; or, because the customers were not at home when actions had to be taken.

Regulators have been reluctant to expose customers, particularly residential and small commercial customers, to the full range of real time electricity price fluctuations. They have thus limited the potential success of their experiments; and, later judged the achieved success inadequate. Essentially, they have unwittingly created self-fulfilling prophesies, rather than truly tested the potential range of customer responses.

The simple ability for the customers to preselect electricity prices, above which certain energy consuming activities would be terminated automatically, has eluded most of the designers of the experiments. Perhaps the limited experiments which have offered this ability, and have thus proven more successful, will encourage these experimental designs to be tested more broadly. Then, perhaps, we might actually learn something about the potential of the “smart grid”.

5 Ed Reid { 02.25.09 at 8:46 am }

Larry Dickman,

“Gas-fired capacity is used to meet peak demand, and it rarely operates.”

Natural gas is used as a fuel for electric power generation in three types of plants: simple cycle turbines (SSTs), used for peaking, as you suggest; combined cycle turbines (CCTs), used typically as intermediate load generators; and, repowered coal plants, also used to meet intermediate loads.

“The debate on long-distance transmission, and more to the point, on building an extra-high voltage overlay (765 kV), centers more on base load energy sources–coal, nuclear, wind.”

That is the first time I have ever seen wind referred to as a base load power source. It might be someday, with appropriate storage, but we are not there yet.

“A problem with restructuring, or privatization, is that the load serving entity has no long-term obligation to serve customers, and cannot count on even having customers long-term, so they are reluctant to invest in any long-term generation plants.”

Most power generation in the US, with the exception of hydro, has always been investor owned. Restructuring of the electric utility industry, where it has occurred, has separated the previously vertically integrated utilities into their fundamental components: generation, transmission and distribution. This separation has provided customers the opportunity, as well as the obligation, to arrange for the purchase and delivery of the power which they consume. In some cases, regulators have required the distribution utility to function as the provider of last resort (POLR) for customers who fail to contract for power, or whose power providers fail to perform. This requirement has been onerous in several cases, frequently as the result of ethically questionable activities by power providers, most of which have now “left the scene”.

Many customers, particularly industrial customers, were dissatisfied with the vertically integrated utility model and lobbied aggressively for restructuring. The state utility regulators proceeded, “each in their own words, each in their own way”, as is their wont. Some succeeded; some did not.

However, I don’t believe there is any way that the previously vertically integrated utilities can “unsell” their generating capacity. We will have to move forward from our current situation.

6 Tom Tanton { 02.25.09 at 9:26 am }

We don’t need a “smart grid” as the forum described it–which is really no more than the old grid on steroids. What we ‘need’ is an intelligent, adaptive, self healing and learning grid–much more like the ‘Net than the Highways. (and why is GE selling the smart grid using the Scarecrow–the ultimate brainless entity?)
The problem is the grid wannabe-controllers are forcing a concept of the grid around the requirements of certain generation types (wind) rather than the other way around–which would actually be an inlelligent approach. The cognescenti are using the same design paradigm as got us in this pickle (historically grid design was used to accomodtae powerplants that have to be far away from people cuase they were ‘dirty’–now only because the resource is far away.)
Further, the electric grid is an interdependent and interconnected infrastructure–dependent on and required for–natural gas delivery, phones and communications, banking, and transportation (ever been at an airport when the ‘net goes down due to a power glitch?) We all need to keep electricity in the proper persepctive: it is an enabling resource not a good, service or goal in itself.

7 davod { 02.25.09 at 9:28 am }

On a related topic – Windpower peak loading.

Does anyone have the answer to absorbtion of Windpowers Peak Power.

8 Larry Dickman { 02.25.09 at 2:05 pm }

Ed Reid,

What is the capacity factor of gas-fired generation? At least in the MISO, it’s very low.

Wind is an intermttent resource, but baseload in the sense of going to the bottom of the stack–it’s fuel price is zero, after all. It lowers the spot price for everyone.

The restructured states are a mess. They can’t assure long term resrouce adequacy at all, which is why schemes like PJM’s RPM have had to be invented.

There is a huge amount of generation that is not owned by IOUs. There is an easy way for IOUs to “unsell” their generation. Illinois, for example.

9 Tom Tanton { 02.26.09 at 6:04 am }

For the last two commenters. First, wind seldom operates in correlation with the peak of most grids. In California, for example the peak is driven by summer air conditioning and wind operates ~5% during that period. In Denmark, the situation is someewhat different (seasonally) but the output/demand profile is still not in correlation, hence large give=away to other Northern EU states with transfers back when DK needs it (ironically from coal and hydro.). Wind does not provide “peak” cpacity anywhere in the world, nor does it contribute much to dependable on-peak capacity (also known as capacity credit or value–different than capcity factor) It does not provide baseload power–that is a different thought than “cheapest.” Wind also does not provide, necessarily, the “cheapest” resource simply because of the forced ramping of other connected units which decreases their efficiency. By being dispatched in a flittering manner, they impose (although do not experience) costs on others.
Natural gas fired generation can have capcity factors of over 65% easily, at least for some units, depending on the availability of resources within a grid. The important thing is to have capacity available when needed for dispatch.

10 davod { 02.26.09 at 8:40 am }


Please read the link included with my post. The issue is what to do with the windpower when it is at its peak loading. Where do you send it. How does your grid accomodate the spike in power.

The Danes are lucky that the Norwegians can switch off their hydro generated power and so absorb the Danish excess capacity.

The UK problem is that they have no high capacity lines able to take away the excess loading.

New York City is applying for permission to build windpower capacity out far in the Atlantic (Far enough that they have to apply to the Federal government for permission). What is the NYC answer to the peak feed of power from the wind mills.

11 Tom Tanton { 02.26.09 at 1:41 pm }

Davod–The term “peak loading” generally is used to refer to the period of highest demand on the grid. Nevertheless, the period of time when wind turbine output peaks is not coincident. What to do? Simple–the governments (in the U.S. and in DK and elsewhere) should stop paying exhorbitant amounts to build even MORE wind. If there is too much wind at any specific moment, just shut the turbines down–don’t add stress to the transmission system.

12 martha thacker { 02.27.09 at 2:49 am }

Wind cannot be stored. It cannot be located close to the general population as it is a health hazard. It produces power in spits and spurts. So why would the US want to build miles and miles of transmission lines for such an unreliable source of power? The answer might be with Obama’s chief economic advisor, Larry Summers. He is heavily invested in a wind farm corporation, First Wind. When Enron went down…First Wind went up. Even some of the executives. First Wind has lawsuits against them just about everywhere they go. The NY AG is investigating them for bribery of public officials, anti trust violations and fraud. I would not be surprised if Larry and his gang didn’t write the legislation for transmission lines. It will save him money. The Left does not protect the environment. Larry Summers has been known among environmentalists for dumping toxic waste in third world countries.

13 TokyoTom { 02.27.09 at 3:00 am }

Jerry, what kind of deregulation do you think we need? What do you think of Paul Joskow’s call for a Federal Power Act of 2009?

14 Larry Dickman { 02.27.09 at 3:15 pm }

The solution to tapping the vast amounts of wind resources is transmission lines and storage. By deploying more wind, and over a larger area, the fleet’s contribution at peak is likely to be more stable.

15 Progress Report: MasterResource (1Q–2009) — MasterResource { 03.21.09 at 12:27 pm }

[...] Smart Grid, Dumb Economics (Jerry Taylor) [...]

16 150,000 and Counting –Thank You Viewers! — MasterResource { 07.25.09 at 7:55 am }

[...] next five most popular posts have been: Smart Grid, Dumb Economics 2,361 [...]

17 Worth Re-reading Weekend (MasterResource works for lazy summer days too) — MasterResource { 08.01.09 at 8:09 am }

[...] Smart Grid, Dumb Economics 2,361 Views (Robert Michaels) [...]

18 Andrew McKillop { 04.21.10 at 8:37 am }

Cut the [crud] and face up to reality is the message for smart grid dreamers, whether big government or save-polar-bear ecologists.

Say we wanted even 10% of current OECD car fleets shifted to all electric by the unrealistically early dates some claim, for example “about 2015-2020″.

That makes around 20 million in USA, 21 million in Europe, and plenty in Japan, South Korea and Taiwan.

Say each electric car will need an average 4 kW charge rate, and we all know what times of day they will want to charge at home, back from work, and ready to get out there to the nearest 30-mile tailback on urban highways, the next morning.

Do the math !

Now build the power plants let alone the wire and cable needed for this Hero Quest. Then bolt on the smart grid gimmicks and gadgets to amuse the investor crowd.

Leave a Comment