Electricity Crises around the World: Short Term Economics not Inadequate Generating Capacity

Mark B. Lively* ( MBeLively@aol.com This e-mail address is being protected from spambots. You need JavaScript enabled to view it )

Japan was severely ravaged by an earthquake on 2011 March 11.  From half the world away, it seemed as if what the earthquake didn’t devastate, the ensuing tsunami did.  In addition to the immediate damage, the long term suffering has been made worse by nominally inadequate supplies of electricity, so inadequate that there have been rotating blackouts.  Rotating blackouts involve intentionally disconnecting large areas from the grid for an hour or more at a time.  Some friends whose son lives near Tokyo worry about him being stuck in the subway if a rotating blackout hits the subway.

Certainly the earthquake and tsunami disabled a large amount of central station generating capacity.  But, if Japan is anything like California, there should still be a surplus of operable generating capacity, not necessarily central station generators but a large number of small consumer-owned backup generators.  These small consumer-owned backup generators, often diesel fired, could provide electricity to the grid and obviate the need for rotating blackouts.  But for these small consumer-owned backup generators to provide electricity to the grid, the utilities would need to

  • Connect the small consumer-owned generators to the grid;
  • Pay the owners of the small generators for their emergency operation; and,
  • Charge consumers for emergency power.

The first issue involves dealing with low voltage interconnections, including allowing the distribution grid to operate as a collection grid.  The other two issues are economic and will be discussed in a greater detail in this paper.

Connecting Small Generators to the Grid

When California experienced the Enron debacle of 2000/2001, the California ISO instituted rotating blackouts, similar to the Japanese experience in the wake of the March 11 earthquake and tsunami.  Ten years ago[1] I pointed out that California had tens of thousands of small consumer-owned diesel generators that could be used to provide electricity to the system.  My summary of the data was that these small consumer-owned diesel generators had a combined capacity equal to about 80% of the California peak demand.[2] But these consumer-owned generators only produced electricity when their owners were disconnected from the grid.

Customer connections would need to be reconfigured so that these small backup generators could be operated synchronously with the remaining central station power plants.  Connecting thousands of small generators would be easier and quicker than building large new power plants.

That California had a combined capacity in consumer-owned backup generators that rivaled the capacity of its central station power plants suggests that Japan also would have sufficient capacity in consumer-owned backup generators to avoid rotating blackouts.  Thus Japan could avoid rolling blackouts if Japan would reconfigure customer connections and pay for the power delivered by the consumer-owned backup generators.

Paying small generators for their emergency operation

System operators continuously match the supply and demand for electricity.  This matching of supply and demand is generally accomplished by controlling the output of generators, since they are often owned by the utility or committed to the utility.  System operators do occasionally control the demand for electricity.  Rotating blackouts is an extreme form of controlling demand.  A more acceptable form of controlling demand is cycling electric water heaters and air conditioning systems, a concept referred to as Load Management (LM), Load Control (LC), or Demand Side Management (DSM).  Some utilities also offer large industrial customers discounts in exchange for a promise by the customer to reduce consumption during emergencies.  For this paper I will use DSM as a collective reference to these similar concepts.

When there is a surplus of electricity, the system operators notice that the frequency increases above its nominal value: 60 Hertz (cycles per second) in the US and about half the world; and, 50 Hertz in other parts of the world.  I understand that Japan has parts of its systems at each of these standard frequencies.  When there is a shortage of electricity, the system operators notice that the frequency decreases below its nominal value.  This is often experienced when a large generator suddenly shuts down.  While other generators are responding, the system frequency is substandard, slowly rising back to the standard.

System operators have standard ways to pay for central station generation and for DSM, often under long term contracts.  Also, central station power plants are often included in an elaborate computer program, “system dispatch,” that determines the projected short run marginal cost (SRMC) of operating the central station generators.  This SRMC is used in setting the payment to the central station generators.  SRMC is often determined hourly and sometimes on an intra-hour basis.

System operators often do not have a way to pay for the emergency operation of small backup generators.  I developed an approach over twenty years ago[3], which I have been refining since then.[4] I call the concept WOLF, for Wide Open Load Following, in that the price follows the load up, at least when the load goes up without a concomitant increase in generation.

When system frequency is the standard (50 or 60 Hertz), the WOLF price is the same as the price the system uses to pay central station power plants, such as SRMC.  When the system frequency is below standard, the WOLF price is higher than the SRMC following a formula that uses the system frequency as an independent variable.  Conversely, when the system frequency is above standard, the WOLF price is lower than the SRMC.

In the context of this paper, the WOLF price would be used during emergencies to provide an incentive for consumer-owned backup generators to operate synchronously with the central station power plants.  The WOLF price could also be extended to price variances at the central station power plants.  A central station power plant producing more than the dispatch amount would be paid the WOLF price for that differential.  A central station power plant producing less than the dispatch amount would pay the WOLF price for that differential.

The cost to operate a small backup diesel generator can be several times SRMC.  For instance, very small generators offered for sale on a hardware chain’s web site had heat rates of about 30,000 Btu/KWH.  This compares unfavorably to heat rates in the range of 8,000Btu/KWH to 9,000 Btu/KWH for central station power plants.  Larger diesel generators, such as those in my ten year old study or at the Japanese nuclear power plants, would have heat rates somewhere between these extremes.

Since small generators require premium fuels, the cost multiple is likely to be significantly more than the 3+ multiple suggested by a heat rate comparison.  Further, there is also the issue of providing compensation to the owner of the backup generator for the ownership and maintenance costs.  This suggests that the WOLF price formula should be at least a factor of 10 or 100 by the time the frequency gets so low that the system operator would be considering rotating blackouts.

Limiting WOLF to emergency generation, either by consumer-owned backup generators or for the production variance on central station power plants, reduces the exposure of the electric system to price variance on SRMC.  Many utility pricing plans include a provision that all generators get paid the same price, such as SRMC.  A generalized application of WOLF could greatly increase the payments made by the electric system to central station generators during the emergencies that have been leading to rotating blackouts.

The California Enron debacle of 2000/2001was especially financially harmful to the utilities participating in the California ISO market because the prices Enron demanded for its fraction of the energy into the market were also paid to all other generators providing energy into the market at the same time.  The California utilities might have been able to survive prices that increased by a factor of 10 or 100 if those prices only applied to Enron’s 10% of the market.  But the utilities were unable to survive when the Enron prices were applied to all energy entering into the market.[5]

Charging consumers for emergency power

Rotating blackouts ration electricity by physical fiat.  Economists claim that price is the best way to ration a commodity.  However, charging consumers the WOLF price for all consumption would produce revenue far in excess of the revenue needs of the utility, at least if the utility limits the WOLF price to backup generators and to variances by central station generators.  An alternative approach is to announce an emergency rationing level and to price variances from the rationing level at the WOLF price.  Such a pricing mechanism limits the excess revenue during the emergency and gives consumers an incentive to reduce consumption even more than the rationing amount.

For example, consider the case where the utility has an emergency such that the utility expects to be 10% short of power.  A customer using 15 KW would be expected to reduce its consumption back to 13.5 KW.  For the 13.5 KW the customer would pay the standard price.  For any variance from 13.5 KW, the customer would experience the WOLF price, whether that variance is positive, representing the concept that the customer exceeded the 13.5 KW, or whether the variance was negative, representing the concept that the customer reduced consumption below the 13.5 KW rationing level.

Table 1. Inelastic Customer Example







$   1.49




$   1.80




$   3.29


Table 1 illustrates the case when the customer declines to ration electricity, effectively showing an inelastic demand curve.  The standard price is $0.11/KWH.  This produces an hourly charge of $1.49/hour.   The 10% of load that was supposed to be curtailed, or the variance, is priced at the WOLF price, assumed to be $1.20/KWH.  This costs the consumer $1.80/hour.  The total cost is $3.29/hour or $0.219/KWH for the 15 KW total load.  The effect is that the customer experiences an average price of electricity that is twice the standard price.

Table 2. Elastic Customer Example







$     1.49




$   (1.80)




$   (0.32)


Table 2 illustrates the contrasting case when the customer responds to the rationing request twice as much as was specified.  Now the variance amount is a negative 1.5 KW producing a credit to the customer of $1.80/hour.  The result is that the customer receives a net credit during the emergency period of $0.032/hour even though the customer is consuming 80% of its standard load.  The average price is a nonsensical negative $0.0263/KWH.

Paying customers for negative variances played a big role in the California ENRON debacle of 2000/2001.  Aluminum smelters had contracts under which they bought electricity at fixed prices.  These smelters shut down and sold their “negawatts” into the California market, reducing the shortages that California would otherwise have been experiencing.


Most areas of the world have a plethora of small consumer-owned backup generators.   These generators are installed by retail consumers as a contingency against a failure of utility supply.  Though many of those supply interruptions are related to problems on the distribution wires, sometimes there is a supply interruption due to an insufficiency of central station generation supplies.  Under such emergency conditions the small consumer-owned backup generators can be used to back feed the network.  Instead of the low voltage wires distributing electricity, the low voltage wires collect electricity.  But converting a distribution system into a collection system requires connecting the small consumer-owned backup generators to the network instead of the standard approach of having them only feed the consumers load.

The financial cost of operating small consumer-owned backup generators can be many times the cost of operating central station power plants.  An emergency pricing mechanism needs to be used to pay the small consumer-owned backup generators.  WOLF provides a variable price for emergencies that can achieve these radically high prices during the emergency and automatically fall when the emergency is over.  Limiting the WOLF price to consumer-owned emergency generators and schedule imbalances reduces the financial drain on the utility.

The finances of the utility can be balanced by implementing an emergency rationing plan, one that does not involve rotating blackouts.  Instead, variances from the emergency rationing plan are priced using WOLF.  This allows elastic customers to surpass the rationing requirements and earn the WOLF price for that action.

The WOLF plan for emergency situations was described for the California ENRON debacle of 2000/2001, when it was shown that California had sufficient consumer-owned backup generators to replace most of its central station power plants, though that would be a costly process.  However, paying for the operation of consumer-owned backup generators might be preferable to imposing rotating blackouts on the citizens.  The concept proposed in “Saving California With Distributed Generation” could help Japan avoid the rolling blackouts that it is experiencing after the earthquake and tsunami of March 11.  The concept could also be used in other parts of the world, such as Iraq, where the loss of supply is due to political upheaval such as wars.


* Utility Economic Engineers, Gaithersburg, MD, USA, MbeLively@aol.com, 301-428-3618.


[1] See “Saving California With Distributed Generation: A Crash Program To Use Small, Standby Diesel Generators To Keep The Lights On,” Public Utilities Fortnightly, 2001 June 15.  Available for free download at www.LivelyUtility.com

[2] Utilities also own backup generators such as the ones at the nuclear plant sites that are to be used during emergencies to provide electricity at the nuclear plant.  These also can be used to forestall rotating blackouts at least when they are not needed for nuclear emergencies.

[3] See "Tie Riding Freeloaders--The True Impediment to Transmission Access," Public Utilities Fortnightly, 1989 December 21, Available for free download at www.LivelyUtility.com

[4] See “A Pricing Mechanism To Facilitate Entry Into The FCAS Market: Comments Of Mark B. Lively, Utility Economic Engineers,” Investigation Of Hydro Tasmania’s Pricing Policies In The Provision Of Raise Contingency Frequency Control Ancillary Services To Meet The Tasmanian Local Requirement, Office of the Tasmanian Economic Regulator, 2010 July 9.  Available for free download at www.LivelyUtility.com, as well as the OTTER web site.

[5] For a discussion of the pernicious effects of using SRMC to price all energy in an electricity market, see “Ratemaking To Facilitate Demand Response, Both ISO Dispatchable And ISO Non-Dispatchable: Comments Of Mark B. Lively, Utility Economic Engineers,” FERC Docket RM10-17-0000 Demand Response Compensation In Organized Wholesale Energy Markets, 2010 October 11.  Available for free download at www.LivelyUtility.com, as well as the FERC web site.

Click to view a printable version of this article.