Sioshansi F.P. Smart Grid: Integrating Renewable, Distributed & Efficient Energy
Подождите немного. Документ загружается.
offerings. Conventional financial models may be employed,
but electricity prices do not conform to some of the
assumptions these models require, which means that they may
not produce consistent and therefore reliable results. The level
of hedging premiums therefore remains the subject of
speculation.
Comparing the posted prices of competitive retailer products
with the cost of paying spot prices for that load is one avenue
for establishing risk premiums, albeit a somewhat flawed one.
Such a comparison uses already known spot market prices
and retail prices that were based on the retailer's price
expectations. However, it is at least a rudimentary indicator of
implied risk premiums. Applying that reasoning to
competitive markets in the Northeast yields implied
premiums of 15–40% for a fully hedged service. The
difference among retailers’ rates for equivalent service
reflects their forward market view (each's expectations of
prices), along with other transactional considerations, like the
cost of operating a retail business (acquiring and servicing
customers).
Auctions and RFPs for default service provide another hedge
cost indicator. The results of the auction for default service in
Illinois caused some to conclude that the implied risk
premium was 20–40%. Recent studies of price response by
ISO-NE utilized risk premiums that are graduated in the
degree of risk of the pricing plan; RTP has the lowest (3–5%),
TOU even higher (8%), and the uniform rate had the highest
(15%). Under these risk premiums, the analysis concluded
that the majority of benefits of price response redound to
those that adopt that behavior.
241
Estimating the Hedging Cost Premium in Flat
Electricity Rates
How can the hedging cost premium be quantified? In one
approach, the hedging premium is considered to be
exponentially proportional to the volatility of loads, the
volatility of spot prices, and the correlation between loads and
spot prices. This can be represented as follows:
where:
π = Risk Premium
σ
L
= Load Volatility
σ
P
= Spot Price Volatility
ρ
L,P
= Correlation Between Load and Spot Price
For example, if price volatility was assumed to be 0.6, load
volatility was 0.2, and the correlation between load and the
spot price was 0.4, the resulting estimate of the hedging
premium would be 5%. In other words, on average, customers
are paying 5% more than they would if they were simply
exposed to spot prices.
With an assumption about the distribution of these three
variables, a Monte Carlo simulation can be used to
approximate a distribution around this premium. Assuming
that the variables are all triangularly distributed with a
minimum of 0 and a maximum of 1, a Monte Carlo
simulation of 1,000 iterations produces the hedging premium
distribution shown in the following figure.
242
Simulated Distribution of Hedging Cost Premium
The mean, median, and mode of the premium are 11%, 8%,
and 5%, respectively. The standard deviation is 10%.
References
[2] http://documents.dps.state.ny.us/public/Common/
ViewDoc.aspx?DocRefId={4FA9C260-CE37-4FA983CA1D7904409
[3] Of course, benefits over multiple years would be much
higher. In a full cost-benefit analysis, all relevant costs
would also need to be factored in.
[4] Faruqui, A.; Hledik, R.; Tsoukalis, J., The power of
dynamic pricing, The Electricity Journal (April 2009).
[5] Its applicability to residential customers is prevented by
state legislation that has frozen portions of residential
rates in order to recover the costs of the energy crisis of
2000–2001 from the unfrozen portions.
[6] FERC Staff, A National Assessment of Demand Response
Potential, Washington, DC, June 2009.
243
[7] Faruqui, A.; Hledik, R.; Newell, S.; Pfeifenberger, J., The
power of five percent, Electr. J. Vol. 20 (October 2007).
[8] Revenue neutrality means that the revenue collected from
the class to which the new rate is being applied would
not change from the revenue collected under the old
rates. In the case of dynamic pricing, this means that the
customer who has a load factor equal to the class average
would see no change in her or his bill. Load factor is the
ratio of a customer's average demand to her or his peak
demand.
[9] Several of the test results are discussed in A. Faruqui, R.
Hledik, S. Sergici, “Rethinking pricing: The changing
architecture of demand response,” Public Utilities
Fortnightly, January 2010.
[10] http://news.smh.com.au/breaking-news-national/
vic-govt-to-review-smart-meters-20100203-nd88.html.
[11] Vickrey, W., Responsive pricing of public utility
services, Bell J. Econ. Manag. Sci. 2 (1971) 337–346.
[12] Schweppe, F.C.; Caramanis, M.C.; Tabors, R.D.; Bohn,
R.E., Spot Pricing of Electricity. (1987) Kluwer
Academic Publishers.
[13] Hirst, E., Price-responsive demand in wholesale markets:
Why is so little happening?Norwell, MA. Electr. J. (May
2001); ISBN: 0-89838-260-2.
[14] Bolton, D.J., Costs and Tariffs in Electricity Supply.
(1938) Chapman & Hall, London.
[15] B. Alexander, “Dynamic pricing? not so fast! a
residential consumer perspective,” Electr. J., Vol. 23,
39–49. July 2010 and S. Brand, “Dynamic pricing for
residential electric customers: A ratepayer Advocate's
Perspective.” Electr. J., Vol. 23, 50–55. July 2010.
244
[16] http://www.mtc.ca.gov/news/info/toll_increase.htm.
[17] Cross, Robert G., Revenue Management. (1998)
Broadway Books, New York, NY.
[18] J. Upton, “Giants make a dynamic move: Team to
implement pricing strategy for nonseason tickets.” The
Examiner, February 9, 2010.
[19] http://sabres.nhl.com/club/page.htm?id=39501.
[20] See the appendix at the end of this chapter for a
discussion of the hedging premium.
[21] F.A., Wolak, An Experimental Comparison of Critical
Peak and Hourly Pricing: The PowerCentsDC program,
In: Prepared for The 15th Annual Power Conference,
The Haas School of BusinessU.C. Berkeley, March 13.
(2010).
[22] Additional details are available in Ahmad Faruqui and
Ryan Hledik, “Transition to Dynamic Pricing.” The
Public Utilities Fortnightly, March 2009.
[23] Chao, H., Price-responsive demand management for a
smart grid world, Electr. J. Vol. 23 (February 2010).
245
246
Chapter 4. The Equity Implications of Smart
Grid
Questioning the Size and Distribution of Smart Grid
Costs and Benefits
Frank Felder
Chapter Outline
Introduction 85
Smart Grid Is Not Consistently Defined, Which Leads
to a Mismatch of Its Costs and Benefits 88
The Compulsory Nature of Smart Grid Raises
Fundamental Equity Concerns 93
Smart Grid Raises Other Equity Concerns Besides
Compulsion 95
Conclusions 98
References 99
Important equity questions remain unanswered with respect to smart grid and center on both the size
and distribution of its costs and benefits Smart grid is an ambiguous term with multiple meanings It
captures the general notion that the electric power system could be upgraded with new measurement,
data, and computational technologies to provide consumers, system operators, and transmission and
distribution utilities with additional functionalities that are purported to deliver significant benefits to
utilities, ratepayers, and society in the form of lower costs, improved reliability, and lower
environmental impacts Unfortunately, by using this term, some advocates of smart grids may be
making the assumption rather than the case that these technologies work, that their benefits to society
will exceed their costs, and that these technologies are equitable Regulators asked to approve smart
grid investments must explicitly consider multiple definitions of equity, require utilities to provide
more choices to consumers, and probe the cost-benefit analyses that are used to justify large smart grid
investments
247
Cost effectiveness of smart grid, cost allocation, equity
Introduction
The many and various combinations of technologies
encompassed by the term smart grid are large and uncertain
investments. A major recent report, but by no means the only
one, places the cost of smart grid for the United States alone
between $338 and $476 billion over a twenty-year period
with net benefits estimated in the range of $1,294 to $2,028
billion as shown in Table 4.1[1]. Stated differently, the cost
estimates vary by almost 41% and benefit estimates vary by
almost 57%. There is no guarantee that the potential net
societal benefits will occur, either for society as a whole or
for particular segments of society, particularly low-income
families. The above cited study, for example, excludes the
costs of generation and transmission expansion needed to
support additional renewable resources, which would provide
much of the environmental benefit associated with smart grid
and also excludes the customer costs for smart-grid-ready
appliances needed to capture fully the benefits of smart grid
for consumers [1]. Chapter 7 of this text provides rather rough
“guesstimates” of the benefits of smart grid, on a global scale
and suggest that EPRI may be under-estimating the benefits.
Table 4.1
Summary of Estimated Costs and Benefits of the Smart Grid
20-Year Total ($ billion)
Net Investment Required (Cost) 338–476
Net Benefit 1,294–2028
Benefit-to-Cost Ratio 2.8–6.0
Source: EPRI [1].
Besides questions of the relative size of costs and benefits, a
host of equity questions surround smart grid. For the most
248
part, because these smart grid investments would be
recovered through mandatory charges assessed in utility rates,
they thus raise substantial equity issues that must be
considered and analyzed. Whether future utility bills would
decrease or not increase as much without smart grid is much
less certain. Smart grid advocates are making an analogy with
smart phones, computers, and other electronic devices that
have taken off over the last twenty years but fail to
acknowledge the fundamental difference between smart grid
and consumer electronic devices. With smart grid, consumers
are being asked to guarantee much if not most of these
investments. There is little discussion of the issue and
importance of choice. Instead, arguments are made, for
example in the case of advanced meters, that one size must fit
all. The meter purchase and selection are not left up to the
consumer but to the utility and the regulators who must
approve the investments. The power of the technological
revolution in computers, communications, and electronic
devices is that it gives consumers a wide range of choices,
including the choice not to purchase a particular device or use
a service.
The compulsory nature of cost recovery is critical to any such
equity analysis. Even if an individual consumer would be
better off as a result of contributing to the recovery of smart
grid investments, the requirement to do so may be considered
inequitable. There are also other fundamental and critical
questions regarding the different equity concerns with smart
grid and the conditions under which smart grid would result
in equitable outcomes. That being said, the fact that equity
issues can be raised should not be a reason to halt debate.
Instead, equity considerations must be defined precisely,
articulated clearly, and analyzed thoroughly as applied to
249
smart grid. A lot more analysis and discussion are needed to
answer key questions, both related to the equity of smart grid
investments and to their costs and benefits, before regulators
should proceed.
In addition to the compulsory nature of cost recovery is the
fact that lower income consumers spend a greater percentage
of their income on energy, including electricity, than higher
income consumers: the lowest income category in the United
States spends about 15% of its income on electricity, whereas
the highest income bracket spends approximately 1.5%, ten
times less on a percentage basis [2]. Thus, raising the cost of
electricity is regressive: it harms lower income consumers
more than higher income ones (see Table 4.2). Given how
vital electricity is to the health and safety of families, it is not
considered to be a commodity but a merit good [3] and [4].
This means, if one accepts this definition, that its availability
and pricing should be based upon need for some, not ability to
pay. These three factors—compulsory recovery of costs,
energy costs disproportionately affecting the poor, and the
fact that electricity is a merit good—combine to raise
important equity issues with respect to smart grid.
Table 4.2
U.S. 2005 Electricity Consumption and Costs by Household Income
U.S. Households Annual Electricity
Household Income Number (million) % kWh Cost % of Income
Less than $10,000 9.9 8.9% 7,854 $785 15.7%
$10,000 to $14,999 8.5 7.7% 8,710 $871 7.0%
$15,000 to $19,999 8.4 7.6% 9,506 $951 5.4%
$20,000 to $29,999 15.1 13.6% 10,040 $1,004 4.0%
Source: 2005 Residential Energy Consumption Survey, Energy Information
Agency
250