20040220Power Direct.pdf

Brad M. Purdy
Attorney at Law
BarNo. 3472
2019 N. 17th St.
Boise, ID. 83702
[phone]
FAX: [phone]
bmpurdy~hotmail. com
ISB #3472
Attorney for Petitioner
Community Action Partnership
Association ofldaho and
AARP.
FTCEIVED 0- ED
200:; FEB 20 PN 3: 30
urn; i iF~; 'i~'
("'
-- - -u -Iv,)
BEFORE THE IDAHO PUBLIC UTILITIES COMMISSION
IN THE MATTER OF THE APPLICATION
OF THE IDAHO POWER COMPANY FOR
AUTHORITY TO INCREASE ITS RATES
AND CHARGES FOR ELECTRIC SERVICE
TO ELECTRIC CUSTOMERS IN THE
STATE OF IDAHO
CASE NO. IPC-O3-
AARP
Direct Testimony of
Thomas Michael Power
1. Introduction and Summary
Please identify yourself for the record.
My name is Thomas Michael Power. I am Professor of Economics and
Chairman of the Economics Department at the University of Montana, Missoula
Montana , 59812.
Have you testified as an expert witness before this and other commissions
in the past?
Yes. I have testified before this Commission on numerous occasions since
1978 including testimony in the last Idaho Power Company general rate case. I have
attached Appendix B to this testimony that discusses my professional experience and
expertise.
What topics will your testimony cover?
I will focus on those aspects of IPC's cost of service study and its
proposed rate design that inappropriately burden the residential class and conflict with
good public energy policy.
Could you please summarize your conclusions on IPC cost of service and
rate design?
Yes. Let me simply list the conclusions supported by the body of my
testimony here:
i. One of the basic principles guiding IPC's rate design is the assumed
need to collect fixed costs by levying fixed charges. Neither business nor economic
principles support such a strategy. Businesses regularly and appropriately collect fixed
costs on the basis of customers' use of the services those businesses provide.
Power: Direct
AARP
P.
ii. Collecting fixed costs in fixed charges conflicts with the cost of service
principles this Commission has used for decades. That "principle" would lead all of the
capital costs associated with hydroelectric generation and base-load generation to be
collected in either demand charges or customer charges.
iii. Utilities tend to favor high fixed charges not because of any economic
principle but because of their private interests: Such charges stabilize their cash flow in
the face of loads that fluctuate with weather. Such charges also allow the reduction of
usage charges and allow a variety of promotional pricing policies for the most price
sensitive customers.
iv. IPC seeks to classify over 36 percent of the costs of distribution
system lines and transformers as "customer costs." Customer costs are costs that vary
with the number of customers, but IPC makes no effort to demonstrate that these
distribution costs vary in this way.
v. The allocation of a significant part of the distribution system costs on a
per customer basis and the remainder on the basis of peak demand effectively charges
small users twice for the use of the distribution system. Because the "customer
component" can handle 85 percent of the average residential customer s load, a
substantial additional allocation on the basis of residential load is inappropriate.
Adjustments to solve this double-charge problem effectively lead back to a pure
demand allocator.
vi. IPC's assertion that the proposed $10 per month customer charge is
modest compared to the $25 per month charge the cost of service analysis indicates is
appropriate is based on an error. In calculating the $25 charge IPC divides all
Power: Direct
AARP
P. 2
distribution costs by the number of customers even though IPC itself classifies most of
them as demand costs.
vii. IPC's "collect fixed costs in fixed charges" principle could justify a
monthly customer charge close to $50 and an energy charge of 1.2 cents per kwh. That
would be an irrational outcome when IPC is currently facing full marginal costs
expressed in kwh terms of 9 cents during the summer.
viii. The distribution system does not only provide peak hour services. It
also provides a variety of valuable services to customers throughout the year. The fact
that these services are not peak-load-related does not mean by default that they are
customer-related.
xi. IPC's cost of service analysis shows that the rates paid by the
irrigation class are so low compared to irrigation class costs that a huge revenue deficit
exists. About 40 percent of the rate increase IPC proposes for the residential class is
associated with the residential class paying part of the irrigation class' costs. This led
IPC to raise the rate increase to the residential class from 13 to 19 percent. This large
irrigation revenue gap has existed for at least 20 years. I recommend that the
Commission find a way to assure that it is systematically closed over a reasonable time
period and not allowed to burden other customers for another 20 years.
x. Rate design should focus on getting price signals correct. Given the
high and rising marginal and incremental costs associated with providing electrical
service , this means that rate design should focus on the usage charges, not the fixed
monthly charges.
xi. IPC's proposal to introduce seasonal and time-of-use rates can
Power: Direct
AARP
P. 3
improve the accuracy of the usage charges that customers face. For demand-metered
customers, better balancing demand and energy charges to reflect relative peak
capacity and energy costs is also appropriate.
xii. IPC is incorrect in assuming that because the residential class is not
demand metered , demand costs cannot be conveyed to residential customers and
might as well be collected in the fixed monthly charge. There is a high correlation
between energy usage and peak demand. For that reason, high energy charges can
effectively convey the high costs of peak usage.
xiii. Higher summer kWh charges for the residential customers are not the
only or necessarily the best way to convey the higher costs of peak load use.
Residential peak loads come in the winter when IPC's marginal costs are also high. The
proposed summer rates do nothing to confront customers with the costs associated with
winter peak usage.
xiv. A block rate structure that provides an initial block of electricity at a
low rate and then charges a higher rate for all usage in excess of that initial block is also
a seasonal rate in the sense that during high consumption seasons, summer and winter
more of the load moves into the higher tail-block and is billed at the higher rate.
xv. Such a blocked rate design combined with a low monthly customer
charge would benefit the vast majority of residential customers compared to IPC'
proposed rate design. It is also likely to protect households with relatively lower levels
of consumption, including those on low and fixed incomes.
Power: Direct
AARP
P. 4
2. THe Allocation of Fixed Costs on a Per Customer Basis
IPC witness John Gale has indicated that IPC's "cost-based approach" to
rate design "has led to rate design proposals that better align fixed costs with fixed
prices and variable costs with variable prices." (Page 13 at 4) Is there any economic
principle that indicates that fixed costs should be collected in fixed charges?
A. Absolutely not. All businesses have fixed costs. Most businesses do not try to
cover those fixed costs with fixed charges.
Consider large retail stores and shopping centers. They have to provide
extensive parking, floor space , and personnel to handle peak demands for their
services. Those stores could charge a parking fee and an entry fee to each customer
and could charge higher fees on peak days and .peak hours. But almost none do that
quite the opposite. Often when you have to pay to park in private parking facilities, the
stores will pay your parking costs for you as long as you make almost any purchase.
They certainly do not charge entry fees to their stores. They include those fixed costs in
the charges they make to customers and those charges tend to be proportional to the
purchases that the customers make.
Very few businesses collect their fixed costs through fixed charges unrelated to
usage. Competitive markets, in general , simply do not allow them to do that. Instead
businesses have to collect their fixed costs in usage-related charges. This is not a sign
of market failure or inefficiency.
How does IPC use the "fixed costs should be collected through fixed
charges" principle in this case?
Power: Direct
AARP
P. 5
As Mr. Gale says: "The emphasis on moving fixed and variable prices to
be more reflective of fixed and variable costs led to the Company s proposals to
increase the monthly service charge for residential and small general service
customers." (Page 13 at 9)
It was IPC's application of this incorrect "principle" that led to its proposed
quadrupling of the monthly customer charge for residential customers.
Do utilities have a history of favoring the type of fixed charges competitive
firms rarely can impose?
Yes. In the past, electric utilities have tended to favor classifying as many
of their costs as possible as "capacity" or "customer" costs. They have pursued the type
of "fixed-variable" division of costs Mr. Gale proposes in this case, where energy costs
are defined as those that vary with usage in the short run and all other costs are either
capacity or customer costs. This approach has two advantages from the utility'
perspective. First, it allows the utility to seek to recover as large a part of its costs as
possible in relatively fixed and reliable charges. This stabilizes utility revenues. Second
it allows the utility to pursue a variety of promotional ventures. With low estimates of
marginal energy costs , the utility can justify very low rates to customers who expand
their energy usage while burdening those customers whose use is less flexible with high
charges.
The point is that utilities might have a private interest in distorting cost analysis in
a particular direction, towards low energy charges and high fixed charges. Regulators
need to be vigilant in resisting this tendency when it conflicts with the public interest.
Has this Commission resisted efforts to collect all fixed costs as demand
Power: Direct
AARP
P. 6
or customer charges?
Yes. That is clear in the cost of service approach that this Commission
has approved for use by IPC. Although most of the costs associated with hydroelectric
and coal-fired generation are fixed, this Commission has not approved the collection of
all of those fixed costs on the basis of class contribution to the annual coincident peak
load. Instead this Commission has recognized that many of the fixed costs associated
with generation are incurred in order to produce energy. Burning coal in the open or
letting water run downstream does not generate electric energy. Huge amounts of
capital must be used to transform that natural energy into useful electric energy. That is
the reason that the system load factor is used to classify the majority of the fixed costs
associated with generation as energy costs, not capacity costs.
This Commission has also rejected the use of class contributions to a single
annual peak as the basis for allocating the fixed costs that are classified as capacity
costs. Rather than loads at a single hour on a single day dictating the allocation of
capacity costs, a weighted average of the twelve monthly coincident peaks is used. In
addition , for most customers, it is the peak usage during each month that determines
the demand charges that are due, not peak usage at one hour during the year. Both
these decisions by the Commission turn the demand allocation and charges into a type
of usage charge as opposed to a fixed charge.
Q. Mr. Gale appears to characterize demand charges as fixed charges. Is this a
correct description of demand charges?
A. Mr. Gale explains the proposed quadrupling of the residential monthly
customer charge in the following way: "Since (residential) customers are not demand
Power: Direct
AARP
P. 7
metered , the service charge is the only fixed rate component available to adjust and
thus becomes more important as a tool for fixed cost recovery." (Page 13 at 13)
This is a doubly confusing assertion. Demand charges are a usage-related
charge; they are tied to a customer s peak usage during a particular time period. Like
energy charges, demand charges are tied to the customer s pattern of use.
Second, simply because a customer is not demand metered, it does not mean
that the only appropriate way to collect peak demand costs is through a per customer
charge. Such an approach has no cost-based logic to it.
Customer costs are costs that vary primarily with the number of customers
served. Stated differently, they are costs that could be avoided if a customer ceased
taking service. That is the cost causal principle behind customer costs. That definition
is parallel to energy costs that focus on the costs associated with expanding the energy
producing capacity of the system or demand costs thatfocus on expanding the capacity
of the system to meet peak loads.
If demand costs cannot be billed directly to residential customers because
demand meters are not cost-effective for loads that small, there is no logic to concluding
that the obvious way to collect those demand costs is through a customer charge. To
the extent that residential peak usage tends to track residential energy usage, collecting
those charges through the energy charge might be quite appropriate from a cost casual
point of view.
You seem to be implying that IPC is proposing to collect demand costs
through a customer charge. IPC says that it is seeking to collect distribution costs that
are not demand-related through a fixed monthly "service charge." Aren t you misstating
Power: Direct
AARP
P. 8
IPC's position?
I don t think so. Ms. Brilz in her cost of service study calculates a monthly
customer service charge" of almost $25. That calculation explicitly involves collecting
all distribution demand charges on a per customer basis. (Exhibit 38 , Excel
spread sheet at K300) It is this high "monthly service charge" figure that both she and
Mr. Gale cite as indicating that the proposed increase in the monthly charge from $2.
to $10 is just the first step in closing a huge gap between fixed costs and fixed charges.
As Ms. Brilz says in justifying the quadrupling of the customer charge: "The $10.00-
Service Charge represents approximately 40 percent of the cost-of-service result on
$24.61 shown at line 300 on page 1 of Exhibit No. 42." (Page 36 at 2)
It is important to understand that the $24.61 is not a cost of service result. It
explicitly takes costs that even IPC classifies as demand-related and collects them as a
customer cost.
Has IPC's "fixed costs should be collected through fixed monthly charges
principle been applied correctly in IPC's calculation of a $25 monthly customer charge?
No. Inexplicably, IPC has under-calculated the monthly customer charge
that this "principle" calls for. Although IPC divided all distribution demand-related costs
by the number of customers, it failed to do the same to the fixed demand related costs
associated with electric production and transmission. Since these, too, are fixed costs
that cannot be collected through a demand charge from residential customers , the logic
of the $25 calculation would demand that production and transmission demand charges
also be collected in this fashion. Mr. Brilz s Exhibit No. 42 suggest that doing so would
add another $13.50 per month to the monthly charge, bringing it to over $38 per month.
Power: Direct
AARP
P. 9
There is no more logic to collecting distribution demand-related costs in a monthly
customer charge than there is to collecting production and transmission demand-related
costs in this manner.
A $38 per month fixed charge is not the limit of what one could calculate. IPC
faces significant overhead costs that are difficult to associate with customer usage. All
of them could be treated as "fixed overhead costs." Then there are the fixed costs of
production that this Commission has characterized as energy-related. Those fixed costs
also could be collected on a per customer basis. Under IPC's "principles " the only
costs that are clearly legitimate to collect in the residential kWh usage charge are the
variable fuel and purchased energy costs. These represent about a fifth of the total
residential revenue requirement. Such a "principled" approach would lead to a monthly
customer charge of about $50 and a kWh charge of 1.2 cents.
This may sound like a gross exaggeration, but some utilities that have stayed
focused on revenue stability and the promotion of electric usage have implemented rate
designs that are not far removed from this model.
Q. What are your objections to basing residential and small commercial electric
rates on a fixed-variable cost approach?
A. Such an approach has no economic logic to it since it ignores cost causality
and focuses merely on whether a cost is fixed or varies in the short term with usage.
The reason that the fixed costs were incurred in the first place simply gets ignored.
Such an approach is not a cost of service approach since cost causality is largely
ignored.
From an energy policy point of view, unless one believes that the long run
Power: Direct
AARP
P. 10
incremental cost of electric energy is and will remain very low, such a pricing policy will
give grossly distorted price signals. Since some of the major costs associated with
producing electric energy are the fixed investment costs associated with the generating
facilities , focusing only on the variable costs can be very misleading. Hydroelectric
base load thermal-electric, wind-electric, etc. have very low or zero fuel costs , but that
does not mean that they produce electricity at near zero economic cost. Cost allocations
and rate design that seriously distort price signals and encourage inefficient
consumption behavior are very costly in the longer term.
Do IPC's estimates of the incremental costs of electricity suggest that
additional electric supplies will be available in the future at very low cost?
No. The marginal cost study that IPC prepared for this rate case estimates
total marginal energy, capacity, and transmission costs for the summer months in the
7 to 9.2 cents per kwh range and 6.6 cent range for the peak winter months. (I
have used monthly system load factors to convert the monthly generation and
transmission marginal capacity costs to a kWh basis.) (IPC Response to Idaho Irrigation
Pumpers Association Second Production Request, No. 30.
IPC's 2002 Integrated Resource Plan and the preliminary figures being
developed for the 2004 Integrated Resource Plan both indicate future incremental costs
of supply in the 4 to 8 cent per kWh range on a levelized , life-cycle cost basis.
Depending on the fuel source , fuel costs would represent a quarter to two-thirds of
those costs. (2002 IRP, Tables 11 and 12; 2004 IRP Advisory Council December 18
2003 presentation, pp. 66-67.
Does IPC recognize that the incremental costs of serving its customers in
Power: Direct
AARP
P. 11
the present and in the future will exceed the embedded costs?
Yes. IPC President and CEO , J. Lamont Keen , in his prefiled direct
testimony said: "Clearly, growth has not paid for itself. The incremental costs of adding,
operating, and maintaining generation , transmission and distribution plant are greater
than the embedded costs associated with generation , transmission and distribution
plant that have been the basis of Company rates over the last ten years." (Page 16 at
16)
In this setting, it is important to find ways of signaling to customers the higher
incremental costs associated with providing them with the electric services they seek.
Those price signals need to take a form to which customers can respond by modifying
their electricity usage behavior. IPC's proposals for seasonal and time-of-use rates
move in this direction. Efforts to increase fixed monthly charges that customers can do
nothing to avoid while at the same time reducing usage charges are a move in the
wrong direction.
3. The Calculation of a "Customer Portion" of the Distribution System
IPC has calculated that 36.25 percent of the costs of distribution lines and
transformers are "customer-related" and should be allocated on the basis of the number
of customers being served. What data does IPC provide to indicate that these are
customer-related costs?
None. IPC does not even attempt to make such an argument. Instead it
again makes a type of "fixed-variable" argument. IPC estimates what the minimal load
is that is always on its system. It does this by analyzing its load duration curve. It labels
Power: Direct
AARP
P. 12
this minimal load that is always present its "fixed" load. It compares this to its peak load
and labels the difference its "variable" load. The minimal load is 36.25 percent of the
peak load and therefore IPC concludes , 36.25 percent of the distribution lines and
transformers should be labeled "customer costs" and allocated on a per customer basis.
(Brilz page 9 at 18; Brilz Work Papers, pages 10-11)
Because it has been able to label this level of demand as "fixed " IPC seems to
believe that the costs associated with it are not demand-related. It is hard to follow the
logic here. In the short run all of the distribution costs are "fixed " not just this portion. In
addition, if the load were constant at what is now 36.25% of peak level, distribution lines
and transformers would still have to be sized to meet that load and put in place at
considerable cost. This portion of distribution system is no different than any other
portion of the distribution system. Both are designed to deliver electricity to customers.
Why do you not find this approach to defining a "customer component" of
the distribution system convincing?
The analysis has nothing to do with analyzing the part of distribution
system costs that vary directly with the number of customers. A causal connection with
the number of customers is never explored. This approach would al$o lead to over a
third of the costs associated with generation and transmission to be classified as
customer costs.IPC used its load duration curve to identify the minimum load
regularly put on its system. IPC needs only 36.25 percent of its total electric capacity to
meet those loads. IPC uses only 36.25 percent of its transmission capacity to meet
those loads. In that sense 36.25 percent of IPC's generation and transmission capacity
are also "fixed." That, following lPG's logic, would justify allocating those costs on a per
Power: Direct
AARP
P. 13
customer basis. In general, all investments that serve IPC's base load would be
classified as customer costs if this approach were taken.
Is this IPC approach to identifying a "customer component" of the
distribution costs a variation of a "minimum system" approach to distribution costs?
IPC may have something like that in mind , but it has not provided any
evidence that it has conducted such an analysis. That is not how the 36.25 percent
figure was calculated. Just as important, 36.25 percent is not a "minimum distribution
system." The entire residential class is the source of only 35.8 percent of the load on
the distribution system. In that sense, this "customer component" of the distribution
system could serve the entire residential load. 36.25 percent of total load represents
over 1,400 megawatts of capacity. Some utilities are no larger than this.
Are there other conceptual problems with the way IPC has applied this
demand-customer" division of distribution costs?
Yes. IPC's approach places an unreasonable double cost burden on
those classes of customers whose loads are relatively small.
Please explain the double cost burden that IPC's "customer component"
of distribution costs causes for smaller customers.
The 36.25 percent of the distribution system that IPC labels customer-
related is obviously capable of serving a substantial part of average distribution load.
For customers with relatively light loads , that part of the distribution system is capable of
serving almost their entire load.
The average load per customer across all classes at the primary system level is
7.4 kW. IPC's "minimum system" is capable of serving 36.25 percent of that load or
Power: Direct
AARP
P. 14
67 kW. But the average residential load is only 3.16 kW. For that reason, IPC'
minimum system" (at the primary level) can serve 84.5 percent of residential
customers' loads. For non-residential customers , the average load is 29 kW. IPC'
minimum system" is capable of serving only 9 percent of those loads.
The double charging of residential customers comes from this difference in the
share of class loads the "minimum system" is capable of carrying. When IPC allocates
responsibility for the "non-customer" component of the distribution system on the basis
of demand (63.75 percent ofthe system costs), IPC does not take into account the part
of each class s load that is already being met by the minimum system. Even though only
15.5 percent of the residential customers' loads still need to be served by the primary
distribution system but 91 percent of the other customers' loads still have to be served
IPC implicitly assumes that an equal percentage of all classes' loads remain to be
served. As a result, the residential customers are asked to pay for the distribution
system serving their "minimum loads" again. This leads residential customers to be
double charged for the loads served by the "minimum system " a completely
unacceptable result.
Is this a widely recognized problem with the approach that IPC has taken?
Yes. The NARUC "Electric Utility Cost Allocation Manual" that IPC cites
as its guide specifically warns about this problem:
Cost analysts disagree on how much of the demand costs should be allocated
to customers when the minimum-size distribution method is used to classify distribution
plant. When using this distribution method , the analyst must be aware that the
minimum-size distribution equipment has a certain load-carrying capability, which can
Power: Direct
AARP
P. 15
be viewed as a demand-related cost.
When allocating distribution costs determined by the minimum-size method
some cost analysts will argue that some customer classes can receive a
disproportionate share of demand costs. Their rationale is that customers are allocated
a share of distribution costs classified as demand-related. Then those customers
receive a second layer of demand costs that have been mislabeled customer costs
because the minimum-size method waS used to classify those costs." (Page 95)
Can this double charging of customer classes with small average loads be
avoided?
Yes. What needs to be done to avoid this double assignment is to subtract
from the distribution demand allocation that part of the load that can be served by the
minimum distribution system. This would substantially reduce the distribution peak load
allocator for the residential class. In fact, given that IPC calculated its "minimum
system" as a percentage of total load , it would carry the allocation back to what one
would get if a purely demand-based allocation had been used in the first place. That is
accounting for the load the minimum system is capable of carrying would make IPC'
approach pointless. The correct way to allocate distribution system costs is simply to
use the demand allocator.
But is it not true that some of the distribution costs do not vary directly with
the level of demand and therefore are not really demand-related costs?
Yes and no. The distribution system, like the entire electrical system, is
capital intensive and therefore involves large amounts of fixed costs. But fixed costs are
not customer costs. Those fixed costs are incurred in the pursuit of a broad variety of
Power: Direct
AARP
P. 16
objectives and the amount of fixed costs incurred vary due to a wide variety of
circumstances. It true that the current peak loads on the distribution system do not
explain all of the costs associated with the current system. In that sense , one could
argue that some of those costs do not vary primarily with current peak loads.
If current peak loads do not explain some of the costs of an electric
distribution system, what other design criteria help explain those costs?
Among the most important are the following:
The location and density of customers.
The topography, geology, and character of surface occupation.
The desire to have a resilient and stable system that can continue
to provide service under a variety of possible contingencies.
The economies of scale at the time the distribution system is built
or upgraded.
The expectation that profits can be earned on volumes of electricity
sold to the customer.
How does the location and density of customers affect the costs of the
distribution system?
The closer the customers are located to the company s transmission
system, the lower are the distribution costs. The more densely settled the customers
are , the lower the costs since poles and transformers can be more intensely utilized and
few miles of line have to be strung.
What role does topography, patterns of surface occupancy, and geological
conditions play in determining the costs of the distribution system?
Power: Direct
AARP
P. 17
It is cheapest to set poles and string conductors on flat, unoccupied terrain
with firm but easily excavated under-surface materials. When the area is already
densely settled and used , the cost of building lines rises. When there are many river or
stream crossings or steep changes in topography, the costs will also be higher. When
the poles or underground cable has to be set in bedrock or unstable material , the costs
are going to be higher.
How does the pursuit of a resilient and stable distribution system affect the
costs of the distribution system?
Distribution systems in densely settled areas are planned so that the loss
of one particular line to a particular area or unusual load conditions do result in
customers losing service or the quality of service declining. In densely settled areas the
utility seeks to construct a network with sufficient reserve capacity so that it can meet a
variety of potential contingencies. This increases the quality and reliability of service.
These are valuable attributes of electric service that can be obtained only through larger
distribution system investments. These are also services that are provided to customers
throughout the year, not just at time of system peak.
How does the existence of major economies of scale in the capacity of a
distribution system at the time it is constructed or upgraded affect the relationship
between current peak load and costs?
Because additional increments of capacity can be had at very low
incremental costs, excess capacity is regularly built into the distribution system. The
carrying costs of that excess capacity are lower than frequent upgrades. As a result
there often is not a close relationship between current loads, the capacity of the system
Power: Direct
AARP
P. 18
and the costs of that system.
Why did you mention above the fact that the pursuit of profitable volumes
of sales may lead the utility to incur additional distribution system costs?
Because that is in fact the over-riding "design" criteria. The distribution
system is extended and the costs associated with that extension are incurred in the
pursuit of profitable sales. That is the causal force driving the utility to take on those
costs.
What is the point of listing all of these determinants of distribution costs?
To underline the fact that it is not just one design criteria, e.g. meeting the
peak hour load, that determines distribution system costs. There are many other
determinants that have little or nothing to do with peak hour loads. In that sense IPC is
correct: One cannot show that all of these costs are tied only to peak demand. IPC is
also correct that many of these costs are incurred simply in order to stand ready to
serve customers throughout the year. From that, however, IPC jumps to the false
conclusion that a substantial portion of these costs is "customer costs." We know that
that is not the case. Even IPC does not attempt to show that these costs vary with the
number of customers. Knowing that the costs of the distribution system were incurred
so that customers can receive energy and meet their peak loads throughout the year
simply points out that these investments were made to sell customers electricity and the
costs associated with serving this market should be met, as most other businesses
meet them , through usage charges, not per customer charges.
Has the Commission in the past commented on the conceptual weakness
of IPC's efforts to categorize a significant part of the distribution system as "customer
Power: Direct
AARP
P. 19
costs
Yes, 17 years ago in the 1987 IPC general rate case (U-1006-265A), the
Commission s order (Order No. 1984) commented as follows:
... (T)he parties attempted to break down distribution-related costs into demand-related
energy-related and customer-related costs... .The division of costs into these three
components is weakest and least relevant for distribution plant. Simply put, distribution-
related costs do not neatly fall into any of these categories to the exclusion of the
others. Formulas and conventions separating them are nothing more than that--
formulas and conventions. The W12CPs reasonably partition these costs among the
customer classes even though their attempt to force costs into demand-related , energy-
related or customer-related components is largely meaningless." (Section 119B)(3).
IPC's efforts to describe over 36 percent of the distribution system costs as
customer costs" are still "largely meaningless.
What is your recommendation on the allocation of distribution costs?
Ideally, they should be allocated on the basis of a mix of energy and
demand. As a practical matter they can be appropriately allocated on a demand basis.
IPC's per customer allocation of a significant part of the costs of the distribution system
should be rejected.
Following the logic in your discussion above of the allocation of the costs
associated with the distribution system, could not the same argument be made for the
allocation of transmission costs on the basis of a mix of energy and demand?
Absolutely. In the last two IPC general rate case I have made exactly that
argument. I still feel that a careful analysis of the cost-causal logic those costs provides
Power: Direct
AARP
a strong technical justification for allocating transmission in a manner similar to the
current allocation of production plant. Appendix A to this testimony presents that
analysis again.
Why have you put that material to an Appendix?
Although I still find that analysis technically correct, this Commission has
not found it persuasive in previous rate cases. In order to focus on other issues of
importance to residential customers, I have chosen not to emphasize transmission
allocation.
Why then did you include Appendix A at all?
As I say at the beginning of Appendix A, I expect other parties to be trying
to persuade the Commission to change the cost of service allocations in ways that may
burden residential customers. The point of the Appendix is to remind the Commission
that persuasive arguments can be made to shift allocations in a direction that would be
more beneficial to residential customers.
4. The Burden of the Irriaation Class on Other Customers
What is your concern about the way in which IPC proposes to handle the
revenue deficiency associated with the irrigation class?
IPC's cost of service analysis indicates that the irrigation class should face
a $40.5 million increase in rates. Almost 50 percent of the $85.6 million total rate
increase that IPC seeks should be coming from the irrigation class. IPC , however
proposes to collect only $15.1 million of their $40.5 million revenue deficiency from the
irrigators. The rest of the increase the cost of service study indicates $hould come from
Power: Direct
AARP
P. 21
that class would come, instead, from other customers.
How would this affect the residential class?
About half of the irrigation revenue deficiency that is not collected from the
irrigation class would be collected from the residential class, an additional $12.1 million.
This $12.1 million increase to residential customers associated with the irrigation class
revenue deficiency would be on top of the $28.7 million increase IPC's cost of service
study indicates is due from the residential class. Thus, IPC's proposed revenue spread
boosts the rate increase to residential customers by over 42 percent. Instead of facing
a 13.5 percent rate increase , residential customers would face a 19 percent increase.
This appears to be a particularly extreme additional burden to put on residential
customers.
Is this a new problem with respect to the irrigation class?
No. Twenty years ago, in 1982, IPC's cost of service study indicated that
irrigation rates should have been increased 60 percent or $25 million. In 2003 dollars
this would be a $47.75 million increase. The current IPC cost of service study indicates
that gap is still $40.5 million and that a 67.1 percent increase is still required.
In 1982 IPC was seeking a larger system increase (26.3 percent) than it is
seeking in the current rate case (17.7 percent). If the irrigation deficit is expressed in
terms of the increase required over and above the system-wide increase sought, the
deficiency gap has grown from 47.5 to 57 percent of irrigation rates.
It would appear that some fairly dramatic steps need to be taken to
systematically close this very large gap. If only a small step is taken and IPC does not
come back in with another general rate case for ten years, this substantial burden on
Power: Direct
AARP
residential and other customers could continue for another 20 years.
Do you believe that the cost of service analysis should be used
mechanically to determine each class s revenue requirements?
Certainly not. I agree with this Commission s long-standing position that
cost of service analysis is just the starting point for determining class revenue
requirements. Reflecting cost responsibility is one important objective in rate design , but
customer impact and rate stability are also important considerations. However, this
Commission has usually tried to move systematically to close very large revenue
deficiencies rather than let them continue or grow. My concern is with the size of the
burden on residential customers and the likelihood that it could continue indefinitely.
Can the burden being carried by residential and other customers
associated with the low rates irrigators pay be justified in economic development terms?
By keeping irrigation rates low, is the agricultural sector stimulated and the rest of the
economy encouraged to expand, benefiting all workers and businesses?
No. Cross-subsidization of one type of business by all residences and
other businesses cannot have that sort of effect. "Robbing Peter to pay Paul" is not a
viable economic development strategy. Shifting purchasing power from residential
customers to irrigators reduces the expenditures the households can make in local
businesses, depressing the local economy. In addition, all other businesses are
burdened by higher costs so that one type of business, the irrigators , can have lower
costs. That discourages business expansion across the board , depressing the
economy, not stimulating it.
But in rural areas, where irrigated agriculture is a more dominant part of
Power: Direct
AARP
the local economy, don t lower irrigation costs stimulate the local economy?
That is unlikely. "Rural" no longer means "agriculturaL" Most rural
residents are not irrigators. Most rural businesses are not irrigated farms. The rural
economy is increasingly diverse and non-agricultural. The economic connection
between the rural economy and agriculture has increasingly reversed so that the
diversity in the non-agricultural local economy, including the urban economies within
commuting distance supports farm families rather than the other way around. Farm and
ranch families increasingly supplement their household income and stay engaged in
agriculture by taking jobs in the non-agricultural sectors of the surrounding economy. In
that sense, promoting the non-agricultural economy is crucial to the survival of family
farms and ranches.
Is it likely that the problem of the revenue deficiency of the irrigation class
will solve itself over time?
No. The relative importance of the summer peak seems very likely to
continue to grow on IPC's system. As IPC has emphasized in its testimony, serving this
growing peak summer load is going to require large investments in production
transmission, and distribution that are focused on the summer load. As IPC has pointed
out, the incremental costs are going to be larger than the embedded costs as gas-fired
peaking units are used to supplement the relatively inexpensive hydroelectric power and
base-load coal-fired generation. Market prices of electricity at time of peak summer
demand are also likely to be high since the region to the south and west is also
summer-peaking. When the costs associated with upgrading generation, transmission
and distribution to continue to meet the summer peak are added to future cost of service
Power: Direct
AARP
analyses, the cost assignments to the irrigation class are certain to continue to rise.
What do you recommend the Commission do about the irrigation class
revenue deficiency?
Since I do not know what this Commission will determine the ultimate IPC
system-wide revenue deficiency is, let me phrase the revenue deficiency in terms of the
gap relative to the overall system deficiency. That gap is 42 percent. IPC proposes a
percent rate increase for the irrigation class. That would be a 6.2 percent increase
relative to the overall increase. If IPC has general rate cases every five years , as
opposed to the 10 years since the last general rate case, it would take 35 years to close
the gap. That would be a much faster rate of closure than the zero closure over the last
20 years. But such a rate of closure would burden all other CU$tomers and the south
Idaho economy for a third of a century. I do not believe that is reasonable.
I would propose that the tariff approved in this case set an irrigation rate that
adjusts upward annually over the next five years. Since any pragmatic person would
have to admit that cost of service analyses have a wide range of uncertainty and error
associated with them, let us assume that 1PC's cost of service study is accurate to only
within plus or minus 33%. That would suggest that we could only be confident that the
irrigation class revenue increase over and above the system increase should be
between somewhere in the 28 to 56 percent range. If we choose the lower bound to be
conservative, the increase to irrigators over and above whatever the overall increase is
should ultimately be 28 percent. IPC proposes to increase irrigators' rates 6.2 percent
on top of the system-wide increase. That would leave a 20.5 percent deficit if the target
is an ultimate 28 percent increase for irrigators.
Power: Direct
AARP
If the tariff indicated an irrigation rate that increased 4 percent each year over the
next five years and was accompanied by an accounting order that allowed these
increased revenues to accumulate and be returned to all other classes of customers at
the time of the annual power cost adjustment, the gap in irrigation rates could be
systematically closed in a reasonable period of time.
The point is to find a way to systematically close a significant part of this gap in a
way that is neither sudden nor disruptive but assures that a good part of it closed in
the near future.
III. Rate Design Issues
IPC's objectives in this case for the residential class is to move more of
the costs to the customer charge and keep the only usage charge the residential
customers face, the kilowatt hour charge, relatively low. Does that make sense from the
point of view of public energy policy?
No. As IPC's President and Chief Executive Officer has pointed out
incremental costs exceed embedded costs on the IPC system and with ongoing growth
that is driving costs and rates upward. In that setting, rather than scaling back the rates
that can encourage consumers to modify their electricity usage patterns, it is more
important than ever to accurately convey to customers the costs associated with their
electricity use in a way that encourages them to modify that usage.
The point is not to penalize unavoidable electricity use but to provide an
economic incentive that encourages electricity users to look for cost effective ways of
Power: Direct
AARP
reducing their usage, thus helping their families and IPC to avoid the high and rising
incremental costs.
In what way is shifting more of IPC's costs to fixed monthly charges
incompatible with improving price signals?
Shifting more costs to fixed monthly charges is appropriate only if one
thinks that current usage charges are too high and are irrationally discouraging the use
of electricity. If one were to believe that encouraging increased electricity use is good
public policy, then it would be appropriate to adopt IPC's rate design proposals since
IPC explicitly does propose changing rate design so that energy charges can be kept
lower.
Aren t you distorting what IPC is proposing? IPC is shifting costs to fixed
monthly charges for the residential and small commercial class only because, as it says
there is no other way to collect these fixed costs.
IPC is mistaken in that assertion. As discussed above, there is no principle
whatsoever, either economic or business , that says that fixed costs should be collected
in fixed charges. It is true that residential and small commercial customers are not
demand-metered because such meters at this point are not cost effective. But that does
not mean that usage-related costs such as demand costs cannot be conveyed to them.
There is a high correlation between peak demand and energy usage. As one
these moves up or down, the other tends to follow. The correlation , of course, is not
perfect, but it is high. For instance , the correlation between the highest monthly
coincident peak and average load on the IPC system over the last 12 years has been
70 percent. (IPC Response to AARP Request 14) More relevant to the residential class
Power: Direct
AARP
for the test year the correlation between monthly coincident demands and energy usage
was 81 percent.
This high correlation between energy use and peak demand indicates that IPC is
not helpless in conveying to customers the costs associated with peak demands.
Conveying the information through high energy charges at the time of peak demand can
be effective. That certainly would be more effective than reducinQ usage charges while
boosting charges that are unrelated to usage. The latter strategy can do absolutely
nothing to convey information about the high costs of peak demand use; it can only
undermine the communication of that crucial information.
Do you support IPC's proposals to increase demand charges to better
reflect the costs associated with peak demand?
Yes. For those customers that are demand metered, adjusting demand
and energy charges to better reflect the relative importance of those two types of usage
makes sense.
Do you support IPC's proposals to make use of seasonal and time-of-use
pricing?
Yes. that is one way to convey crucial information about how the costs
associated with electrical use varies over the day and over the year.
Is the type of rate design that IPC proposes for the residential class the
only way to convey information about the high marginal costs associated with electricity
use?
No. As pointed out above, IPC's proposal to increase the fixed monthly
customer charge actually tends to reduce usage charges and thus encourage electric
Power: Direct
AARP
usage. IPC has said that that is a conscious part of its rate design. From a public policy
point of view that makes no sense.
Even IPC's proposed seasonal rates for residential customers are problematic.
Residential loads peak during the winter. The residential peak load during the summer
is 16 percent below its winter peak and the energy usage in the peak summer month is
30 percent below the highest winter month usage. Not surprising, marginal supply costs
are high not only during the summer but also during the winter. Two of the winter
months (along with three of the summer months) are part of the Weighted 12 Coincident
Peak. Marginal energy costs are as high or higher in December and January as they
are in June, one of the peak summer months.
IPC's proposals for seasonal rates ignore the impact of winter peak loads.
Combined with the impact of shifts in costs to fixed customer chargers, energy rates
during the winter peak periods are reduced in relative terms. Encouraging electricity use
during an important peak period by setting the rates low is not a good idea.
Is IPC actually proposing to lower residential winter rates?
That will depend on the overall rate increase this Commission authorizes.
If we use IPC's full revenue request as a reference point, the existing rate structure (a
$2.51 monthly service charge and a non-seasonal rate) would lead to a kwh charge of
92 cents for the residential class. For the winter heating season, however, IPC
proposes a 4.91 cent charge.
How could this be avoided?
Instead of exclusively using higher summer rates and fixed charges and
effectively reducing winter rates blocked rates could be used for the residential class.
Power: Direct
AARP
Blocked rates that offered a modest initial block at a low rate but then charged a higher
rate for consumption above that level would automatically be seasonal rates. In all
seasons of the year when residential consumption increased , more consumption would
be billed at the higher tail-block rate. In that sense, the higher marginal cost would be
conveyed during both summer and winter peak seasons.
Could you provide an example tied to the residential revenue requirement
in this case?
Yes. In the following example I have used IPC's residential revenue
requirement for illustrative purposes even though I do not support either the allocation of
costs to the residential class nor the overall revenue increase IPC is seeking.
An initial 400 kwh per month block of electricity could be priced at a rate 25
percent lower than the price for consumption above that level. In the following rate
design I have increased the monthly customer charge from $2.51 to $3.00. I will discuss
this change later. With that monthly customer charge the appropriate rates for the initial
and tail block would be:
Customer Charge:$3.
Initial 600 kwh block:834 cents per kwh
All other kwh:6.445 cents per kwh
This rate design can be compared with that proposed by IPC:
Customer Charge:$10.
Non-Summer Rate:910 cents per kwh
138 cents per kwhNon-Summer Rate:
Note that the tail-block rate I propose is almost the same as (5 percent higher than) the
Power: Direct
AARP
summer rate proposed by IPC. In that sense, it gives the same marginal supply cost
information during both summer and winter peaks that IPC proposes only for the
summer period.
In addition to conveying the high costs of electric consumption during both
the summer and winter period, what other advantages are there to such a blocked rate?
Such a rate provides rate relief to customers with modest levels of
consumption. In effect, it provides a modest block of electricity at a reduced rate.
Customers whose consumption stays within or close to this block face lower rates.
Those whose consumption stretches beyond that block pay higher rates. The further
beyond the initial block consumption goes, the greater the impact on the customer s bill.
This will provide some rate relief for smaller households and those who cannot
afford larger homes and more electric using appliances. I would expect it would benefit
for instance, senior citizens and many low-income households.
Why do you mention low-income households? Do not some of them have
large families living in relatively low quality structures that use large quantities of
electricity?
Some low-income households certainly can be characterized in that way.
But, in general, low and fixed income households consume less electricity, just as they
consume less of most other goods. They live in smaller homes and apartments; they
have a smaller stock of electric-using appliances; and they have to "pinch their pennies
more.
Over the last three decades I have reviewed dozens of studies on the link
between income and electric consumption. I have yet to find a study showing that lower
Power: Direct
AARP
P. 31
income households, in general , consumed more electricity than higher income
households. This includes older, 1981-1983 vintage , analyses of income-electric
consumption patterns on IPC's system and in Idaho as a whole. Disproportionately
raising the bills of smaller residential customers will place an increased burden on those
least able to pay for basic utility service: those with low and fixed incomes. This is
important and relevant information when considering the impact of rates on customers.
But do not the consumption patterns of IPC's LlEAP customers indicate
high levels of consumption by low-income households?
The IPC LlEAP customer consumption data provided in response to
Commission Staff Request 40 does indicate high levels of electric consumption for the
September 2002-May 2003 period. Several things should be kept in mind when
drawing inferences from this LlEAP data for all low and fixed income households.
First, LlEAP customers represent both the lowest income customers IPC has and
are likely to be the highest energy consuming low income households. LlEAP
customers, until recently, had to have incomes below 125 percent of poverty. The self-
selection process for the LlEAP program that provides support in paying utility bills is
likely to attract those low-income households with the highest utility bills, i.e. those with
unusually high electric usage. In that sense the LlEAP households may not be a good
sample of all low and fixed income households.
Second, at least in Montana, LlEAP program participants are a relatively small
sample of the LlEAP-qualified population. In Montana only 25 to 30 percent of LlEAP-
qualified households seek utility bill relief through LlEAP. Despite more than a decade
of efforts to boost the participation rate through utility and private outreach efforts, the
Power: Direct
AARP
participation rate has not increased.
Third, most low income and fixed income households do not think of themselves
as living in poverty and/or do not meet the federal guidelines to be so classified. The low
income population stretches well beyond the federal poverty classification.
The IPC LlEAP customer consumption data should be taken seriously. The high
electric consumption levels it documents highlights the need for expanded low-income
weatherization and bill payment assistance programs to do something permanent about
these high levels of energy consumption.
At the same time , until there is a lot more study that confirms that IPC's low and
fixed income households contradict decades of evidence that low and fixed income
households use less energy than high income households, this data should not be used
against a blocked rate design that allows multiple public policy objectives to be pursued
simultaneously.
In your proposed rate design above, you have raised the monthly
customer charge from $2.51 to $3.00. Why did you propose that?
Given that I believe that it is far more important to have usage charges
that reflect the expected higher incremental costs of the present and future, I do not
really believe that there is any reason to raise the customer charge. However, the
customer charge has not been increased for 10 years and a case could be made to
increase it to reflect some of the effects of inflation.
In addition, the appropriate definition of customer costs is those costs that could
be avoided if the customer ceased to take service. That would include those meter
reading and billing functions that actually vary with the number of customers. Some of
Power: Direct
AARP
the meter reading and billing costs are relatively fixed. A meter reader has to walk by a
residence whether or not the customer is taking service and there are fixed overhead
costs associated with the customer accounting function. A $3.00 a month charge would
cover about 70 percent of the meter reading and customer account costs that IPC
calculates. That would appear to be a reasonable reflection of the variable part of those
costs. Given that there is no urgency to carefully convey such customer costs since they
do not affect customer behavior and by lowering usage charges may encourage
inefficient customer behavior, there is no need to get the monthly customer charge
exactly right.
What would be the impact of your rate design proposals compared to the
rate design that is currently in place but adjusted to meet the revenue requirement IPC
seeks in this case?
That is indicated in the following table. Across the entire year, about two-
thirds of all residential bills would be lower, all those bills less than 1 126 kwh per
month.
Power: Direct
AARP
Comparison of Current IPC Rate Design and AARP Proposal
Consumption Current IPC AARP Difference
Level Rate Design Proposal AARP-Current IPC
(kwh/mo)($/mo)($/mo)($/mo)
200 $14.$12.$1.
400 $26.$22.$3.
600 $38.$35.23 $2.
700 $43.$41.$2.
800 $49.$48.$1.
000 $61.$61.$0.
200 $73.$73.$0.
400 $85.$86.$1.46
600 $97.$99.$2.
800 $109.$112.$3.
000 $120.$125.46 $4.
500 $150.41 $157.$7.28
000 $179.$189.$9.
500 $209.$222.$12.
000 $239.$254.$15.
500 $268.$286.$17.
000 $298.$318.$20.
What would be the impact of your rate design proposals compared to the
rate design proposed by IPC?
Again , accepting for illustrative purposes IPC's overall revenue target and
its allocation of costs to the residential class, the next table compares the relative
impacts of AARP's and IPC's rate design proposals on customers with different levels of
consumption.
During the non-summer period a little more than half of residential bills would be
lower and a little less than half would be higher. The lower the level of electric
consumption , the more beneficial the rate design is to a residential customer. It is the
upper half of residential consumers, those with monthly non-summer consumption
above 876 kwh who would face higher bills. Those customers are responsible for three-
quarters of the residential non-summer energy use.
During the winter heating season , November through March , slightly less than
Power: Direct
AARP
half of the bills would be lower under this rate design.
During the summer, almost all residential bills would be lower even though all
customers who consume more than 400 kwh per month would face tail-block rates
higher than those proposed by IPC. Eighty percent of summer bills containing 96
percent of residential consumption would face such higher rates.
Comparison of IPC and AARP Rate Design Proposals
Consumption IPC Proposal AARP Proposal Difference: AARP-IPC
Level Summer Non-Summer Summer Non-Summer Summer Non-Summer
(kwh/mo)($/mo)($/mo)($/mo)($/mo)($/mo)($/mo)
200 $22.$19.$12.$12.$9.$7.
400 $34.$29.$22.$22.$12.$7.
600 $46.$39.46 $35.$35.23 $11.$4.23
700 $52.$44.$41.$41.$11.$2.
800 $59.$49.$48.$48.$10.$1.
000 $71.$59.$61.$61.$10.$1.
200 $83.$68.$73.$73.$9.$4.
400 $95.$78.$86.$86.$9.$8.
600 $108.$88.$99.$99.$8.$11.
800 $120.48 $98.$112.$112.$7.$14.
000 $132.$108.$125.46 $125.46 $7.$17.
500 $163.44 $132.$157.$157.$5.$24.
000 $194.$157.$189.$189.$4.$32.
500 $224.$181.$222.$222.$2.$40.29
000 $255.$206.40 $254.$254.$1.$47.
500 $286.$230.$286.$286.$0.40 $55.
000 $316.$255.$318.$318.$1.$63.
Are you not afraid that the lower rates for the initial block will encourage
additional consumption during peak periods?
No. Only 4 percent of residential energy is sold in the 18 percent of bills
that stay within the 400 kwh block. That is , almost 96 percent of the residential energy
sales are on bills where the customer faces the higher tail-block rate. In that sense, the
vast majority of consumption is given the right price signal.
Power: Direct
AARP
Does this conclude your prefiled direct testimony?
Yes, it does.
Power: Direct
AARP
Appendix A
The Unavoidable Uncertainty in Cost of Service Analysis and the Case for Allocatina a
Portion of the Transmission and Distribution Systems on the Basis of Eneray Delivery
Why do you propose to discuss the indeterminacy that is unavoidable in
cost of service analysis?
I expect the witnesses for various customer classes to propose
approaches to cost of service analysis that deviate significantly from both IPC's study
and the principles used by this Commission over the last two decades. I simply want to
make clear that there are similar changes that could be made that would favor the
residential class. It is important for the Commission to keep that in mind when
evaluating other cost of service proposals that would burden the residential class with a
greater share of the costs.
Has this Commission recognized the uncertainty associated with cost of
service analysis in the past?
Yes. Seventeen years ago in IPC's 1987 general rate case (U-1006-
265A), the Commission made the following comments in its order (No. 1984) on the
limitations but usefulness of cost of service analyses:
No cost-of-service study presented irt this case accurately defines Idaho Power s costs
of serving its customers. All cost-of-service studies suffer from common defects. They
attempt to reduce a dynamic system to a static one for purposes of study and thereby
omit important considerations, and their results vary according to their originators
subjective assumptions underlying their objective arithmetic. We do not belittle the
Power: Direct
AARP
value of cost-of-service studies for rate setting purposes. But the limitations of the
studies should be stated so that the results can be used with an awareness of their
limitations....
As in (U-1 006)-185, we recognize the subjective assumptions underlying all cost-of-
service studies and the inevitable errors in any undertaking of such magnitude.
However, these studies contain information that is useful in the allocation of revenues
among customer classes. We recognize, however, that allocation of revenues among
customer classes is not entirely a mathematical or scientific process. Cost-of-service
studies provide a useful starting point for allocating revenues , but in the end we must
and do, consider other factors such as rate continuity, equity and proportionality"
(Section II(A))
analysis?
What is the primary problem that leads to controversy in cost of service
The allocation of fixed costs among various classes of customers is the
major problem. When costs are readily seen to YMY with particular usage
characteristics, as for instance with energy production or the mailing of a customer s bill
there is little controversy. Those costs are assigned on the basis of the variable to which
they are directly tied. Unfortunately, however, most utility costs are not variable costs in
that sense. Rather, they are fixed costs that were incurred in the pursuit of multiple
objectives and were affected by multiple forces. Dividing those fixed costs among
customers on some cost causal basis tied to customer electric usage is very difficult.
Is this a new problem?
Power: Direct
AARP
Certainly not. From the point of view of economic analysis, it has been
around for as long as formal economic analysis has been around. For most of this
century students studying economics have worked through the exercise of proving that
there was determinant way of allocating fixed joint costs among joint products based
upon cost analysis alone. It cannot be done. That is the reality that this Commission and
any cost of service analysis must face. Some non-cost consideration, usually a set of
policy judgements or marketing considerations , will have to be called upon to make that
division of the fixed costs among customers.
Does knowing that most of IPC's system has to be "designed to meet
peak loads" provide the basis for concluding how these fixed costs should be allocated?
That is, can we use this basic "engineering fact" to conclude those fixed costs are
peak-load-related" and proceed to allocate them on the basis of peak loads?
Absolutely not. To say that the utility has to design its system to meet
peak loads provides us with almost no economic-engineering information. That
statement has the same standing and conveys the same information as a statement
that IPC has to design its system to withstand the forces of gravity. The latter is certainly
a correct engineering statement. A cost analyst could proceed on that basis to allocate
every customer an equal share of fixed costs because the force of gravity is
approximately the same at every customer s location. Similar statements could be made
about withstanding expected thermal extremes, solar radiation, wind speeds, etc. Such
true" statements about the engineering design of IPC's system would get us nowhere
with respect to rational cost allocation among customer classes. The same is true of
slogans about the system having to be designed to meet peak demands.
Power: Direct
AARP
Do most parties and this Commission already admit that knowledge that
the system has to meet peak loads does not tell us how to allocate the fixed costs
associated with that system?
Yes. When it comes to the allocation of generation, transmission , and
distribution costs, neither this Commission nor IPC accept that "slogan" about meeting
peak loads to dictate how the fixed costs are to be allocated. We know that fixed
generation costs are not all capacity related. Since, as already mentioned , it is a
physical fact that burning fuel in an open field will not produce useable electric energy,
we know that fixed costs have to be incurred in order to obtain electric energy. Most
accept that some substantial part of fixed generation costs are energy-related , not
capacity-related. Similarly with transmission , transmission lines can substitute for unit
coal trains carrying fuel to a generating facility or natural gas transmission charges by
allowing the facility to be located closer to the coal or natural gas source and shipping
the electric energy to load centers. It is for that reason that many commissions classify a
substantial part of the transmission system, that associated with remotely sited plants
as part of "power supply." In some jurisdictions this leads a significant part of
transmission to be classified as energy-related.
We know that customer services and meters must be sized to meet the
customer s peak loads. Yet it is rare for anyone to suggest that these costs be allocated
on the basis of peak demands. They are usually labeled "customer" costs. IPC does not
even classify all of the distribution system poles, lines , and transformers as "peak-load
related." It includes some of them in the customer cost category.
The point is that the mere statement that part of the electric system has to
Power: Direct
AARP
be sized to meet peak loads does not provide a plausible basis for concluding that all of
those costs should be allocated solely on a proportional basis with peak loads.
On what other basis can costs be allocated if this general "all or nothing
approach is not taken?
On the same basis that has been used for generation , transmission , and
distribution costs discussed above: on the basis of cost analysis rather than cost of
service slogans. The cost analyst has to look at more that one or two pre-determined
design criteria and make a sweeping judgment. The cost analyst has to look at the role
that each criterion played in determining the total fixed cost investment and what the
quantitative relationship was between that particular planning criteria and costs. When
one goes beyond the slogan and looks at the multiple functions each investment is
pursuing and the quantitative role each planning criteria plays in determining costs, a
quite different engineering-economic relationship is likely to emerge.
Can one expect all of the utility's fixed costs to conveniently fall into one of
the pre-determined "energy,
" "
capacity," and "customer" categories?
No. Some costs will have to be labeled "none Qf the above" and a
judgment will have to be made about how to distribute responsibility for them.
This is implicitly recognized by many cost analysts but is usually hidden behind
the technocratic mumbo-jumbo that surrounds most cost of service analysis. For
instance, the reason that IPC classifies 36.25 percent of distribution costs to the
customer" category is not that it found that these costs "vary with the number of
customers" but because it determined that those costs did not vary with energy and
peak loads. Since those costs also do not vary with the number of customers, the
Power: Direct
AARP
logical conclusion should have been that they were "none of the above." Instead of
coming to that conclusion , IPC made an argument based on the need to collect fixed
costs on the basis of fixed charges and used the "customer cost" category as a "none of
the above" residual category and dumped those substantial distribution costs there.
Similar things are often done with generation costs. Since the variable fuel and
O&M costs are easily seen as "energy-related " they are classified in that way and then
the "capacity" category is used as a residual for all other generation costs. "Capacity
cost" comes to mean "not-variable-energy cost." This allows the analyst to hide the fact
that she or he does not really have a rational basis for allocating many of the- costs
because they are really "none of the above.
You paint a pretty indeterminant picture when it comes to cost allocation.
Does that leave this Commission with almost nothing to work with in making these cost
allocation decisions?
No. It simply means that there is no correct formula that the Commission
can use. If there were, of course, there would be no need for any decision-making by
the Commission on this important issue. If cost of service and revenue allocation was a
scientific, technocratic problem with a determinant solution, we would leave this issue to
anonymous engineers, accountants, and economists. Because it is not, a commission
appointed by the Governor through the political process is asked to make those
decisions. This makes sense. Someone s interests or policy objectives are necessarily
going to guide these decisions. The point of public regulation is to allow the Commission
to see-that it is public policy objectives that govern these unavoidable judgments.
Can you provide an example of the limits of simple criteria that are
Power: Direct
AARP
presented as engineering design criteria in specifying how costs should be allocated
among customers?
Yes. Let me use an example I have presented to this Commission before:
the relationship between the peak load capacity of transmission and distribution lines
and the cost of those lines.
It is usually taken as a truism that since electric lines, substations, and
transformers are sized to meet peak loads , it is only logical to allocate them on the basis
of peak demand. Any other allocation would appear to violate the design criteria that led
those costs to be incurred. What is wrong with that conclusion?
It assumes two things: First, that that electric delivery system only
provides peak load services and, second , that the costs of electric delivery system rise
proportionately with peak load. We know that neither of these assumptions is correct. I
discussed the first assumption earlier. Here I want to focus on the second assumption.
Do the installed costs of transmission lines vary proportionately with the
size of the peak load they are designed to handle?
Not at all. There are significant economies of scale in the construction of
transmission and distribution lines that at the time of construction allow the capacity of a
line to be increased without a proportional increase in costs. As a result, the cost per
unit of capacity declines dramatically for higher capacity lines.
Does IPC data on its embedded transmission costs confirm this general
proposition?
Yes. The use of historical accounting costs for transmission lines
constructed at various points in the past is not really the appropriate data to use since it
Power: Direct
AARP
mixes dollars of different purchasing power together. More recently constructed lines
will be recorded at higher costs than older lines even though in inflation adjusted terms
that may not be the case. As a result, historical accounting cost data can be misleading.
IPC reported transmission line costs and estimated capacities by line voltage in
response to AARP Data Requests Nos. 4 and 5. For the two transmission line voltages
for which IPC reports considerable mileage of transmission line, 69 kv (1 170 miles) and
230 kv (971 miles), there is clear evidence of capacity growing faster than costs. The
cost per mile for the 69 kv lines was $22 000. The cost per mile for the 230 kv lines was
$87 000, about four times as high. But the capacity of an electric line increases with the
square of the voltage, implying that the capacity of the 230 kv line would be over
times higher than the capacity of the 69 kv line. As a result, the cost per unit of capacity
on a 230 kv line is only about a third (35.5 percent) that of a 69 kv line. IPC data on the
capacity of typical 69 kv and 230 kv lines partially confirms this. The minimum capacity
it reports for a 69 kv line is 20 mva while that of a 230 kv line is 300 kva , 15 times larger
even though the cost is only about 4 times larger. IPC, however, reports a range of
actual capacities for lines at a given voltage, depending on the conductor size used. Of
course the historical data reported also involves lines with different conductor sizes all
averaged together. This makes it difficult to see a clear relationship between voltage
level , cost per mile, and line capacity.
Have you analyzed this relationship between electric line capacity and
costs previously?
Yes. My testimony before this Commission in IPC's last general rate case
analyzed that relationship. It was also the subject of an article of mine published in the
Power: Direct
AARP
Public Utility Fortnightly ("Making Sense of Peak Load Cost allocations,March 15
1995, pp. 43-46). This part of my testimony summarizes that past analysis.
Based on your past analysis , what is the theoretical relationship between
the capacity of an electric line and its cost?
For illustrative purposes the table below presents data on transmission
costs and capacity developed by Montana Power Company transmission engineer
Gene Braun in 1994 for the Montana Power Marginal Cost Collaborative Group of which
I was a member. The dollar values, of course , are now out of date, but the basic
relationship is unlikely to have changed.
~oltage Cost per Relative Relative Relative
Cost permileCostCapacity
59 kv $90 000
115kv $100,000 1 .0.40
1 61 kv ~125 000 5.4
230 kv $150 000
This is the same information presented in my 1994 testimony in IPC'last
general rate case.
Note that considerable additional transmission capacity can be purchased at
costs far below the average cost. For costs 11 percent higher, one can obtain almost
200 percent more capacity when moving from a 69 to a 115 kv line. That represents an
incremental cost that is only 6 percent of the average cost at 69 kv. For costs 67
Power: Direct
AARP
percent higher, one can obtain 1 000 percent more transmission capacity by moving
from a 69 to a 230 kv line. That represents an incremental cost that is about 7 percent
of the average cost per kw for the 69 kv line. Clearly the design capacity of the
transmission line is not the only or the primary determinant of the cost of the lines.
Can you be more specific about how transmission line costs vary with
capacity as indicated in the above data?
Yes. As is clear from just a casual review of the above data, one way to
describe the relationship between transmission costs and capacity is that there is a
large fixed cost element to transmission costs. In the above data , about $86 000 of the
cost per mile appears to be fixed. That means that 60 to 85 percent of the costs do not
vary no matter what the capacity of the transmission line. Put slightly differently, at the
midpoint in this range of transmission facilities , the design load is responsible for only
28 percent of the costs of the transmission line. The other 72 percent of the cost of the
transmission facility is not tied to the design load. That is not surprising. Many of the
costs associated with right-of-way, engineering, and construction are the same
regardless of the capacity of the conductor being hung on the poles. Only a small part of
those costs vary directly with the capacity of the line. That means that a sizeable part of
the cost of the line do not vary with capacity and cannot, if we seek to allocate costs on
the basis of actual causality, be labeled "peak-load-related." With 60 to 85 percent of
the costs of transmission facilities in this category of "not-peak-Ioad-related " clearly
some other allocation method is called for.
An alternative way of describing this relationship between transmission
costs and transmission capacity that does not assume large fixed costs that do not vary
Power: Direct
AARP
with capacity is that the relationship is non-linear with the costs varying with the fifth root
of the capacity to be served. Very substantial changes in capacity lead to only modest
changes in cost. Again, if we wish to allocate costs on the basis of actual cost-usage
relationships, that part of the transmission costs that vary with peak load should not be
allocated in proportion to peak load but in proportion to the fifth root of peak load.
How did you derive the quantitative relationship that you have just
described?
I used a linear regression analysis of the data cited above to derive an
equation to explain the relationship. If one assumes that a linear relationship is
appropriate , a regression of costs on relative capacity will identify the component of cost
that does not vary with the design capacity of the line. For the above data, that is about
$86 000 per mile. The equation that resulted was the following:
Cost = $85 700 + $6 017 * Relative Capacity
Alternatively, if one rejects a large fixed component, an exponential function can
be fit to the data. This was done by regressing the log of the costs on the log of the
relative capacity. The equation that resulted was the following:
Cost = $86 287 * (Relative Capacity).
Both regressions provide a good fit for the data. The R2 for the linear
equation was 0.97 and for the exponential relationship it was 0.95.
What does this economic-engineering relationship between costs and
transmission capacity tell us about the appropriate allocation of transmission costs?
It tells us that a substantial part of the costs are not "peak-load related"
and have to be treated as "none-of-the-above." Alternatively, we have to use an
Power: Direct
AARP
allocator that deviates substantially from the proportional one used by IPC. This is not a
minor issue. If one uses a proportionality assumption to describe the relationship
between transmission costs and design capacity instead of one of these two empirically-
based descriptions, one misstates the impact of additional loads on costs by as much
as a factor of four, an error of 300 percent!
Do these results reinforce the implications of the multiple functions of the
transmission system?
Yes. The purpose of the transmission system, the purpose for which it was
designed and built, was not to just deliver electricity at 12 peak hours per year. It never
would have been built if that were its only purpose. The transmission system s primary
business logic is a year-around one, not a peak hour one. It provides valuable services
throughout the year by delivering lower cost energy and more reliable energy. That was
what caused it to be built. The peak hour service is a relatively minor part of its function
and one that has relatively modest impacts on the system s costs. The part of the costs
allocated on the basis of peak hour loads should also be appropriately modest.
Does any transmission line have to be built to meet peak hour loads?
No. On-site generation is always conceptually a substitute for transmission
and , increasingly, is also a practical substitute. The location of additional generation on
the IPC system, for instance, is largely being dictated by the desire to avoid the costs
associated with major transmission upgrades. Generation is often used to reinforce the
transmission system rather than simply upgrading that transmission system. Given the
tradeoffs between generation and transmission and the recognized energy component
of fixed generation costs, it is clear that transmission should not be allocated solely on a
Power: Direct
AARP
peak hour basis. It is this obvious relationship between transmission and generation that
often leads transmission associated with remotely sited plants to be treated as if it were
generation. But the same considerations come into play for almost every transmission
line since the alternative to that line is generation facilities at the load center.
Do utility transmission planners recognize the very inexpensive nature of
additional peak load transmission capacity?
Yes. It is standard practice to built substantial excess capacity into
transmission systems. Often the intent is to build the line so that it can handle expected
future growth so that the line will not have to be replaced or rebuilt during its useful
lifetime. Such excess capacity would not be acceptable to either the utility or regulators
were it not for the fact that that additional capacity is relatively inexpensive to obtain.
Given the recognized low cbst associated with meeting peak loads when planning the
transmission system , it does not make sense to act as if those costs are high when it
comes to allocating those transmission costs.
This situation is to be contrasted with generation. Even though there may be
some limited economies of scale in generation, the utility is expected to avoid excess
generating capacity by planning the size and timing of additions to match expected load
growth. Utilities are not supposed to purposely build capacity years ahea~i: of time
because the cost associated with that extra capacity is too high. Yet often just the
opposite is true of transmission and distribution planning where the costs of additional,
capacity are so cheap, it is considered uneconomic not to build ahead of time to serve
future loads.
What does this empirical information suggest about how transmission
Power: Direct
AARP
costs should be allocated?
If we were to use the fifth-root relationship to allocate all transmission
costs, transmission allocations would not vary much between customer classes
because a class that had peak usage that was twice another classes would only get
allocated about 15 percent more transmission costs ( 20.2 = 1.15). Such an allocation
would significantly burden small classes with a sizeable share of the transmission costs.
This is not surprising. Since many of these costs are largely fixed , this type of allocation
would assign a large share of costs for simply being on the system and taking some
service. This is almost certainly unacceptable. It would be similar to what IPC has
proposed for the distribution system: charge a significant percentage of the costs of the
system to customers on a per customer basis. Just as this is not appropriate for the
distribution system it is not appropriate for the transmission system either.
An alternative would be to assign responsibility for annual average demand on
the transmission system on an average usage basis, and , for that part of the load above
average levels of demand , allocate responsibility on the basis of peak loads. Such an
approach would be based on the correct assumption that some size transmission
system would have to be constructed even if all customers had perfectly level loads
throughout the year. It is the loads in excess of that average usage that require a larger
sized system. This, of course, is how this Commission classifies generation: The load
factor is used to identify the part of generation that is to be allocated on the basis of
average usage (which is the same as energy usage) and the remaining load is allocated
on the basis of the weighted twelve coincident peaks.
If the fifth root relationship were applied to the peak-related portion of the costs
Power: Direct
AARP
P. 51
the allocation would be very close to an average demand or energy allocation because
of the compressing effect of the fifth root relationship. Although this would be a cost-
based approach tied to the technical relationship between transmission capacity and
costs, it deviates so much from traditional practice that it is probably unacceptable.
As an alternative , the fifth root relationship to the peak-related portion could be
ignored and the peak-related portion of the transmission load could instead be allocated
in proportion to peak loads as is now done for the entire transmission cost. This
approach is the equivalent to applying the same allocator to the transmission serving
IPC's service territory as is used for generation and often is used for transmission
associated with power supply. This gives recognition to the limited role played by peak
loads in determining transmission costs while collecting the transmission costs on the
basis of usage and services received.
This approach recognizes that the transmission cost relationship is one that
involves a large component of fixed costs that are appropriately classified as "none-of-
the-above." In that situation we need to make a policy judgment as to how these fixed
costs should be allocated. I would suggest that the way to do this would be to reflect in
the cost allocation the fact that the transmission system is used to deliver electric
services all year long, energy, reliability, stability, etc. On those grounds a significant
portion of the costs are appropriately allocated on the basis of average usage just as is
usually done with generation and generation-related transmission.
Does the relationship between electric line costs and line capacity also
apply to distribution lines?
Yes. Exactly the same technical , cost-based argument can be made for
Power: Direct
AARP
allocating distribution line costs in the same way: part on the basis of average demand
and part on the basis of peak demand. Of course with distribution plant, only those
customer classes who are served by particular types of distribution plant would be
included in the allocation. Customers who take service at the transmission level would
continue to pay none of the distribution costs and those taking service at the primary
level would pay none of the secondary distribution costs.
Your emphasis on the fact that electric lines involve substantial costs that
do not vary with the size of the load served sounds a lot like IPC's argument that there
is some level of distribution costs that does not vary with peak load. What is the
difference between your approach and lPG'
The difference lies in how IPC and I respond to the recognition that some
fixed costs do not vary with usage. IPC's response is that such costs must be "customer
costs." But that makes no more sense than labeling them energy costs or capacity
costs unless one can show that they vary with the number of customers, the definition of
a customer cost.
Do the costs of poles, conductors, and transformers vary primarily with the
number of customers?
No. This is why they are not customer costs. The number of poles, length
of conductor, number of transformers, etc. all depend on the density of customers, the
character of the terrain , the expected growth in load , etc., not the number of customers.
Compare, for example , a large apartment or condominium complex, with , say,
500 households in it. A single line to the apartment complex from the closest primary
circuit would be the only distribution network needed. Compare that to a suburban or
Power: Direct
AARP
exurban area also having 500 households where residences are located on two to ten
acre lots scattered amidst farm land. In that situation there would be miles and miles of
lines , hundreds and hundreds of poles, and a transformer for almost every household.
The cost of the distribution system is simply not directly related to the number of
customers. There is no cost causal justification for using that measure to allocate
distribution costs. It is not the cost-determining variable.
In a situation where costs cannot be unambiguously labeled "energy,
demand " or "customer " how can the cost allocation rationally proceed?
In that situation , where cost-causal analysis cannot guide the allocation
your policy objectives must be used. I suspect that lPG's guiding policy objective is to
stabilize its revenues so that cost recovery does not fluctuate as much as usage varies
with weather. This is not a matter of ultimate recovery of those costs since rates are
designed to allow that recovery over time as weather patterns average into the normal
pattern on which rates are based. It is a matter of smoothing cash flow from year to
year. IPC may also be interested in promoting energy use, especially during non-peak
periods. Keeping energy rates low would facilitate that.
Although those are understandable objectives from a business point of view, they
may not be the policy objectives guiding this Commission. It is likely that this
Commission is more interested in setting electric rates that provide clear economic
signals to customers so that they are encouraged to use electricity efficiently and
responsibly. Energy consumption has important economic, environmental , and national
security implications that should not be ignored. This Commission is also likely to be
concerned with the fairness of the impact of various cost allocations and rate designs in
Power: Direct
AARP
terms of their impacts on different groups of customers.
Power: Direct
AARP
APPENDIX B
Qualifications of Thomas Michael Power
What is your current employment?
I am a Professor of Economics and Chairman of the Economics Department at
the University of Montana in Missoula, Montana, 59812.
Please describe your formal education and training.
I received my Bachelor s Degree in Physics from Lehigh University in
Bethlehem, Pennsylvania. I graduated with high honors and Phi Beta Kappa. .was
elected a Woodrow Wilson Fellow in national competition and attended Princeton
University where I received by Masters and Doctoral Degrees in Economics.
I taught math and physics at Lehigh University and have taught economics at
Princeton University, Lehigh University, and the University of Montana. I have been
the faculty of the University of Montana since 1968. I have served as Chairman of the
Economics Department since 1978. My specialties are regional economics and
resource economics with an emphasis on energy, water, and environmental resources.
Have you testified as an expert witness before this and other utility regulatory
commissions?
Yes. Since 1974 I have appeared before numerous federal, state, and
municipal regulatory commissions. I first testified before this Commission in the Pioneer
case in 1975. Since then I have testified in about two-dozen cases before this
Commission. I testified in IPC's last general rate case before this Commission in 1994-
Power. Direct
1995 (IPC-94-5)and prepared testimony in the docket that was scheduled to evaluate
lPG's contract to purchase the output of the proposed Garnet facility (IPC-01-42) in
2002. I have also testified before the utility regulatory commissions in the following
states: Arizona, Colorado, Florida, Indiana, Illinois, Kansas, Montana, Nebraska
Nevada, Oklahoma , Oregon , Texas, Utah , and Washington. In addition, I have also
testified before the Federal Energy Regulatory Commission, the Northwest Regional
Power Planning Council , and the Bonneville Power Administration as well as before
various congressional committees.
In addition , I have testified in utility cases before the City Councils of Seattle,
Austin , and Spokane. I have also testified before the Snohomish County, Washington
Public Utility Board and the Springfield , Oregon , Public Utility Board. I have testified in
State District Courts in Idaho, North Dakota, Oregon, and Montana and in Federal Court
in Montana.
I have testified before the Montana Board of Natural Resources and the
Washington Department of Ecology, and the Washington Energy Facility Site Evaluation
Council on the siting of energy facilities.
I have served as lecturer at National Association of Utility Regulatory
Commissioners' Technical Conferences and at annual conferences of the Mid-America
Regulatory Commissioners and the Western Utility Regulatory Commissioners.
Since 1988 I have served on the Montana Power Company (now Northwest
Energy Company) Technical Advisory Committee. For several years I have also
served on the Montana Regulatory Reform Working Group. I currently serve on the
Power: Direct
Montana Governor s Energy Security Task Force. In the past I have served on the
Montana Governor's Citizens' Advisory Council on Energy.
Have you done other studies dealing with energy economics?
Yes. In 1975, I received an NSF/RANN grant to assemble a team of economists
geologists, and energy technologists to study coal development in the Northern Great
Plains. That study led to a series of almost a dozen reports, the final summary being
published as Projections of Northern Great Plains Coal Mining and Energy
Conversion Development 1975-2000 A.D. Several of the other papers dealing with
defining coal markets and energy projection techniques have also been published.
Between 1976 and 1985 I conducted studies of the economics of alternative
energy systems, transmission reliability, the applicability of the PURPA rate making
standards to hydroelectric system "going thermal", utility avoided costs, optimal
operation of storage hydroelectric facilities, development of electric utilities on Indian
reservations, and the impact of energy facility development on local economic
development. In 1995 Public Utilities Fortnightly published an article of my entitle
Making Sense of Peak Load Allocations.
Can you give examples of other studies have you done in the field of resource
economics?
In 2001 Island Press published Post-Cowboy Economics: Pay and Prosperity
in the New American West, which I co-authored with Richard Barrett. In 1996 two
other books of mine were published. Island Press published Lost Landscapes and
Failed Economies: The Search for Value of Place. E. Sharpe published
Environmental Protection and Economic Well-Being: The Economic Pursuit of
Power. Direct P. 58
Quality. The latter book is the rewritten and updated Second Edition of The Economic
Pursuit of Quality which was published by M. E. Sharpe, New York in 1988. I have also
contributed a dozen chapters to various other books. Among the many articles and
reports I have published are "The Wealth of Nature (Issues in Science and
Technology, National Academy of Sciences , Spring, 1996), "Economic Well-being and
Environmental Protection in the Pacific Northwest (lIIahee: Journal for the
Northwest Environment, 11(3 & 4), Fall-Winter, 1995), "Ecosystem Preservation and
the Economy ofthe Greater Yellowstone (Conservation Biology, September, 1991)
and "Urban Disamenities (Journal of Urban Economics June, 1981). In 1980
Westview Press published by book on The Economic Value of the Quality of Life
:.
1979 , the Audubon Society published my study The Central Arizona Project: An
Economic Analysis. I have published papers on almost a dozen federal irrigation
projects in the western states in addition to papers dealing with the value of in-stream
flows for wildlife and recreational uses. I have testified before the State Board of
Minerals and the Environment and the Oahe Conservancy Board in South Dakota as
well as the Alberta Energy Resources Conservation Board and Natural Resource
Conservation Board on topics related to resource development. I have also testified
before Canadian Federal Environmental Review Boards.
Power: Direct