4-2500-19796-2
E-015/GR-08-415
STATE OF
OFFICE OF
ADMINISTRATIVE HEARINGS
FOR THE
PUBLIC UTILITIES COMMISSION
|
In the Matter of the Application of |
FINDINGS OF FACT, CONCLUSIONS AND RECOMMENDATION |
The above-entitled matter came on for
evidentiary hearing before Administrative Law Judge Bruce H. Johnson on November
12 - 14 and 18 - 20, 2008, at the offices of the Minnesota Public Utilities
Commission in
The parties to this proceeding are: ALLETE Corporation d/b/a Minnesota Power
Company (“Minnesota Power,” “MP,” or the “Company”); the Minnesota Department
of Commerce/Office of Energy Security (the “OES”); the Minnesota Office of
Attorney General -- Residential Utilities Division (the “OAG/RUD”); Large Power
Intervenors (“LPI”); the Minnesota Chamber of Commerce (the “MCC”); Energy
Cents Coalition (“ECC” or “Energy Cents”); and Boise, Inc. (“Boise”).
Samuel Hanson, Thomas Bailey, and
Elizabeth Brama, Attorneys at Law, Briggs and Morgan, 2200 IDS Center,
Valerie Means and Julia Anderson,
Assistant Attorneys General, 1400
Ron Giteck and William Stamets, Assistant
Attorneys General, 900 BRM Tower,
Robert S. Lee and Andrew P. Moratzka,
Attorneys at Law, Mackall, Crounse & Moore, 1400
Eric F. Swanson, Attorney at Law,
Winthrop & Weinstine,
Michael Franklin, Director, Minnesota
Chamber of Commerce,
Commission Analysts Robert Harding, Louis
Sickmann, Stuart Mitchell, Michelle Rebholz, and Chris Fittipaldi appeared on
behalf of the PUC Staff.
Notice is hereby given that, pursuant to
Minn. Stat. § 14.61 and the Rules of Practice of the Minnesota Public Utilities
Commission (the “Commission”) and the Office of Administrative Hearings,
exceptions to this Report, if any, by any party adversely affected must be
filed within 15 days of the mailing date hereof with the Executive Secretary,
Minnesota Public Utilities Commission, Metro Square Building, Suite 350, 121
7th Place East, St. Paul, Minnesota 55101-2147.
Exceptions must be specific and stated and numbered separately. Proposed Findings of Fact, Conclusions of Law
and Order should be included, and copies thereof shall be served upon all
parties. Oral argument before a majority
of the Commission will be permitted to all parties adversely affected by the
Administrative Law Judge’s recommendation who request such argument with their
filed exceptions or reply. Exceptions
should be e-Filed with the Commission.
The Commission will make the final
determination of the matter after the expiration of the period for filing
exceptions as set forth above, or after oral argument, if such is requested and
had in the matter.
Further notice is hereby given that the
Commission may, at its own discretion, accept, reject, or modify the
Administrative Law Judge’s recommendations and that said recommendations have
no legal effect unless expressly adopted by the Commission as its final order.
A.
Jurisdiction and
Procedure.
1.
On May 2, 2008,
MP filed a petition to increase its electric rates in
2.
Minnesota
Power’s claimed revenue deficiency in the initial filing was based on a test
year of July 1, 2008, through June 30, 2009; an 11.15% rate of return on common
equity; an equity ratio of 54.79%; and the application of the resulting overall
rate of return of 8.68% to the rate base calculated as $713,096,651. [1]
3.
The Company’s
initial filing, with significant errata filings, came before the Commission on
June 12, 2008. The Commission found MP’s
filing to be incomplete, and further found that the filing would not be in
proper form and substantially complete until the Company re-filed its
application in a form that corrected all errors, omissions and other
deficiencies. [2]
4.
On June 12,
2008, the Company filed its Supplemental Rate Case filing (Supplemental Filing)
restating the request for a general increase in its electric rates. In the
Supplemental Filing, MP sought an annual rate increase of $45,023,320, or
approximately 9.5 percent per year over current rates for firm and non-firm
sales of electricity.
5.
After a hearing
on July 15, 2008, the Commission found the Company’s filing to be substantially
complete as of June 12, 2008, suspended the proposed rates pending a final
decision on the merits, and referred the case to the Office of Administrative
Hearings for a contested case proceeding.
On July 21, 2008, the Commission issued a Notice and Order for Hearing
in this matter, and this contested case proceeding ensued. On the same date, the Commission also entered
an Order Setting Interim Rates.[3] The Interim Rate Order approved the Company’s
interim rate proposal and authorized the Company to put interim rates into
effect for service rendered on and after August 1, 2008, subject to refunding
that portion of the rate not found to be supported in the final rate
determination.
6.
The Commission’s
Notice and Order For Hearing identified the following issues:
(2) Is the rate
design proposed by the Company, including proposed revisions to customer
charges, reasonable?
(3) Are the Company's
proposed capital structure, cost of capital, and return on equity reasonable?
(4) Is the
Company's proposed collection of $18.6 million claimed fuel clause
undercollection reasonable?
(5) Are the
Company's proposed changes to its Rider for Fuel Adjustments reasonable?[4]
7.
On July 21,
2008, the Commission also entered an Order in Docket No. 08-463 granting the
Company's May 2, 2008, petition for a base fuel adjustment rate change, with
the following clarifications:
1. The
following issues will be addressed in the rate case proceedings: 1) MP's
proposal to recover "lagged fuel clause costs associated with the
implementation of the new base cost of fuel"; and 2) its proposed changes
to the Rider for Fuel Adjustments.
2. If any significant adjustments to the cost of energy occur as a result of the general rate case, then the base cost of energy may need to be reconsidered and reflected in final rates subsequent to the Commission's decision in the general rate case.[5]
8.
On September 26,
2008, the non-Company parties filed Direct Testimony.
9.
On October 22,
2008, the Company, OES, and LPI filed Rebuttal Testimony. In Minnesota Power’s rebuttal testimony, the
Company recognized adjustments to its initial filing, including an increase in
operating income by $1.1 million. Other
adjustments decreased MP’s rate base by $11.1 million to $703.7 million. Based on these adjustments, Minnesota Power
revised its claimed revenue deficiency to $41.4 million.[6]
10.
The Company,
OES, OAG/RUD, MCC, LPI, and Energy Cents filed Surrebuttal Testimony on
November 5, 2008. In Minnesota Power’s
surrebuttal testimony, the Company noted that it had mistakenly reversed one of
its adjustments, resulting in a decrease instead of an increase in operating
income under current rates. After
correcting its mistake, MP revised its revenue deficiency calculation to $39.8
million. [7]
B.
Summary of
Public Comments.
12.
In general,
Residential class and General Service class customers objected to the Company’s
proposed increase in rates, particularly the size of the increase. Many retirees living on fixed incomes noted
that recent increases in other expenses they incur have made their energy bill less
affordable, even without the proposed increase.
13.
Some customers
had specific suggestions for addressing MP’s proposed revenue deficiency. Three customers and one shareholder
maintained that Minnesota Power's executive compensation should be reduced
before customer charges are increased.
One customer objected to any portion of the costs for wind power being
covered in rates. Another argued that
ALLETE should use profits from its other enterprises to pay for MP's
infrastructure costs, rather than raising rates.
14.
Regarding rate
design, several residential customers noted that MP’s proposal would result in
a 100% increase in their monthly charge.
One customer had trouble distinguishing between the increase in the
monthly charge and the overall rate increase. Several residential customers
objected to the disparity in percentage increases between classes—particularly,
the proposed percentage increase for Residential customers in comparison with
the percentage increase for Large Power customers. Those Residential customers suggested that
because of the large volume of electricity that Large Power customers purchase,
Large Power customers should bear more of the proposed increase. Several customers described their efforts to
conserve energy and asserted that this rate increase would nullify their
conservation efforts.
15.
A resort owner
indicated that the proposed increase in the General Service class was too great
a burden on small businesses, and the City of International Falls urged the
Commission to strike a balance in MP's new rates between the burdens on
residential and general service customers and the relatively lighter burden on
Large Power customers.
16.
A significant
number of customers objected to the proposed rate increase as excessive when
viewed in combination with the resource adjustment applied to off-peak and
regular meter rates and other riders on the bills they were already having to
pay. Off-peak customers pointed out that
MP’s current off-peak rates offer a savings of around 50% from retail rates,
but the proposed off-peak rate increase would reduce that savings to around
28%. Many of the off-peak and dual fuels
customers emphasized the expense that they have had to incur to obtain their
lower rates. Those customers had
expected that they would be able to realize savings on their electric costs,
and they objected to MP’s dual fuels and off-peak proposals as disincentives to
use these programs and as unfair to those customers who had recently incurred
significant costs to switch their service.
17.
On the other
hand, several customers supported the rate increase as necessary to maintain
the Company as a good provider of electricity.
One customer compared his electricity bill to his tax bill and asserted
the he gets more benefit from Minnesota Power than from the government.
18.
Harvey Schmitt,
Director of Housing Services for Catholic Charities, indicated that as a low
income housing provider, his organization cannot raise the rents it charges to
cover the proposed increase in the General Service rate because Catholic
Charities’ rents are capped by the Department of Housing and Urban
Development. Catholic Charities proposed
that low income housing providers be eligible for the Company’s proposed new
low income rate.
19.
Shareholders
generally supported the rate increase as necessary to maintain Minnesota
Power's efficiency and dependability.
Many of them noted that it had been 14 years since the Company’s last
rate increase. Minnesota Utility
Investors (MUI) urged approval of MP's rate increase as necessary to provide
reliable electrical service. MUI noted
that the Company’s stock is an investment grade security and that it needs to
retain that rating to obtain the capital required for developing the
alternative energy sources required by 2025.
MUI urged approval of an ROE that will allow MP to 1) attract capital at
reasonable rates; 2) ensure reliability of electrical service; and 3) allow
shareholders a reasonable rate of return, now and in the future.
20.
Public comment
relating to Minnesota Power’s economic development activities is described in
Finding 235, below.
C.
Description of the
Company.
21.
Minnesota Power
is an operating division of ALLETE, Inc., which is a
22.
Minn. Stat.
§ 216B.16, subd. 4, imposes on MP the burden of showing “that the rate
change is just and reasonable.” Minn.
Stat. § 216B.03 provides: “Every
rate made, demanded or received by a public utility . . . shall be
just and reasonable . . . .
Any doubt as to reasonableness should be resolved in favor of the
consumer.”
23.
The Minnesota
Supreme Court described the Commission’s role in determining just and
reasonable rates in a rate proceeding as follows:
[I]n the exercise of the statutorily
imposed duty to determine whether the inclusion of the item generating the
claimed cost is appropriate, or whether the ratepayers or the shareholders
should sustain the burden generated by the claimed cost, the MPUC acts in both
a quasi-judicial and a partially legislative capacity. To state it differently, in evaluating the
case, the accent is more on the inferences and conclusions to be drawn from the
basic facts (i.e., the amount of the claimed costs) rather than on the
reliability of the facts themselves. Thus, by merely showing that it has
incurred, or may hypothetically incur, expenses, the utility does not necessarily
meet its burden of demonstrating it is just and reasonable that the ratepayers
bear the costs of those expenses.[9]
24.
In this
proceeding, the Administrative Law Judge’s role is to assess the evidence
presented and make recommendations to the Minnesota Public Utilities
Commission. Whether Minnesota Power has
met its burden of proof is ultimately for the Commission to decide, based on
the record.
25.
In rate cases
before the Commission, utilities determine the extent to which projected revenue
will cover the anticipated costs of operation, including a return on investment
to shareholders. The period used to
measure this revenue and these costs is called the test year. Minnesota Power proposed a projected test
year of July 1, 2008, to June 30, 2009. Over that period, MP estimated total
operating revenues of $535,814,764, and total utility operating expenses of $497,899,486.[10] MP estimated its rate base as $699,711,856
and proposed a rate of return of 8.68%.
To meet that rate of return, MP calculated a required operating income
of $60,734,989. By the Company’s
calculations, MP will experience a required gross revenue deficiency of $38,921,550
over its test year.
27.
Minnesota Power
explained its reasons for using a projected test year as follows:
Although a
projected test year may turn out to be slightly different than those actual
financial results, the question is whether it will be more accurate than the
use of a historical test year, adjusted for known change. Because a projected
test year incorporates the Company's best estimates of utility plant, sales and
financial assumptions for the upcoming year, it is likely a better
representation of what will actually occur during the year than an adjusted
historical year. Either way, the accuracy depends on the assumptions that are
made.[11]
28.
The OAG/RUD
maintains that using a historical test year, adjusted for known changes, is the
superior method for calculating the expenses to be incurred by the
utility. The OAG/RUD contends that since
the activities that generate most expenses in the utility area are consistent
from year to year (as demonstrated by the class customer cost of service study,
for example), use of an adjusted historic test year affords the benefits of
showing costs that have actually been incurred and allowing those costs to be
audited for suitability for recovery from ratepayers.[12]
Basing revenue requirements on financial data from a
test year, a representative slice of the utility’s normal operations, is
intended to base rates on experience instead of conjecture. It is also intended
to replace the fiscal discipline of the market place, which is absent for
monopolies, with the fiscal discipline of prior determination of reasonable
costs.[13]
30.
The OAG/RUD’s
argument for requiring Minnesota Power to use an historical test year is primarily
two-fold. First, it argues that since
the Company will not be able to provide
actual data on a jurisdictional basis for the early months of the test year, the parties will not be able to
determine the accuracy of the Company's
forecasting for that portion of the test year.[14] Second, it argues that the unreliability of
the Company’s projected test year is established by a retail sales forecast for
the projected test year that is unreasonably low—that is, a $54 million
decrease in test year sales in comparison with 2007 retail sales without a
reasonable explanation.[15]
31.
While
acknowledging that its methodology does not provide actual data on a
32.
Whether the
Company’s revenue projections for a test year are unreliable or unreasonably
low does not depend on whether the test year is historical or projected. In either case, the reliability of sales
forecasts depends primarily on the reliability of the forecast methodology and
inputs. For example, in this proceeding,
the ALJ has found that some of Minnesota Power’s sales forecasts were
unreliable because the forecasting methodology is less reliable than other approaches.[18] If the Commission agrees with those findings,
those forecasts would be unreliable regardless of what test year the Company
used.
33.
Finally, the use
of a projected test year, including the use of split test years, has been
common and has been approved by the Commission in other rate cases. Minnesota Power itself used split projected
test years in five of its previous seven cases: GR-87-223 (July 1, 1987 – June
30, 1988); GR–81- 250 (July 1, 1981 – June 30, 1982); GR-80-76 (May 1, 1980 –
April 30, 1981); GR-78-514 (July 1, 1978 – June 30, 1979); and GR-77-360 (May
1, 1977 – April 30, 1978).[19]
34.
Whether there
are deficiencies in some of the Company’s specific cost and revenue forecasts
will be analyzed in other Findings.
Nonetheless, even though an historic test year may be generally superior,
particularly regarding expenses, the OAG/RUD has not demonstrated that the mere
use by Minnesota Power of a projected test year has created a bias toward
maximizing a forecast revenue deficiency.
In view of all of the above, the ALJ therefore concludes that the
Company’s use of a projected test year from July 1, 2008, to June 30, 2009, is
reasonable, and that the Commission should allow it.
1. General
Approach to Forecasting Sales Used by
35.
Minnesota
Power’s test-year sales are forecast for the last six months of 2008 and the
first six months of 2009, which correspond to the utility’s current fiscal
year. The Company employed multiple
forecasting techniques to estimate its test-year sales and revenue projections
for its various customer classes. It
relied in whole or in part on its econometric Advance Forecast Report (“AFR”)
for five customer classes—i.e., the Residential, General Service, Large Light
and Power,[20] Municipal Pumping,
and Dual Fuels classes;[21] it relied
solely on specific monthly load forecasts for its Large Power class and on
historical data for its Lighting class.[22] With the exception of Large Light and Power,
Lighting, and Large Power classes, the usage of all of those classes is
strongly affected by weather patterns.[23]
36.
More
specifically, the test year sales and revenue figures forecast by MP for all
but Large Power customers were estimated using a budget forecasting program
that is primarily based on annual sales and revenue estimates from its AFR,
which MP then allocated to months in the test year using MP’s monthly budget
forecast. However, as previously noted,
Minnesota Power is unique in that roughly 70 percent of its sales are to
customers having steady energy usage patterns that are not affected significantly
by weather.[24] Rather than using econometric modeling or
specific load forecasts, the Company used marketing forecasts for its Large Power
class customers, whose usage is relatively consistent and production-based.[25] In developing test year forecasts of monthly
sales and revenue projections for those customers, MP consulted with each of
them individually to produce marketing forecasts.[26]
37.
To estimate test
year sales for classes for which it relied on its utility budget forecasting
program, MP began by producing two separate and independent forecasts, one
using a monthly econometric model and the other using an annual econometric
model. Both are time series econometric
models that incorporate serial correlation.
One respect in which they differ is that the monthly model uses data
over the 1993 to 2007 time frame, while the data for the annual model uses data
from 1965 through 2007, with the estimation conducted over the time frame
beginning in 1969 forward.[27] The two models arrived at two different results. MP then used the results of the monthly model
to convert the projected annual sales volumes AFR model into monthly estimates
that form the basis for the retail sales forecasts for weather-sensitive
customer classes.[28]
38.
Other
39.
Minnesota Power
projected its test-year sales estimates for individually billed customers based
on specific load forecasts, historical usage data, or marketing information on
prospective energy usage provided by customers to the Company’s marketing
department, although in some cases that information was correlated with
statistical methods.[31] Given Large Light and Power customers’ power
usage characteristics (generally shifting between months) and the amount of
available monthly data, statistical forecasting techniques are not appropriate
and may produce unreasonable test-year forecasts.[32]
40.
In its analysis
of the Company’s forecasts for its seven Large Light and Power customers, the
OES prepared a visual sales analysis for each Large Light and Power customer
from January 2000 through the end of Minnesota Power’s budget forecast in
December 2009. After examining those
historical sales patterns to ensure that sales have been consistent for those
customers, the OES established a three-year average of actual energy usage for
each affected customer. The OES then
charted line graphs of historical usage for each customer and projected usage
during the two budget years (2008 and 2009, which includes the test-year). Based on that graphical analysis, the OES
concluded that the test-year sales forecasts for two LLP customers had been
underestimated, and it recommended an upward adjustment of the sales to those
seven customers.[33]
41.
Approximately 70
percent of Minnesota Power’s sales are to a small group of Large Power
customers in the forest products and taconite industries whose operations are
energy intensive.[34] The usage patterns of those customers tend to
be relatively consistent and unaffected by weather but are significantly
affected by demand for wood products, paper, and taconite, which can
significantly affect annual production levels.
Accordingly, it is reasonable for the Company to primarily rely on
consultations with individually billed Large Power customers to produce
marketing forecasts of test-year monthly sales and revenue projections.
2. Forecasts of Sales to Residential
Service, General Service, Municipal Pumping, and Dual Fuel Classes.
42.
The OAG/RUD
challenged the Company’s forecast of residential and general service revenues
on the grounds that the test year sales to these classes were lower than
historical data would support.[35] It noted that in 2007, Minnesota Power had
“actual sales [that] were 8% greater than what were budgeted for the year
2007. Actual revenues for 2007 were $49
million higher than budget, which is more than the $45 million increase that
the Company is requesting in this case.” [36] The OAG/RUD further noted that the Company’s
own comparisons of its operating revenues, expenses and operating income for
the
Operating
revenue decreased $2.4 million from 2007, primarily due to
decreased fuel clause recoveries and the reduction in revenue from sales to
Other Power Suppliers.
Fuel Clause recoveries decreased $15.0 million in
2008 primarily as a result of decreased power expenses reflecting increased
Company generation and increased hydro availability.[38]
43.
In
rebuttal, Minnesota Power
argued that its sales do
not increase or decrease simply because of the passage of time, but rather
because of specific factors that are determinative of the amount of energy a
class of customers will use.[39] The Company went on to identify four factors that it
believes are more
relevant than
historical sales data in forecasting
sales to the Residential and General Service customer
classes: namely, the increase in
the price of electricity, the decrease in the price of natural gas, per capita
income, and area employment.[40] The Company also addressed how it believed that
each one of those factors
indicated reduced energy consumption among residential and general service
customers for the test year.[41] MP
then argued that the OAG/RUD had
failed to address the significance of those indicators in forecasting class sales.
44.
While
acknowledging MP’s responses, the OAG/RUD still emphasized “that sales exceeded
budget by an amount tantamount to the entire proposed rate increase, whatever
the particular cause.” [42] The OAG/RUD’s approach to test year sales
forecasts, however, presents difficulties.
First of all, since the sales estimates developed for any test year,
regardless of whether it is an historical or projected test year, will be the
basis for the utility’s rates for the future, any historical sales revenue data
incorporated into test year revenues will necessarily impact the estimate or
forecast. The issue, however, is whether
the utility’s revenue projections for the test year are reliable, not
necessarily how they correlate to specific historical data. The OAG/RUD’s position also seems predicated
on the assumption that in order to be reliable, test year revenue estimates
must always correlate with actual historical usage. Implicit in that assumption is that using
econometric models to forecast sales to customer classes whose usage is
dependent on such factors as weather, the price of electricity, the price of
natural gas, per capita income, and area employment,[43] is
conjectural and unreasonable, which is clearly not the case.
45.
Although the OES
was also troubled by the disparity between test year estimates for the
residential and general service classes, its challenge to MP’s retail sales and
revenue forecasts for those classes is more specific and technically based than
the OAG/RUD’s challenge. The OES agrees
with the OAG/RUD that since the econometric forecast method that MP used to
estimate test year sales of non-Large Power classes included an estimate of
2007 annual sales, it should correlate well to the to actual sales from the
Company’s 2007 Annual Jurisdictional Report.
The OES, however, then proceeded to compare the Company’s 2007 AFR and
monthly budget process forecasts to weather-normalized actual sales from the
Company’s 2007 Annual Jurisdictional Report.[44] Based on that comparison, the OES concluded
first that MP’s 2007 AFR forecast did not correlate well to actual
weather-normalized sales. From that
comparison, the OES also concluded that the Company’s AFR method underestimated
the sales of all classes (other than Lighting) for the test year. The OES therefore recommended that the
Company’s AFR model should be rejected for ratemaking purposes because it does
not produce reasonable results and would cause rates for MP’s ratepayers to be
unreasonably high until the Company files a new rate case.[45] Additionally, based on the comparisons that
the OES made, it concluded that the Company’s monthly budget forecast produced
representative test-year sales estimates that were more reliable for setting
just and reasonable rates in this proceeding.
Accordingly, the OES advocates use of MP’s monthly budget process models
to project sales and revenues for these customer classes.
46.
Minnesota Power
takes issue with the OES’s conclusions arguing that although the Company’s AFR
model may have produced a forecasting error for 2007, it fails to establish
that the model is inherently biased. The
Company points out that all forecasting models produce forecasting errors
without necessarily having a forecasting bias. In order to determine the
presence of bias, one would need to analyze the forecasts for a number of
years. For unbiased forecasting models, one would see both over-predictions and
under-predictions over those years, with the average forecasting error close to
zero. The Company argues that the OES’s
analysis does not involve any demonstration of a consistent under-estimation of
sales over a number of years that can only be explained by systemic forecasting
bias.[46]
47.
It appears from
the record that Minnesota Power has only been using its current AFR methodology
for annual sales forecasts since 2006, and that the deviation of actual from
forecast values in 2006 was +.15%, resulting in an average forecast error for
the two years of usage of – 1.14%.[47] On the other hand, over ten years of usage,
the Company’s monthly budget forecast error has only averaged +.29%. In the ALJ’s view, the sales and revenues for
the Company’s residential and general service classes should be estimated using
Minnesota Power’s monthly budget process models, as the OES recommends, rather
than using the Company’s approach of using that monthly model only to calibrate
the annual sales volumes projected by its AFR model. Other
48.
Finally,
3. Forecasts of Sales to Large Light and
Power Classes.
49.
Minnesota
Power’s Large Light and Power customers are individually billed, and the
test-year sales estimates for those customers are projected by the Company’s
marketing department, not solely by statistical methods.[49] Usage by those Large Light and Power
customers tends to be relatively stable and predictable, since their usage
would not normally be significantly affected by either weather[50] or by
changes in market conditions, to the extent of Large Power customers.
50.
After reviewing
the test year sales forecasts of Minnesota Power’s individually billed Large
Light and Power customers, the OES concluded that MP’s test-year estimates for two
of those customers, Ainsworth and Polymet, did not correspond with historical
usage patterns and sales trends— specifically, it concluded that the forecast
test year sales appeared unreasonably low in comparison to the 3- and 8- year
averages of past sales to those customers.[51] The OES
therefore concluded that the test year sales estimates for those customers had
been underestimated.[52] To correct
for that underestimation, the OES recommends a somewhat higher sales forecast
for these customers.
51.
In rebuttal
testimony and based on a recent public announcement by Large Light and Power
customer Ainsworth-Grand Rapids (Ainsworth) of its permanent shutdown of
production, Minnesota Power proposed a downward adjustment of the sales
forecast for Ainsworth that was significantly lower than what the Company had
even previously forecast.[53] Specifically, the evidence indicated that the
customer would be indefinitely shut down along with the other
52.
On the other
hand, the Minnesota Power did not otherwise respond specifically to the OES’s
observation that test year forecasts for Polymet did not correspond with
historical usage patterns and sales trends.
Rather, the Company only responded that it had worked with its Large Light and Power customers
directly to get the best information about their needs so that sales forecasts
will be as realistic as they can be.[56] The Company neither argued with nor offered support for the
OES’s proposed
adjustments to the Company’s forecasted sales for Polymet.[57]
53.
In response to the evidence that the Company had
presented regarding Ainsworth, the OES modified its position by accepting MP’s
original forecast of that customer’s sales, but the OES did not agree to a
further reduction based on the Company’s customer-specific information on the
basis that sales forecasts for a test year should be based
on conditions during a “normal” year and not on the occurrence of unique
events.[58]
54.
The Commission
has described its approach to test year adjustments as follows:
As a general rule, the Commission is reluctant to
adjust revenue requirements to reflect changes, certain or not, unless there is
a compelling need to do so. This is because the test year method by which rates
are set rests on the assumption that changes in the Company's financial status
during the test year will be roughly symmetrical -- some favoring the Company,
others not. Not adjusting for either type of change maintains this symmetry and
maintains the integrity of the test year process. Anomalies are likely to exist
in and beyond any test year.
In keeping with these general principles, the
Commission has adjusted for changes in the past only when their certainty and
magnitude would otherwise make the test year process unreliable. [59]
4. Forecasts of Sales to the Large Power
Class.
56.
The OES objected
to Minnesota Power’s initial test year sales and revenue forecasts for a select
number of individually billed customers, including one Large Power customer
(Customer X), as being unreasonable because the test-year sales and revenues were noticeably
lower than sales and revenues during the 12-month period prior to the test year. The OES
examined Customer X’s historical sales and calculated three-year and eight-year
sales averages. Concluding that its two
average sales calculations were a clear indication that test-year sales for
Customer X had been underestimated, the OES proposed an adjustment
increasing the sales forecast usage for that Large Power customer.[60]
57.
In September 2008 Minnesota Power entered into an agreement
with Cleveland-Cliffs, now known as Cliffs Natural Resources, Inc., providing
for contract extensions and amendments to the Electric Service Agreements with
Hibbing Taconite and United Taconite. Although MP did not
project both companies’ total power usage to
change substantially from what was assumed in the test year sales forecast, the Company
indicated that relative quantities of
Large Power firm and interruptible energy usage have changed as a result of
recent changes in Minnesota Power's incremental/market-based price of
interruptible energy compared to the firm energy price. Minnesota Power asserts that those changes resulted in decreases in projected test year revenues from the
projections in the original case filing.[61] Additionally in June 2008, Enbridge Energy
entered into a new Electric Service Agreement (“ESA”) providing for electric
service under Minnesota Power's Large Light and Power rate rather than the
previous discretionary rate pipeline service. This Agreement was filed with the
Commission for approval on August 18, 2008, in Docket No. E-015/M-08-976. The revenues under Minnesota Power's new Agreement
with Enbridge are projected to be lower than
those assumed under the pipeline discretionary rates that were in place at the
time of the rate case filing. Minnesota
Power is therefore seeking downward adjustments of its initial sales forecasts
for those customers.[62]
58.
The OES also
objects to the proposed downward adjustments to Hibbing Taconite, United
Taconite, and Enbridge. It argues that none of the three ESAs
Minnesota Power lists have been approved by the Commission. Therefore, there is no guarantee that the
revenue effects that Minnesota Power recommends will occur. Second, since the
agreements do not involve changes in sales levels, only modifications in
revenues from those customers, Minnesota Power has a burden to show why all
other customers on Minnesota Power’s system should have to subsidize these
customers by paying higher rates to offset their revenue reductions. In other
words, the OES argues that inclusion of those adjustments on a total system
level would unreasonably burden all other Minnesota Power ratepayers.[63]
59.
As noted above, “the Commission has adjusted for changes in the past
only when their certainty and magnitude would otherwise make the test year
process unreliable.” [64] With regard to the adjustment the OES
proposes for the sales forecast for the single Large Power customer (Customer
X), the adjustment appears based on a reduction in
usage. Business cycles are not regular
and predictable, and near term future market conditions and customer production
levels do not necessarily have a predictable relationship with historical
market conditions and customer production levels, even those in the recent
past. Minnesota Power’s forecasts of a
reduction in that Large Power customer’s sales for the test year, and also for
the life of a three-year rate, if that is what the Commission orders, are
consistent with the evidence that was presented concerning market conditions
for each of the Large Power customers during the test year and extending into
the near future. The ALJ therefore
recommends that the Commission approve that adjustment.
60.
On the other
hand, the Company’s proposed downward adjustments in sales forecasts for
Hibbing Taconite, United Taconite, and Enbridge present a different situation;
they are based on a reduction in rates, and not usage. It is therefore less clear that those
proposed revenue reductions are based on economic conditions. Moreover, their certainty and magnitude are
currently less predictable and dependent on future actions of the Commission,
which will be based, in part, on their impact on other ratepayers. The ALJ therefore recommends that the
Commission approve the Company’s sales forecasts for its Large Power customers,
but disapprove MP’s proposed adjustments for Hibbing Taconite, United Taconite,
and Enbridge.
61.
The OAG/RUD’s
more general objection to the Company’s sales forecasts for Large Power
customers assumes that marketing forecasts are necessarily less reliable than
relying on past usage adjusted for inflation.
However, that position does not account for the possibilities that large
power customers may experience reasonably foreseeable long term or permanent
changes in requirements or changes in production during the life of the rate
increase being sought. The ALJ therefore
recommends that the Commission not rely on the OAG/RUD’s approach to downward
adjustment of Large Power customer sales forecasts, which is based on
historical sales subject to an annual adjustment for inflation.
62.
In the
competitive market environment, prices and operating incomes are determined by
the free interaction of market forces, such as supply and demand. Those market forces define the optimal levels
and mix of the variety of goods and services that are produced in the
economy. In rate-regulated industries,
prices (described as rates) and operating incomes (returns) are determined by
regulatory agencies. Those agencies must
set reasonable rates to ensure such utilities are financially able to provide
an adequate supply of satisfactory services.
Providing those services is dependent, in part, on a utility’s ability
to compete for necessary funds in the capital markets. The utility must earn enough to offer
competitive returns to investors in order to attract those funds. In the regulated utility context, a fair
return enables the utility to attract sufficient capital to conduct business,
at reasonable terms. [65]
63.
The Commission
has consistently followed a number of principles in determining appropriate
rates of return (ROR) in utility rate setting proceedings. Those principles can be briefly stated as:
· The rate of return should be sufficient to enable the
regulated company to maintain its credit rating and financial integrity.
· The rate of return should be sufficient to enable the
utility to attract capital.
· The rate of return should be commensurate with
returns being earned on other investments having equivalent risks.[66]
64.
One determines
ROR by calculating the weighted average cost of the various sources of capital
used by a company. The weighting converts
the sources of capital (debt or equity) to the percentages reflected in the
company’s capital structure. Capital
structure generally refers to the mix of long- and short-term debt, preferred
stock, and common equity that constitute those sources of capital. To reflect the impact of the different cost
of various types of capital, each component is weighted by its relative
proportion in the overall mix of capital to determine the overall cost of
capital. Calculation of the overall ROR
requires a determination of costs and types of capital used by the company.[67]
65.
Minnesota Power
has no legal existence separate from its parent company ALLETE and therefore
has no publicly-traded common stock.[68] Since there is no market-driven balance
between debt and equity securities to assess MP’s capital structure, other
approaches must be employed to strike that balance.
66.
Minnesota Power followed Standard & Poor’s (S&P) and
Moody’s benchmarks for debt to capital (equity) ratio ranges to retain ALLETE’s
current BBB+ financial rating, adjusted by debt equivalents.[69] The Company did not conduct any comparison of
its proposed capital structure with the utility operating divisions or subsidiaries
of comparable companies. Rather, the
Company compared its estimate of debt equivalents to those held by other
investor-owned utilities.[70]
67.
Minnesota
Power’s assessment concluded that its capital structure should be comprised of
54.79% common equity and 45.21% long-term debt.[71] The Company maintains that its proposed test
year capital structure was reasonable and appropriate for the following
reasons:
The Company’s objective is to maintain adequate
investment grade credit ratings in order to continue to access the capital it
needs at reasonable costs. The substantial capital expenditure requirements
facing Minnesota Power make this objective both more difficult and more
important. The Company’s recommended test year capital structure produces
capital ratios somewhat inferior to ratios consistent with ALLETE’s current
S&P credit rating. ALLETE will need to issue additional equity in the test
year and beyond to generate adequate credit metrics while it funds Minnesota
Power’s capital requirements, and the Company has communicated to rating
agencies its plans to finance these capital requirements with a combination of
debt and equity issuances that will produce capital structures supporting
ALLETE’s current S&P credit rating.[72]
68.
The OES compared
MP’s proposed capital structure to the average capital structure for other similar
utilities. The OES noted that the
average 2007 equity ratio for the comparable group of electric companies (used
to determine the OES’s proposed ROE) is 47.88 percent. Using a standard deviation of the equity
ratios for that group results in a range of 41.58 percent to 54.18
percent. Similarly, the average 2007
equity ratio for the OES-determined comparable group of combined gas and
electric companies (less one outlier company) is 51.14 percent. Based on these comparisons, the OES concluded
that MP’s proposed equity ratio of 54.79 percent is too high.[73]
69.
The OES
conducted an analysis of MP’s appropriate equity ratio, using the S&P
factors that were identified in MP’s analysis.
OES arrived at a common equity figure of 52.11 percent.[74] The parties’ competing proposed capital
structures are as follows:
|
|
MP Proposal |
OES Proposal |
|
Long Term Debt |
45.21% |
47.89% |
|
Common Equity |
54.79% |
52.11% |
|
Total |
100.0% |
100.0% |
70.
The OES proposal
adjusts the imputed capital structure for MP to put more of the structure into
long-term debt, which is currently lower in cost than equity. This adjustment results in a reduction of the
overall revenue required to meet the ROE figure. The OES further recommended that MP be
organized as a separate subsidiary that is wholly owned by ALLETE (rather than
an operating division).[75]
71.
It is not
appropriate to consider ALLETE’s needs in imputing an appropriate capital
structure to MP. The capital structure
must be reflect the economic structure of the utility operations for which ROR
and ROE are being calculated. In this
instance, the paramount consideration is striking an appropriate balance
between the cost of capital that a business should reasonably incur while
maintaining a respectable rating for the issuance of securities. The OES has shown that its proposed mix of
debt and equity meets these dual considerations better than MP’s proposal.[76] The OES has therefore demonstrated that its
proposed capital structure is appropriate for the ROR and ROE calculations in
this proceeding.
C.
Competing Determinations
of ROE.
72.
In its
rate-setting orders, the Commission has balanced ratepayer and utility
interests. This balancing is required to
carry out the Commission’s statutory responsibility to set rates that are just
and reasonable. A reasonable rate
enables an investor-owned utility to recover its operating expenses,
depreciation, and taxes, as well as compete for funds in capital markets. Allowing a fair and reasonable return upon
the utility’s investment in property used to provide the utility service is a
factor in setting just and reasonable rates.
This return on investment in property is more commonly referred to as
return on equity (“ROE”).[77]
73.
For
publicly-traded companies, ROE is determined by the actual performance of that
company’s stock in the marketplace.
Since ROE is a market-based concept and Minnesota Power does not exist
in the marketplace except as an operating division of ALLETE, it is necessary
to establish the ROE figure for Minnesota Power by other means. The Commission has historically relied upon
the Discounted Cash Flow (“DCF”) analysis to derive ROE for rate cases. This is the most widely accepted model and
one that has been used consistently as a starting point for establishing the
cost of equity in public utility cases before the Commission.[78]
74.
The basic
standards for the determination of ROE are set forth in Hope[79] and
75.
The Commission’s
order should provide the Company with the opportunity to earn an ROE that is:
(1) adequate to attract capital at reasonable terms; (2) sufficient to ensure
the financial soundness of the Company’s operations; and (3) commensurate with
returns on investments in utilities of comparable risks.
76.
Based on its
proposed capital structure, Minnesota Power, through Dr. Roger A. Morin, made
the following recommendations for ROR and ROE:[81]
|
|
Percent of Total |
Cost |
Weighted Cost |
|
Long Term Debt |
45.21% |
5.68% |
2.57% |
|
Common Equity |
54.79% |
11.15% |
6.11% |
|
Total |
100.0% |
|
8.69% |
77.
The OES, through
Dr. Eilon Amit, also made recommendations on both the Company’s ROR and ROE:[82]
|
|
Percent of Total |
Cost |
Weighted Cost |
|
Long Term Debt |
47.89% |
5.68% |
2.72% |
|
Common Equity |
52.11% |
10.74% |
5.59% |
|
Total |
100.0% |
|
8.32% |
D.
The Discounted
Cash Flow (“DCF”) Model.
78.
The DCF model is
based on the theory that a stock’s price represents the present value of all
future expected cash flows. The DCF
model is widely used to determine ROEs for utilities. The DCF model expresses
the ROE as the sum of the expected dividend yield and long-term growth rate.[83]
79.
The most common
form of the DCF model is the “Constant Growth” form. Under the Constant Growth DCF model, the
price of a stock is a function of the collective ROE required by investors,
which is determined as the sum of dividend yield and growth.[84]
80.
The Commission
has relied upon the “constant-growth” form of DCF (rather than any one of
several variations of the DCF method) in a number of recent rate cases. The
“two growth rates” DCF (TGDCF) model that Dr. Amit principally relied on in
this rate case is an extension of the constant growth rate DCF model.[85] That two-stage DCF is a reasonable method for
determining ROE where the short-term projected dividend growth rates for a
company may not be expected to continue in the long-run.[86]
81.
TGDCF analysis
uses a short term growth rate for the first five years of the period of
analysis, and a long-term growth rate for years six to infinity. The long-term growth rate reflects the
sustainable growth in value.[87] TGDCF assumes that for a relatively short
period earnings and dividends may grow annually at a different rate than the
long-term sustainable growth rate, and at the end of this short period, both
earnings and dividends will grow at a constant, sustainable annual rate.[88] Use of the TGDCF is a more accurate method of
determining present value when companies in the comparison group have projected
short-term growth rates that are unsustainable in the long run.[89]
E.
The Company’s
ROE Recommendation.
1.
Summary of the
Company’s ROE Recommendation.
82.
Minnesota Power
proposed an ROE of 11.25% based on Dr. Morin’s analysis. He conducted a number of studies, relying
primarily on a Constant Growth DCF analysis, which initially resulted in mean
ROE figures of 10.78% and 10.82%. Dr.
Morin also incorporated the results of his Capital Asset Pricing Model (“CAPM”)
analysis to arrive at his ROE figure.[90] MP contends that Dr. Morin’s analysis corrected
flaws in the various modeling approaches and accounted for a significant
increase in investor risk based primarily on the Company’s customer mix.
83.
Dr. Morin
examined a group of investment-grade dividend-paying utilities designated as
“integrated” utilities by S&P. These
companies all possess electricity generation, distribution, and transmission
assets, based on Value Lines’ SIC code. Dr.
Morin applied screens to exclude foreign companies, private partnerships,
private companies, non dividend-paying companies, companies with market
capitalization of less than $500 million, and companies below
investment-grade. At least 50% of the revenues
of the remaining 29 companies (Integrated Electric Utility Group) came from
regulated electric utility operations.[91]
84.
Dr. Morin
applied the DCF analysis, rather than the TGDCF analysis, to the Integrated
Electric Utility Group using the growth forecast published by Value Line and
arrived at an estimate of equity costs of 10.2% for the group. With recognition of flotation costs, Dr.
Morin’s cost of equity estimate rose to 10.4%.[92]
85.
Dr. Morin did
the same analysis using the consensus analysts’ earnings growth forecast
published by Zacks. His result was a
cost of equity for the adjusted Integrated Electric Utility Group of 11.3%,
unadjusted for flotation cost. Dr. Morin
then added flotation costs to bring the cost of equity estimate to 11.6% under that
analysis.[93]
86.
Dr. Morin also applied
the Value Line and Zacks forecasts to the electric utilities that make up
Moody’s Electric Utility Index (less one member of the Moody’s Group for which
no forecast was available). Those DCF
analyses resulted in an estimated cost of equity of 10.9% for the Moody’s
Group. When he adjusted the Value Line
forecast for flotation costs, the cost of equity rose to 11.1%. When he adjusted the Zacks forecast for the
Moody’s Group for flotation costs, the cost of equity rose to 11.0%.[94]
87.
The following
table summarizes Dr. Morin’s ROE results for his two groups:
|
|
|
|
Study Type |
Derived ROE |
|
CAPM |
11.2% |
|
Empirical CAPM |
11.5% |
|
Risk Premium Electric |
10.5% |
|
Allowed Risk Premium |
10.1% |
|
DCF - Vertically Integrated
Elec Utilities - Value Line Growth |
10.4% |
|
DCF - Vertically Integrated
Elec Utilities - Zacks Growth |
11.6% |
|
DCF - Moody’s Elec Utilities -
Value Line Growth |
11.1% |
|
DCF - Moody’s Elec Utilities -
Zacks Growth |
11.0% |
88.
The mean and
midpoint of Dr. Morin’s analyses was 10.9%.
From that number Dr. Morin added an additional 25 basis points (.25%) to
the ROE figure to account for his perception of increased risk based “on
utility bond yield spread differentials between A-rated and Baa-rated bonds, on
observed beta differentials, and on [his] professional judgment.” The resulting ROE that he proposed was 11.15%.[96] Based on his capital structure analysis, Dr.
Morin proposed an overall ROR for Minnesota Power of 8.69%. CITE.
F.
The OES’s ROE
Recommendation.
1. Summary of the OES’s ROE Recommendation.
89.
The OES proposed
an ROE of 10.74% based on Dr. Amit’s analysis. Based on his capital structure
analysis, the OES proposed an overall ROR for Minnesota Power of 8.32%.[97]
2. The OES’s
Comparable Groups.
90.
Dr. Amit, testifying
on behalf of the OES, prepared two comparison groups to analyze MP’s ROE
requirement. One group was comprised of
electric companies[98] and the
other comprised of electric/gas companies.[99] For the electric company group, Dr. Amit
selected domestic electric utilities that: a) were listed in the Compustat
database of April 2008 (provided by S&P) and b) met two conditions: their
primary Standard Industrial Classification (SIC) code was 4911 (electric
utilities), and their shares were publicly traded on one of the stock
exchanges.[100]
91.
Dr. Amit
screened the electric companies that remained to eliminate those that did not
have regulated retail electric services as their primary business, those
lacking a bond rating, those with bond ratings outside the range BBB- to A
(ALLETE’s rating is BBB+), those without dividends or a reliable dividend
history, those whose regulated revenues were less than 60 percent of total
company revenues, and those whose beta and standard deviation varied by more
than one standard deviation from the group’s mean. Those screens left 21 companies that Dr. Amit
identified as the Initial Electric Comparison Group (“IECG”).[101]
92.
Dr. Amit applied
several risk measures to the IECG to assess the similarity of each company to
MP. He applied a DCF analysis to each
company to identify those whose ROE deviated significantly from the average
required rate for the group. To address
companies whose ROE was demonstrably different from MP’s likely ROE, Dr. Amit
applied an ROE screen. He described his
rationale for that screen as follows:
Under basic
financial and economic principles, companies with similar investment risks are
expected to have similar required rates of return. Therefore, after performing
a DCF analysis on my IECG group, I eliminated from this group the companies
with required rates of return that deviated significantly from the group’s average
required rate of return. [102]
93.
The screen that
Dr. Amit had used was “any company for which the DCF analysis resulted in a
required rate of return that deviated by more than one standard deviation from
the IECG’s average required rate of return.”[103] MP objected to this screen as having “biased
the selection of comparable companies by prejudging their DCF results.”[104] But the ALJ finds that criticism to be
unjustified. That screen excluded
companies whose rate of return is either overly high or overly low, since an
unusual ROE would reflect conditions that made that company fundamentally
dissimilar to MP. Fourteen companies
were left after that screen was applied, and those companies were identified as
the Final Electric Comparison Group (FECG).
Dr. Amit applied similar screens to arrive at his Final Combination
Comparison Group (“FCCG”).[105]
94.
Dr. Amit
concluded that the companies in his comparison group required the use of TGDCF
to accurately determine present value. In
the recent Otter Tail Power rate case
proceeding, the Commission accepted both the OES’s use of the TGDCF method and
the results the OES obtained.[106] Applying
that analysis to his two comparison groups in this proceeding, Dr. Amit arrived
at the following ranges of ROEs:[107]
95.
The ALJ finds
that Dr. Amit’s comparable groups are appropriate for use in calculation of an
ROE for Minnesota Power. By contrast, the
ALJ finds that Dr. Morin’s comparable groups were not closely tailored to match
MP’s financial profile. Dr. Amit’s
approach to comparable groups is also consistent with the Commission’s
longstanding approach to calculating ROE.
96.
Although Dr.
Morin’s and Dr. Amit’s ROE analyses differ in a number of particulars, Dr.
Morin considered that most of those differences were “minor” and that the only
thing about which they fundamentally disagreed was that Dr. Morin “added an
additional risk premium” to his estimated ROE “in order to recognize MP's
peculiar risk circumstances.”[108] One of the major issues in setting an
appropriate ROE is therefore whether or not an additional risk premium is
warranted.
97.
The DCF model is
based on long-term growth and assumes
cash flows in perpetuity and a constant dividend payout ratio. Dr. Amit addressed the constant growth bias
in the DCF model by using the TGDCF approach.
But Dr. Morin maintained that the DCF model would undervalue ROE
requirements because “we have entered an era where investors are repricing
risk. We are witnessing a fundamental
shift in risk aversion on the part of investors.”[109]
98.
Minnesota Power argues
that its risks are higher than the normal electric utility for several reasons. One risk that Dr. Morin identified was the
Company’s reliance on sales to a few
extremely large industrial customers concentrated in the volatile taconite and
paper industries. He also
believed that Minnesota Power’s planned construction program magnified that
risk. Finally, Dr. Morin also identified
MP’s reliance on coal-based generation as another risk factor because of
uncertainty about the potential for future regulations to reduce greenhouse gas
emissions.[110]
99.
Minnesota Power
also relied the fact that in its 1994 ratemaking proceeding, the Commission had
expressly adjusted for MP’s unusual risk factors.[111] But Dr. Morin acknowledged that the increase
in the Company’s equity ratio from its 1994 capital structure significantly
mitigated the risk that may have existed in 1994.[112] Dr. Morin compared the beta of ALLETE to that
of the companies in Dr. Amit’s comparison groups. Dr. Morin maintained that he adjusted those
betas for capital structure differences, to “purge” the financial risk from
total risk and found that Minnesota Power’s unlevered beta (which reflects
business risk only) remained considerably higher than the average of Dr. Amit’s
comparison groups.[113]
100.
Beta is basically
a measure of the volatility of a stock relative to the volatility of the market
as a whole.[114] Dr. Amit compared the volatility of returns
of MP and ALLETE and found that ALLETE volatility was much higher than MP. On the other hand, Dr. Morin’s analysis does
not incorporate that step. Rather, Dr.
Morin concludes, in substance, that since ALLETE is riskier than a typical
utility electric utility, then it follows that MP is riskier. In other words, Dr. Amit disagreed with Dr.
Morin’s failure to isolate MP’s risk from ALLETE’s in his calculations.[115] Moreover, Dr. Amit used, as a reasonable
measure of volatility, the standard deviation of revenues and rates of return
to substitute for beta (which MP lacks).
He made that calculation with an adjustment for differences in annual
revenue between ALLETE, MP, the FECG, and the FCCG. That comparison indicated that MP is less
risky than both comparison groups that Dr. Amit used.[116] Dr. Amit also noted that ALLETE’s BBB+ bond
rating (itself a measure of risk) is higher than that of any company in the
FECG. This necessarily means that MP has
less risk than any of the FECG companies.[117]
101.
Dr. Morin
identified concentration of large customers as another reason to adjust MP’s
ROE to account for greater risk. On the
other hand, Dr. Amit noted that Minnesota Power’s cost of electricity is among
the lowest in the country.[118] He concluded
that the loss of any particular customer load would result in MP having more
power to sell on the Midwest Independent System Operator (MISO) Day 2 market,
which did not exist at the time of the Commission’s 1994 MP Order. With an available market for MP’s excess
low-cost power, the potential for loss of a large customer due to economic
conditions does not constitute a basis for increasing MP’s ROE.[119]
102.
The ALJ
concludes Dr. Amit’s DCF analysis accounts for an appropriate level of any risk
posed by MP’s customer mix and other potential risk factors, particularly since
the Company has the option of selling available power through MISO. The ALJ also concludes that Dr. Morin’s
adjustment for risk unreasonably emphasizes investor risk in his ROE calculation.
H.
Flotation Cost
Adjustment.
103.
Dr. Morin indicated
that when a company issues additional shares of common stock, the increased
supply of common stock normally causes a downward pressure on the price per
share, and that based on various academic studies, a reasonable measure of the
relative price decline is about 1.5 percent.
Dr. Morin therefore maintained that an additional 1.5 percent in
flotation cost adjustments should be made to address that reduction in price.[120]
104.
Dr. Amit disagreed
with Dr. Morin’s additional flotation cost adjustment of 1.5 percent. While Dr. Amit acknowledged that numerous
financial studies indicate that market pressure costs may generally exist when
additional shares of common stocks are being issued, those market pressure
costs vary across companies and across market conditions. Dr. Amit concluded that an adjustment to the
flotation costs to reflect market pressure for Minnesota Power "is
warranted only if MP shows that such costs have existed for its public
issuances."[121]
105.
Minnesota Power
did not offer any empirical data to support that issuing new shares of ALLETE
stock has created the downward pressure on the price per share described by Dr.
Morin. Dr. Amit analyzed ALLETE’s price
behavior in comparison with the Dow Jones’ price behavior for 30 days prior to
ALLETE's public issues in 1993, 1998, and 2001.
On the days of ALLETE's public issues, its stock price went down by less
than the closing price of the Dow Jones. From that, Dr. Amit concluded that
there is no market pressure impact on ALLETE (and therefore on Minnesota Power). Dr. Amit also observed that the average price
of ALLETE stock declined by only an average of 0.15 percent during the 30 days
prior to ALLETE’s previous three stock issues.
Based on that data, Dr. Amit concluded that no market pressure
adjustment to flotation costs is warranted for Minnesota Power because historically
no significant market pressure on ALLETE's stock price has been observed.[122]
106.
In calculating flotation costs, Dr. Morin relied on empirical
studies of several different companies’ stock
issues in the range of $60 to $500 million that had shown the average direct flotation
costs for stock issues in that range to be between
3.5 percent to 5 percent. It was also Dr. Morin’s opinion that allowing for market pressure costs, as described
above, raises the flotation cost allowance to well above 5%.[123] On that basis,
Dr. Morin used a 5 percent flotation
cost figure.
107.
While the OES
agreed with much of MP’s DCF methodology, the OES contends that the flotation
costs that the Company used were too high.
Dr. Amit based his analysis on his conclusion that the correct flotation
costs were those associated with the historical issuance costs for ALLETE,
which averaged
3.61 percent. Using ALLETE’s costs in Dr. Amit’s ROE
calculation resulted in a reduction of seven basis points from the results
obtained by Dr. Morin for ROE.[124] The ALJ
concludes that Dr. Amit’s approach to calculating flotation costs is more
reasonable and reliable than Dr. Morin’s and further supports the OES ROE
result for calculating Minnesota Power’s revenue requirements.
108.
An important
part of the DCF analysis is determining an appropriate dividend yield. Dr. Morin and Dr. Amit differed in their
respective approaches to calculating dividend yield. Dr. Morin argues that Dr. Amit’s approach
understates the proper dividend yield,[125] while Dr.
Amit argues that Dr. Morin’s approach overstates dividend yield. Both
economists agree that the annual DCF model states that the expected dividend to
be used is Do
(1+g), where Do is the current annual dividend rate and g is
the annual expected growth
rate, and that expected dividend
assumes that Do, the annual dividend rate at the time the DCF is performed, would increase by g a year
from the date at which the DCF was
performed.[126]
109.
Dr. Morin maintains that the normal methodology of the annual DCF
model should be employed, that the appropriate expected dividend to be used in
that analysis is the current dividend times (1 + expected growth rate)[which is
(Do
(1+g)], and used an example with the
result is 4%(1 + .06) = 5.24 percent (differing from the
results of Dr. Amit’s method by 12 basis points).[127]
110.
On the other
hand, Dr. Amit maintained that the appropriate dividend calculation would use
one-half of the expected growth rate (Do (1+g/2)). It is his opinion that the timing of dividend
payments results in an overstatement of expected dividend yield using the Do (1+g) formula as proposed by Dr. Morin, and that
overstatement translates into a higher ROE award than is warranted by the
actual financial results forecast by the DCF model.[128]
111.
Dr. Amit reasons that conceptually the
appropriate dividend to be used in the DCF analysis may be the annual dividend
rate at the beginning of the next period (year). However he argues that if all
the companies in his group were to increase their dividend in the
second quarter of 2008, the appropriate dividend rate
to be used in the DCF analysis would be the annualized dividend based on the
second quarter of 2008, increased by the growth rate, “g.” But the companies in his group might raise
their dividend rates in different quarters. Moreover, for some of those
companies the current dividend rate might not change over the next one or two
quarters but then increase in subsequent quarters. For
other companies, the dividend yield may remain constant for three or four
quarters. Thus, in Dr. Amit’s opinion, a reasonable estimate of the expected
annual dividend yield for the IECG can be derived as the most current
annualized dividend yield x (1 + 0.5g), where “g” is
the expected growth rate.[129]
112.
The ALJ concludes that Dr. Amit’s more nuanced approach to
dividend yield is a more accurate estimation of the Company’s dividend yield
and that use of the formula that Dr. Amit proposes for estimating expected
dividend—i.e., the Do
(1+g/2) formula —is therefore
appropriate.
J.
Updating of Stock
Price Data.
113.
In his initial
DCF study, Dr. Amit calculated the dividend yield by using thirty-day average
closing stock prices for his comparison groups from the period May 8 through
June 9, 2008.[130] Dr. Amit emphasized the importance of using
the most current price per share because it “incorporates all publicly
available information.” He also stated that “non-recent historical prices
should be avoided in calculating the dividend yield.”[131] Dr. Morin testified that the DCF analysis that
Dr. Amit employed generally presents difficulties because utility company
historical data have become less meaningful for an industry in a “state of
change” and past earnings and dividend trends are not necessarily indicative of
the future earnings.[132] It is in the context of that belief that the
Company argues that Dr. Amit’s failure to update the prices of the stocks he
relied on in his surrebuttal testimony is fatal to his DCF results and results
in artificially low dividend yields, which are completely unrealistic in
today’s markets.[133]
114.
In response, Dr.
Amit acknowledged that his uniform practice in other rate cases has been to
update his dividend yield analysis with his surrebuttal testimony, but in this
case he declined to do so because of the abnormal conditions in equity markets
that have prevailed since June 9, 2008.[134] It is Dr.
Amit’s opinion that although stocks of utilities are less sensitive to market
volatility, they are not immune from that volatility, and that rather than
using data on dividend yield from a period when markets are volatile, it is
more prudent to rely on an historical period for data.[135]
115.
In Dr. Amit’s
opinion, the unpredictable ways in which market conditions have moved are not
likely to continue throughout the test year (and even less likely to last
through the effective life of the rates set in this proceeding).[136] As Dr. Amit
also noted, the federal government has taken action toward adopting direct
stimulus spending of over $700 billion to address the current conditions in the
national economy.[137] In this proceeding, Minnesota Power has not
shown that the recent market volatility is likely to continue over the long
term or shown what particular data would more accurately reflect future market
conditions. One can only speculate about
the data in a fluctuating market that might correctly predict the direction of
the market when the dramatic swings stop.
When balancing the interests of shareholders and ratepayers, the burden
is on MP to demonstrate that its ROE proposal results on just and reasonable rates. In this instance, using the stable market
information of the recent past is superior to relying more recent data in a
highly volatile market. The ALJ
therefore concludes that Dr. Amit’s conclusion concerning a just and reasonable
ROE is not less reliable because it is not based on updated market information.
|
|
Percent of Total |
Cost |
Weighted Cost |
|
Long Term Debt |
47.89% |
5.68% |
2.72% |
|
Common Equity |
52.11% |
10.74% |
5.59% |
|
Total |
100.0% |
|
8.31% |
117.
The following treatment
of wholesale margins for rate making purposes is divided into three subject
areas: asset-based margins, non-asset-based
margins, and ancillary service market (“ASM”).
Each of those areas raised issues that will be addressed separately.
118. Asset-based margins result from the Company’s sale of
energy generated by its own facilities that is not needed to serve its retail
needs. The cost of assets used to
generate the energy sold into the market must be included in rates, since
ratepayers are incurring the cost of the assets generating those margins. The principle is well-accepted that treatment
of asset-based margins must benefit the ratepayers.[138]
119.
MP maintains a
total of 1,408 MW generating capacity; 1,246 MW of the total is
coal-fired. The remainder is a mix of
purchased steam, biomass, and hydro-generated electricity.[139] Minnesota Power described how it now sells
energy it generates in the wholesale market:
Minnesota Power’s participation in the wholesale
electric market changed significantly on April 1, 2005, with the beginning of
the MISO Day 2 Energy Market. While Minnesota Power’s customers still receive
the benefit of the Company’s low-cost generation, the methods used to allocate
the costs of energy within the MISO footprint are now significantly different.
Prior to the MISO Day 2 Market, Minnesota Power generated energy to serve its
retail load, selling any excess energy to wholesale energy purchasers directly
or through traders, and buying any energy shortfall it experienced from
wholesale energy sellers directly or through traders. After the advent of MISO
Day 2, however, Minnesota Power and all other generators must offer all the
energy they generate into the Day 2 Market for sale, and must purchase all of
the energy needed to serve its retail load out of the Day 2 Market. This market
structure is intended to allow more efficient and effective use of generation
and transmission resources, and eliminate the higher, and hidden, costs
associated with discriminatory and self-dealing energy transactions.[140]
1.
Asset-Based
Margin Forecast.
120. In its initial filing, Minnesota Power forecast
Minnesota Power’s customer requirements are projected
to continue to expand. In addition to
continued commercial and residential growth, about 100 MW in industrial
expansions is forecast to occur. A key resource “addition” will be the
termination of a 175 MW off-system sale to GRE (see Table on page 6 of this
testimony) in April 2010;
121.
The OES objected
to the Company’s $22 million forecast as too low. It based its objection on the fact that the
energy for those wholesale contracts came from Minnesota Power’s Boswell and
Taconite Harbor generation units, and that Minnesota Power therefore still has
the low-cost power from those plants available to continue selling at wholesale
to other utilities and energy purchasers through bilateral contracts or the
MISO Day 2 wholesale energy market.[144]
122.
Minnesota Power indicates
that it executed the SMMPA and GenSys contracts when its purchase of the
123.
Minnesota Power further
indicates that it entered into the SMMPA and GenSys contracts sales during
times of extended industrial downturns that made it prudent to sell power at
wholesale to offset retail customer revenue requirements as much as
possible. The Company asserts that when
the regional economy and Minnesota Power's industrial customers experienced a
dramatic recovery and expansion beginning in 2004-05, the energy requirements
to serve the Company’s customers also increased significantly. When those committed off-system sales ended in
late 2007/early 2008, Minnesota Power was routinely purchasing 100-150 MW of
energy on a daily basis through both bilateral and MIS0 Day 2 market
transactions to meet its retail customers' needs. With off-system sales ended, the Company
contends that those 100-150 MW of purchased energy are now used to meet retail
demand. Minnesota Power argues that this increased use of self-generation for
retail reduces the need for Minnesota Power to access the energy markets for
system needs, lowering costs for ratepayers.[147]
124.
Additionally, the
Company asserts that, of the energy market purchases it had been making, it
needed 75 MW to replace the 75 MW capacity of the Company’s Boswell 1 and 2
units that had been dedicated to meet the supply requirements under the Alliant
wholesale contract.[148] MP also indicates that it needed the balance
of the purchased power to meet a portion of the requirements of the SMMPA and
GenSys wholesale contracts, as well as the Company’s retail load demands.[149]
125.
In addition to
its existing generating capacity, MP has also identified a number of renewable
sources of electricity generation that are expected to be available by
2010. To support those generation
sources, MP plans to add a 170 MW peaking plant powered by natural gas. These resources are intended to help MP meet
its statutory obligation to increase its renewable generation sources.[150] MP is required to provide 25 percent of its total
retail electric sales to retail customers in
126.
Minnesota Power contends
that under “relevant statutes, Commission orders, and prudent utility practices”
the Company is required to dispatch its generation on an economic basis that
serves native load first.[152] Minnesota Power maintains that this obligation
is met by “stacking” its generation from the lowest cost to the highest cost,
using the lowest cost available sources from the stack first to serve retail
customers.[153] MP has not explained how its practice is
consistent with the 25% renewable requirement which is in place now with lower
compliance levels.
127.
Minnesota Power
presented retail sales forecasts that were, in some cases, significantly lower
than historical customer usage in 2007.
The ALJ has recommended that the Commission accept the OES’s higher
forecasts for residential and general service classes[154] and, with
one significant exception, the OES’s recommendation for somewhat higher
forecasts for Large Light and Power customers.[155] But the ALJ agreed with the Company’s forecast
for lower test year usage by Large Power customers than what those customers
used in 2007.[156] As previously noted, the Company’s large
power customers have accounted for as much as 70 percent of its retail sales,[157] and it was extended
industrial downturns and resultant usage reductions by those customers that
prompted the company to enter into the power purchase agreements which are now
expiring. In other words, Minnesota
Power has not identified any retail customer sales growth that would preclude
energy from its expiring wholesale contracts from being sold in other wholesale
transactions that will generate significant asset-based margins consistent with
MP’s historical sales.
128.
The ALJ
therefore concludes that it is unlikely that the Company will need all of the
power made available by the expiration of the SMMPA and GenSys contracts to
serve its retail customers. Minnesota
Power has not identified any basis for its lower forecast of asset-based
margins beyond the possibility that the three identified wholesale contracts
might not be renewed. This contrasts
sharply with the consistent history of MP’s asset-based margins since the MISO
Day 2 market began operation.[158] The ALJ therefore concludes that Minnesota
Power’s asset-based margin forecast is too low, and that the Commission should
accept the OES’s test year estimate of asset-based margins.
2.
The OES’s
Approach to Calculating a Fixed Credit for Asset-Based Margins.
129.
The OES has proposed
calculating a fixed credit for asset-based margins to be applied through base
rates. To calculate the credit, the OES
recommends using the Company’s average actual asset-based margins, within the Minnesota
Jurisdiction, from 2005 through 2007.
During that period, Minnesota Power realized positive margins of
$20,529,898, $42,609,138 and $41,736,879, respectively, using Minnesota
Jurisdictional amounts. The resulting
average would therefore be $34,958,638.[159]
130.
Both the OES and
MP have proposed that the fixed credits be recognized through base rates. MCC proposed that a fixed credit to base
rates not be used, suggesting instead that these margins could be passed to
rate payers through the fuel clause adjustment (FCA). MCC maintains that this approach addresses
variablity in the amounts of these margins and directly credits customers based
on actual use of electricity, thereby eliminating cross-class subsidies.[160] However, Minnesota Power contends that this
approach would be inconsistent with the Company's current practice, which the
Commission has already approved. The
Company notes that if the Commission were to adopt the FCA credit proposal,
then an exception would be necessary to address a loss of load, and the details
of that exception would require further development.[161]
As the Commission recently held, “In sum, the Commission will set Otter Tail’s
base rates on the assumption that Otter Tail’s costs are offset by $5.41
million in revenues from Otter Tail’s asset-based wholesale margins.”[162] The ALJ concludes that recognizing the credit
through base rates is appropriate.
131.
The OES acknowledged
that adoption of the $34,958,638 three-year average, less the $22,057,477
amount MP proposed in its test year, would result in a $12,901,161 adjustment,[163]
and it offered the following justification for such an adjustment:
·
In OTP’s recent
rate case the Commission approved a four year average of asset-based margins.
For MP, the OES considers a three year average to be more reasonable largely to
insure a representative amount of asset-based margins is included in the
test-year. Also due to the discretion MP had, and OTP did not have, in deciding
when to file its rate case. The Commission required OTP to file a rate case by
October, 2007, whereas MP had a choice regarding when to file its rate case.
·
In assessing the
effects of MP’s additional flexibility in deciding when to file its rate case
on its proposed costs and revenues in this rate case, the OES notes that MP
chose a forecasted test year which appears to be designed to minimize MP’s
expected revenues and thus result in higher rate increases. For example, some
of MP’s wholesale contracts have recently expired at the end of 2007 and early
in 2008, however as noted above those
revenues from MP’s generators will still be available to MP through new sales.
In addition, as shown on MP’s Schedule A-3, page 1 of 2, Other Operating
Revenue, which went from $116 million in 2007 based on actual information to
$74 million in the test year. By contrast, OTP used a historical test year in
its most recent rate case. Thus, the
Commission needs to ensure that MP’s ratepayers are not prejudiced by MP’s
choice of a test year, and ensure that the test year amount of asset-based
margins is a reasonable representative amount.
·
The Commission
noted in the OTP rate case that there should be less reliance on pre-MISO Day 2
asset-based margins. OES has noted in the Annual Automatic Adjustment Reports
(AAA Reports or Annual Electric Fuel Reports) that asset-based margins
increased with the start of the MISO Day 2 market. OES has also noted in the most recent Xcel
rate case, in the most recent OTP rate case and in the current MP rate case
that asset-based margins clearly increased after the start of the MISO Day 2
market in April 2005
·
One of the
benefits of the MISO Day 2 market is the efficient sale of excess generation
not needed for retail into the MISO Day 2 market. Since ratepayers are paying
for all of the administrative and other costs of the MISO Day 2 market, they
should also be given a reasonable level of the benefits for the sale of excess
generation.[164]
132.
The ALJ
concludes that using a three-year average of actual asset-based margins is the
most reasonable approach for determining a representative level of asset-based
margins in the test year setting the Company’s rates. Moreover, selecting 2005, 2006, and 2007, as
the period for averaging historical asset-based margins is a reasonable and
appropriate method for determining asset-based margins for the test year. By contrast, Minnesota Power’s test year
forecast is $6 million below the actual margins for the last full year for
which data is available. In summary, the
OES approach to forecasting asset-based margins for the test year is reasonable
and results in an appropriate level of forecast revenue to apply as a fixed
credit to be applied to base rates.
133.
Non-asset-based
margins result from the unregulated purchase and sale of energy for non-retail
purposes. Typically, the same utility marketers, sharing common equipment,
handle both asset-based margins and non-asset-based purchase and sale
transactions. The Commission
requires some attribution of wholesale non-asset-based margins to ratepayers to
the extent that ratepayers funded services allowed the utility to receive those
wholesale margins.[165] It is therefore appropriate for non-asset-based
margins to cover their incremental costs and provide a reasonable contribution
towards common costs. However, Minnesota
Power is not performing non-asset-based trading transactions except for limited
virtual transactions to hedge prices between Day-Ahead and Real-Time markets on
behalf of retail customers.[166] In other
words, the Company’s only participation in those markets is to ensure that it
can provide least-cost supply to its customers. Where those purchases exceed requirements, the
excess is sold to the market and the margins obtained are allocated to the FAC.[167]
134.
Nonetheless, the
OES proposes that a $300,000 cap on ratepayer responsibility be imposed on any net
losses that might arise from those virtual transactions.[168]
Because the Company reported net losses
of $160,000 from virtual transactions in its FYE 2007 Annual Automatic Adjustment report, the OES
believes that imposition of a $300,000 cap would function as a protective
measure to ensure that MP’s cost of
hedging does not become too high.[169] The proposed cap would apply to net losses arising
from hedging transactions over a one-year period. Any losses in excess of the cap would not be allowed
for recovery via the FCA. Therefore, any
losses in excess of the cap would be incurred by the Company’s shareholders. The OES also recommends that language
specifying that no costs from speculative trading will be allowed in the FCA or
otherwise be charged to ratepayers be incorporated in the Company’s rates
tariff to ensure that its future virtual trading will never be speculative.[170]
135. Minnesota Power opposes the proposed cap, arguing
that it could possibly impede transactions that are beneficial to ratepayers.[171] Where energy is sold in day-ahead virtual
transactions and purchased back in real time, a margin is created. In the Company’s experience, the real-time
market is lower priced than the day-ahead market more than 50% of the time, and
therefore most of the time the margin created is positive. The Company indicates that it only uses virtual
transactions in the MISO Day 2 energy market on a limited basis and these
transactions are a very effective way to move energy from the day-ahead market
to the real-time market (or vice versa) to minimize ratepayer costs.[172] Minnesota Power treats the margins from
virtual transactions in the same manner it treats margins created by excess
purchases, that is, all margins plus or minus are allocated on an energy (MWh)
basis to all sales. Because of the opportunity to generate positive margins
from prudent virtual transactions, Minnesota Power argues that its continued
use should not be limited.[173]
136.
The ALJ
concludes that a cap on losses from the Company’s virtual transactions is a
reasonable measure to prevent ratepayers from having unlimited exposure to such
losses, and that the proposed $300,000 cap, which is nearly twice the amount of
the Company’s documented historical losses, provides the Company with ample
room to engage in transactions that are beneficial to ratepayers. The ALJ also concludes that tariff language
specifying that no costs from speculative trading will be allowed in the FCA or
otherwise be charged to ratepayers is a reasonable measure to prevent any
future speculative trading.
C.
Ancillary
Service Market Margins.
137. Both Minnesota Power and the OES noted that MISO
expects to open a new market, called the Ancillary Services Market (“ASM”). Ancillary services “help ensure that there is
sufficient generation to match loads on the transmission system instantaneously
to preserve service reliability.” The
ASM market is intended to assist utilities to respond quickly to system
fluctuations in output as well as load by providing generation assets. [174]
138.
ASM margins
include margins from spinning reserves, regulation reserves and supplemental reserve
requirements. MP described its expected
involvement with ASM margins as follows:
Minnesota Power will be participating in a joint filing to address ASM
in early May. With the start of the ASM,
139.
Because ASM has not begun operationally and because the ASM
proceeding is currently before the Commission,[176] the OES
recommends that the issues raised in the ASM
Docket should be addressed in that separate docket before developing an ASM
margin sharing proposal for Minnesota Power. In that regard, the OES recommends
that the Commission expressly state in its final decision in this rate case
that the Company’s rates may be revised in the future to address the Company’s
ASM costs and revenues.[177]
140.
Since Minnesota
Power has not addressed the ASM issue in this proceeding, the ALJ recommends that
the Commission adopt the OES’s ASM recommendations.[178]
141. Following the enactment of the 1990 Clean Air Act
Amendments of 1990 (CAA),[179] the U.S
Environmental Protection Agency (EPA) adopted both the Federal Clean Air
Interstate Rule (CAIR),[180] requiring SO2 and NOx
reductions, and the Clean Air Mercury Rule (CAMR),[181] requiring
reduction of mercury emissions. Under
CAIR and CAMR,
142.
Minnesota
Power’s plans for trading emissions allowances are:
Minnesota Power’s SO2 allowance
allocations from the EPA are expected to exceed actual emissions through 2015,
creating a surplus of allowances that could be sold to market or carried
forward to future years to the benefit of ratepayers. The Company’s emission allowance strategy
will seek to maximize the value of its SO2 credit inventory on
behalf of ratepayers. In contrast, the
Company expects its annual NOx emissions to exceed the annual EPA
allocation in the same timeframe, creating annual deficits. The Company will seek to minimize the
compliance costs associated with the projected NOx deficits by
continuing to evaluate the addition of emission reduction equipment to
Minnesota Power’s system, by strategically purchasing allowances from the
market, and by maximizing the value returned to ratepayers for other emission
credits such as SO2. [182]
143.
SO2
allowances produce two potential revenue streams for the Company. First, there
can be proceeds from the EPA when it withholds allowances from utilities and
sells those allowances in the market.
Second, a utility like Minnesota Power can sell some of its allowances
either directly or through a broker.
Minnesota Power projected $2,695,000 in test-year revenues from
anticipated sales of surplus SO2 emission allowances. That amount represented half of the projected
2009 sales of $5,195,000. The Company
indicated that it did not include any budgeted revenues for the first six
months of the test year because it did not plan to sell any surplus allowances
in 2008.[183] Of the total projected figure, $195,000 of
the budgeted amount represented anticipated proceeds from the EPA, and the
remaining projection represented potential direct sales.[184]
144. Since 1994, Minnesota Power has received $4,033,237
from surplus SO2 allowance sales, but none of these revenues have
been applied to benefit ratepayers.[185] In this proceeding, Minnesota Power proposes
to eliminate all revenues and expenses associated with its SO2 and
NOx allowances from the test-year but to include them in a future
separate cost recovery rider.[186] The Company maintains that the market for allowances
is volatile, and that allowance prices have been fluctuating significantly,
making establishing the proper amount of test year revenues difficult.[187] The Company argues that those difficulties are
obviated if revenues and expenses simply flow to ratepayers when incurred. MP therefore proposes to address SO2
allowance revenues and expenses through a rider, rather than through base
rates, and it filed a proposed rider with its initial Petition.[188]
145.
The OES argues that over the last fourteen years, the Company has
been predictably receiving some net revenues from the EPA’s sales of its SO2
allowances.[189] The OES considers the $195,000 that the
Company expects to receive from the EPA in 2009 to be representative of that
annual component of the Company’s allowance income stream; it therefore
recommends a test-year
adjustment increasing MP’s test-year revenues by $195,000, or $165,538 on a
146.
The ALJ
concludes that the OES proposal is a reasonable approach to ensuring that
ratepayers receive the benefit of revenues the Company obtains from sales of
its SO2 allowances—that is, revenues and expenses that are
reasonably predictable are given rate base treatment, while revenues and
expenses that are likely to be volatile are covered in a rider. The ALJ expresses no opinion on whether the
Commission should address allowance brokerage sales in an existing rider or in
a new one.
MISO divides its operations into categories, including “Day 1”
operations (dealing with security, outages, tariffs, transmission-line
congestion and energy imbalances, billings and settlements, and market
monitoring) and “Day 2” operations (implementing a competitive wholesale market
for electricity, including locational marginal pricing and financial
transmission rights).[190]
148.
MISO charges
member utilities like Minnesota Power various administrative costs associated
with the MISO Day 2 market. The
Commission has determined that utilities, including Minnesota Power, can
recover MISO Day 2 costs through the FCA, with the exception of MISO Schedule
16 and 17 charges. Schedule 16 and 17
charges were determined to be administrative and not energy in nature. For that reason, Schedule 16 and 17 costs are
recovered through base rates rather than through the FCA. In an Order entered on December 20, 2006 (“MISO Day 2 Order”), the Commission
prescribed the following treatment for MISO Schedule 16 and 17 deferred costs:
3. In its
next rate case a utility may seek to recover Schedule 16 and 17 costs at an
appropriate level of base rate recovery. The utility may not increase rates to
recover MISO administrative costs unless the costs were prudently incurred,
reasonable, resulted in benefits justifying recovery and not already recovered
through other rates. However, a utility may seek to recover Schedule 16 and 17
costs and associated amortizations through interim rates pending the resolution
of a rate case, subject to final Commission approval. [191]
149.
In this rate
case, the Company’s filing seeks test-year recovery of MISO Schedule 16 and 17
costs relating to two different periods:
1) The test-year period of
July 1, 2008, to June 30, 2009, in the amount of $1,326,227 (
2) Deferred
costs for the period April 2005, to June 30, 2008 in the amount of $1,490,664.[192]
150.
The Commission
described this cost recovery mechanism as follows:
Each petitioning utility may use deferred accounting
for MISO Schedule 16 and 17 costs incurred since April 1, 2005 [the start of
Day 2]. Each utility may continue
deferring Schedule 16 and 17 costs without interest until the earlier of the
utility’s next electric rate case or March 1, 2009.[193]
151.
The OES agrees
that MP has demonstrated ratepayer benefits arising from participation in the
MISO Day 2 market and that MP has allocated the current test-year period
administrative costs between shareholders and retail ratepayers in a fair and
reasonable manner.[194] The OES therefore accepts MP’s proposal for
treatment of current test-year MISO Schedule 16 and 17 costs, and the test year
amount of $1,326,227 is not in dispute between MP and the OES.
152.
MP calculated
the recovery amount for deferred costs by taking its budgeted total deferred
costs of $4,471,991 on a Minnesota jurisdictional basis and proposing that the
deferred costs be amortized on a three-year schedule and be given rate base
treatment.[195] The OES noted that the amount of deferred
cost that the Company originally sought was its budgeted amount, not the
historical amount; the OES also argued that the Company should use a five-year
amortization schedule and that deferred MISO Schedule 16 and 17 costs should
not be given rate base treatment.[196]
153.
The Company
subsequently agreed that the total sum of the deferred costs under
consideration should be the amount that MP actually incurred, resulting in a
reduction of $48,511 and total deferred cost of $4,423,480. The Company and the OES do not agree on
whether a five-year or three-year amortization period should be used, or
whether the deferred amount should be afforded rate base treatment.[197]
A.
Schedule 16 and
17 Deferred Costs Amortization Period.
154.
Selecting a
reasonable amortization period is important because ratepayers will continue to
pay the deferred MISO Day 2 costs in base rates until MP files its next rate
case and may overpay significantly if MP does not file another rate case within
three years. For example, the Company
filed its last rate case fourteen years ago, well beyond the amortization
periods for rate case expenses that were approved in the 1994 rate case.[198]
155.
The OES
maintains that the amortization period for deferred MISO Schedule 16 and 17
costs should be set at a level which reasonably reflects when MP is likely to
file its next rate case.[199] The OES used
MP’s history of filing rate cases to determine that MP’s average time between
rate cases is five years. The OES
concluded that the appropriate amortization period for these expenses is five years.[200]
156.
MP contends that
the Commission already has established a three-year amortization period for deferral
of MISO Schedule 16 and 17 costs by ordering that by March 1, 2009, utilities “shall
begin amortizing the balance of the deferred Day 2 costs through March 1, 2012, unless and until the utility has a
rate case addressing the utility's proposal for recovering the
balance.”[201] However, the
OES contends that the Commission was not
establishing the amortization period to be used in utility rate cases, nor was
it binding itself in this or other future rate cases.[202]
157.
The Commission’s
MISO Day 2 Costs Order does not
establish an amortization period for these costs. The ALJ has recommended elsewhere in this
report that the Commission accept Minnesota Power’s proposal to file another
rate case within three years by directing the Company to do so. If the Commission accepts that
recommendation, it should allow
B.
Propriety of
Adding Schedule 16 and 17 Deferred Costs to Rate Base.
158. In addition to recovering deferred MISO 16 and 17
costs through an amortization schedule, as discussed above, Minnesota Power has
proposed to include in its rate base the unamortized balance of those deferred
MISO 16 and 17 costs. Based on
information supplied by the Company, this would increase the rate base by
$2,948,987.[203]
159.
Minnesota Power
argues that because those costs originate as an expense, the delay between
payment and recovery warrants treating the unamortized amount as an investment,
like any other prepaid expense. The
Company therefore argues that it should be able to earn a return on the amounts
not paid in the test year.[204] Additionally, MP argues that the Commission
has not ruled out placing Schedule 16 and 17 Deferred Costs in a utility’s rate
base. It relies on the fact that
although the MISO Day 2 Cost Orders did not allow utilities to include interest
during the period when the costs were being deferred, the Order did not
preclude the inclusion of those unamortized deferred costs in rate base in a
rate case.[205]
160.
The OES argues
against rate base treatment of deferred Schedule 16 and 17 deferred costs. It argues first that the Company’s proposal
to place those costs in the rate base is inconsistent with ratemaking
principles in that returns are
generally allowed only for capital costs and that deferred MISO Schedule 16 and
17 costs are expenses and should be treated as traditional expenses are
treated, without a return. The OES also
asserts that that the Commission already addressed this issue in Otter Tail
Power Company’s most recent rate case and did not allow Schedule 16
and 17 deferred costs to be place in that utility’s rate base. Finally, the OES argues that the Commission’s statement in the MISO Day 2 Costs Order that “[e]ach
utility may continue deferring Schedule 16 and 17 costs without interest until the earlier of the utility’s next electric
rate case or March 1, 2009,” expressly precludes rate base treatment of those
costs. The OES therefore recommends that
the Commission require MP to remove the deferred MISO Schedule 16 and 17 costs
from rate base and treat these expenses as expenses.[206]
161.
The ALJ
concludes that the proposition that investors should be given a return (rate
base treatment) on deferred Schedule 16 and 17 costs is unreasonable and should
be rejected. First, the OES and the
Company both agree that although ratepayers will pay for both the current MISO
Schedule 16 and 17 costs as well as the deferred costs from the period of April
1, 2005 to June 30, 2008, the Company has been allowed to retain (not defer in
order to return to ratepayers) any revenues associated with its activities in
the MISO Day 2 energy market. Moreover,
the Commission has already decided that the utilities cannot recover interest
expenses on the deferred MISO administrative costs. Paragraph 2 of the MISO Day 2 Costs Order
explicitly states that deferral of Schedule 16 and 17 costs was to be “without
interest.” Placing the costs in rate base would be in direct contradiction with
that decision. Finally, as Ms. Campbell
pointed out, these costs are expenses and are generally only allowed interest,
not a return. The MISO administrative
costs are clearly not capital costs to which a return would be applied, if
allowed. For the reasons set forth
above, the ALJ recommends that the Commission direct Minnesota Power to remove
deferred MISO Schedule 16 and 17 costs from the rate base.
C.
162.
“Other Wheeling
Revenues” are revenues generated from MP’s transmission facilities through the
MISO tariff, when utilities other than MP use, or wheel, the Company’s
transmission lines to move energy across MP’s system. As with MP’s generation assets, MISO requires
that utility’s transmission assets be available for use by other utilities, if
the lines are not fully used to serve MP’s retail customers. Other utilities pay MISO for their wheeling
activities, and MISO, in turn, pays those revenues to MP. Because transmission
assets are funded by ratepayers as part of MP’s rate base, ratepayers must be
credited with wheeling revenues from MISO that are associated with such assets.[207]
163.
The OES first
noticed that Minnesota Power had made a very large, unexplained reduction in
test year Other Wheeling Revenues of $5.8 million on a total Company basis,
when compared with the Company’s levels from previous years.[208] In response
to the OES’s discovery, MP admitted in July 2008 to having omitted $4 million
in revenue from MISO, which it had derived from wheeling energy across its
transmission system. The Company’s
explanation was that it had incorrectly assumed that MISO would change its
transmission revenue sharing allocation method, which would have resulted in a
reduction in test year Other Wheeling Revenues. However, MISO never made that change. Because of that, both MP and OES agree that
an additional $4,070,155 in revenue needs to be added into rates for Other
Wheeling Revenues from MISO.[209]
164.
When the Company
recognized the increase in Other Wheeling Revenues, MP noted that it also did
not include in the test year any expense associated with MISO point-to-point
transmission service for the delivery of power the Company purchases from
165.
The OES objected
to the inclusion of those costs, particularly since the Company did not make
documentation available to support them until the evidentiary hearing. The OES maintains that MP's documentation is
untimely and inadequate to support the claimed expenses and that the claimed
$2.8 million in costs should not be allowed.
In the alternative, the OES proposes that cost recovery should be
limited to the $1.9 million that was documented as on-going transmission costs
by MP, based on invoices that the Company provided at the hearing.[211]
166.
Before the evidentiary
hearing, MP expressly advised the OES that the expenses associated with the
Ontario Path revenues had been omitted from rate recovery.[212] In response,
the OES advised Minnesota Power of the minimum information OES would need for
consideration of MP's claimed Ontario Path wheeling expenses:
·
MP’s purchase
power contract with
·
Proof that the
purchase power contract is being used to serve retail customers;
·
Information to
support that the transmission expenses are real (actual invoices);
·
Information to
show that these transmission expenses are not already embedded in current
rates;
·
Any other
information MP considers appropriate to support their case for cost recovery
purposes.[213]
167.
MP expressly
indicated that the expenses associated with the Ontario Path revenues had been
excluded from rate recovery.[214] Prior to the
hearing, MP reviewed the test year expenses again and concluded that the
Ontario Path expense had not been previously included.[215] MP noted that
the original test year budgeted total O&M expense for Transmission of
Electricity of Others was only $1,349,610. MP indicated that this amount is much smaller
than the total Ontario Path transmission expense of $3.6 million. This supports the conclusion that the Ontario
Path expense was not included in the test year budget.[216]
168.
The OES has
expressed legitimate concerns that expenses could have been “double counted”
through MP’s failure to identify these expenses from the intital filing of this
proceeding. MP bears the burden of proof
to show what costs are incurred in providing electricity to customers. On this issue, MP has provided both an
adequate explanation and a sufficient factual basis for inclusion of the
Ontario Path expenses in calculating rates.
The ALJ recommends that the Commission include the Ontario Path costs in
calculating base rates.
169.
On October 14,
2005, Minnesota Power filed an application with the Commission in Docket No.
E-015/M-05-1678 seeking its approval to proceed with projects designed to
reduce the pollution emitted from some of its electric generators in the
Arrowhead region of
If the
revenues collected by the rider differ from the AREA Plan’s costs, parties may
propose compensating rider adjustments once all of the AREA Plan components are
in service or in the Company’s next general rate case, whichever is sooner. But in no event shall the Company calculate
the amount it recovers through the rider on the basis of more than $53.9
million for capital costs and $4.07 million for annual operating and
maintenance expenses. If the actual costs of the AREA Plan exceed these caps,
the Company shall report this fact to the Commission to enable a Commission
review [of] the matter. [219]
170.
On March 31,
2008, Minnesota Power filed another application with the Commission in Docket
No. E-015/M-05-1678 in which, among other things, the Company sought to recover
AREA Plan annual O&M expenses in excess of the $4.07 million cap that the
Commission imposed in its AREA Order.[220] By Order
entered on August 5, 2008, the Commission disposed of that petition in the
following way:
Dismissed
the petition without prejudice.
The
Commission will address these issues in the Company's rate case.[221]
171.
In this
proceeding, Minnesota Power initially proposed to transfer the recovery of
costs related to Laskin Units 1 and 2 and Taconite Harbor Units 1 and 2 from
the AREA Rider to its base rates, thereby including $13,866,854 in AREA costs
in the rate case, including $4,749,256 of O&M expenses.[222] The
$4,749,256 in requested annual O&M expenses exceeds the cap established by
the Commission on June 13, 2006, by $679,256.
172.
The OES objected
to inclusion of annual O&M costs that exceeded “the initial operation and
maintenance (O&M) cap of $4.07 million that was established in the
Commission’s Order on June 13, 2006,” [the AREA
Order] and proposed a downward adjustment of $679,256 ($568,533 Minnesota
jurisdiction) to account for the amount by which Taconite Harbor O&M costs
exceed $4.07 million.[223] To address
the OES’s objection, Minnesota Power proposes to move its request for O&M
costs that exceed than $4.07 million out of the rate case and into the AREA
Rider docket for further review by the Commission but leave $4.01 million in
the test year to be incorporated into the rate.[224]
173.
In the ALJ’s
view, the positions that the parties’ have expressed on this issue, to some
extent, miss the mark. In a rate
proceeding, Minn. Stat. § 216B.03 provides that "[e]very rate made,
demanded or received by a public utility...shall be just and reasonable...Any
doubt as to reasonableness should be resolved
in favor of the consumer."
Costs that are uncertain, unpredictable, or unreliable, are unreasonable
and should not be recovered in base rates.
The issue here is not whether this rate case or the AREA rider docket is
the appropriate forum for Minnesota Power to request the Commission to approve
O&M expenses in excess of the $4.01 million cap set in its AREA Order. This issue here is a ripeness issue—that is,
whether at this point in time—i.e., during this rate proceeding—the $679,256
($568,533 on a Minnesota jurisdictional basis) in annual O&M costs in
excess of the cap that the Company is seeking to recover are certain,
predictable, and reliable. The ALJ
concludes that they are not. In response
to questioning by the ALJ, Company witness, Ms. Hodnick, offered the following
testimony:
And we
installed -- because
In short, Minnesota Power has not yet established
that $679,256 ($568,533 on a Minnesota jurisdictional basis) of annual O&M
costs in excess of the $4.07 million cap are certain, predictable, and reliable
and therefore reasonable to include in the rate.
174.
In its AREA Order, the Commission did not
definitively conclude that AREA Rider O&M costs of $4.07 million were
reasonable. It left open the possibility
of a true up that might establish actual costs below the cap of $4.07 million
cap. The ALJ therefore recommends that
if the Commission concludes that a true up is no longer necessary, it allow
AREA O&M costs of $4.07 million to be included in the rate, without
prejudice to allow the Company to request a compensating rider adjustment upon
a showing that there are certain, predictable, reliable, and otherwise
reasonable O&M costs in excess of the $4.07 million cap.
VII.
Incentive
Compensation.
175.
Minnesota Power
proposes to include approximately $6,823,793 of employee incentive compensation
expenses in the test year. That budgeted
amount is divided between three programs: Results Sharing ($4,242,510); Annual
Incentive Program (AIP)( $1,180,844); and Long Term Incentive Program (LTIP)
($1,400,439).[226] All MP employees participate in the Results
Sharing Plan, the AIP applies to 90 management employees, and the LTIP applies
to 43 management employees of MP and ALLETE’s corporate operation.[227]
A.
Incentive
Compensation Levels.
176. The Results Sharing incentive is paid out when and if
ALLETE meets a target level financial performance (including the cost of paying
a target level award). The minimum
amount of the payout is 3 percent of base compensation. If Key Result Area (KRA) goals are accomplished
in the three areas of employee safety, environmental compliance, and system
reliability, the incentive payout is increased to 5 percent of base
compensation. If financial performance
is sufficiently above the target level and all KRA goals are achieved, the
Company increases awards proportionately up to a maximum of 15 percent of base
compensation.[228] Since inception of the Results Sharing
program in 1991, Results Sharing awards have averaged 6 percent of base
compensation.[229]
177.
Participants in
the AIP are eligible for target level awards ranging from 10 percent of base
compensation to 50 percent of base compensation. As described by MP, “A participant’s (i) base
compensation plus (ii) the AIP target award opportunity plus (iii) the 5
percent target level Results Sharing award opportunity equates to total
compensation for employees participating in AIP and Results Sharing.” MP included in the test year budget an amount
reflecting 62.5 percent of the total possible incentive compensation (AIP and
Results Sharing), since that was the amount actually paid out in 2007. [230]
178.
MP indicates that the total compensation available to employees eligible
for the LTIP is “[a] participant’s (i) base compensation plus (ii) target award
opportunities under Results Sharing and the AIP plus (iii) the target award
opportunity from LTIP equates to total direct compensation.” Target performance to trigger LTIP awards is
weighted, with the greatest emphasis on earnings. The Company maintains that the overall amount
of incentive compensation is set so that the total compensation “is near the
midpoint of the competitive market level.”
MP indicated that “[g]oal achievement below the target performance level
would result in no or lower incentive awards being paid and below market
compensation for management.” In this proceeding, MP seeks to recover 100
percent of the LTIP target compensation in rates.[231]
179.
Minnesota Power
argues that its incentive compensation proposals are needed to provide
sufficient compensation to attract, motivate, and retain talented employees,
and are necessary to provide high quality service to customers. To attract and retain employees with the
necessary talent and ability, MP argues that its total compensation package
must be competitive. In 2007, the
Company commissioned a study by Hewitt Associates which determined that the
base salaries of MP’s executive officers were near the 50th percentile of
energy services company benchmarks. That
study also indicated that annual incentive compensation targets fell near the
50th percentile of energy services company benchmarks. The only exception was ALLETE’s Chief Executive
Officer (CEO) position, which fell below the 50th percentile of those
benchmarks.[232]
180.
The Company
further argues that it “is differently situated from other utilities, based
largely on the age of its workforce and its geographic location in the state. With the changing demographic of the state’s
workforce as a whole, which is more pronounced for Minnesota Power … incentive
compensation is becoming increasingly necessary to attract new workers.” [233]
B.
OES Proposed
Limits on Results Sharing and AIP.
181.
In
response to the Company’s incentive compensation proposals, the OES noted that in
its prior general rate case orders, the Commission has followed a practice of
limiting incentive compensation to no more than 25 percent of base
compensation. Accordingly, the OES proposed
a cap of 20 percent for AIP costs, together with the 5 percent cap for results
sharing costs. The OES maintains that
these limits would provide the Company with recovery of combined incentive
compensation up to 25 percent of base salary.
The OES also recommends that $167,143 in test-year AIP incentive
compensation expenses of be excluded to keep the total within a 25 percent
cap. The OES argues that its
recommendation is consistent with Commission precedent.[234]
182. In response, MP argues that its proposal would
recover no more than 36.5% of base compensation for the Results Sharing and AIP
programs combined.[235] The Company maintains that “the high end
would only be achieved if the Company reaches its AIP targets, which would not
only keep the Company sound, but also ensure such specific benefits to the
utility such as ensuring the Company’s ability to increase its supply of
renewable energy and providing leadership development and succession planning
for Minnesota Power.”[236] MP also argues that it must offer a plan such
as the AIP to offer competitive salaries in its “difficult market,”[237] and that “denying recovery of these costs
sends the message that either a larger percentage of salaries should be
allocated to base compensation (which is a potentially more expensive method of
employee compensation, and would be earned regardless of Company performance),
or that market-level compensation should not be paid.” [238]
183.
In reply, the OES
noted that the Commission’s Order in MP’s last rate case indicated a concern with
the wide range of the incentive compensation awards available to MP management
employees:
Regarding incentive compensation, the Commission has
found in previous rate cases that incentive compensation plans can be effective
management tools when properly designed and administered. In this case, the Commission finds that the
two incentive compensation programs proposed by MP (Results Sharing and
Incentive Compensation for Officers and Management) have been appropriately
designed.
The Commission will approve them with one
modification. Perception of favoritism
or inequity in the administration of the incentive program could have negative
repercussions for employee morale and consequently negatively affect their
productivity. Such a wide range of award
(up to 60 percent of base pay) substantially increases the possibility of such
perceptions. In addition, as the
Commission has previously found, offering key decision makers large financial
rewards for producing short-term shareholder benefits does not promote
regulatory efficiency or the longterm fortunes of the Company. The Commission, therefore, will limit annual
incentive payments to 15 percent of an individual’s base pay, the same maximum
level available to all other employees.[239]
C.
OES Proposed
Disallowance of LTIP.
184.
As the OES observed,
the Company proposes to pay incentive compensation under the LTIP in addition
to that obtainable under both Results Sharing and AIP. Referring to the Commission’s
practice of excluding from test year expenses any that incentive compensation
that exceeds 25 percent of base compensation, the OES proposes that 100% of the
LTIP expenses be disallowed.[240]
185.
The Company’s
proposed incentive compensation plan is strikingly similar to that rejected in
the Company’s last rate case. As
justification, the Company asserts that limiting the total incentive
compensation to 25 percent of base compensation will not meet the Company’s
need to attract skilled employees and motivate them to provide excellent
service. However, the Company has offered
little evidence that it suffers from a sufficient competitive disadvantage in
attracting skilled and talented employees to warrant a departure from the
limits the Commission has previously imposed on incentive compensation. Evidence that its executive compensation annual
incentive compensation targets are at the midpoint of energy services company
benchmarks does not necessarily establish a competitive disadvantage. Neither is there persuasive evidence
supporting the Company’s assertion that its geographic position within the
state places it at a competitive disadvantage.
Although location in a major metropolitan area may provide amenities
that are attractive to some prospective employees, the reverse may also be true
of a northern
186.
In summary, the
ALJ concludes that the OES proposal to allow the cost
of the Results Sharing with a 5 percent cap and the AIP with a 20 percent cap,
but removing the cost of the LTIP in its entirety, will both benefit ratepayers
and result in rates that are just and reasonable. The ALJ therefore recommends that the
Commission approve that proposal.
187.
Minnesota Power
had acknowledged that its incentive compensation plans do not require the
Company to pay the target amounts in any particular year, that the Company has
reserved the right to discontinue all its incentive compensation plans, and that
there is no obligation for the Company to adopt new plans, even when rates
charged to customers that are based on having that expense. As a consequence, the OES is also
recommending that any incentive compensation that is included in base rates but
that is not actually paid to MP’s officers and employees be refunded to
ratepayers. Although Minnesota Power takes
the OES’s proposal to mean that refunds will take place if the incentive
compensation plan is changed to some different form of incentive plan,[242] that does not appear to be what the OES is
proposing. The ALJ understands that the
proposed refund mechanism would require tracking of the amounts actually paid
in incentive compensation and only require refunds of any amount not paid to
employees up to the total amount of expense allowed by the Commission.
188.
A refund
mechanism for incentive compensation would be consistent with Commission
precedent. As the Commission stated in a
prior rate case:
In the original Order, the
Commission expressed strong disapproval of the Company’s retention of the right
not to make incentive payments earned under the plan. The Commission continues
to view this as an inappropriate transfer of risk from shareholders to
ratepayers and as inconsistent with the test year concept on which rates are
based. The Commission will therefore require the Company to record all earned
but unpaid incentive compensation recoverable in rates under this Order for
future return to the ratepayers. This will adequately protect ratepayers’
interests and prevent erosion of the test year concept.[243]
The Commission concurs
with, accepts, and adopts the ALJ’s recommendation on this issue, which was to
cap individual incentive compensation payments at 25% of an employee’s base
salary; to base total, company-wide incentive compensation on amounts actually
paid out between 2002 and 2005; and to continue the tracking and refund
mechanism established in the Company’s 1992 rate case.[244]
190.
The Commission
also followed this approach to incentive compensation included in rates but not
paid in its NSP Gas Rate 2007 Order:
The Commission finds that
the Company’s proposed level of incentive compensation in this proceeding is
reasonable and will approve it. The Commission also adopts the ALJ’s finding
and will require Xcel to refund amounts included in the test year for incentive
compensation that were not actually paid.[245]
191.
In its most
recent rate order, the Commission stated:
The
Commission also concurs with the Administrative Law Judge that the Company
should be required to establish a mechanism to refund to ratepayers any incentive
compensation included in rates that is not actually paid. While it is probable that the Company will
continue to make payments under the incentive compensation plan throughout the
period that rates will be in effect, the terms of the plan do not require the
Company to do so. In fact, the plan explicitly grants the Company the right to
discontinue it at any time.[246]
193. In its rate proposal, the Company identified a total
of $1.3 million in ALLETE aircraft costs during the test year. Of that total, 92% or $1.2 million was
allocated to Minnesota Power.[247] This amount is approximately $600,000 lower
than the Company’s 2007 expenses on a
194.
On March 8,
2007, the Commission issued an Order approving Minnesota Power’s acquisition of
an aircraft that MP had previously shared with a spun-off subsidiary (Aircraft
Ownership Order). At that time, the
Commission required Minnesota Power to include in its next rate case filing a
cost benefit analysis addressing whether (i) the benefits of allocating 100% of
an aircraft to Minnesota Power exceeds the costs; (ii) whether the aircraft
ownership allocated to Minnesota Power is necessary in the provision of utility
service; and (iii) whether the arrangement is beneficial compared to using
alternative transportation.[249]
195.
Previously, in
2006, MP had hired an independent assessor, Avicor Aviation (Avicor), to
determine whether possession and use of a private corporate aircraft was
appropriate. Avicor determined that the
corporate aircraft provides a cost benefit to the company compared to
commercial airline flights and “strongly recommend[ed]” that “ALLETE continue
to utilize a corporate aircraft for its business needs.”[250] Regarding specific issues, Avicor determined:
Based on the travel patterns assessed, ALLETE
utilizes its Hawker 700A to ensure the efficiency of its employees[.]…
One small note about productivity in relation to
ALLETE and where it is located – consistent and timely commercial lift is
clearly a problem at DLH [
In its
report, Avicor concluded that it was evident from the flights that were
undertaken that the aircraft are being used to move key individuals where they
need to be when they need to be there, without wasting their time and
productivity.[252]
196. The OAG/RUD conducted a trip-by-trip analysis of the
usage of the aircraft and identified several trips that it considered to be problematic. One such trip was a flight from
197.
The OAG/RUD questioned
the business purpose of the
198.
The OAG/RUD also
questioned the regulatory necessity of using the corporate aircraft for
multiple trips to
199.
In response, Minnesota Power asserted that flights
to
200.
Based on its
review and findings, the OAG/RUD argues that the evidence it has presented calls
into question Avicor’s 2006 conclusion that ALLETE uses the corporate jet to
fly to destinations that are difficult to access via commercial flights. The OAG/RUD also asserts that "ALLETE’s
corporate aircraft is used as an executive perk, which Minnesota Power
ratepayers struggling through an economic crisis should not be required to
fund."[258] Rather than requiring the Company to deduct
from the expense figure the cost of individual corporate aircraft flights, for
which there was no apparent benefit to ratepayers, the OAG/RUD has recommended
that all expenses associated with ALLETE’s corporate aircraft be excluded from
Minnesota Power’s rates.
201.
Minnesota Power has
objected to the OAG/RUD ‘s proposal to disallow all test year corporate
aircraft costs. In so doing, the Company
notes that the OAG/RUD is not recommending that if those costs are disallowed,
the Company should be allowed to recover as an offset the costs of commercial
flights that Minnesota Power would have incurred if the corporate aircraft was
never used. The Company also argues that
there should also be an offset to any disallowance for the productivity that that
is lost when its employees must use commercial flights.[259]
202.
Minnesota Power further
maintains that ALLETE has carefully limited the use of its corporate
aircraft. It cites the following
procedures to demonstrate that it has limited use of the corporate aircraft to
appropriate situations:
(1) commercial travel rates and schedules must be
carefully considered first; (2) more than one employee generally must be
traveling on the corporate aircraft for its use to be approved; (3) an employee
requesting use of the aircraft must prepare a Company Aircraft Travel
Request/Approval form before flying; (4) travel must be approved in advance by
a Vice President; (4) ALLETE CFO Mark Schober reviews the forms monthly to
ensure appropriate use; and (5) ALLETE’s Internal Audit Department audits
aircraft usage annually.. In addition, flight logs are completed for each trip.[260]
203.
Under the Aircraft Ownership Order, the Commission
required Minnesota Power to include in its next rate case filing a cost benefit
analysis demonstrating that the benefits of allocating 100% of an aircraft to
Minnesota Power exceeds the costs; that the aircraft ownership allocated to
Minnesota Power is necessary in the provision of utility service, and that the
arrangement is beneficial when compared to using alternative transportation.[261] In addition to that, the Commission also
ordered that:
ALLETE shall show and provide adequate support for these allocations of
aircraft costs in its next rate case. Additionally, ALLETE shall explain and
support in the context of its next rate case whether or not the entire aircraft
should be included in rate base for regulated purposes or whether some
proportional share of the aircraft equal to regulated use should be included in
rate base. [262]
204.
Although
Avicor’s report does not represent a cost-benefit analysis, as that term is
commonly understood, it does meet at least one of the criteria that the
Commission established in the Aircraft
Ownership Order. It does establish
that use of a corporate aircraft is beneficial in comparison to using
alternative transportation in many situations. Moreover, the Aircraft Ownership Order does not appear
to require the Company to submit a formal benefit-cost analysis unless it proposes
to allocate 100% of the aircraft’s cost to Minnesota Power. However, the Company has not chosen to do
that. Rather, it has offered trip by
trip information on the proportional share of the aircraft’s cost that it
believes equates to the aircraft’s regulated use to support inclusion of the
allocated costs in base rates. In other
words, the Company has complied with the alternative method of allocating the
aircraft’s costs that the Commission appears to have afforded in its Aircraft Ownership Order. It also has offered explanations of why each
was necessary in the provision of utility service when that was otherwise not
readily apparent.
205.
The OAG/RUD
discovered some apparent discrepancies in Minnesota Power corporate aircraft
cost allocations, but those discrepancies were not numerous in the context of
dozens of flights. Whether the frequency
and magnitude of the discrepancies that occurred are a sufficient basis to
disallow all of the proposed corporate aircraft test year costs is ultimately
for the Commission to determine.
However, the ALJ does not consider the discrepancies that OAG/RUD’s
investigations revealed to be a sufficient basis and further finds that the
Company has made a good faith effort to comply with how it interprets the
Commission’s Aircraft Ownership Order.
Moreover, since neither the OAG/RUD, the OES, nor any other party has
recommended adjustments to the test year corporate aircraft costs to reflect a
lesser benefit to ratepayers than that claimed, the ALJ recommends that the
Company’s proposed corporate aircraft test year costs be allowed.
IX.
Corporate Cost
Allocations.
206.
As previously
discussed, Minnesota Power is an operating division of ALLETE, which also owns
businesses that are not rate-regulated by the Commission. In this proceeding, the Company proposes to
allocate certain costs from ALLETE to MP for inclusion in the test year
expenses. In its Docket 1008 proceeding,[263]
the Commission adopted a four-part hierarchical methodology to govern such
allocations:
1) Tariffed
rates shall be used for tariffed services provided to nonregulated activity.
2) Costs shall
be directly assigned whenever possible.
3) If costs
cannot be directly assigned, they shall be allocated based on an indirect
cost-causative linkage to another cost category or group of cost categories for
which direct assignment or allocation is available.
4) When neither
direct nor indirect cost causation can be found, the costs are to be allocated
using a general allocator.[264]
In the same Order, the Commission also adopted a
default general allocator that uses the ratio of all expenses directly assigned
or attributed to regulated and unregulated activities, excluding the cost of
fuel, gas, purchased power, and the cost of goods sold.[265]
207.
In the Docket 1008 Order, the Commission also
recognized that cost allocations should be sufficiently flexible to reflect
differences between utilities and differences in the characteristics of the
unregulated entities:
The
Commission understands that utilities differ in many essential respects,
including their participation in affiliated operations. The Commission believes that the hierarchical
principles offer sufficient flexibility for each utility to develop appropriate
allocation methodologies based on the principles.[266]
208.
In its
subsequent Order Closing Docket 1008,
the Commission reaffirmed that a utility is allowed to deviate from the default
approach in future rate cases when the utility establishes that:
… its cost
allocation principles produce similar results as would allocations following
the recommended cost allocation principles,
* * *
or the
public interest is better served by another method.[267]
209.
The Commission’s
Docket 1008 Order establishes the
burden that utilities must meet when they employ allocation principles that are
different from those established by the Commission:
Should a
utility wish to base its cost separations on different principles, the burden
of proof would be on that utility to prove that its cost allocation principles
arrive at fully allocated costs, free of any cross-subsidization. The utility would have to show that the goals
of fully allocated costing, as expressed in this and other Orders, are fully
realized. The utility would have the
burden of demonstrating that it considered all of its costs and that they are
allocated to share burdens and benefits equitably between the regulated and
nonregulated operations.[268]
210. As a division of ALLETE, Minnesota Power uses the
same accounting methodology as ALLETE.
At the time of its last rate case in 1994, Minnesota Power utilized the
standard FERC Uniform System of Accounts for its chart of accounts (COA). That system gathered costs by responsibility
center and FERC account, but did not allow for functional cost accumulation or
activity cost reporting.[269] In 1998, Minnesota Power implemented the
accounting system for allocating costs (BIS system) that it uses today. Its BIS system provides for (i) direct
charging to specific "Lines of Business;" (ii) direct charging to
non-regulated activities; (iii) direct charging and billing of costs incurred
on behalf of subsidiaries and other entities; and (iv) proper allocation of
unassigned charges for which it is not feasible to direct charge. The BIS system accumulates costs by function
and segregates regulated and non-regulated activities and expenditures.[270]
211.
On September 17,
2001, Minnesota Power filed a petition with the Commission that, among other
things, requested the Commission’s approval of the Company’s accounting
methodology for asset accounting and functional cost assignment for use in all
future asset-related or cost-based proceedings before the Commission.[271] The Commission issued an Order in
that docket on August 8, 2002, that
approved MP’s accounting methodology petition with certain conditions (August
8, 2002 Order). Among those conditions
was a requirement that the Company meet with the Department of Commerce (now
OES), the OAG/RUD, and Commission staff to review allocation methods and
overall accounting methodology. The
Commission also required MP to update the OES and Commission staff annually of
changes to allocators.[272] As to the future effect of that meeting, the Commission
stated:
Finally, the Commission clarifies that this
consultation and reporting is no substitute for the Commission’s review of the
merits of MP’s allocators in MP’s next rate case, where the Company will bear
the burden of proof that its allocators are cost causative and fairly allocate
costs.[273]
212.
The meeting that
the Commission specified in its Order of August 8, 2002, occurred on September
2, 2002. At that time, the Company
advised the OES, OAG/RUD, and Commission staff that it did not have a general
or default corporate allocator, but rather that it directly charges corporate
costs or allocates.[274] Since September 2, 2002, Minnesota Power has
made annual compliance filings in Docket No. E-015/M-01-1416 that have
documented changes the Company made to its cost accounting systems during the
preceding year.[275] The Company’s most recent compliance filing
was made on January 10, 2008.[276]
213.
Although the
evidence presented by Minnesota Power indicated that the Company has no general
or default allocators,[277] based on
evidence supplied by the Company, the OES concluded that:
It appears that ALLETE specifically allocates certain
costs to its non-regulated operations and subsidiaries, with any remaining
costs left in MP’s regulated operating division. This allocation method means that MP’s
regulated operations serve as the default for ALLETE’s corporate costs. This
process is not consistent with Commission directives and precedent as detailed
earlier, and may assign to MP’s regulated operations a disproportionate share
of ALLETE’s costs.[278]
214.
The OES thus has
raised the issue of the Company using Minnesota Power as a default for allocating
ALLETE’s corporate costs. The Company
denies doing that. The ALJ was unable to
find sufficient detail in the prefiled testimony of either party to
definitively resolve the issue one way or another,[279] and at the
evidentiary hearing there was only perfunctory cross-examination of the
witnesses with knowledge of that issue.[280] Unless Commission staff can establish from
the record that the Company has, in fact, been using Minnesota Power as a
default for allocation of ALLETE corporate costs, the ALJ recommends that the
Commission approve the Company’s corporate cost allocators, as proposed.
215.
In any event, as
the remedy for what the OES considers to be an unreasonable allocation of
corporate costs, the OES is not recommending that the Commission require the
Company to allocate corporate costs in this proceeding in a different way. Rather, the OES recommends that the
Commission require Minnesota Power to legally separate its Minnesota regulated utility
operations—Minnesota Power Company—from its parent company, ALLETE, by the time
the Company files its next electric rate case, advancing three arguments to
support that recommendation.[281] First, the OES believes that a legal
separation provides a greater level of protection to ensure that ratepayers are
not subsidizing any part of the parent company’s unregulated activities.
Second, it also believes that the creation of a holding company structure would
provide a greater level of transparency to ALLETE’s corporate costs
allocations. Third, the OES argues that a legal separation would reduce the
resources required to investigate any irregularities between the utility and
non-utility entities.[282]
216.
The ALJ
disagrees with the OES recommendation of legal separation. Even if the Commission has the authority to
require Minnesota Power to do business in a different legal form, the problem
that the OES seeks to correct is a cost accounting problem that would not
necessarily be resolved by legally separating Minnesota Power from ALLETE. The problem that the OES seeks to correct is
ensuring that services exchanged between employees of the two entities are
properly documented and accounted for using generally accepted accounting
practices. Presumably, even if the two
entities were legally separated, ALLETE employees would continue to perform
work for Minnesota Power, and Company employees would continue to perform work
for ALLETE; and issues of proper cost accounting are likely to remain.
B.
Amount of Allocated
Corporate Expenses.
217.
The Company estimated $89,526,995 for ALLETE’s total corporate
costs in the test year.[283] MP assigned $70,994,154 in ALLETE corporate
costs to Minnesota Power in 2007. The
Company has assigned $75,423,988 in ALLETE’s corporate costs (84%) to Minnesota
Power in the test-year. This represents $4,429,834 or an increase of
6.2% over the ALLETE corporate costs that were assigned to Minnesota Power in
2007.[284]
218.
The OES contends
that Minnesota Power has failed to establish that the increase in corporate
costs proposed by MP for the test year is reasonable. In support of its contention, the OES points
out that ALLETE’s
total corporate costs have generally been trending down from $96,002,471 in
2003 to $85,187,470 in 2007, and that the rate of inflation for the period July
2007 to July 2008 was 2.5 percent according to the most recent Consumer Price
Index data from the U.S. Department of Labor Bureau of Labor Statistics.[285] The OES relied on the CPI because it excludes
food and energy prices and because, historically, corporate costs do not
include significant amounts of food and energy costs but consist primarily of
wages and other corporate overhead.[286] OES therefore recommends that Minnesota Power recover $73,678,620 in the test year for
corporate costs. That recommendation is
based on allowing the Company to recover 2007 corporate costs adjusted by an
annual inflation figure of 2.5 percent over 18 months. If approved by the Commission, this
adjustment would result in a reduction to MP's corporate costs by $1,528,502 on
a
219.
In response,
Minnesota Power asserted that just under half of the $4.4 million increase in
corporate costs is attributable to salary increases, while the balance is related
to new staff and increased service costs for Information Technology, Project
Engineering, Internal Audits, increased expenses for the Company’s
Conservation Improvement Program,
and its membership in the Electric Power
Research Institute (EPRI). The company
supplied a table indicating the amounts by which each of those items was
projected to increase.[288] MP also
relied on the cross- and re-direct examination of OES’s analyst, in which he
accepted that the costs were being incurred for the purposes
stated, that the Company would actually incur those costs during
the test year, and that expenses of that kind would not only be incurred during
the test year, but also going forward after the test year.[289] The Company
also relied on the fact that OES was not disputing that new staff or IT,
project engineering, and internal auditing services, or CIP, EPRI, and salary
expenses were appropriate or prudent items to include in a test year.[290] In the
absence of such challenges, MP asserts that it is unreasonable to make an
adjustment to its proposed costs and that “these costs should not be rejected
in favor of a theoretical expense amount that has never actually been
experienced by the Company.”[291]
220.
As discussed
above in the analysis of the test year, utilities typically use historical
costs, adjusted by known and measurable changes, to arrive at the appropriate
level of test year expenses. This method
replaces the impact of market forces with “the fiscal discipline of prior
determination of reasonable costs.”[292] In this proceeding, there has been no “prior
determination of reasonable costs” relied upon by MP. The kinds of costs that the Company relies on
in projecting increases for the test year are the kinds of costs that MP has
incurred in the past. The issue is not
whether those kinds of costs are reasonable and prudent; the unanswered
question is why the percentage increase of those costs that Company is
projecting for the test year, is so much higher than in past years and why the
Company expects that higher level of costs to continue throughout the life of
the rate. Moreover, the Company has not
explained why it has proposed budgets showing increased expenses at the same
time it is forecasting reduced sales. In
short, Minnesota Power has the burden of showing that any increases in cost
from its most recent quantifiable expenses are justified and result in rates
that are just and reasonable. As
determined by statute, “Any doubt as to reasonableness should be resolved in
favor of the consumer.”[293] The ALJ concludes that Minnesota Power has
not met that burden with regard to its projected increase of corporate costs
from 2007 to the test year.
221. The OES has proposed a reduction to MP's corporate
costs by $1,528,502 on a
222.
Minnesota Power
allocated energy costs using kWh sales for both a jurisdictional separation of
costs and its class cost of service study (“CCOSS”), and it used the E8760
allocator to allocate energy costs among customer classes. The E8760 allocator
takes into account the costs of energy based on the time of day the energy is
used.[295] If a customer
class consumes proportionately more of its energy during periods of peak demand
when the market price for electricity is higher, the E8760 will assign that
class its proportionate share of this high cost energy.[296] The Company described how it uses the E8760
allocator as follows:
The E8760 is based on Minnesota Power’s system Locational Marginal Price
(“LMP”) hourly cost and the hourly energy use of each class. It is derived by
multiplying the hourly energy usage of each class by the system’s LMP cost by
hour, summing and taking the ratio of the sum of each class to the total.
Applied as a cost allocator, the E8760 will yield class-specific
responsibilities that take into account class use-patterns and time-variant
system costs. In contrast to a straight non-weighted energy allocator, the
E8760 results in a slight shift of class-specific responsibilities away from
classes that take proportionately more of their energy during off-peak periods,
to classes that take proportionately more of their energy during more expensive
on-peak periods.
[297]
223.
The name E8760
reflects that there are 8760 hours in a year and that the different energy
costs in each hour are used in developing a different energy factor for each
customer class. MP indicates that its
E8760 allocator is based on the methodology used by Xcel Energy in its CCOSS in
a recent electric rate case (Docket No. E002/GR-05-1428).[298] The OES supports MP’s use of the E8760
methodology for CCOSS purposes, stating that “[t]he OES agrees with the Company
that using the energy allocator E8760 would allow the CCOSS to reflect class
cost responsibilities more precisely since energy costs vary, sometimes
significantly, from hour to hour.”[299]
224.
However, the OAG/RUD
argues that the E8760 allocator is a poor method of determining costs in the
CCOSS because: 1) MP used different allocators for other items, such a fuel
costs; 2) LMP from 2007 does not reflect the LMP costs from the test year
(2008-2009); 3) LMP pricing is a function of other utilities’ cost of power, not
just MP’s cost; and 4) the treatment of line losses in the E8760 allocator
differs from MP’s own reported costs through line losses.[300]
225.
The OES responded
to the OAG/RUD with a description of how the LMP is calculated. Based on that analysis, the OES concluded
that using the E8760 energy allocation to assign energy costs is reasonable because
the LMP is the market price that MP must pay for the energy provided to its
customers.[301] Minnesota Power also produced a comparison
between 2007 LMP prices and 2007 megawatt hours, showing an increase in the
percentage of costs that the CCOSS would allocate to the residential and
general service classes.[302] That
comparison established that making adjustments to match costs of fuel to the
LMP would decrease the assignment of costs to the Large Power class and would actually
increase the share of the costs assigned to Residential and General Service
classes.[303]
226.
The ALJ
concludes that the Company’s use of the E8760 allocator is a reasonable method
of allocating costs in the CCOSS and that the Commission should not require MP
to use a different methodology for that purpose.
XI.
Proposed FCA
Matching Adjustment.
227.
As previously
indicated, on July 21, 2008, the Commission entered an Order in Docket No.
08-463 that directed issues relating to a proposal by the Company to recover
lagged fuel clause costs associated with the implementation of the new base
cost of fuel, as well as associated changes to the Company’s Rider for Fuel
Adjustments to be addressed in this
general rate proceeding.[304]
228.
The essence of
the Company’s proposal was to change the lagged fuel cost adjustment to a
forecast adjustment. Minnesota Power
contemplated the change would be accomplished by providing it with a $19
million adjustment to account for the two and a half months that would drop out
of the fuel clause calculation when the shift occurred from a lagged to a
forecast adjustment. The OES and the LPI
opposed the adjustment as unsupported in the FCA structure and not being of
benefit to ratepayers.[305]
229.
During the
evidentiary hearing, MP, OES,
230.
Despite the
OAG/RUD’s objections, the ALJ concludes that none of the Stipulation’s terms
are contrary to the public interest. The
ALJ therefore recommends that the Commission accept the Stipulation and
consider the Company’s proposal to establish an FCA true-up mechanism when that
tariff filing is made.
XII.
economic
development.
231.
The OES supports
recovery of Minnesota Power’s economic development expenses, but only to the
extent that those expenses are shown to be cost-effective by application of a
“rate payer impact test” demonstrating that the Company’s economic activities
have a quantifiably favorable impact on its rates. The OES does not propose any quantitative
rate payer impact test. Rather, it
asserts that Minnesota Power bears the burden of developing such a test and
demonstrating that the results of that test establish a favorable impact on
customer rates. Finally, the OES argues
that since Minnesota Power has failed to produce a reliable quantitative ratepayer
testing methodology and therefore failed to establish which of its economic
development activities have had a quantifiably favorable impact on its rates,
all of its economic development costs should be disallowed.[308]
232.
First of all,
Minn. Statutes § 216B.16, subd. 13, provides:
In other
words, the Legislature has given the Commission statutory discretion to allow a
utility to recover economic development expenses in a rate proceeding.
233.
The ALJ observes
that economic development activities are like planting seeds. While some may grow to maturity and bear
fruit in the form of new rate paying customers, others may fail to
germinate. Even the economic development
activities that do bear fruit may do so at different rates and over different
time periods. In short, the impact of a
utility’s economic development activities on ratepayers does not lend itself to
quantitative analysis. The Commission
recognized this in its Otter Tail Power decision:
As this
Commission has previously concluded, any link between economic development
expenditures and benefits to rate payers will of necessity be indirect. This
indirect impact of necessity means that such costs are not easily translated
into hard data analysis.[309]
234.
The fact that
the OES has not suggested a methodology for quantitative analysis of the
financial impact of economic development activities on ratepayers is further
evidence of its lack of susceptibility to quantitative analysis.
· The Northland Foundation urged support for MP's
economic development programs as a key part of numerous beneficial programs for
· The Minnesota Community Development Fund (MCCF)
supported MP's request for recovery of its economic development costs. MCCF noted that through MP's founding
membership in MCCF, the Company has been instrumental in obtaining more than
$10 million for northeastern and central
· Walt Prahl, President of Eventis/Hickory Tech, urged
approval of MP's economic development costs due to MP's "very active and
critical role in the economic development efforts within
· Sansio (a Duluth-based software company) described
the impact of MP's economic development programs on its own business and the
opportunities for its employees. The
Brainerd Lakes Area Development Corporation (BLADC) described the significance
and importance of Minnesota Power’s partnership in the region’s economic
development efforts, including financial support, sharing information, and
fostering entrepreneurial development.
· The Hubbard County Regional Economic Development
Commission (Hubbard County EDC) described its efforts with the Greater
Minnesota Housing Fund, the Minnesota Housing Partnership, and the Minnesota
Housing Finance Agency on a pilot/demonstration project to significantly reduce
energy consumption and provide significant dollars for residential rehab
efforts. Within the distressed
neighborhood where this project was conducted, Minnesota Power provided energy
audits to assist lower income customers an opportunity to reduce their energy
costs. Hubbard County EDC supported
recovery of MP's economic development costs to encourage this sort of
participation in the future.
· Cirrus Design Corporation [the aircraft manufacturer]
described how MP assisted in relocating Cirrus to
· Brian Graff, Vice President, Marketing & Public
Relations, SMDC Health Systems, noted that Minnesota Power has been an
effective leader on economic development issues (at times the sole effective
leader) in many areas and has been a reliable partner and catalyst across the
board. MP helped found Area Partnership
for Economic Expansion (APEX), a partnership for improving business
opportunities in
· Marlene Pospeck, Mayor of Hoyt Lakes, described MP's
economic development programs as:
"an extremely important element our in efforts to diversify the
economy of the
236.
On the other
hand, the assertion of the OES that a utility’s economic development activities
should also benefit its shareholders is a valid observation. In its Otter Tail Power decision, the
Commission recognized the dual nature of potential benefits when it allowed
that utility to recover 50% of its economic development expenditures.[310] Therefore, in
the absence of a reliable quantitative method for apportioning the financial
impact of Minnesota Power’s economic development activities between ratepayers
and shareholders, the ALJ recommends that the Commission apportion those
expenditures equally between ratepayers
and shareholders and allow Minnesota Power to recover 50% of those expenses in
this rate proceeding.
237.
Minnesota Power
originally proposed to recover $1,191,789 as rate case expenses, with none
of those costs allocated to the non-regulated company, and to amortize those
expenses on a three-year schedule;[311] it also
proposes to include the unamortized balance of rate case expenses in the rate
base.[312] Although the OES agreed with the amount that
the Company had proposed as rate base expenses, the OES disagreed with the
three-year amortization schedule and with including the unamortized balance of
rate case expenses in the rate base; it also argued that a portion of the rate
case expenses should be allocated to the non-regulated company. [313]
238.
As previously
discussed, selecting a reasonable amortization period for rate case expenses is
important because ratepayers will continue to pay those costs in base rates
until MP files its next rate case and may overpay significantly if MP fails to file
its next rate case by the end of the amortization period. The amortization period should therefore
represent as accurate a prediction as possible of the period of time that will
elapse until the Company’s next rate case.[314]
239.
In support of
its proposal to recover rate case expenses on a three-year amortization
schedule, the Company asserts that it is “willing to put into a Commission
order at the end of this case the requirement that the Company file another
rate case on or before December 31, 2011.”[315] The OES takes
a more conservative approach to predicting an appropriate amortization
period. It argues that while MP’s
current belief is that the Company will file a rate case in 3 years, many
factors can impact the need for a utility to file a rate case, including but
not limited to inflation, cost-of-money, the currently allowable rate of
return, construction activity, and changes in the amount of customers’ usage,
along with accounting and policy changes.[316] The OES proposes a five-year amortization
schedule for rate case expenses, arguing that it corresponds to the average
period between rate cases that the Company has filed since 1976.[317]
240. The ALJ has previously recommended in this report
that the Commission accept Minnesota Power’s proposal to file another rate case
within three years by directing the Company to do so. If the Commission accepts that
recommendation, it should allow
241.
The OES argues
that the fact that ALLETE includes not only Minnesota Power but also several
non-regulated operations complicates regulatory review. Because of the Company’s non-regulated
activities, the OES contends regulatory assessments of its rate proposal necessarily
include considerable time devoted to issues of allocating costs and revenues
between the utility and non-regulated operations. For example, if ALLETE did not have
non-regulated activities, there would be no need for corporate cost
allocations. Therefore, the OES contends
that it is reasonable for the Company to also allocate to non-regulated
activities a portion of the regulatory assessment costs, as well as the Company
costs.[318] The OES further argues that it had
recommended and the Commission had approved an allocation of rate case expenses
to the non-regulated activities in at least nine previous case proceedings.[319] The OES therefore recommends an allocation of
5.76 percent of the rate case costs, or $68,647, to the non-regulated
operations of Minnesota Power.[320] The OES arrived at that percentage by dividing
the Company’s test-year non-regulated corporate support services costs by the
sum of its test-year regulated and non-regulated corporate services costs.[321]
242.
In response,
Minnesota Power accepts in principle the proposition that some rate case
expenses should be allocated to non-regulated activities, but disagrees with
the amount that the OES is proposing to allocate.[322] Although the Company agrees that 5.76% of
certain rate case expenses may be allocated to non-regulated activities, it
maintains that regulatory assessments are expenses that should not be included in
that allocation.[323] It therefore proposes an adjustment of $34,087,
which it computes by calculating 5.76% of $1,191,789 less $600,000 in regulatory
assessments for total allocable expenses of $591,789.[324] The net effect of Minnesota Power’s approach
is to reduce the amount of allocable rate case expenses to 2.86 percent of the
total rate case expenses.[325] The Company argues that a lower adjustment is
warranted because “it is not logical to allocate any portion of the regulatory
assessment (the costs of the Commission and the OES) to the non-regulated
activities of Minnesota Power.”[326]
243.
The OES’s
decision to include all rate case expenses in its analysis has a rational
basis. There appears to be no
discernable reason for excluding regulatory assessment from the total. Computing the allocation percentage by
dividing the Company’s test-year non-regulated corporate support services costs
by the sum of its test-year regulated and non-regulated corporate services costs
also has a rational basis. It appears to
the ALJ that what the Company appears to consider illogical is prescribing an
allocation percentage of 5.76 percent in the proceeding in comparison with the
0.35 percent allocation percentage that the Commission approved in 1994. However, the Company offers no insight into
the Commission’s reasoning for approving a lower allocation percentage in
1994. Any number of things may have entered
into the Commission’s 1994 decision. In
the absence of a reasoned explanation of why a lower allocation percentage was
approved in the earlier proceeding, the ALJ recommends that the Commission
accept the OES’s analysis and approve an allocation of 5.76% of $1,191,789 in
rate case expenses to non-regulated activities.
244.
Finally, like
the Company’s proposal regarding unamortized deferred MISO 16 and 17 costs,[327] the Company
also seeks to include unamortized rate case expenses in the rate base.[328] Again, the Company contends that because
those costs originate as an expense, the delay between payment and recovery
warrants treating the unamortized rate case expenses as an investment, like
other kinds of prepaid expenses. The
Company therefore argues that its shareholders should be able to earn a return
on the amounts not paid in the test year.
Moreover, in further support of rate base treatment of unamortized rate
base expenses, Minnesota Power cites the fact that the Commission approved a
three-year amortization of rate case expenses, including the unamortized
balance in the rate base in Company’s 1994 rate case.[329] Minnesota Power requests the identical
treatment of rate case expenses in this case.[330]
245.
In response, the
OES argues against rate base treatment of unamortized rate case expenses for
the same reasons it argued against rate base treatment of deferred MISO Schedule 16 and 17 costs. The OES asserts that the Company’s proposal
to place those costs in the rate base is inconsistent with ratemaking
principles in that returns are
generally allowed only for capital costs, and that rate case expenses and
should be treated as traditional expenses are treated, without a return. Additionally, the OES notes that since the
Company’s 1994 rate case, the Commission has consistently denied recovery of
rate case expenses in rate base in twelve more recent rate cases.[331]
246.
It appears to the ALJ that the Commission has adopted another
policy about rate base treatment of rate expenses since the Company’s 1994 rate
case. Like legislatures, past
Commissions cannot bind the action of future Commissions, and in this case the
Commission has given utilities thirteen years’ notice of its change in
policy. The ALJ therefore recommends
that the Commission not allow the Company to recover unamortized rate case
expenses in its rate base.
A.
Agreed-upon Adjustments
to Rate Base.
247.
In setting rates
for a public utility, the Commission must determine the total level of
investment by the utility in its “utility property used and useful in rendering
service to the public.”[332] In utility rate cases, those investments are
referred to as the utility’s rate base. MP’s
initially filed revenue requirement of $45,023,320 included a proposed rate
base of $713,096,651.[333] The OES and MP have agreed on the following
adjustments to the rate base as initially filed:
• Cash Working Capital Methodology; calculated by applying the
OES’s lead/lag days to the OES O&M expense adjustments; results in a
decrease to the test-year cash working capital requirement by $648,863 (This
particular amount assumes Commission approval of all OES proposed adjustments;
otherwise the amount will need to be recalculated).[334]
•
• BPUC Transmission Asset Sale; decrease test-year rate base by
$228,420.[336]
• Badoura Pine River Surface Project; decrease test-year rate
base by $3,913,595.[337]
• Mesabi Nugget Service Extension;
increase test-year rate base by $1,120,378.[338]
• Taconite Ridge Wind Project; increase
test-year rate base by $825,327.[339]
• Customer Advances and Deposits; decrease
test-year rate base by $2,526,812.[340]
• Deferred Taxes; decrease test-year rate
base by $6,198,049.[341]
• BEC-4 Boiler Project; decrease test-year rate base by $323,922.[342]
• E015/D-08-422 Depreciation Expense; decrease test-year rate base by $2,186,066.[343]
B.
Rate Base
Treatment for Deferred Rate Case Expenses.
249.
MP proposed
inclusion of the deferred portion of rate case expenses in the rate base. The OES objected to this proposal as
inappropriate. As discussed in the
general treatment of rate case expenses, the ALJ recommends that MP not be
allowed to include these deferred expensed in the rate base.[344]
C.
Rate Base
Treatment for Deferred MISO Schedule 16 and 17 Costs.
250.
MP proposed
inclusion of the deferred portion of MISO Schedule 16 and 17 costs in the rate
base. The OES objected to this proposal
as inappropriate. As discussed in the
general treatment of deferred MISO Schedule 16 and 17 costs, the ALJ recommends
that MP not be allowed to included these deferred expensed in the rate base.[345]
D.
Asset Retirement
Obligation Depreciation Methodology.
251.
The concept of
“decommissioning” involves the assumption that a generation facility will
eventually reach the end of its service life and will have to be shut down and
replaced. However, in practice decommissioning rarely occurs. The parties have cited no example of a
252.
Although
decommissioning may be rare, the Commission allows electric utilities to
incorporate the possibility of decommissioning plants into their rates by
allowing them to recover costs associated with decommissioning over time. In the past, the Commission has consistently
approved a cost-based method for accounting for decommissioning expenses and
net salvage value. That method (the
Decommissioning Method) involves estimating the plant’s negative salvage value
at the end of the plant’s service life and then amortizing that negative
salvage value, along with associated decommissioning expenses, on a straight
line schedule over the plant’s service life.[347] The asset
retirement obligation approach (the ARO Method) accounts for asset retirement
costs somewhat differently. It is based
on market value, and asset retirement costs are amortized on an accelerated
schedule over the plant’s estimated service life.[348] Thus, under the ARO method, ratepayers
receiving service from the asset at the beginning of its service life pay asset
retirement costs that are greater than ratepayers receiving service at the end
of its service life.[349]
253.
Generally
accepted financial accounting standards recognize both the Decommissioning
Method and the ARO Method as acceptable accounting methods. However, in an Order dated July 11, 2003, the
Commission placed the following conditions on Minnesota Power’s use of the ARO
method in future retail rate cases:
The Commission will accept MP’s accounting for
adoption of FASB 143 and the resulting regulatory asset, with the ultimate issue of rate recovery to be determined in
MP’s next rate
case proceeding. Additionally, regarding future rate recovery of this
regulatory asset MP will be required to show that the ARO
accounting method is a superior method for purposes of rate
recovery over the current salvage value method. Although
the Commission recognizes the importance of FASB 143 to provide consistent
accounting for asset
retirement obligations in Company’s financials, this function does not necessarily require a change
in our current rate
recovery method for salvage/decommissioning costs in future rate cases.[350] [Emphasis supplied.]
254.
In this rate
case, Minnesota Power seeks to recover asset retirement costs for its generation plants using the ARO accounting method. At the outset of
this proceeding, the Company included a total of $3.5 million in asset
retirement obligations in the test year.[351] The Company subsequently
reduced that amount by $826,211 because previously approved negative net
salvage value had been included in both the ARO amount of $3,507,674 and in its
depreciation expense amount for purposes of this rate case. Therefore, the amount that the Company is now
seeking to include in the test year for asset retirement obligations is
$2,681,463.[352] But the
issue in this rate case is not whether the Company’s ARO method is an
acceptable accounting method. Rather, as
the Commission previously directed, it is whether MP has shown ARO to be
superior to the net salvage value method and therefore, whether it is a
reasonable accounting method in this proceeding for ratemaking purposes.
255.
Minnesota Power advances three arguments in support of superiority
of the ARO Method over the Decommissioning Method:
(i) The ARO Method was developed, vetted and adopted
by two sophisticated accounting rule setting authorities, namely the FERC and
the Financial Accounting Standards Board;
(ii) The ARO Method is a more systematic and ratable
method of allocating asset retirement costs over the life of the asset. Under
the Decommissioning Method, more asset retirement expense is recognized near
the end of an asset’s life, thereby placing a disproportionate burden on future
ratepayers; and
(iii) Perhaps most importantly, the FERC approved
Minnesota Power’s ARO method in the Company’s most recent wholesale rate case.[353]
256.
The Company also
argues that the ARO Method is superior because it is measured at fair value
while the Decommissioning Cost Obligation is measured at current cost, without
taking into account the impact of inflation when the obligation is ultimately
settled in future years.[354]
257.
First, as the
OES and the LPI point out, the effect of Statement of Financial Standards
(“SFAS,” also called “FASB”) 143 is to establish the ARO Method as an acceptable way of accounting for the
future decommissioning of a generating plant in retail rate cases; it does not
purport to establish the ARO Method as superior to the Decommissioning Method
for that purpose.[355] That is the way the Commission regards FASB
143.[356]
258.
Second, like
FASB 143, FERC Order 631 does no more than prescribe accounting standards
whenever a utility happens to use the ARO Method. In fact, FERC Order 631 expressly states that
it does not purport to establish the ARO Method as superior to the
Decommissioning Method for the purpose of setting retail rates, and it
expressly does not require the Commission to adopt the ARO method for that
purpose:
The Commission [FERC] will decline to make policy
calls concerning regulatory certainty for disposition of transition costs,
external funds for amounts collected in rates for asset retirement obligations,
adjustments to book depreciation rates, and the exclusion of accumulated
depreciation and accretion for asset retirement obligations from rate base;
these are matters that are not subject to a one size fits all approach and are
better resolved on a case-by-case basis in rate proceedings. The
Commission is of the view that utilities will have the opportunity to seek recovery of qualified costs for asset
retirement obligations in
individual rate proceedings. This rule should not be construed as pregranted authority for rate recovery in a
rate proceeding.[357] [Emphasis supplied.]
259.
Accordingly, FERC approval of Minnesota Power’s use of the ARO
Method in its most recent wholesale rate case does not establish that it is
superior to the Decommissioning Method in this retail rate case.[358]
260.
Third, the ALJ agrees with the OES, LPI, and AGO/RUD that the ARO
Method is not substantively superior to the Decommissioning Method and
therefore more reasonable for retail ratepayers. First, Minnesota Power argues that by
correlating cost recovery to the declining value of generation assets, the ARO
Method provides a better matching of cost with the ratepayers who receive the
benefits.[359]
However, the ALJ is unable to discern any intrinsic societal value that
would be adversely affected by having future ratepayers pay a share of
decommissioning costs that was disproportionate to the depreciated value of the
asset providing them with electrical service.
261.
The Company also
argues that the ARO Method more accurately reflects the retirement obligation
and related annual expense because they are measured at fair value while the
Decommissioning Cost Obligation measures them at current cost. It argues that the Decommissioning Method
therefore does not take into account the impact of inflation when the obligation
is ultimately settled in future years.[360] But as the OES correctly observes, by
allowing the Company to recover those costs well in advance of the end of a
generating plant’s service life, it provides the Company with funds to cover
future decommissioning costs, thus allowing the Company to receive the time
value of money. On the other hand,
looking at the time value of money from the ratepayers’ perspective, the ARO
Method appears to confer an additional economic benefit to future ratepayers
that current ratepayers will not receive, since an accelerated amortization
schedule appears to distort the principle of time value of money to a greater
degree than a straight line amortization schedule.
262.
After
considering the Company’s arguments, the ALJ concludes that, in theory, the ARO
Method appears to be neither superior nor inferior to the Decommissioning
Method and, on that basis alone, appears to fail the test that the Commission
established on July 11, 2003. However,
as the OES observed, the ARO Method in practice is actually inferior to the Decommissioning
Method from the standpoint of its economic impact on ratepayers. In theory, the total costs recovered under
the Decommissioning Method and the ARO Method would be the same.[361] The accelerated nature of ARO means that the
amount of depreciation/decommissioning expense at the beginning of an asset’s
life is by definition too high for use in later years when it should be greatly
decreased. However, since the depreciation/decommissioning amount is fixed in
base rates at the time of a rate case, that amount will continue in rates until
the next rate case. Therefore, if ARO
were allowed for ratemaking, that “too high” amount will be fixed in rates, and
will continue in rates until the utility files its next rate case.[362]
263.
Put another way,
with straight line amortization, there is no possibility that the
decommissioning costs established for the test year will be unrepresentative of
the costs recovered in subsequent years throughout the life of the rate. However, that is not the case with the ARO
Method, since the costs assigned to the test year will have the effect of
creating a “step” during the life of the rate that would exceed what the normal
ARO amortization schedule would specify for subsequent year under that rate. The ALJ therefore concludes that the
Commission should not approve Minnesota Power’s proposal to use the ARO Method,
rather than the Decommissioning Method, for recovery of the future costs of
decommissioning its generation facilities, and that appropriate adjustments be
made to its test year and rate proposal.
E.
Non-Rate-Based
Generators.
264.
Minnesota Power
operates two generation facilities, with a combined output of 50 MW, that are
not currently included in its rate base—the Rapids Energy Center (“REC”)
adjacent to the Blandin paper mill in Grand Rapids and the Sappi/Cloquet
Generator No. 5 at the
265.
After
UPM-Kymmene (UPM or Blandin) purchased Blandin Paper’s operations in
266.
Under the terms
of the parties' Steam Service, Operation and Support Agreement, which the
Commission previously approved, the REC is dedicated to one customer— Blandin
Paper. For a set price, Minnesota Power
provides all the steam for Blandin's paper making, as well as an amount of
power for paper production equivalent to that produced by the on-site
generation units.[366] For the most part, the relationship between
the REC and the Blandin mill physical plants is such that most of the
electricity generated by the REC can only be produced when the mill is making
paper. There is, however, some ability
of the REC to generate a small amount of electricity when the paper mill is not
running and when the Blandin production process does not need it.[367]
267.
Section 6 of the
Steam Service, Operation and Support Agreement between MP and Blandin provides
that if the parties do not reach an agreement by March 2010 to extend the
current cogeneration agreement or have Blandin purchase back the cogeneration
facilities at the production site, then Minnesota Power will remove the
facilities consistent with the terms of its site lease with Blandin. Minnesota Power and Blandin are currently
engaged in discussions regarding whether their cogeneration agreement should be
extended past 2010 or whether Blandin should purchase back some or all of the
cogeneration facilities.
268.
The record therefore
establishes that UPM currently has a contingent property interest in the REC
that is likely to be either perfected or extinguished on or before March
2010. As previously, discussed, The ALJ
has recommended that the Commission order the Company to file another rate case
by December 2011.[368] Because of the possibility that the REC will
no longer be owned by Minnesota Power in December 2011, or even exist, the
Commission should delay consideration of the inclusion of REC into the rate
base until Minnesota Power’s next rate case.
269.
In 2000,
Minnesota Power entered into an agreement with Sappi (formerly Potlach),
another paper mill operator, under which Minnesota Power installed a 25 MW
turbine generator at Sappi’s Cloquet Mill site (Sappi 5).[369] Under that agreement, Minnesota Power retains
its ownership interest in the turbine generator while Sappi owns the boilers
and other infrastructure needed for the generation of electricity at the Sappi
5 site.[370] The agreement also provides that Minnesota
Power must pay for the fuel and O&M related to operation of Sappi 5 and
must also make a monthly Infrastructure Payment to Sappi for using Sappi’s
boilers and infrastructure to produce the steam used to generate
electricity. The electricity produced
using Sappi 5 is not dedicated to Sappi, but rather is available to meet the
needs of all Minnesota Power’s retail customers. The parties’ agreement also provides that in
May 2016 Sappi has the right to purchase the turbine generator from Minnesota
Power for $1.[371] Since the power produced at Sappi 5 is not
dedicated to Sappi’s Cloquet mill, Sappi obtains the electrical power needed to
operate that mill from Minnesota Power’s general retail distribution system,
like any other retail customer.[372]
270.
The OES argues
that both the REC and the Sappi 5 generation facilities should be placed in
Minnesota Power’s rate base in this proceeding and cites five reasons. First, since even the Blandin mill is a
retail customer, both facilities are used to serve the Company’s retail
customers. Second, the generation from
both is included in the most recent Mid-Continent Area Power Pool Load and
Capability Report.[373] Third, the Company included generation from
the two facilities in its 2004 Integrated Resource Plan (IRP) as generation
required to serve retail customers.[374] Fourth, MP counts these generators towards
its Renewable Energy Standards and in the Midwest Renewable Energy Tracking
System; and fifth, the generators are owned by Minnesota Power, the regulated
utility. In the ALJ’s view, apart from
considerations of administrative consistency, whether the REC and Sappi 5
facilities should be included in the Company’s rate base at this time turns on
the answers to two questions: (1) to what extent are the two facilities
dedicated cogeneration facilities? and (2) to what extent would rate base
treatment at this time interfere with the contractual rights of third parties?
271.
There are
material differences in function and status between the REC and the Sappi 5
generation facilities. The REC and the
Blandin mill are physically codependent; the REC can generate electricity only
when the mill is making paper. Blandin
has a contractual right to all of the output of REC and normally does use all
of REC’s output. The energy that
Minnesota Power’s other retail customers obtain from that facility is, at best,
minimal and sporadic. Moreover, the REC
not only provides the Blandin mill with electric energy, it also provides the
mill with steam necessary for the papermaking process. Finally, if not exercised, UPN’s contingent
contract right to regain ownership of the REC expires in March 2010—that is, the
ownership status of the REC will be resolved before Minnesota Power files its
next rate case. In short, the REC
generation facility represents a true cogeneration facility, and placing it in
the Company’s rate base now could result in legal complications. It therefore makes sense to wait until
Minnesota Power’s next rate case to determine whether that facility, if still
owned by the Company, should be placed in the rate base.
272.
The Sappi 5
generation facility’s situation is materially different. Its output does not go directly to Sappi’s
Cloquet mill, and it is therefore not physically a cogeneration facility. Rather, all of Sappi 5’s output goes into
Minnesota Power’s retail distribution system and is therefore available to all
of the Company’s retail customers.
Conversely, Sappi obtains all of the electrical power needed to operate
its Cloquet mill from Minnesota Power’s retail distribution system, like other
retail customers. It was Minnesota Power
that supplied the turbine and has operated the Sappi 5 facility since it was
constructed. Although Sappi owns the
property on which the facility was constructed and some of the infrastructure,
it has never been directly involved in its operation. Sappi also possesses an option to purchase
the facility that is exercisable in 2016, but that will not occur before
Minnesota Power files its next rate case.
Unlike REC, there is no evidence that Sappi is currently involved in
negotiations regarding exercising its contractual right to purchase the
facility, and if Sappi indicates an intent purchase it in 2016, there would
have to be a proceeding before the Commission during which removal of Sappi 5
from the rate base could also be considered.
The ALJ therefore concludes that, unlike the REC, the Sappi 5 facility
should be accorded rate base treatment now.
273.
If the Commission
were to place either the REC or Sappi 5 facility, or both, into the rate base
in this proceeding, the corresponding revenues, depreciation, and O&M
expenses for the facilities associated with them must be recognized at the
appropriate level. With regard to
revenues, the OES recommends that a pro rata share of total system revenues
should be assigned to the facilities on the basis of megawatts produced. On the other hand, Minnesota Power argues
that the Commission should accept the facility-specific budget revenue
estimates for the test year that the Company produced for each of the two
facilities. Neither the OES nor any
other party raised specific criticisms of the reliability of the
facility-specific revenue estimates that the Company made for the REC and Sappi
5 facilities. In the absence of such
criticisms, the Commission should accept Minnesota Power’s facility-specific
estimates rather than the OES’s more general allocation based on system-wide
averages.
274.
On the other
hand, the OES does take issue with some of the expenses that Minnesota Power
proposes to allocate to those two facilities—specifically, depreciation and
O&M expenses. In its direct
testimony, the OES expressed concern that the Company’s composite depreciation
rates of 11.76% and 9.64% appear to be based on depreciation lives that were
unreasonably short, and it recommended that the Company address this
depreciation life issue in their reply comments.[375]
275.
Minnesota Power
indicated that depreciation for the REC was based on a March 2010 termination
date, and that Sappi 5 was based on a
May 2016 termination date, the dates on which the respective contracts with UPN
and Sappi expire. In response, the OES
pointed out that the life of the underlying contracts did not bear any
necessary relationship with the service lives of those assets, and OES
continues to object to the Company’s depreciation expense for the REC and Sappi
5 facilities.[376] Minnesota Power has never come forward with
proposed depreciation expenses for the two facilities for the test year based
on reasonable estimates of their service lives.
The ALJ therefore suggests that there are three options available to the
Commission: (1) deny any depreciation expense for REC or Sappi 5 to the extent
that either is included in the rate base; (2) allow depreciation expense based
on the average service life of Minnesota Power’s other generation facilities;
or (3) provide the Company with the opportunity to come forward with reasonable
estimates of the service lives of either or both of the facilities with
appropriate documentation.
276.
With regard to
the O&M expenses associated with the REC and Sappi 5 facilities, the OES
also believed that $17.1 million in O&M expenses that Minnesota Power
allocated to the two facilities in the test year were unreasonably high and
unsupported. Similar to its
recommendation for revenues, the OES therefore proposed to assign a pro rata
share of total system O&M expenses to the facilities on the basis of their
respective capacities in megawatts.[377] In rebuttal testimony, the Company came
forward with a more specific listing of the O&M expenses that it had
budgeted for the two facilities for the test year.[378] In its surrebuttal, the OES still expressed
concern that in the aggregate, the O&M costs of the REC and Sappi 5 were
nearly four times higher than their pro rata shares of system wide O&M
costs.[379]
277.
During the
hearing, Minnesota Power provided the other parties with detailed listings of
its projected test year O&M expenses for the REC and Sappi 5 facilities,
together with comparisons with the actual O&M expenses for the two
facilities in 2006 and 2007.[380] More specifically, the Company provided that
information off the record to the other parties after the hearing recessed on
Monday, November 17, 2008. It subsequently
introduced that information into the hearing record as Exhibit 101 on
Wednesday, November 19th, the next to the last day of the hearing.[381] Although that exhibit contains detailed
information on the Company’s proposed test year costs for the REC and Sappi 5
in comparison with the actual O&M costs for 2006 and 2007, there was
relatively little time for other parties to analyze the additional data
contained in that exhibit, and the hearing record therefore contains little
testimony or further information about its contents.
278.
Nonetheless,
that data raises some significant issues that need to be resolved. Exhibit 101 indicates that certain O&M
expenses for the Sappi 5 facility, which the Company aggregates as
“Miscellaneous Expenses” in 2006 and 2007, are shown to increase about 16% from
2006 to 2007 and, again, from 2007 to the test year, and there is no
explanation of what accounts for those rather large annual increases. The O&M expenses for the REC increased about
2% from 2006 to 2007 but about 8˝% from 2007 to the test year. While many of the test year O&M expenses
appear reasonable in comparison with 2006 and 2007, at least two categories
show unusually high and unexplained increases in comparison with the actual
expenses for 2007—namely, Salaries and Wages (9%) and Contract Services
(63%). Those two expense items alone
represent $1,032,577 in increased test year costs for the REC. Lacking reasonable explanations, the ALJ
cannot conclude that Minnesota Power has demonstrated that its proposed test
year O&M expenses for the REC and Sappi 5 facilities are reasonable. Again, the ALJ suggests that there are three
options available to the Commission: (1)
deny any test year O&M expenses for whichever of the two facilities is
included in the rate base; (2) allow O&M expenses based on allocating each
a pro rata share of system wide O&M costs; or (3) provide the Company with
an opportunity to come forward to demonstrate reasonable bases for those costs.
279.
Minnesota Power
opposes putting either the REC or S/C5 into rate base. In the event they are placed into the rate
base, the Company argues that the revenues for the facilities must be facility-specific
budgeted revenues documented by MP in is testimony and schedules.[382] On the other hand, the OES argues that the
assigned revenues should be based on system wide average revenues per MWh
multiplied by the number of MWh generated by each of the generators in
question.[383] However, unlike the Company’s projected
O&M expenses, the OES cites no reason for questioning the accuracy of MP’s
facility-specific revenue estimates.
Lacking support or a reason to reject the actual revenue figures, the
ALJ concludes that the Commission should accept Minnesota Power’s
facility-specific revenue estimates.
A.
Class Revenue Apportionment.
280.
When setting
rates, the Commission is responsible, in part, for determining how much each
customer class should contribute to meeting the utility’s revenue requirement. In making that determination, the Commission
considers the following factors:
The Commission requires utilities to file a CCOSS because the cost a
utility incurs to provide service is one factor the Commission considers in
determining how much each customer class should contribute to meeting the
utility’s revenue requirement, and how to recover each class’ share of the
revenue requirement from the members of the class. Other factors include economic efficiency;
continuity with prior rates; ease of understanding; ease of administration;
promotion of conservation, ability to pay; and ability to bear, deflect or
otherwise compensate for additional costs.[384]
281.
Minnesota Power
apportioned its total revenue responsibilities among rate classes based on its
CCOSS and its rate design objectives, which included cost based rates, maintaining
reasonable rate continuity, mitigating rate shock, and encouraging the
efficient use of resources. The Company
proposed the following allocations among customer classes:
|
Class Revenue Responsibility — Proposed Increase by
Class[385] |
||||||
|
Customer Class |
|
Increase by Class (as originally proposed) |
Class Responsibility for Percent of Total Revenue |
Percent Increase in Revenue |
|
Percent Revenue Responsibility Differs from Cost Responsibility |
|
Residential |
|
$17,041,158 |
17.4% |
23.8% |
|
-12.3% |
|
General Service |
|
10,281,072 |
10.8% |
23.0% |
|
-1.4% |
|
Large Light and Power |
|
4,527,801 |
13.9% |
6.8% |
|
-0.1% |
|
Large Power |
|
21,324,514 |
56.4% |
4.5% |
|
4.3% |
|
Municipal Pumping |
|
803,775 |
0.9% |
20.1% |
|
24.6% |
|
Lighting |
|
-0- |
0.5% |
0.0% |
|
35.7% |
|
|
|
|
|
|
|
|
|
Totals |
|
$44,978,320 |
100% |
9.7% |
|
0.0 |
282. The Company maintains that its proposal for rate
design will move residential customers much closer to cost-based rates. After an initial residential rate increase of
14% in 2009, Minnesota Power is proposing two additional phased-in increases of
5% in 2010 and 2011. This phase-in is
balanced by a comparably larger initial increase in the Large Power class, with
corresponding reductions of that Large Power rate increase in tandem with the
Residential class phase-in.[386]
283.
The
MCC would prefer a rate design that reflected customers’ full cost of service,
but it supports Minnesota Power’s proposal for rate design as a step in that
direction.[387]
284.
The OES agrees
with MP that the existing residential class subsidy needs to be reduced. The OES supported the revenue apportionment
among classes that the Company initially proposed, as well as the costs of
serving each class identified in MP’s CCOSS.[388] However, the OES later added a qualification
to that support. Minnesota Power identified three Large Power
customers who have recently signed contract amendments that impact MP’s sales
and revenue forecasts, and the Company is seeking adjustments to its revenue
deficiency based on those amendments.
The Company is also seeking adjustments resulting from changes to
contracts with two Large Light & Power customers.[389] The
OES objects to those adjustments and takes the position that if the Commission
were to grant the adjustments, they should be recovered from the affected
customer class.[390]
285.
The OAG/RUD denies
the existence of any current subsidy of residential class customers. It argues that no fully distributed, embedded
CCOSS can be used to support the finding of a subsidy because it assigns joint
and common costs to customer classes,[391]
and it asserts that the Company’s CCOSS has been skewed to overemphasize costs
to the Residential class. As an example,
the OAG/RUD notes that 11% of the kWh sales were allocated to the Residential
class, while the CCOSS assigns 12.7% of the rate base cost of generation and
transmission to that class. By contrast,
the OAG/RUD points out that the Large Power class purchases 69% of the power,
but is only assigned 65% of the generation and transmission cost.[392] The OAG/RUD further argues the Peak and
Average method that the Company used to allocate production and transmission
costs was a subjective choice and inferior to other possible methods of
allocation.[393] Assuming that, as Company argues, its unique
concentration of large customers poses a risk to investors and increases the
cost of capital, the OAG/RUD also argues that that factor should be taken into
consideration in the allocation of costs, with the large power class
responsible for costs in proportion to the risk it poses to the Company.[394]
286.
The ECC
maintains that both Minnesota Power and the OES have ignored Commission
precedents establishing that non-cost factors should be taken into account when
setting class revenue requirements that govern rate design. More specifically, the ECC argues that both
the Company and the OES have given insufficient consideration to ability to
pay. The ECC argues that nearly forty
percent of MP’s residential customers will experience 42-55% rate increases,
and that since the Company’s proposal adversely affects a significant number of
MP’s customers, particularly low usage and low and fixed income customers, the
proposal is both inequitable and unreasonable.[395]
287. The LPI’s position is that the inter-class subsidy
that residential customers are currently receiving from Large Power customers
is inherently inequitable and should be eliminated.[396] The goal should be
to fully reflect the results of the Company’s cost of service study, which requires
eliminating all inter-class subsidies.[397] To the extent the
Commission approves an increase that is smaller than the Company has requested,
the LPI argue that the reduction should be first used to reduce or eliminate
the subsidy paid by the Large Power class.[398] If the Commission approves
a fairly small reduction of the Company's requested increase, the LPI recommends
that the phase-in of residential rates be expanded to include further 5%
residential increases in 2012 and 2013 and that there should be corresponding
reductions of the subsidies paid by rate classes whose rates are above cost of
service.[399]
288.
In analyzing MP's
CCOSS, the OES considered the extent to which revenue apportionment assigned to
each customer class a percentage share of the Company’s revenue requirement in
a way that satisfies four rate design principles: (1) provide the utility a
reasonable opportunity to recover its revenue requirement (thus, 100 percent of
cost responsibility is assigned to customer classes as a total); (2) ensure, as
much as possible, that each class recovers all of the costs identified by the
CCOSS as caused by that class, without subsidization or subsidy, in order to
enhance efficiency and encourage conservation; (3) avoid sudden dramatic rate
changes that may cause “rate shock;” and (4) establish rates that are
understandable.[400] The OES’s
proposed revenue allocation reasonably meets criteria 2 and 3, above, by reducing
inter-class subsidies while avoiding rate shock. The OES’s proposal results in moderate
percentage increases in the allocation of cost responsibility to the
Residential and General Services class customers, which under MP’s current rate
design are assigned substantially less revenue responsibility than their
respective costs of service.[401]
289.
The Commission
has historically considered a variety of cost and non-cost factors when
designing rates. As the Minnesota
Supreme Court explained in St. Paul Area
Chamber of Commerce v. Minnesota Public Service Commission:
Once
revenue requirements have been determined, it remains to decide how, and from
whom, the additional revenue is to be obtained.
It is at this point that many countervailing considerations come into
play. The commission may then balance
factors such as cost of service, ability to pay, tax consequences, and ability
to pass on increases in order to achieve a fair and reasonable allocation of
the increase among the consumer classes.[402]
290.
The Commission
has also identified a number of cost and non-cost factors to consider when
determining customer class revenue responsibility. Both types of factors are important to determine
just and reasonable rates. The factors
identified by the Commission include avoidance of rate shock for individual
customer classes, low-income customers’ ability to pay, a company’s ability to
recover the rate increase from others, the ability of companies to decrease the
burden of a rate increase through tax deductions, and the recognition of the
historical continuity of rates and rate increases.[403]
291. The intervenors have offered differing proposals for customer
class revenue responsibility. The LPI’s
proposals reduce inter-class subsidies, but could result in increases for some
classes that are large enough to result in rate shock.[404] The OAG/RUD and the ECC argue, in effect,
that subsidies to the residential and general service classes should be
increased.[405] The OES supports the Company’s customer class
revenue allocation, subject to the Commission denying certain adjustments which
the Company is seeking.
292.
Of the various
revenue allocation proposals, the ALJ concludes that a modification of the
MP/OES allocation proposal best reflects and balances the relevant cost and
non-cost factors. The ALJ has recommended elsewhere that the Commission grant
the downward revenue adjustment that MP is seeking for LLP customer Ainsworth,
but that the Commission adjust the revenue forecast that MP is proposing for Polymet
upward, as recommended by the OES.[406] The ALJ has
further recommended that the Commission deny MP’s request to make downward
adjustments to the sales forecasts of Large Power customers Hibbing Taconite,
United Taconite, and Enbridge based on proposed contractual rate changes.[407] If the Commission
accepts those recommendations, then the ALJ further recommends that the
adjustments made to the revenue forecast of the Large Power class be allocated
to that customer class for recovering required revenue, as proposed by the OES.
With those exceptions, the ALJ concludes that
the revenue apportionment proposed by MP minimizes the effects of rate shock,
while modestly addressing subsidies between customer classes and therefore
recommends that revenue apportionment.
B.
MP’s Composite
Allocation Methodology.
293.
Minnesota Power
used an allocation method called the Peak and Average (“P&A") method
to allocate production and transmission fixed costs in its CCOSS.[408] Under that method, the demand related
classification of fixed costs is calculated by dividing a class annual
coincident peak ("CP") or demand by the sum of the system annual CP
or demand plus the average demand or energy.
The coincident peak is the demand that a rate class experiences at the
time of a system peak or maximum demand.[409]
294.
The LPI argued
that the P&A method is flawed,[410]
and that the coincident peak method, which uses the peak for each class at
system peak demand to allocate the fixed costs of production and transmission,
best reflects cost-causation.[411] However, the LPI indicated that if the Commission
were to determine that average demand (usage) should be reflected in the
allocation of fixed production and transmission costs, then the average and
excess demand (A&E) method of fixed production and transmission costs would
be a more reasonable method than the P&A method for giving weight to
average demand or usage.[412] Nonetheless, the LPI recommended the
Commission base the revenue allocation in this case on the cost of service
study presented by the Company, but that for MP’s next rate case, it would be more
appropriate for the Company to base the allocation of revenues on the more
traditional and cost reflective A&E method.[413]
295.
Although the OES
agrees that the Peak and Average method proposed by MP and the A&E method
proposed by the LPI are among those recommended by NARUC,[414]
the OES indicated that a third method—the Equivalent Peaker method, which the
Commission has approved in its two most recent rate cases— was the most
appropriate method for classifying and allocating production plant costs. Rather than change the allocation method in
this rate case, the OES recommended that Minnesota Power use the Equivalent
Peaker method in its next rate case, or explain why it had chosen to use a
different method.[415]
296.
Since none of
the intervenors has objected to Minnesota Power’s use of the P&A method to
allocate production and transmission fixed costs in this rate case, the ALJ
recommends that the Commission approve the Company’s use of that method and
direct MP to use a different method in its next rate case, if the Commission
considers that to be appropriate.
C.
Residential and
Dual Fuel Interruptible Residential Customer Charges.
297. Customer billings are typically comprised of a
monthly customer charge, paid by any customer connected to a utility’s system,
and usage charges for the electricity consumed.
The monthly customer charges are set by class and may differ by zones
within a utility’s service area. MP’s
existing Residential Service Charge is $5.00 and includes the first 50 kWh of
electricity usage. MP’s CCOSS indicates that
residential customer-specific costs are $24.79 per customer per month. Based on that analysis, the Company proposes
to increase the Residential Service Charge to capture more, but not all, of
that cost through the fixed monthly charge.
MP indicated that because it recognizes “the imprecise nature of cost of
service studies and the need to avoid extreme rate shock,
298.
To support its
proposed increase in Residential Service Charge, Minnesota Power compared its
proposal to the monthly service charges of several distribution cooperatives
adjacent to Minnesota Power’s service territory, which also provide electric
service to residential customers. The
Company argues that those rates would be “a good proxy for the level of service
charge Minnesota Power ratepayers could reasonably tolerate because the
customers/members of cooperatives live in the same region as Minnesota Power
customers and are subject to similar economic conditions and financial challenges.”[417] However, distribution cooperatives are
member-owned, and their rate design is not necessarily comparable to that of
investor-owned utilities, particularly one with the customer mix of
299.
As the OES
observes, customer service charges set below cost represent an intra-class
subsidy. Intra-class subsidies arise when some
customers within a class pay more than the cost to serve them and subsidize
other customers within the same class who pay less than the cost to serve them.[419]
Intra-class subsidies occur when customer charges are set below
costs. These intra-class subsidies occur
because revenue responsibility apportioned to the class must be recovered
either through the customer charge or through the energy charge. To the extent customer charges do not recover
the full cost of connecting and keeping a customer on the system (including
connecting to the system along with ongoing metering, billing, customer service
and repair), the costs associated with these services will be recovered through
the energy charge. As a result,
customers with higher monthly usage pay through their energy charges not only
for energy costs, but also for the customer costs added to the energy charge. High-usage customers are therefore
responsible for the revenue that would otherwise have been collected in a
monthly customer charge from low-usage customers.[420]
300. Because the
Company’s current $5.00 per month Residential Service Charge does represent an intra-class subsidy, the
OES agrees with the principle of moving customer services charges closer to
cost over time, but it believes that MP’s proposal both to double the customer
charge and to eliminate the first 50 kWh from the customer charge could result
in rate shock for residential customers.
The OES therefore recommends moderating the increase in the proposed
customer charge from the proposed $10 to $8 per month as a means of reducing
rate shock.[421]
301.
On the other
hand, both the OAG/RUD and the ECC object to the Company’s proposed increase in
the Residential Service charge for residential classes and propose that MP
retain the existing customer charge of $5.00 for all residential classes.[422] They maintain
that a 100% increase of the Residential Service charge constitutes rate shock,
is based on an imprecise and defective CCOSS, and is inconsistent with
residential service charges that the Commission has recently approved for other
publicly owned utilities.[423] The OAG/RUD
and the ECC also assert that any increase in the customer charge contravenes
the directive in Minn. Stat. § § 216B.03 to promote conservation.[424]
302.
Additionally,
both the OAG/RUD and the ECC argue that even the lesser charge recommended by
the OES still amounts to a 60% increase in the customer charge and is
unreasonable.[425] Noting that
an increase of that magnitude was rejected by the Commission in the 2005 CenterPoint Energy Rate Order,[426] the OAG/RUD and the ECC maintain that the
Commission’s reasons for rejecting the increase in that matter apply with equal
force in this proceeding.[427] The ECC also
points out that the Commission’s decision in that case was based, in part, on
the desire to promote conservation.[428]
303.
The Commission
has described its approach to customer service charges as follows:
The customer charge has two main functions, one practical and one grounded
in ratemaking policy. Its practical
function is to help stabilize utility revenues and reduce the risk that the
utility will over- or under- recover its revenue requirement due to
weather-related fluctuations in gas usage and sales. Its ratemaking function is to ensure that
each customer bears responsibility for a certain level of the Company’s fixed
costs regardless of usage.[429]
After
acknowledging that Residential customer charges cause customer dissatisfaction,
the Commission went on to state:
[C]ustomer charges play an important role in the rate
structure. They reduce utilities’
capital costs by ensuring baseline levels of revenue, thereby reducing
consumers’ rates. They help mitigate
rate volatility between seasons by recovering some fixed costs during the
low-usage, summer months. They promote
equity by ensuring that the rate structure does not shift the full system-costs
imposed by low-usage and seasonal customers to normal-usage, high-usage, and
year-round customers.[430]
304.
In this rate
proceeding, the ALJ concludes that the OES has demonstrated that an increase in
the Residential customer charge to $8.00 appropriately assigns costs to that
class, while avoiding customer confusion or rate shock. The ALJ therefore recommends that the
Commission reduce the Company’s proposed Residential Service Charge from $10.00
to $8.00 per month.
305.
The Company also
proposes to increase the Residential Service Charge for Dual Fuel Interruptible
residential customers from $5.00 to $10.00 to correspond with its proposal to
increase the service charge for residential customers.[431] Although none of the Intervenors specifically
addressed that issue, the ALJ also recommends that the Commission reduce that
service charge from $10.00 to $8.00 per month to bring it into conformity with
the ALJ’s recommendation on the residential customer charge.
D.
Seasonal Residential
Customer Charge.
306.
Minnesota Power proposed that the Seasonal Residential Service
Charge be set 10 percent higher than Residential Service Charge. It therefore proposed a Seasonal Residential
Service Charge of $11 per month, or 10 percent higher than its proposal for the
Residential Service Charge. The Company
also proposes to begin billing Seasonal customers on a monthly rather than an
annual basis.[432]
Following the Company’s suggestion that the Seasonal Residential
customer charge should be 10 percent higher than the Residential Service Charge
and with the OES’s proposed Residential customer charge of $8.00, the OES
recommends a Seasonal Residential Service Charge of $8.80 per month. It also
recommends approval of MP’s proposal to implement monthly billing for its
Seasonal customers.[433]
Again, none of the other intervenors addressed the Seasonal Residential
Customer Charge.
307.
The ALJ concludes that a Seasonal Residential Service Charge of
$8.80 and billing seasonal customers on a monthly, rather than annual, basis
are both reasonable and recommends that the Commission approve those proposals.
E.
General Service Energy
and Customer Charges.
308.
Minnesota Power
proposes an increase in the energy charge for General Service customers with
demand meters to 6.75˘ per kWh, with a demand charge to $6 per kW per month. The Company proposes an increase of the energy
rate for General Service customers without demand meters to 8.5˘ per kWh. Finally, the Company proposes an increase in
the monthly service charge for General Service customers from $3.81 to $8.50
per month.[434]
309.
The OES
expressed no opposition to the Company’s proposed demand and energy rates for
the General Service class but maintains that the proposed $8.50 per month service
charge is set too low in comparison to that class’ relative cost of
service. The OES argues that a service
charge of $10.50 will recover the same ratio of class-specific costs, as
measured under the CCOSS, as the OES proposed $8.00 service charge will recover
for the Residential class.[435] Minnesota
Power did not oppose the OES proposal and noted that increasing the Service Charge
for the General Service class would result in a corresponding reduction in the
General Service class energy rates, and that the total revenue requirement for
the General Service class would therefore not change.[436]
310.
The OAG/RUD
objects to both the Company and the OES proposals for higher service charges
for the general service class. It points
out that the Company’s current General Service Charge is $3.81, and that an
increase to $8.50 per month is excessive and unreasonable burden on small
businesses in these difficult economic times.[437] None of the
intervenors had specific positions or recommendations on the increases in
energy rates for General Service class customers.
311.
The ALJ
concludes that a service charge of $10.50 for General Service Customers is
commensurate with the recommended increase for Residential customers and is not
unreasonable, particularly since it will be offset by a corresponding reduction
in energy rates for that class. The ALJ
therefore recommends that the Commission approve that service charge. The ALJ also recommends that the Commission
approve the Company’s proposed increase to 6.75˘ per kWh in the energy charge
for General Service customers with demand meters, and that the demand charge for
those customers be increased to $6 per kW per month. Finally, the ALJ recommends that the energy
rate for General Service customers without demand meters be increased to 8.5˘
per kWh.
F.
Residential Rate
Restructuring (Lifeline vs. Low Income Rider).
312. Currently the Company does not have an energy
assistance program targeted to low income customers. Rather, all of its residential customers benefit
from an increasing block rate structure known as the Lifeline Rate. Under the existing Lifeline Rate, residential
customers are not charged for their first 50 kWh of usage; the cost of that
energy usage is included in their monthly customer charge. Residential customers then pay for the next
300 kWh of monthly usage at a discounted rate of 4.773˘ per kWh, plus an
additional 1.058˘ per kWh for the fuel adjustment for the test year.[438] The energy rate for monthly usage over 350 kWh
is 7.218˘ per kWh, plus an additional 1.058˘ per kWh for the fuel adjustment
for the test year.[439]
313.
Minnesota Power
proposes to eliminate the Lifeline Rate block structure and replace it with a
flat Energy Charge of 8.3˘ per kWh for all energy usage, based on MP’s initial
calculation of revenue deficiency.[440] The Company
contends that its proposed residential rate structure “works better to collect
the portion of customer-related costs not collected through the monthly Service
Charge, as compared to a discounted rate for the first energy block.”[441]
314. On the other hand, the Company is proposing a new Residential
Low Income Assistance Program (“Low Income Rider”) to meet the needs of the
Company’s low income residential customers.
One feature of that program is retaining
the existing Residential Service Charge of $5.00 for customers enrolled
in the program, although that reduced customer service charge would no longer
include any energy usage component. Low
Income Assistance Program participants
would then pay a discounted energy charge of 7.25˘ per kWh (based on MP’s
claimed revenue deficiency). MP argues
that this charge “is very close to the rate that currently applies to standard
firm Residential customers for energy usage above 350 kWh per month.” [442]
315.
Customers
eligible for the Residential Low Income Assistance Program and its discounted
service charge and rate will be customers who qualify for the federal Low
Income Heating and Energy Assistance Program (LIHEAP). To identify qualified customers, the Company
will rely on information obtained from those outside agencies that now qualify
low-income residents for heating assistance programs. MP indicates that this approach will minimize
the administrative burden and presumably program implementation costs.[443]
316. The ECC cites statistics that approximately 111,266
households within MP’s service territory live at or below 50% of the State
Median Income and are income-eligible for LIHEAP. The ECC contends that the majority of eligible
low-income households in MP’s service territory do not receive LIHEAP because
they do not apply for the program. Believing
that the majority of MP customers who are income-eligible do not apply for
LIHEAP, the ECC argues that LIHEAP qualification is “not a good proxy for
identifying low-income customers or for ascribing attributes to them.”[444] In support of its position the ECC relies on
the results of MP’s last rate proceeding and a more recent gas rate matter for
the proposition that LIHEAP does not constitute an acceptable substitute for
identifying low-usage customers.[445]
317.
The ECC notes
that of the 111,266 households in MP’s service territory that are
income-eligible for LIHEAP, only between 9,716 and 12,695 received a LIHEAP
grant. Of those who received grants, ECC
argues that only approximately 6,000 households applied any portion of the
grant money to their MP electricity bill.[446]
318.
The ECC also
asserts that one-quarter of MP’s residential customer base (24.8%, or 27,197
customers) use less than 350 kWh (Lifeline level). Therefore, the ECC argues that MP is proposing
the largest percentage rate increases in the lowest usage tiers. It further maintains that the combined
elimination of the Lifeline Rate, the increase in the customer service charge
(to those not in the Low Income Rider), and the volumetric charge increase in
the lowest usage tier, will adversely affect that group of customers the most. The ECC asserts that while low usage
customers are more likely to be low and fixed income customers, only 2,100 of the
Lifeline level customers receive LIHEAP.[447] ECC maintains that the combination of these
factors results in rates that are not just and reasonable. Further, ECC argues that the proposed rate
design regarding low income customers violates Minn. Stat. § 216B.16, subd. 15,
which requires the Commission to consider a customers’ ability to pay in
setting rates.[448]
319.
The OES argues
for limited retention of an increasing block rate structure, to the extent
incorporated in the Low Income Rider.
The OES agrees with limiting participants to those Residential customers
who qualify for heating assistance under LIHEAP. The OES notes that, as proposed, the Low
Income Rider program will result in an intra-class subsidy, since
non-qualifying Residential customers will pay more for electric service to make
up the difference in class cost responsibility.[449] The OES maintains that this subsidy would be
most pronounced with high usage Residential customers subsidizing low usage
Residential customers. However, the OES
observes that the amount of intra-class subsidy will be limited through
constraints on the number of LIHEAP income-eligible customers. The OES concludes that MP’s proposal, “by
lowering both the customer charge and the energy charge for qualifying low
income customers, is intended to provide some assistance as to lower bills for
low usage, low income customers (to the extent the customer charge is lowered),
and will give some help for high usage, low income customers (to the extent the
energy charge is discounted).”[450] The MCC also supports the Company’s proposed
Low Income Rider.[451]
320.
In the ALJ’s
view, Minnesota Power’s current Lifeline Rate is significantly over-inclusive
in that it provides an intra-class subsidy based on usage that benefits
residential customers of all income levels rather than being targeted to low
income customers. While the
Company’s proposed Low Income Rider still involves an intra-class subsidy by
discounting the customer and energy charges paid by low income customers, the
benefits are more narrowly targeted to low-income customers. The ALJ concludes that some degree of
intra-class subsidy targeting low income customers is warranted, given
indicators that a greater number of residents in the Company’s service area
have low incomes than those in other parts of the state.[452] MP’s proposed Low Income Rider addresses both
low- and high-usage low income customers, by providing some assistance with
both the customer and energy charges.
The ALJ therefore recommends that the Commission approve the Company’s
discontinuance of its Lifeline Rate and replacement of that rate with the
Company’s Low Income Rider.
G.
Dual Fuel
Interruptible Residential Service Tariff.
321.
The Dual Fuel
Interruptible Residential Service rate is available to residential customers
with electric heating. To qualify for
the rate, a customer must have a non-electric backup heating system, which must
provide up to 30 percent of the customer’s heating during the year. When demand for power is high, Minnesota
Power can interrupt electric heating, whereupon the backup heating system takes
its place. Converting to dual fuel
interruptible service involves significant investment by the homeowner.[453] During the
public hearings, dual fuel customers emphasized the need for them to have an
energy rate lower than the rate for other residential customers to enable dual
fuel customers to recover their investment within a reasonable time.[454]
322.
The Company is
proposing an increase in the energy rate from 3.7˘ to 6˘ per kWh, an increase
of 2.3˘ per kWh.[455] Minnesota Power based its proposed increase
in the Residential Dual Fuel rate on an analysis of the Company’s incremental
cost of providing Dual Fuel service, as well as comparisons with current prices
of other home heating fuels such as propane, fuel oil, and natural gas. Nevertheless, the OAG/RUD objects to the
increase as being too high, noting that the difference between the Residential
and Dual Fuel service rates is currently 3.5˘, while under the Company’s
proposal the difference will be 2˘.[456] The OAG/RUD
believes this reduction in the difference of the Residential and Dual Fuel
service rates may actually discourage customers from using dual fuel service.[457] No other
intervenor raised objections to the proposed duel fuel energy rate.
323.
Since the rate that Minnesota Power is proposing for dual fuel residential
customers has an empirical basis and since the OAG/RUD has not presented evidence to support its concerns, the ALJ
recommends that the Commission approve the Company’s Dual Fuel Interruptible
Residential Service tariff.
H.
Triple E and
Residential Heat Pump Service Tariffs.
324.
Minnesota Power
is proposing a new Triple E and Residential Heat Pump Service with discounted tariffs
for Residential class customers having either a Triple E certified home or a
Ground Source or Air Source Heat Pump for heating or cooling. That new tariff would involve a discounted
energy charge for Triple E customers of $0.073 per kWh, and for heat pump
customers of $0.063 per kWh. The Company
argues that these discounts will provide incentives for additional Residential
customers to invest in technology that results in lower energy use.[458]
325. The OES urged denial of the Company’s proposed Triple
E and Residential Heat Pump Service tariff changes for several reasons. First,
although the new tariffs are designed to promote energy efficiency and
conservation technologies, the OES believes that the discounted energy charge
sends the wrong price signal for conservation by telling customers to use more
electricity. Second, since heating and
cooling usage increase during the peak times of the day, discounting the energy
charge under both tariffs provides no incentive to reduce on-peak usage. Third, the Company itself indicates that it
expects a number of its customers with either Triple E certified homes or heat
pumps to move from Residential Dual Fuel Service, an interruptible service, to
the new tariffs, which provide firm service.
While Dual Fuel Service customers receive a discounted energy charge,
they must have an alternative fuel source available and allow the Company to
interrupt their energy when needed. In
contrast, customers on either the Triple E or Residential Heat Pump service
tariffs will receive a discounted energy charge without any interruption to
their service or reduction in the underlying cost to serve them. Finally, the OES contends that the proposed energy discounts in the Triple E and
Residential Heat Pump Service rates would result in an unwarranted intra-class
subsidy for those customers, a subsidy that would be available to existing
customers, as well as new customers.[459]
326.
In response,
Minnesota Power argues that while the dual fuel interruptible service option
encourages customers to reduce on-peak consumption by offering a discount rate
in exchange for taking interruptible service, the Triple E and Heat Pump service
options encourage customers to reduce overall energy consumption by offering a
discount rate for employing these energy efficient technologies. MP asserts that the efficiency of these
technologies “can dramatically reduce the energy demand placed on Minnesota
Power’s system, and the discount rate is a critical element in reducing the
payback period for the customer’s investment in the new technology.”[460]
327.
Although the
Company’s proposed new Triple E and Residential Heat Pump Service tariffs would
have the salutary effect of encouraging residential customers to invest in new,
energy efficient technologies, the ALJ concludes that the potential adverse
consequences of establishing those new tariffs outweigh that potential
benefit. The ALJ therefore recommends
that the Commission not approve those new tariffs.
328.
Minnesota
Power’s existing Large Light and Power tariff (LLP) is available to customers
with total power requirements of less than 10,000 kW. Minnesota Power proposed to
increase the limit for the tariff to all customers with total power
requirements of less than 50,000 kW. MP
explained that some large customers have load profiles that do not fit well
with the Company’s Large Power class requirements. This change is proposed to give those
customers another service option. The
LLP demand charge would be $8.00 per kWmonth, with an energy rate of 4.5˘ per
kWh.[461] The OES supports this proposal, and none of
the other intervenors opposed it.[462]
The ALJ therefore recommends that the
Commission approve the Company’s proposed demand charge of $8.00 per kWmonth and
energy rate of 4.5˘ per kWh for LLP customers.
a) Increase
the Demand Charge for the first 10 kW or less of Billing Demand to $161,385.
b) Combine
the Demand Charges for Firm Power and Excess Power into one Firm Power Demand
Charge of $15.10/kw-month.
c) Add
time-of-use energy rates, proposed to be 2.8˘/kWh on–peak and 1.2˘/kWh off-peak
for Firm Energy.
d) Clarify
the applicability of the Large Power Surcharge, including increasing the
threshold for its application to 50,000 kW.
e) Add a
non-curtailable option to the Rider for Large Power Incremental Production
Service.
f) Expand
the application of the Rider for Expedited Billing beyond the taconite
customers to all Large Power customers.
g) Eliminate
the Rider for Implementing “Best Efforts” Marketing Policy- Large Power Class.
h) Increase
the Service Voltage Adjustment to $1.50/kWh-month.[463]
330.
LPI and
A.
Results Sharing
Compensation.
331. MP indicated that an issue regarding Results Sharing
Compensation was agreed to with no adjustment needed.[469]
B.
332. With regard to the Hibbard Energy Center there was no
dispute over an adjustment of $(27,183) to include depreciation expense and
associated taxes.[470]
C.
Brainerd Public
Utilities Commission Asset Sale.
333. With regard to the Brainerd Public Utilities
Commission Asset Sale there was no dispute over an adjustment of $(4,289) to
include depreciation expense and associated taxes.[471]
D.
Conservation
Improvement Plan Expenses.
334. There was no dispute over an adjustment of $219,810
to reflect the average of 2008 and 2009 Conservation Improvement Plan expenses
for the test year, and associated taxes.[472]
335. With regard to Property Taxes there was no dispute over an adjustment of
$(384,338) to reflect actual test year property tax expenses.[473]
F.
Service Life
Petition for Transmission and Distribution.
336. There was no dispute over an adjustment of $(306,653)
to include test year plant-in-service balance and related depreciation.[474]
G.
Interest on LP
Expedited Billing.
337. There was no dispute over an adjustment of $107,077
to include all costs in retail rate case (as opposed to allocating a portion to
FERC jurisdiction) because interest on LP expedited billing relates only to
retail rate class.[475]
H.
Fuel and
Purchased Power Deferral (Miscellaneous and General Expenses).
338. There was no dispute over an adjustment of
$(3,017,465) to fuel clause lag costs, to reflect Minnesota Power's request for
consideration in fuel clause docket.[476]
I.
Badoura-Pine River
Project.
339. There was no dispute over an adjustment of $ (19,204)
to include depreciation expense and associated taxes relating to the
Badoura-Pine River Project.[477]
J.
BEC4 – Boiler
Surface Project.
340. There was no dispute over an adjustment of $(4,870)
to include depreciation expense and associated taxes relating to the BEC4 –
Boiler Surface Project.[478]
341. There was no dispute over an adjustment of $(324,532)
to account for depreciation expenses under a recent Commission decision.[479]
7. The proposed
changes in tariff provisions, with two exceptions, are reasonable and should be
approved. The proposed Triple E and Residential Heat
Pump tariff changes have not been shown to be reasonable and should not be
approved.
9. For
non-asset-based margins, the OES proposal of a $300,000 cap on ratepayer
responsibllity for losses arising from virtual transactions to hedge against
price shifts in the Day-Ahead and Real Time markets is reasonable and should be
adopted.
10. For ancillary service
market margins, the OES proposal that these issues be addressed in the ASM Docket is reasonable and should be
adopted.
11. Regarding SO2
and NOx allowances, a credit of $195,000 should be made to base
rates for the EPA sales expected during the test year. For the remaining allowance sales, those
amounts should be returned to ratepayers through a cost recovery rider.
12. MP has demonstrated that
the test year expenses for MISO Schedule 16 and 17 costs of $1,326,277 are
appropriate for recovery through base rates.
MP has demonstrated that deferred Schedule 16 and 17
costs of $4,423,480 should be recovered on an amortized expense basis. The appropriate amortization period is five
years, unless the Commission orders MP to file another rate case within three
years, in which case the amortization should occur over three years. MP has not shown that any of the amortized
expense amounts are approroprite for inclusion in its rate base.
14. MP has demonstrated that AREA Plan expenses up to the $4.07
million cap imposed by the Commission are reasonable and should be included in
MP’s test year O&M expenses. MP has
not demonstrated that the Minnesota jurisdictional amount of $568,533 is
sufficiently certain, reasonable, or reliable for inclusion in base rates.
15. MP has not demonstrated that its incentive compensation methodology
and amounts are reasonable. The
modifications proposed by the OES are consistent with prior Commission
treatment of incentive compensation and result in rates that are just and
reasonable. Imposing a tracking
mechanism for actual amounts paid and a refund of unpaid incentive compensation
already included in rates is reasonable and should be adopted.
17. MP has
not affirmatively shown that its corporate cost allocation process conforms to
the Commission requirements in the Docket
1008 Order. MP has not affirmatively
shown that its test year corporate costs are appropriate for calculating base
rates. The OES has demonstrated that its
test year corporate cost calculation for MP of $73,678,620 is reasonable and
should be adopted. The OES has not shown that requiring legal
separation of Minnesota Power and ALLETE is needed or reasonable to address
ongoing issues of corporate cost allocation.
19. The stipulation between MP, OES, Boise, MCC, and LPI
regarding the proposed FCA adjustment and other related billing issues is
reasonable and should be approved by the Commission.
21. MP has demonstrated that it incurred reasonable rate case
expenses in this matter of $1,191,789.
The OES has shown that the overall expense should be reduced by 5.76
percent to reflect the portion of the expense that is allocable to nonregulated
activities. The resulting amount should be recovered on an amortized expense
basis. The appropriate amortization
period is five years, unless the Commission orders MP to file another rate case
within three years, in which case the amortization should occur over three
years. MP has not shown that any of the
amortized expense amounts are approroprite for inclusion in its rate base.
22. The agreed-upon adjustments to MP’s rate base are
reasonable and should be adopted. MP has
not shown that the Asset Retirement Obligation methodology (ARO Method) is superior to the
Decommissioning method for calculating depreciation. MP’s proposal to use the ARO Method in
determing test year expenses should not be adopted.
23. The OES
has shown that the Sappi 5 generation facility should be included in MP’s rate
base. MP has shown that the Rapids Energy
Center should not be included in the Company’s rate base. Should
the Commission choose to put either or both facilities in the rate base,
revenue and O&M expenses will need to be adjusted to reflect the facility
put into the rate base. MP has shown
that its revenue estimates are reasonable and should be adopted. The OES has shown that its O&M cost
calcuation is reasonable. The Commission
may also deny test year O&M expenses as not demonstrated or afford MP the
opportunity to supplement the hearing record on this issue.
25. MP has demonstrated that it will experience a revenue
shortfall. MP is entitled to recover
this revenue shortfall through an adjustment of its electric rates to increase
its revenues.
27. MP has not
demonstrated that an increase in the Residential Basic Charge from $5.00 per
month to $10.00 per month is an appropriate adjustment to balance the need to
recoup the costs of serving the residential class of customers without
interclass subsidies, with the need to encourage conservation, avoid rate
shock, and account for other factors between rate classes. Based on the
record in this proceeding, OES has demonstrated that an increase in the
residential basic charge to $8.00 per month is an appropriate adjustment that meets
the Commission’s standards for changes in rates.
29. MP has demonstrated that an increase in the General Service
Customer Charge from the existing $3.81 per month is appropriate. The proposal from MP to increase the charge
to $8.50 per month does not sufficently maintain the relationship between the
charges on the Residential and General Service classes of customer. The ALJ concludes that the OES recommendation
to increase the monthly charge to $10.50 is an appropriate adjustment to
balance the need to recoup the costs of serving the General Service class of
customers without interclass subsidies and with the need to encourage
conservation, avoid rate shock, and account for other factors between rate
classes. Based on the record in this
proceeding, an increase in the General Service basic charge to $10.50 per month
is an appropriate adjustment that meets the Commission’s standards for changes
in rates.
30. MP has demonstrated that its proposed replacement of the
Lifeline Rate with a Low Income Rider meets the needs of low-income residential
customers, while striking the best balance between the various rate design
principles of the Commission.
31. Modifying MP’s electric rates in the manner described in the
Findings and Conclusions above results in just and reasonable rates that are in
the public interest within the meaning of Minn. Stat. § 216B.11.
Based on the
foregoing Findings and Conclusions above, the Administrative Law Judge makes the
following:
IT IS
RECOMMENDED that the Public Utilities Commission order that:
1. Minnesota
Power is entitled to increase gross annual revenues in the manner and in an
amount consistent with the terms of this Order.
2. Within
30 days of the service date of this Order, Minnesota Power shall file with the
Commission for its review and approval, and serve on all parties in this
proceeding, revised schedules of rates and charges reflecting the revenue
requirement for annual periods beginning with the effective date of the new
rates, and the rate design decisions contained herein. Minnesota Power shall include proposed
customer notices explaining the final rates.
Parties shall have 14 days to comment.
3. (If
the Commission orders an Interim Rate Refund) within 30 days of the service
date of this Order, Minnesota Power shall file with the Commission for its
review and approval, and serve upon all parties in this proceeding, a proposed
plan for refunding to all customers, with interest, the revenue collected
during the Interim Rate period in excess of the amount authorized herein. Parties shall have 14 days to comment.
|
Dated: February 19, 2009 |
/s/ Bruce H. Johnson ________________________________ |
|
|
BRUCE H. JOHNSON Administrative Law Judge |
Reported: Shaddix and Associates
Transcripts
Prepared (Six Volumes)