Kansas City Power & Light Co. v. State Corp. Commission

No. 114,781
IN THE COURT OF APPEALS OF THE STATE OF KANSAS
KANSAS CITY POWER & LIGHT COMPANY,
Petitioner/Appellant,
v.
THE STATE CORPORATION COMMISSION OF THE STATE OF KANSAS,
Respondent/Appellee.
SYLLABUS BY THE COURT
1.
Judicial review of a utility rate appeal must be filed directly in the Court of
Appeals. K.S.A. 66-118a(b).
2.
The Kansas Judicial Review Act, K.S.A. 77-601 et seq., controls how a court
reviews utility rate cases.
3.
A court reviews such cases to see if the agency action is based upon a
determination of fact, made or implied by the agency, that is not supported to the
appropriate standard of proof by evidence that is substantial when viewed in light of the
record as a whole. K.S.A. 2015 Supp. 77-621(c)(7).
4.
The phrase, "in light of the record as a whole," means the adequacy of the
evidence in the record before the court to support a particular finding of fact which shall
be judged in light of all the relevant evidence in the record cited by any party that detracts
1
from such finding as well as all of the relevant evidence in the record cited by any party
that supports such finding. This includes any determinations of veracity by the presiding
officer who personally observed the demeanor of the witness and the agency's
explanation of why the relevant evidence in the record supports its material findings of
fact. In reviewing the evidence in light of the record as a whole, the court shall not
reweigh the evidence or engage in de novo review.
Appeal from Kansas Corporation Commission. Opinion filed March 9, 2016. Affirmed.
Robert J. Hack and Roger W. Steiner, of Kansas City Power & Light Company, of Kansas City,
Missouri, and Glenda Cafer and Terri Pemberton, of Cafer Pemberton LLC, of Topeka, for appellant.
Brian G. Fedotin, of Kansas Corporation Commission, of Topeka, for appellee.
Before HILL, P.J., GREEN and LEBEN, JJ.
HILL, J.: The Kansas Corporation Commission granted Kansas City Power &
Light Company a rate increase of $48.67 million. KCP&L wants more.
The company appeals, claiming the Commission's findings of fact are not
supported by substantial competent evidence. Additionally, the company contends that
the Commission's decision to exclude flotation costs when calculating its return on equity
was unreasonable, arbitrary, or capricious. After reviewing the entire record and not
reweighing the evidence, we hold the Commission's rate increase indeed is supported by
substantial competent evidence. The Commission weighed conflicting expert opinions,
accepted some, rejected others, and adopted a middle ground approach in computing this
new rate. Additionally, the Commission has met its goal of balancing the interests of all
parties concerned when it formulated this rate. We affirm.
2
The case history reveals the administrative process.
KCP&L is a vertically integrated electric public utility that produces, transmits,
and delivers electrical power to customers in designated areas of northeast Kansas and
parts of Missouri. KCP&L is a wholly-owned subsidiary of Great Plains Energy, Inc.,
(GPE) based in Kansas City, Missouri. GPE is a public utility holding company which
also owns another utility, GMO, formerly Aquila, Inc. KCP&L's Kansas customers
include more than 200,000 residences, 28,000 commercial firms, and 1,000 industrial and
municipal facilities. KCP&L provides retail electrical services and also sells electrical
power wholesale and in bulk power transactions.
In January 2015, KCP&L filed its application seeking to increase its Kansas rates
by $67.3 million; this amounted to a 12.5 percent increase for KCP&L customers.
KCP&L's prefiled testimony in support of the rate increase was based on test year data
from July 2013 through June 2014, plus known and measurable changes projected
through March 31, 2015. KCP&L submitted prefiled testimony in support of its
application. A number of parties intervened in the proceeding, including the Citizen's
Utility Ratepayer Board (CURB) and Walmart, Inc. The Commission's Staff and various
intervenors also submitted prefiled testimony challenging different aspects of KCP&L's
rate calculations.
Prior to any evidentiary hearing, KCP&L, the Commission's Staff, and several of
the intervenors entered into settlement agreements on some of the issues raised by the
application. As a result, the Commission's hearing was limited to testimony on four
issues, one of which was the appropriate return on equity (ROE)—the cost the company
needed to pay a reasonable return to holders of common stock. The ROE is used as a
multiplier when calculating KCP&L's overall revenue requirement.
3
This appeal's focus is on one issue.
The issue of this appeal is did the Commission use the appropriate return on equity
in calculating KCP&L's cost of capital? Obviously, the cost of capital, in turn, affects the
company's revenue requirements—the fundamental base of a utility rate.
Appropriate returns on equity are complex calculations that involve the weighing
of many arcane economic factors and are the subjects of expert analysis. In this case there
were four such experts: one for KCP&L, one for the Staff, one for CURB, and one for
Walmart.
KCP&L presented the testimony of Robert Hevert, an expert who has provided
consulting services to many energy and utility clients. Using various economic models
and forecasts, Hevert opined that 10.3 percent was a reasonable estimate of investors'
expectations of returns from KCP&L's stock. Similarly, a KCP&L executive testified that
the ROE set in the Commission's approved rates in January 2013—9.5 percent—had not
been achieved, nor had a higher ROE from earlier cases been achieved as a result of the
company's operations. In addition, the executive testified that KCP&L had been reducing
dividends over the last few years because of decreased earnings, required capital
expenditures, and the lag in recovery of those expenses.
In response, Staff presented the testimony of Adam Gatewood, the Managing
Financial Analyst of the Commission's Utilities Division. Gatewood did not dispute
KCP&L's calculations of its cost of debt and preferred stock, but he disagreed with the
utility's requested 10.3 percent ROE for common equity. Gatewood asserted that
KCP&L's calculations were based upon overly optimistic growth estimates of corporate
earnings. Gatewood testified that based on his use of the same analytical and forecasting
models, he believed a range of 9.0 and 9.50 percent ROE was reasonable. Gatewood's
4
calculations were based, in part, on evidence that most measures of capital costs had
declined since the prior rate case in 2013.
Two intervenors also presented testimony regarding the appropriate percentages.
CURB presented the testimony of J. Randall Woolridge, Ph.D. Woolridge's analysis was
based upon various estimating models using his own selected proxy group as well as the
proxy group of KCP&L's expert. Woolridge opined that the appropriate cost of equity for
KCP&L was 8.55 percent.
Walmart's expert, Steve Chriss, testified on various subjects including KCP&L's
requested ROE. Chriss testified that based upon KCP&L's proposed ROE, its revenue
requirement would increase approximately $14 million—about 25 percent of the
company's requested rate increase. Chriss opined that the utility's ROE request was
excessive because of the large impact on its revenue requirement and because it was
higher than ROEs approved by state regulators nationwide. Chriss did not conduct an
independent calculation beyond testifying about nationwide trends.
After taking evidence on the matter, the Commission issued its initial order on the
application on September 10, 2015, that summarized the procedural aspects of the case
and approved the two settlement agreements. With respect to the ROE issue, the
Commission summarized the testimony of the various experts and discussed the legal
authorities establishing standards applicable to returns authorized for regulated utilities.
In analyzing the testimony, the Commission found that KCP&L's request for an ROE
over 10 percent was disputed by the other experts, none of whom recommended an
increase over the 9.5 percent previously allowed in the company's last prior rate case. The
Commission found that KCP&L's proposed ROE ran counter to trends in Kansas and
nationwide of historically low costs of capital. The Commission also took issue with
Hevert's testimony because he failed to adjust his ROE calculation after he reduced his
projected growth rate by nearly 25 points (0.25 percent). The Commission found that
5
Hevert's growth rate was much higher than estimates of growth made by the Federal
Reserve Board, the Social Security Administration, and other established predictors of
the economy. The Commission also rejected the testimony of CURB's expert, Woolridge,
as too low.
The Commission also denied KCP&L's request that the ROE include an
adjustment for flotation costs associated with the issuance of common stock, based on the
company's failure to provide evidence of the amount, if any, of unrecovered costs
associated with the issuance of its common stock during the test year. Changing its prior
practice, the Commission concluded that the better practice in treating flotation costs was
to make pro forma adjustments to the test year operational costs to include any actual
costs associated with the issuance of common equity. (The other issues in the
Commission's order need not be addressed, as KCP&L does not challenge those
determinations either in its petition for review or its appellate brief.)
KCP&L argued that the Commission's decision relating to the treatment of
flotation costs violated the principle set forth in Home Telephone Co. v. Kansas
Corporation Comm'n, 31 Kan. App. 2d 1002, 1011, 76 P.3d 1071 (2003), rev. denied 277
Kan. 923 (2004). KCP&L asserted the Commission had changed its procedure for
treating flotation costs without adequately explaining the basis for that change. KCP&L
further argued that the Commission's altered treatment of flotation costs was arbitrary and
capricious because KCP&L had no prior notice of this change before it filed its
application. Finally, KCP&L argued that the overall ROE decision rendered the final
order outside the "zone of reasonableness" and, therefore, was unlawful.
KCP&L seeks judicial review with this court according to K.S.A. 66-118c. The
utility contends that based upon the record as a whole, the Commission's ROE
determination is not supported by substantial competent evidence. The company argues
that the Commission failed to properly weigh the evidence supporting and detracting
6
from its decision and based its ruling on weak, assailable testimony. KCP&L asks us to
reverse the Commission's decision.
We repeat our standard of review.
Judicial review of a utility rate appeal must be filed directly in the Court of
Appeals. K.S.A. 66-118a(b). The Kansas Judicial Review Act, K.S.A. 77-601 et seq.,
controls how we review such cases. In the context of this case, our scope of review
focuses on the facts as found by the Commission. It is the familiar substantial competent
evidence standard but with an expanded field of inquiry—the entire record:
"the agency action is based on a determination of fact, made or implied by the agency,
that is not supported to the appropriate standard of proof by evidence that is substantial
when viewed in light of the record as a whole . . . ." (Emphasis added.) K.S.A. 2015
Supp. 77-621(c)(7).
The legislature modified the statute in 2009 to clarify that the phrase "in light of the
record as a whole" means:
"the adequacy of the evidence in the record before the court to support a particular
finding of fact shall be judged in light of all the relevant evidence in the record cited by
any party that detracts from such finding as well as all of the relevant evidence in the
record . . . cited by any party that supports such finding, including any determinations of
veracity by the presiding officer who personally observed the demeanor of the witness
and the agency's explanation of why the relevant evidence in the record supports its
material findings of fact. In reviewing the evidence in light of the record as a whole, the
court shall not reweigh the evidence or engage in de novo review." (Emphasis added.)
K.S.A. 2015 Supp. 77-621(d).
In addition, the court must, when exercising judicial review, take into account the
harmless error rule. See K.S.A. 2015 Supp. 77-621(e).
7
Although the amended Act requires judicial review of all the evidence supporting
and detracting from the agency's decision, nothing in the statute indicates the legislature
intended to alter the traditional standards applicable in Commission decisions regarding
public utility rates.
Simply put, under the Act, the Commission's order may only be set aside by the
court if it is not supported by substantial competent evidence based upon the record as a
whole; is without foundation in fact; or is otherwise unreasonable, arbitrary, or
capricious. See Kansas Industrial Consumers v. Kansas Corporation Comm'n, 30 Kan.
App. 2d 332, 335, 42 P.3d 110 (2002). In this case, KCP&L—the party claiming the
Commission's actions are invalid—bears the burden of proving that invalidity. K.S.A.
2015 Supp. 77-621(a)(1); see Clawson v. Kansas Dept. of Agriculture, 49 Kan. App. 2d
789, 795, 315 P.3d 896 (2013).
We make some observations on the unique role of the KCC.
The Kansas Supreme Court has pointed out that the Commission's work is
legislative in nature.
"Under the constitutional separation of powers doctrine, the regulation of public
utilities is legislative in nature. The legislators created the Kansas Corporation
Commission and granted it full and exclusive authority and jurisdiction to supervise,
control, and regulate the public utilities of this state and, when acting in the exercise of its
delegated powers, the Commission is not a quasi-judicial body. [Citations omitted.]
"Thus, public utility rate making is a legislative function, whether it is regulated
by an administrative body or by the legislature itself." Kansas Gas & Electric Co. v.
Kansas Corporation Comm'n, 239 Kan. 483, 491, 720 P.2d 1063 (1986).
The legislature has vested the Commission with broad discretion in executing its
functions. Because of this extensive discretion, we have no authority to substitute our
8
judgment for that of the Commission. More importantly, the Commission's decisions
involve complex problems of policy, accounting, economics, and other special
knowledge that go into fixing utility rates. As a result, the court may not set aside a
Commission order merely because the court would have arrived at a different conclusion
had it been the trier of fact. The court may reverse or nullify a Commission order only
when the decision is so wide of the mark as to be outside the realm of fair debate. Kansas
Industrial Consumers, 30 Kan. App. 2d at 336.
The Commission is also unique because it is the only factfinder. It does not review
evaluations of credibility made by subordinate hearing officers or panels. It makes those
credibility decisions itself. Even when this court reviews the agency's decision in light of
the record as a whole, it is not our responsibility to reweigh the credibility of the
witnesses or reweigh the evidence relied upon by the Commission. K.S.A. 2015 Supp.
77-621(d); see Mobil Exploration & Producing U.S. Inc. v. Kansas Corporation Comm'n,
258 Kan. 796, 815, 908 P.2d 1276 (1995). Even though the legislature altered the court's
scope of review under the Act, it has not altered the broad discretion it has previously
delegated to the Commission.
We examine the battle of the experts.
Three experts presented detailed opinions on the appropriate ROE for KCP&L.
Each expert compared the performance of different proxy companies by using the
Discounted Cash Flow (DCF) model and Capital Asset Pricing Model (CAPM) of
economic analysis. These calculations were performed to reach a reasoned estimation of
the amount of return on common stock required to enable KCP&L to remain financially
sound and attract new investors, while balancing the interests of the public in paying
reasonable rates. KCP&L argues that its expert, Hevert, provided the most reliable and
credible evaluation of the ROE. In other words, "our expert's estimations are better than
your expert's estimations."
9
The starting point for ROE calculations under these models is the selection of
proxy companies to use as comparisons. In using the DCF and CAPM models, each
expert selected proxy companies to compare the current ROE the companies receive.
While there was overlap between the experts' proxies, both Staff and CURB used
significantly more companies in their proxy comparison. The following chart illustrates
the variations:
Proxy Companies Used for Comparison
Hevert/KCP&L Gatewood/Staff Woolridge/CURB
American Elec. Pwr Co. Allete, Inc. Allete, Inc.
Duke Energy Corporation Alliant Energy Corp. Alliant Energy Corp.
Empire Dist. Electric Co. Ameren Corp. Ameren Corp.
Hawaiian Elec. Industries, Inc. American Elec. Pwr Co. American Elec. Pwr Co.
IDACORP, Inc. Avista Corp. Avista Corp.
NextEra Energy, Inc. CMS Energy Corp. Black Hills Corporation
Northeast Utilities Consol. Edison, Inc. CMS Energy Corp.
Otter Tail Corporation Dominion Resources, Inc. Consol. Edison, Inc.
Pinnacle West Capital Corp. Duke Energy, Inc. Dominion Resources, Inc.
PNM Resources, Inc. Edison International Duke Energy Corp.
Portland General Elec. Co. El Paso Electric Co. Edison International
Southern Company Empire Dist. Elec. Co. El Paso Electric Co.
Westar Energy, Inc. Great Plains Energy, Inc. Empire Dist. Elec. Co.
IDACORP, Inc. Entergy Corp.
NorthWestern Corp. Eversource Energy
OGE Energy Corp. FirstEnergy Corp.
Pacific Gas & Elec. Co. Great Plains Energy, Inc.
Pinnacle West Capital Corp. IDACORP, Inc.
Portland General Elec. Co. MGE Energy, Inc.
TECO Energy NorthWestern Corp.
Westar Energy, Inc. OGE Energy Corp.
Xcel Energy Inc. PG&E Corp.
Pinnacle West Capital Corp.
PNM Resources, Inc.
Portland General Elec. Co.
SCANA Corporation
Southern Company
Westar Energy, Inc.
Xcel Energy, Inc.
The use of proxy groups in ROE analysis provides a significant benefit because it
ameliorates short-term events that could bias market values of individual companies. At
10
the time KCP&L's rate case was filed, its parent corporation, GPE, had comparable
ratings from both Standard & Poor's (BBB+) and Moody's (Baa2) investors' services and
was rated as "outlook: Stable." The proxy companies selected by the experts possessed
similar ratings. Hevert used only companies with BBB+ and Baa2 ratings. Gatewood
used electrical utilities with those ratings as well as the ratings immediately above and
below BBB+ and Baa2. Woolridge testified he used proxy companies with similar ratings
as KCP&L.
In order to illustrate the issue confronting the Commission, we summarize the
experts' testimony. This is a distillation of the many boxes containing the record on
appeal. We have taken pains to illustrate many details from the experts' testimony
because KCP&L contends substantial facts do not support the Commission's findings.
The many details that follow show why that is not true.
We look first at the KCP&L expert, Hevert.
In selecting his proxy group, Hevert first chose 13 companies with comparable
ratings and risk. In using the DCF model to estimate an appropriate ROE, Hevert
included information from these proxy companies together with projected long-term
earnings growth estimates from various nongovernmental investment sources. Using the
mean measurements of ROE for the proxy companies along with the average of the
growth rate estimates, Hevert calculated a high mean, a mean, and a low mean estimation
of ROE. Hevert then adjusted these figures to include flotation costs. This methodology
resulted in a mean ROE of 9.36 to 9.62 percent and a high mean of 10.34 to 10.60
percent. With no explanation, Hevert did not include the low mean range—8.32 to 8.57
percent—in any of his detailed analysis.
Next, Hevert used a Multi-Stage DCF model, which defines the cost of equity
using a discounted rate. In this model, current price is equated with the discounted value
11
of future cash flow measured at three different time stages. Purportedly this model is able
to focus on near-, intermediate-, and long-term growth rates, recognizing a company will
not necessarily grow at a constant rate. It also can take into account likely increases or
decreases in expected capital spending and the transition to long-term expected payout
levels. In using this model, Hevert assumed a long-term growth rate of 5.61 percent,
asserting that it was reasonable to assume that over time gross domestic product (GDP)
will revert to its long-term (80 year) mean. This analysis, based on Hevert's inputs,
resulted in a mean range of 9.79 to 10.07 percent ROE and a high mean range of 10.07 to
10.36 percent ROE. Again, Hevert's testimony excluded his low mean range of 9.53 to
9.79 percent.
The third type of analysis Hevert conducted involved the use of the CAPM model.
This model includes a risk premium factor to compensate investors for "the non-
diversifiable or 'systematic' risk of that security." The formula includes an estimate of the
relative volatility of the company's stock as compared to the overall stock market. Using
inputs from his initial DCF model and nongovernment market sources, Hevert calculated
a "market capitalization weighted average" under the DCF model and then subtracted the
current 30-year Treasury bond yield. Hevert then factored in a premium because equity
investors bear greater risk than bondholders and require additional compensation for that
risk. Taking into account all of these factors, Hevert opined that the ROE under this
model ranged from 10.24 to 10.92 percent.
Finally, Hevert testified that in addition to the projections obtained from the
various calculation models he used, he considered individual factors which affected
KCP&L's risk profile. Hevert said that KCP&L's energy generation was highly dependent
on coal-fired and nuclear-fired generation capacity, both of which were capital intensive.
This dependency increased KCP&L's risks because of the cost of potential plant failures,
shut downs, and replacement of facilities. Similarly, increased environmental and safety
regulations required additional capital investment. Hevert also testified about significant
12
plans of GPE—the parent company—for capital expenditures, although his testimony
fails to reflect how much of the planned budget involved KCP&L versus GPE's other
subsidiary. These factors, according to Hevert, established a higher risk factor that had to
be considered in calculating the ROE. In light of all the models used and factors
considered, Hevert testified that based upon all the data, an ROE in the range of 10.00 to
10.60 percent was a reasonable estimate of KCP&L's cost of equity. He recommended a
midpoint ROE of 10.30 percent be used. Based upon KCP&L's undisputed cost of debt
and capital structure, a 10.30 percent ROE resulted in an overall weighted rate of return
of 7.94 percent.
We review the work of Staff's expert, Gatewood.
In response to the ROE issue, Staff presented the testimony of Adam H.
Gatewood, a long-time Commission employee. In discussing KCP&L's rate of return,
Gatewood agreed with the company's proposed cost of debt and cost of preferred stock,
but he challenged the company's request for a 10.30 percent return on common stock.
Gatewood acknowledged Hevert's discussion of recent ROE decisions from other
regulatory commissions for similar electrical companies. Gatewood stated that
recognizing the ROEs granted to other utilities in other jurisdictions was relevant but not
controlling in determining an ROE for an individual company. Gatewood testified that
limiting the ROE analysis to other jurisdictions' decisions risked overlooking (1) changes
in capital markets; (2) differences in the ratemaking policies of other states' commissions;
and (3) political pressures or other state-specific factors.
Similarly, focusing on the end-results of other regulators' determinations ignored
the fact-specific information in those other rate cases. According to Gatewood, ROEs
authorized in other states could only serve as a "rough benchmark" of an average ROE.
Gatewood also opined that a utility's ROE should not provide a "cushion" for potential
13
future market changes because rates and factors impacting on rates are periodically
reviewed in order to protect both consumers and investors. Gatewood testified that the
Commission should not speculate—and courts should not require the Commission to
speculate—about the future ups and downs of the economy and capital markets as
opposed to making decisions on current information.
Gatewood used 22 proxy companies as compared to Hevert's 13 companies when
making his computations. Gatewood's proxies were all publicly traded electric utilities
with similar credit ratings to KCP&L's ratings with Moody's and Standard and Poor's or
one step above or below that rating. Gatewood then eliminated companies with pending
mergers or acquisitions; companies that did not have a stable dividend; and companies
that derived less than 70 percent of their revenues from electric utility services. These
limits had been adopted in recent filings with the Federal Energy Regulatory Commission
(FERC). Gatewood opined the parameters provided a group of utilities with
commensurate investment risk compared with KCP&L.
Like Hevert, Gatewood used the DCF model with information from his 22 proxy
companies. Gatewood testified that regulatory commissions consistently used the DCF
model "because, with reasonable inputs, it is a tool that meets the legal standards and it is
a tool that investors have used" to value many types of investments. Gatewood obtained
public information on the current stock prices and dividends paid for each of the proxy
companies as well as their projected 2016 annual dividend rates to input in his models.
Gatewood differed, however, in his view of anticipated growth rates to use in the
DCF model. Gatewood testified that the growth rate has a significant impact on the ROE.
Determining the estimated growth rate is subjective as it is an estimate investors use to
determine stock price. Most investment firms forecast estimated growth for a 3- to 5-year
time horizon. However, the DCF model deals with a forecast for growth beyond that
horizon. Those looking for a long-term perspective of potential growth, according to
14
Gatewood, rely on estimates of the country's GDP that extends out more than 30 years,
normalized due to inflation (nGDP). FERC uses the nGDP to estimate the cost of equity
and returns for interstate pipeline companies as well as electric transmission companies.
Including these factors, Gatewood presented his DCF calculations using dividend
and growth data from his 22 proxy companies. Gatewood determined the average of the
short-term growth rates from the proxy companies and the nGDP. The latter was based
on forecasts through 2090 from the Energy Information Agency and the Social Security
Administration. Looking at this and other sources, Gatewood testified there was a
consensus that long-run economic growth in real terms would be a nominal growth rate
of 4.3 to 4.5 percent. Gatewood testified that Hevert's growth forecast of 5.61 percent
exceeded the consensus from a number of seasoned, professional services and was
unsupportable. Gatewood also performed an Internal Rate of Return (IRR) analysis,
which is a form of a discounted cash flow analysis which allows one to apply growth
forecasts for intended 5-year dividends with the remaining years growing at the long-term
nGDP forecasted growth rate. Using this method with his proxy group, Gatewood found
a mean IRR of 8.25 percent and a maximum IRR of 9.52 percent for KCP&L.
Gatewood likewise used the CAPM model in developing his opinion on an
appropriate ROE. After detailing the elements and calculations using the CAPM model,
Gatewood opined that the information showed that institutional investors and
professional forecasters did not expect future returns to be as great as returns experienced
in the past. Gatewood recommended basing the growth rate on a lower long-term
economic growth estimate. His estimate was based upon widely accepted projections and
expectations of investment advisors, investors, and public policy makers such as the
Federal Reserve Board, the Social Security Administration, the Energy Information
Administration, and corporations such as ExxonMobil. Gatewood testified that lower
expectations of growth were consistent with the Commission's views expressed in recent
rate cases. This included KCP&L's position in a prior docket where the Commission
15
found the company's long-term growth rate was "'unreasonably optimistic.'" Gatewood
criticized Hevert's long-run GDP forecast as unsupported by evidence in light of these
other publicly available forecasts.
Gatewood also testified about the Commission's decision in a KCP&L rate case
filed 3 years earlier, referred to as the 764 Docket. Gatewood testified that during the 764
Docket, KCP&L had requested a ROE of 10.40 percent. However, since the 764 Docket,
Gatewood testified that capital costs are measurably lower and stock market indices and
the yield on long-term debt were at lows not seen in 50 years. Specifically, KCP&L's
interest rates on corporate debt dropped from 5.20 percent to approximately 4.80 percent.
During the same period, the dividend yield from Hevert's proxy group dropped from 4.45
percent to 3.80 percent. Using this DCF model, lower dividends equate to lower cost of
equity.
Gatewood also criticized Hevert's calculations in several other respects. First,
Hevert's measurement of credit ratings was inappropriate, resulting in the narrowing of
his proxy group of comparable companies. Likewise, Hevert's growth rate was based on
various sources that limit earning growth on a 3- to 5-year horizon and assumed the
utilities will grow at that rate in perpetuity. This assumption is well beyond what is
expected of the overall economy. Even FERC staff, in recent federal proceedings, used a
lower long-term growth estimate of 4.39 percent for nGDP. Hevert's growth estimate is
well above that of any credible published estimate and well above measurements adopted
by the Commission in the recent past. Because Hevert's growth estimates were
unreasonable, his analysis under the multi-stage DCF calculation incorrectly concluded
there was little difference between his anticipated growth rates for the short-term as well
as the long-term.
The same issues of over-optimistic, double-digit growth projections also tainted
Hevert's CAPM analysis, according to Gatewood. Hevert should have recognized that his
16
growth estimates were unreasonable as the numbers he relied upon were several times
greater than the growth of the entire U.S. economy; "an earnings growth rate forecast that
is so far above that of the economy should alert an analyst that it is not a realistic growth
estimate to use in perpetuity."
Finally, Gatewood testified that Hevert's risk premium model—comparing the
ROE granted to other utilities in other jurisdictions—is not a reasonable way to control an
ROE decision. Gatewood had noted that this model could be based on outdated
information. In his view, the risk of the electric utilities found in the historic data did not
necessarily compare to the risk faced by KCP&L. According to Gatewood, just as one
must select a comparable proxy group to perform a DCF analysis, comparing KCP&L to
all electric utilities regardless of comparable risk is circular in reasoning. To him, the risk
premium model also ignored changes in the industry over time. Currently, rate design,
pass-through charges, and other rate mechanisms have evolved, lowering the risk of
utilities. Again, Hevert's analysis failed to make any adjustment for changes in rate
design and cost-recovery policies that shifted risk from shareholders to rate payers.
Gatewood also rejected Hevert's theory that KCP&L has unique risks as a result of its
generation portfolio and capital expenditure plan. Although Gatewood acknowledged
those risks existed, modern financial theory did not support the view that shareholders
should be compensated for that risk; shareholders are not entitled to be compensated for
every risk faced by a utility when investors can reduce their risk by diversifying their
portfolios.
According to Gatewood, the appropriate ROE ranged from 9.0 to 9.50 percent. In
making this determination, Gatewood rejected the lower end of his economic findings,
which would have recognized an ROE range of 8.50 to 9.50 percent. Gatewood referred
to this action as gradualism, recognizing that although KCP&L's cost of debt has
decreased, it had not decreased significantly. Moreover, Gatewood acknowledged that
authorized ROEs below 10 percent were a relatively recent development in electric rate
17
cases, and Gatewood was hesitant to recommend an ROE below 9.0 percent until there
was more evidence that current lower capital markets would continue.
We offer CURB's expert opinion.
After reviewing KCP&L's rate application and the testimony presented in support
of it, Woolridge testified about the ROE issue on behalf of CURB. Woolridge, like the
other experts, selected a proxy group of similar utilities and used the DCF and CAPM
models using his selected proxies to determine a reasonable ROE. He also used Hevert's
proxy group using the same models. Woolridge testified that based on his analysis, the
appropriate ROE for KCP&L was 8.55 percent.
In reviewing Hevert's DCF calculations, Woolridge testified there were three
primary errors in Hevert's analysis: (1) He essentially eliminated 1/3 of his results by
ignoring the low-mean figures in his analysis; (2) he relied excessively on overly
optimistic forecasted growth rates; and (3) the GDP growth rate he used was excessive,
did not reflect economic growth in the U.S., and was about 100 basis points above
projections of GDP growth. Recalculating Hevert's DCF results using a more appropriate
nGDP growth rate established that the proper equity cost rates mirrored Woolridge's.
Woolridge also challenged Hevert's CAPM analysis. Again, Woolridge found
Hevert's calculations were based on unreasonably high assumptions about future
economic and earnings growth. Woolridge testified he based his CAPM risk premium of
5.50 percent based on market risk reflected in recent academic studies by leading finance
scholars, on premiums used by leading investment banks and management consulting
firms, and on surveys of companies, financial forecasters, and financial analysts.
Woolridge testified that Hevert's risk premium CAPM approach focused on regulators'
behavior rather than investors' behavior.
18
He testified that capital costs are determined through the financial decisions of
investors, while regulators evaluate not only ROE but also take into account other utility-
and case-specific information. Woolridge found that the risk premium was inflated since
electric utility companies have been selling at a market-to-book ratio in excess of 1.0,
meaning that the authorized rates of return have been greater than the return investors
require. Woolridge testified his recommendation is below the average state-level
authorized ROEs in light of the historically low capital cost rates in the markets. He
testified that regulators have been slow to incorporate the lower ROE, but that the clear
trend with regulators now has been toward lower ROEs, with the norm at less than 10
percent. Woolridge also detailed his disagreement with Hevert's analysis of the impact of
potential Federal Reserve actions.
Hevert filed rebuttal testimony challenging the testimony of Gatewood,
Woolridge, and Chriss. Hevert recognized in his rebuttal testimony that economic
conditions had changed slightly and he reduced his growth rate from 5.61 to 5.37 percent.
But he did not believe the size of that reduction warranted a change in the ROE.
The following chart is a recapitulation of the experts' testimony on ROE:
Discounted Cash Flow Analysis
EXPERT LOW MEAN MEAN HIGH MEAN
Hevert* 8.32 – 8.57 9.36 – 9.62 10.34 – 10.60
Gatewood 8.09 8.41 8.73
Woolridge N/A 8.40 – 8.55 N/A
*Figures include upward adjustment for flotation costs.
Capital Asset Pricing Model Analysis
EXPERT LOW MEAN MEAN HIGH MEAN
Hevert* 9.53 – 9.79 9.36 – 9.62 10.34 – 10.60
Gatewood 6.61 - 6.96** N/A 10.03 - 10.76***
Woolridge N/A 8.1 N/A
*Figures include upward adjustment for flotation costs.
** Based upon forecasted growth data
***Based upon historical (1929-present) growth data
19
How the Commission resolved this controversy.
In its order, the Commission made its objectives clear.
"In determining rates, the Commission first establishes a revenue requirement
and then designs a rate structure. The revenue requirement includes rate base, operating
expenses, and rate of return. The rate of return is simply an opportunity to earn that rate,
not a guarantee. Rate design includes allocating costs among and within the customer
classes.
"In setting rates, the Commission's goal is to balance the interests of all
concerned parties and develop a rate within the 'zone of reasonableness.' The parties
whose interests must be considered and balanced include: (1) the utility's investors vs.
the ratepayers; (2) present vs. future ratepayers; and (3) the public interest.
"In allocating the revenue requirement among the customer classes, the
Commission follows cost causation principles, so 'that one class of consumers shall not
be burdened with costs created by another class.'"
In rejecting the KCP&L requested ROE, the Commission thought the request too
high and explained why.
"KCP&L's proposed 10.3% ROE represents an increase of 80 basis points from
its currently approved ROE of 9.5%. Both Gatewood and Woolridge testified KCP&L's
capital costs have declined since that Order was issued in December of 2012. Wal-Mart's
witness, Steve Chriss testified the national average authorized ROE for vertically
integrated utilities in 2014 was 9.91%, and as of June 2015 it is 9.71%. KCP&L is the
only party advocating an increase to its 9.5% ROE. Despite both Staff and CURB
recommending ROEs that are five basis points higher than what they recommended in the
12-764 Docket, there is no support for an ROE above 10%. KCP&L's proposed ROE
runs counter to the trends in Kansas and nationwide towards lower ROEs in recognition
of historically low costs of capital.
"Despite lowering his projected growth rate from 5.61% to 5.37%, Hevert does
not lower his recommended ROE range. The failure to adjust his recommended ROE
range to account for a lower projected growth rate causes the Commission to question the
20
validity of Hevert's ROE recommendation. To reach KCP&L's requested ROE, the
economy must grow at a similar pace to the growth experienced over the previous eighty
years. KCP&L presumes the U.S. economy will grow at an annual rate of 5.37%. This
prediction is in sharp contrast to the estimates of the Federal Reserve Board, Social
Security Administration, Energy Information Administration, and major corporations
such as ExxonMobil, each of whom produce their own economic growth forecasts. Based
on the current economic climate, the Commission believes a projected annual growth rate
of 5.37% is unreasonably optimistic. The current economic conditions do not support the
10.3% ROE advanced by Hevert.
"At the same time, CURB's recommended ROE of 8.55% strikes the Commission
as too low. Woolridge's recommended ROE is well below the average rates of return
being allowed to electric utilities similar to KCP&L. As explained above, Chriss testified
the national average authorized ROE for vertically integrated utilities in 2014 was 9.91%,
and as of June 2015 it is 9.71%. Similarly, Gatewood testified 9.1% represents the lowest
current ROE in the nation. In Docket No. 14-ATMG-320-RTS, the Commission
approved an ROE of 9.1% for Atmos. Atmos is a natural gas public utility. In setting
Atmos's ROE, the Commission recognized natural gas distribution companies are
generally considered less risky than electric utilities, and therefore Atmos's ROE should
be lower than that of KCP&L. Consistent with Commission policy, CURB's
recommended ROE below 9.1% must be rejected.
"The Commission finds the nGDP growth estimates of 4.38% advocated by
Gatewood, and consistent with the nominal forecast by the Social Security
Administration and Energy Information Administration, to be more credible than the
5.37% suggested by Hevert. Gatewood testified that any ROE within his recommended
range of 9.0% to 9.5% is reasonable. On cross-examination, Gatewood acknowledged
that the yield on KCP&L's bonds is an important consideration in determining an
authorized ROE."
We examine KCP&L's arguments.
KCP&L claims the testimony of Staff's expert, Gatewood, had been "seriously
undermined" and therefore should have been rejected. KCP&L complains that the
Commission failed to explain why it found Gatewood's testimony more reasonable.
21
KCP&L's complaints about Gatewood's ROE analysis are three-fold. First,
KCP&L argues that Gatewood's analysis is unfounded because he failed to incorporate
the greater risk faced by KCP&L by using a broader range, thus less comparable, proxy
group. Second, KCP&L disputes testimony that there was a statewide and national trend
toward lower ROEs because of alleged lower costs of capital. Third, KCP&L
characterized Gatewood's testimony regarding long-term growth rates as "weak" and
undermined by Gatewood's attempt to limit his range from going too low.
In addressing KCP&L's concerns as to the proper treatment of its high-risk factors,
the utility contends that Staff's recommendation of a ROE well below the average returns
allowed by other commissions was arbitrary, especially in light of KCP&L's higher level
of risk. KCP&L asserted that Gatewood's testimony about the risk rating for KCP&L (the
Beta coefficient) failed to take into account the risks identified by Hevert's testimony.
Moreover, only one of the companies in Gatewood's proxy group had a Beta coefficient
as high as GPE.
Gatewood testified, however, that the risk premiums used by Hevert were based
upon historic data that was not reflective of current circumstances. Gatewood also relied
on the Commission's current use of rate design, pass-through charges, and other rate
mechanisms that lowered the risk for utilities; Gatewood claims Hevert's testimony
ignored these mechanisms. Gatewood agreed that the Beta coefficient of GPE/KCP&L
was higher (its stock was more volatile) than almost all of the companies in his proxy
group, but he stated that the differences were not drastic. Finally, Gatewood noted that
shareholders were not entitled to be compensated for every risk because investors can
reduce risk by diversifying their portfolios.
KCP&L also discusses Gatewood's reliance on a nationwide trend toward lower
ROEs and lower costs of capital. But, we note that Walmart's expert, Chriss, cited data
from SNL Financial that reflected that the average ROEs allowed to 129 electrical
22
companies in rate cases between 2012 and 2015 was 9.88 percent. For vertically
integrated electrical utilities (such as KCP&L), the ROE average was 10.10 percent in
2012 and decreased every year through Chriss' testimony in 2015 to 9.71 percent.
Although KCP&L argues in its reply brief that the Commission did not cite this evidence
in its orders, the Judicial Review Act directs this court to consider any evidence in the
record "cited by any party that detracts . . . as well as . . . supports such finding." K.S.A.
2015 Supp. 77-621(d).
In addition, Gatewood testified that it was undisputed that since KCP&L's prior
rate case (764 Docket) filed at the end of 2011, the interest rates on corporate debt
dropped from 5.2 to around 4.8 percent while dividends from his proxy groups dropped
from 4.45 in 2011 to 3.80 percent in the current docket. On cross-examination at the
hearing, Gatewood acknowledged that a FINRA report—issued after the rate case and his
initial testimony were filed—showed that KCP&L's bond yield had increased to 4.461
percent. After 5 years of decreasing, indeed, the bond yield had increased one quarter.
Gatewood testified that the rates fluctuate consistently throughout a rate case and that is
well-known by the Commission. There were no indications during cross-examination that
interest rates on KCP&L's debt had changed.
Finally, KCP&L challenges Gatewood's testimony regarding long-term growth
and his use of gradualism in his analysis. Gatewood testified that he did not come across
any substantial research that would lead him to believe that growth rates and returns
earned in the past are necessarily representative of what investors expected in the future.
Gatewood relied upon growth projections by a number of entities for the short term and
long term. Gatewood also testified that FERC concluded that using long-term nGDP was
the proper basis to estimate long-term growth for interstate pipeline and electric
transmission companies.
23
As for gradualism, Gatewood admitted in cross-examination that his models
included an ROE range including figures of less than 9 percent, even though Gatewood
had recommended 9.2 percent in his testimony. However, he ignored that portion of the
model outcomes below 9.0 percent because he preferred to wait to see if current market
conditions continued before he would recommend an ROE below 9.0 percent. Gatewood
recognized that regulatory commissions are slow to recognize changes in capital markets,
but if 2 to 3 years from now the capital markets remained low, he could see
recommending an ROE below 9 percent. Despite his choice to use gradualism, Gatewood
was comfortable with his inputs and believed his calculations were reasonable and
reflected the realities of the marketplace.
KCP&L attacks Gatewood's gradualism approach by asserting it reflected an
inadequate analysis. Significantly, KCP&L overlooks the fact that its own expert ignored
the bottom third of the outcomes of his various calculations (the low mean figures) and
made recommendations based solely on the high mean averages of his calculations,
which all included flotation costs.
KCP&L further challenges the Commission's finding that Hevert erred in relying
on a growth rate based on the average growth in the economy over the last 80 years.
KCP&L also challenges the Commission's finding that Hevert was not a credible witness
because he failed to adjust his ROE after reducing his growth rate from 5.61 percent to
5.37 percent. KCP&L claims there was no evidence in the record that this "slight change"
in the growth rate would have affected the ROE. However, this constitutes a 24 basis
point reduction in the growth rate. Hevert previously testified that the growth rate was a
very important and weighty factor in an ROE calculation. KCP&L's complaints that the
Commission's rejection of 12 basis points in the ROE for flotation costs was arbitrary,
while arguing that its expert's failure to adjust for a 24 basis point reduction in the growth
rate is insignificant, is disingenuous.
24
KCP&L also repeatedly challenges the Commission's failure to give weight to
Hevert's Chart 3 (the ROE Chart) that listed the ROE established in recent regulatory
rulings across the country in vertically integrated electrical utilities.
The Commission rejected this chart because it failed to take into account other rate
mechanisms such as ECAs, TDCs, EERs and other interim rate changes permitted in
Kansas. Those acronyms stand for energy cost adjustment, transmission delivery charges,
and energy efficiency riders. These are pass-through mechanisms granted by the
Commission that allow the utility to charge its customers for these costs by passing them
through onto their bills. This means the utility can make these charges without the need
of asking for a new rate to be set. They are, however, subject to audit by the Commission.
Some states do not allow such pass-through charges. Hevert's chart did not indicate if
those other states listed on the chart took into account such pass-through mechanisms.
KCP&L argues, however, that the Commission ignored Hevert's rebuttal
testimony that Kansas was rated only "average" by the Regulatory Research Associates
(RRA) and his Schedule RBH-18 provided information regarding other regulatory
commissions' RRA rankings. However, that schedule failed to tie the proxy groups in
Hevert's ROE Chart to this RBH-18 to show how it would support his analysis. In
addition, KCP&L's arguments fail to account for Chriss' evidence from an entity related
to RRA which issued information showing an ongoing decline in ROEs in electrical
companies, including vertically integrated companies such as KCP&L.
As in any case of battling experts, every witness was cross-examined with the
minutiae of their calculations and "inputs" challenged. Both Hevert and Gatewood agreed
that the growth rate was a key element in the DCF models and had a significant impact
on the ROE determination; likewise, they agreed there was an element of subjectivity to
selecting a growth due to the uncertainty about future earnings. Hevert admitted that the
economic models used to determine ROEs are based on certain assumptions, which may
25
be more or less applicable depending upon differing market conditions. He admitted the
choice of economic models, the inputs into those models, the choice of proxy companies,
and the interpretation of model results are based on "reasoned judgment." He also
admitted there was no "strict mathematical solution" when determining ROEs. Instead,
the models used produce a range of results from which the ROE is estimated after a
comprehensive review of relevant data and information. Calculating growth rates is an
inherently uncertain endeavor and usually contested in rate calculations.
Our review of the record reveals that the Commission was confronted with
deciding which expert estimation was closest to the mark. It had to consider the results of
the application of data into sophisticated economic models. It had to answer for itself
whether the inputs each expert used were appropriate. Were the proxy companies proper
in number as well as qualities? Suffice it to say, the testimony of Gatewood and
Woolridge supports the Commission's conclusions. Additionally, the Commission clearly
gave its reasons for not giving Hevert's work more credit. In order to reverse this holding
by the Commission, we would have to hold that Hevert's testimony was correct and the
other two experts were wrong. Naturally, such a holding would require us to reweigh the
evidence. This we cannot do. Thus, we hold the Commission's findings are supported by
substantial competent evidence.
Based on this level of subjectivity and the Commission's expertise, KCP&L's
vigorous arguments simply boil down to a dispute about credibility on an issue requiring
subjectivity and reasoned judgment. It is not this court's role to reweigh the credibility of
witnesses. See K.S.A. 2015 Supp. 77-621(d). The Commission did that.
We turn to the flotation costs issue.
Flotation costs are the costs associated with the sale of new issuances of common
stock. They include out-of-pocket expenditures for preparation, filing, underwriting, and
26
other costs of stock issuance. KCP&L complains that the Commission changed its
practice regarding the treatment of flotation costs without adequate notice to the company
and failed to adequately explain the reasons for this change. Basically, KCP&L contends
this sudden shift in Commission practice was a violation of Home Telephone Co. v.
Kansas Corporation Comm'n, 31 Kan. App. 2d 1002, 76 P.3d 1071 (2003), rev. denied
277 Kan. 923 (2004).
The expert spoke about these costs.
Hevert testified that if a company is to attract adequate capital, a utility must be
given the opportunity to recover prudently incurred flotation costs. Hevert testified that
flotation costs are part of capital costs and should be treated as such, rather than simply as
operating expenses. Because flotation costs are recovered over time, inclusion of flotation
costs is appropriate—even if no new stock issuances are planned in the near future—in
order to ensure the utility has the opportunity to earn its required rate of return for the
future.
In Gatewood's testimony, he reported that Staff did not support recovery of
flotation costs through the ROE. Gatewood testified that KCP&L had not attempted to
qualify the amount, if any, of unrecovered costs associated with its issuance of common
equity. Gatewood testified that although KCP&L proposed to collect over $3.7 million as
flotation costs, KCP&L did not quantify the effect of this element on the revenue
requirement or why recovering this expense through the cost of equity was efficient;
Gatewood believed it was not efficient yet the cost of equity was increased to recover
associated income taxes. Gatewood opined it is simply more efficient for the company to
track the actual flotation costs and include a pro forma adjustment to the test year
operation to include those costs as an expense in the rate case.
27
CURB's expert, Woolridge, criticized Hevert's adjustment for flotation costs and
noted that Hevert made an upward adjustment of 0.12 percent to the equity cost rate to
account for flotation costs. However, Hevert did not identify any test-year flotation costs
for KCP&L. Because there are no identified flotation costs, there is no reason to increase
KCP&L's revenues for unspecified costs.
This case differs from Home Telephone.
The factual situation in this case significantly differs from that found in Home
Telephone, 31 Kan. App. 2d 1002. In that case, both Staff and the utility were working
under the assumption during the entire proceeding that the company's income tax expense
would be based upon the taxes that would be assessed as if Home Telephone was a
Subchapter C corporation. This assumption was made even though Home Telephone was
actually a Subchapter S corporation whose shareholders paid income taxes, not the
company. The Commission had established this practice even though one panel of this
court had previously held that the income tax expenses for Subchapter S corporations
should be the lower of (1) the cost of income taxes the Subchapter S shareholders paid; or
(2) the cost of income taxes that would be assessed if the utility was a Subchapter C
corporation. Greeley Gas Co. v. Kansas Corporation Commission, 15 Kan. App. 2d 285,
807 P.2d 167 (1991).
Despite the Greeley Gas decision, it was undisputed that the Commission
routinely used only the Subchapter C tax imputation (perhaps because it was consistently
the lower of the two calculations). See Home Telephone, 31 Kan. App. 2d at 1006. In any
event, the Commission never mentioned the Greeley Gas case, even after its initial order
which requested the company and Staff work together to obtain and provide necessary
information to determine the correct imputed tax expense. 31 Kan. App. 2d at 1008.
Again, Staff and Home Telephone worked together to determine an agreed Subchapter C
income tax expense. In its final order, however, the Commission denied any income tax
28
expense because Home Telephone had failed to provide full information on the amount
of income taxes from the Subchapter S shareholders, citing Greeley Gas for the first time.
31 Kan. App. 2d at 1008-09.
The panel in Home Telephone found that the Commission had initially relied on its
long practice of relying on the income tax equivalent for a Subchapter C corporation. It
failed to advise either Staff or Home Telephone that it required the type of information as
stated in the Greeley Gas case until after its initial order was issued. Our court found that
the Commission's last-minute change was arbitrary and capricious. 31 Kan. App. 2d at
1011-12. The panel found that the Commission was obligated to inform both Home
Telephone and Staff that it would be deviating from its prior practice. 31 Kan. App. 2d at
1012.
In this case, KCP&L was put on notice as soon as Gatewood filed his testimony
that Staff was challenging the inclusion of flotation costs in the ROE unless KCP&L
established how much of its costs from the 2006 and 2009 issuances remained
unrecovered. It was also aware that Staff was taking a position that flotation costs should
not be included in the ROE as a matter of practice but should be treated as an expense in
the revenue requirement calculation. KCP&L cross-examined Gatewood on this position,
and he agreed that Staff's position on flotation costs was evolving.
While Hevert filed prefiled rebuttal testimony challenging the reasoning for
treating flotation costs as part of ROE rather than an expense, KCP&L made no effort to
provide financial details in response to Staff's assertion that they should be treated
differently. KCP&L had ample opportunity to provide additional information to respond
to Staff's assertions that since 2009, the ROE adjustment had more than paid off the
flotation costs from 2006 and 2009.
29
Even though KCP&L contends that providing additional information could have
been treated as an amendment to its application and, thus, delay the timelines in the rate
case, it cites no prior Commission case where a company providing additional
information in response to an opponent's challenge has caused such a delay. Utility
companies have consistently responded to data requests from Staff and intervenors in
cases, as well as provided supplemental calculations in response to arguments made in
the prefiled testimony of opponents without the Commission restarting the 240-day
statutory rate case timeline.
Nationwide, there appears to be a recent trend in applications filed with public
regulators for challenges to be made to the inclusion of flotation costs in ROEs. Some
regulators have permitted an ROE adjustment, but that adjustment is amortized over a
specific time period, rather than being allowed to be a perpetual cost. See In Re Potomac
Electric Power Company, 314 PUR 4th 165, 205 (Md. PSC 2014). Some regulators have
rejected ROE adjustments for flotation costs based upon claims of need for future capital,
since it was unknown whether the capital would be raised by stock issuances or through
other sources such as retained earnings. See In Re Southwestern Electric Power
Company, 309 PUR 4th 116, 130, 139 (Tex. PUC 2013). Other regulators have allowed
ROE adjustments only when the utility has established that its parent company's stock
issuances have directly benefitted the subsidiary. In Re The United Illuminating
Company, 307 PUR 4th 1, 130 (Conn. PURA 2013). Still, another allows flotation cost
adjustments in ROEs only if actual flotation costs are identified or new common stock
issuances have occurred in the test year and the company has demonstrated that its stocks
are trading at, or below, its book value. In Re Southern California Edison Company, 302
PUR 4th 312, 324 (Cal. PUC 2012).
Some advocates argue that flotation costs should be treated as a transaction cost,
i.e., operating expense, rather than an adjustment to an ROE, similar to the manner in
which bond expenses are measured. See In Re Connecticut Natural Gas Corporation, 310
30
PUR 4th 327, 403 (Conn. PURA 2014); In Re Potomac Electric Power Company, 314
PUR 4th at 203. It has also been asserted that flotation costs should be recovered only if it
is shown that the market-to-book ratios established that the new stock will trade for
below book value. Connecticut Natural Gas Corporation, 310 PUR 4th at 403.
At least one state commission has adopted the concept that flotation costs were
more accurately classified as an expense. See In Re SourceGas Distribution LLC, 297
PUR 4th 529, 548 (Neb. PSC 2012). The Nebraska Commission rejected making an
adjustment to the ROE, finding that utilities should not receive a perpetual return on a
one-time or intermittently occurring cost. Likewise, it found that ratepayers should not be
required to compensate shareholders for any potential, unsubstantiated downward
pressure on their stock prices due to issuances of new stock. 297 PUR 4th at 548.
How to treat these expenses is a matter of policy to be determined by the
Commission and not this court.
In this case, KCP&L had ample opportunity to provide data and testimony in
response to Staff's proposed changes; therefore, the facts in this case do not rise to the
level of the "moving target" that occurred in Home Telephone. See Home Telephone, 31
Kan. App. 2d at 1010.
Finally, KCP&L argues that the Commission has failed to adequately explain the
basis for its change in treating flotation costs. Indeed, the Commission's explanation was
sparse on this point. Basically, it merely relied on KCP&L's failure to identify the amount
of unrecovered costs and reiterated its conclusions that a pro forma adjustment to test-
year operations to include actual costs was a more sensible way in which to recover such
costs. In its order on reconsideration, the Commission merely rejected KCP&L's reliance
on Home Telephone and made no other rationalizations for the altered course.
31
The Commission argues in its brief that it did not adopt a new policy on flotation
costs because it did not have a prior policy on such costs. Despite this assertion,
Gatewood clearly testified that Staff's view of flotation costs had changed and that
flotation costs had been included in certain cases in the past, including KCP&L's prior
cases. On appeal, the Commission argues that while Staff changed its position, the
Commission itself had no established practice.
While these arguments are interesting, we must remember the context. The overall
rate of return adopted by the Commission was supported by substantial evidence from the
record as a whole and was not arbitrary or capricious. Hevert's ROE recommendation
relied only on the high mean of his various ROE calculations. Hevert's low mean range
using the DCF model—8.32 to 8.57 percent—was within the range recommended by
Woolridge. The mean from his DCF analysis was close to that recommended by
Gatewood: 9.36 to 9.62 percent. Even Hevert's multi-stage DCF analysis—using his high
estimated growth rate—did not justify an increase in the 9.5 percent awarded in the prior
docket; it produced a low mean range for an ROE of 9.53 to 9.79 percent. What is even
more significant is that all of Hevert's calculations included an upward adjustment for
flotation costs. Thus, even if the Commission's actions were arbitrary in any shift
regarding the treatment of flotation costs, the ultimate ROE adopted was close to Hevert's
low mean ROE calculations that included flotation costs.
The Commission's decision on KCP&L's ROE was not arbitrary or capricious.
On this point, we are mindful of our judicial task. We must evaluate the overall
rate order in determining whether it is reasonable and provides KCP&L with the
opportunity (not a guarantee) to earn sufficient funds, if it operates reasonably and
efficiently, to pay its operating costs and earn a sufficient return on its rate base to remain
financially stable. Southwestern Bell Tel. Co. v. State Corporation Commission, 192 Kan.
39, 58-59, 386 P.2d 515 (1963). The court must avoid the risk of converting what should
32
be an overall review of the total rate to a narrow lens isolating only one or two issues and
ignoring the Commission's aggregate decision. See Citizen Utility Ratepayer Bd. v.
Kansas Corporation Comm'n, 47 Kan. App. 2d 1112, 1134-35, 284 P.3d 348 (2012).
The United States Constitution prohibits legislatures or their surrogates, the
regulatory commissions, from determining rates for public utilities that are
"confiscatory." Bluefield Co. v. Pub. Serv. Comm., 262 U.S. 679, 690, 43 S. Ct. 675, 67
L. Ed. 1176 (1923). Still, in Bluefield, the Supreme Court recognized the level of
discretion and judgment given to the regulatory agency.
"What annual rate will constitute just compensation depends upon many circumstances
and must be determined by the exercise of a fair and enlightened judgment, having
regard to all relevant facts. A public utility is entitled to such rates as will permit it to
earn a return on the value of the property which it employs for the convenience of the
public equal to that generally being made at the same time and in the same general part of
the country on investments in other business undertakings which are attended by
corresponding risks and uncertainties; but it has no constitutional right to profits such as
are realized or anticipated in highly profitable enterprises or speculative ventures. The
return should be reasonably sufficient to assure confidence in the financial soundness of
the utility and should be adequate, under efficient and economical management, to
maintain and support its credit and enable it to raise the money necessary for the proper
discharge of its public duties. A rate of return may be reasonable at one time and become
too high or too low by changes affecting opportunities for investment, the money market
and business conditions generally." (Emphasis added.) Bluefield Co., 262 U.S. at 692-93.
KCP&L cites Bluefield Co. in support of its claim that the ROE granted by the
Commission is unreasonably low. The Kansas Supreme Court recognized and adopted
these standards in Southwestern Bell Tel. Co., 192 Kan. at 58-59; Kansas Gas & Electric
Co. v. Kansas Corporation Comm'n, 239 Kan. 483, 488-89, 720 P.2d 1063 (1986).
KCP&L limits its focus on the ROE/flotation costs issues. The utility fails to explain how
these decisions, taken together with the Commission's adoption of the settlement
33
agreements and rulings on other contested issues, renders inadequate the overall rate set
by the Commission. The rate approved by the Commission here is hardly confiscatory.
We note that KCP&L initially requested a rate increase of $67.3 million overall.
Staff recommended an increase of $44 million, and CURB recommended just under $17
million. Prior to the hearing, the Commission approved a unanimous partial settlement
agreement that provided for a partial rate based finding of $2,114,033,286 subject to
contested issues of unrecovered reserve for AMR Meters and for fossil fuel inventory
issues.
In addition, it is clear that the approved partial settlement agreement made several
other adjustments requested by KCP&L. The Commission adopted KCP&L's requested
depreciation rates, approved decommissioning cost accruals relating to the Wolf Creek
nuclear plant, and allowed amortization for various costs, including deferred depreciation
of the La Cygne facility. The Commission included rate case expenses not to exceed
$1.83 million to KCP&L's operating expenses and gave KCP&L permission to file an
abbreviated rate case under K.A.R. 82-1-231(b)(3) within 14 months after the order in
this case to address various issues regarding various costs relating to the La Cygne and
Wolf Creek facilities.
The Commission further approved a Transmission Delivery Charge rider with an
estimated annual value of over $33 million and adopted relevant changes to KCP&L's
Energy Cost Adjustment rider. These two permit the company to pass through its
changing costs of transmission and energy without having to seek new tariffs or rate
proceedings.
Finally, the Commission approved a Critical Infrastructure Protection
(CIP)/Cybersecurity Tracker that allows KCP&L to include carrying costs for complying
with federal oversight standards mandatory for the security of the electric grid and critical
34
infrastructures. The order also permits KCP&L to track these ongoing costs, not yet
known, for review in the next rate case.
Even with all of these pass-through mechanisms that allows KCP&L to minimize
its risk, it still claims the ROE is inadequate. KCP&L's own expert's calculations belie
that assertion. Hevert's ultimate recommendation essentially relied only on the high mean
of his various ROE calculations. Hevert's low mean range using the DCF model—8.32 to
8.57 percent—was within the range recommended by Woolridge. The mean from his
DCF analysis was close to that recommended by Gatewood: 9.36 to 9.62 percent. Even
Hevert's multi-stage DCF analysis—using his high estimated growth rate—did not justify
an increase in the 9.5 percent awarded in the prior docket; it produced a low mean range
for ROE of 9.53 to 9.79 percent. What is even more significant is that all of Hevert's
calculations included an upward adjustment for flotation costs.
In summary, all three experts used the same general well-recognized economic
models for estimating the ROE. Their opinions simply differed. Hevert's proxy group of
companies was much smaller, whereas the larger proxy groups selected by Gatewood and
Woolridge arguably could be viewed as providing a broader picture of appropriate ROEs.
Hevert selected growth rates based upon information from private investment analysts,
ignoring estimates from the Social Security Administration, the Energy Information
Agency, and other sources. Although KCP&L and Hevert challenged the inputs and
assumptions relied upon, the Commission's decision is not arbitrary or capricious.
In our review of the record, we hold that the Commission has taken a middle-
ground approach in determining this ROE multiplier, including its treatment of flotation
costs. We cannot fault the Commission for being reasonable.
We affirm the Kansas Corporation Commission.
35