20170215Gaske Rebuttal.pdf

Ronald L. Williams,ISB No.3034
Williams Bradbury, P.C.
l0l5 W. Hays St.
Boise,ID 83702
Telephone: [phone]
Email: [e-mail]
Attorneys for Intermountain Gas Company
BEFORE TIIE IDAHO PUBLIC UTILITIES COMMISSION
IN THE MATTER OF THE APPLICATION OF
INTERMOUNTAIN GAS COMPANY FOR
THE AUTHORITY TO CHANGE ITS RATES
AND CHARGES FOR NATURAL GAS
SERVICE TO NATURAL GAS CUSTOMERS
IN THE STATE OF IDAHO
Case No. INT-G-16-02
REBUTTAL TESTIMONY OF STEPHEN GASKE
FOR INTERMOUNTAIN GAS COMPANY
February 15,2017
)
)
)
)
)
)
)
1
2
J
4
5
6
7
8
9
a. Please state your name, position and business address.
A. My name is J. Stephen Gaske and I am a Senior Vice President of Concentric
Energy Advisors Inc., 1300 19th Street NW, Suite 620, Washington, DC 20036.
a. Are you the same J. Stephen Gaske who liled Prepared Direct Testimony earlier
in this proceeding?
A. Yes.
a. What is the purpose of your rebuttal testimony in this proceeding?
A. I am responding to the Direct Testimonies of Mark Rogers and Terri Carlock on
behalf of the Staff of the Idaho Public Utilities Commission ("Staff') regarding
the retum on common equity capital and portions of the Direct Testimony of
Michael P. Gorman on behalf of the Northwest Industrial Gas Users ('NWIGU")
related to the retum on equity capital and capital structure. These witnesses
recorlmend an allowed retum on common equity of 9.25 percent and 9.3 percent,
respectively, for Intermountain Gas' Idaho natural gas distribution operations.
However, as shown in my Prepared Direct Testimony, and as discussed herein, a
return on common equity of g.gpercent is required for Intermountain Gas to be in
a position to raise capital on reasonable terms. I disagree with several areas
presented in the testimonies of Mr. Rogers, Ms. Carlockl and Mr. Gorman that
lead them to recommend an inadequate return, including:
While Staffs ROE analysis and much of the discussion regarding methodologies and models is
contained in the Direct Testimony of Mr. Rogers, Staffs ROE recommendation appears in the Direct
Testimony of Ms. Carlock.
Gaske, Reb. 1
Intermountain Gas Company
l0
ll
t2
13
t4
15
t6
t7
l8
19
I
2
J
4
5
6
7
8
9
I . Failure of Staff and Mr. Gorman to adequately reflect that Intermountain
Gas' Idaho natural gas distribution operations face greater overall risks
thanthe typical company in the proxy group;
2. Failtre of Staffand Mr. Gorman to consider the entire rzmge of results
produced by the DCF model, and in the case of Mr. Gorman, failure to use
the expected oosustainable" growth rate in his multi-stage DCF;
3. Mr. Gorman's use of an understated utility Risk Premilrm approach that
gives partial weight to an out-of-date market risk premium from 1987-1991
rather than giving fulI weight to the market risk premium for the most
recent five-year period;
4. Mr. Gorman's capital asset pricing model ("CAPM") estimates based on
historical market risk premium data ttrat understate investors' requirements
under current market conditions;
5. Failure to take into consideration investors' expectations for higher interest
rates as the Federal Reserve continues taking steps to normalize monetary
policy after an extended period of artificially-low interest rates;
6. The recommendation of an inadequate a flotation cost adjustrnent; and,
7. Mr. Gorman's recommendation to reduce the common equrty ratio in the
capital structure to a level that is less than the actual and target ratio of
Intermountain and well-below the median common equity ratio for the
proxy group companies.
Gaske, Reb. 2
Intermountain Gas Company
10
11
t2
l3
t4
15
t6
t7
l8
19
2T
20
I I. REASONABLENESS OT ROE RECOMMENDATIONS
Please provide an overview of Ms. Carlock's and Mr. Gorman's ROE
recommendations in this proceeding.
Ms. Carlock recommends a range of return on equity for Intermountain Gas of 8.5
percent to 9.5 percent and a point estimate of 9.25 percent. In addition, Ms.
Carlock recommends a reduction of 25 basis points in the authorized ROE for
Intermountain Gas to 9.0 percent if the Commission approves the Company's
proposed Fixed Cost Collection Mechanism ("FCCM"). Staffwitness Rogers
does not perform his own ROE analysis, but rather makes certain adjustments to
the DCF analysis presented in my Prepared Direct Testimony to include the
blended growth rate (i.e., an average of the analysts' eamings growth estimate and
the sustainable growth rate, calculated using Value Line data) for each proxy
group company and to reduce the flotation cost adjustment.
Mr. Gorman recommends a very narrow range of return on corlmon equity of 9.2
percent to 9.4 percent, and recommends a cost of equity of 9.3 percent based on
his Risk Premium analyses, the results of his constant growth DCF model using
analyst growth rate estimates, and his CAPM analysis. Mr. Gorman also
performs a constant growth DCF analysis using sustainable growth rates and a
multi-stage DCF analysis using long-term GDP growth, but he does not appear to
rely on those results in establishing his range or making his recommendation. In
performing his DCF and CAPM analyses, Mr. Gorman used the same proxy
group of gas distribution companies that I used in my Direct Testimony.
Gaske, Reb. 3
Intermountain Gas Company
2O.
A.
J
4
5
6
7
8
9
l0
t1
12
13
t4
l5
t6
t7
18
t9
2t
20
22
I
2
J
4
5
6
a. Please assess the reasonableness of both Ms. Carlock's 9.25 percent and Mr.
Gormants 9.30 percent recommended returns on common equity.
A. Figure 1, below, is a histogram of all retums on common equity authorized in
natural gas distribution company rate proceedings covered by Regulatory
Research Associates between 2012 and20l6.
Figure 1: Authorized Returns on Equity for Gas DistributionQ0l2-2016)2
The ROE recommendations of Ms. Carlock and Mr. Gorman are at the lower end of
equity retums that have been authorized for gas distibution companies since 2012.
Of the 120 rate case decisions with explicit ROE awards, only 14 (or I1.7 percent)
have been lower than9.25 percent and 17 (or 14.2 percent) have been lower than
Gaske, Reb. 4
Intermountain Gas Company
7
8
9
l0
l1
35
30
25
20
15
10
5
0
"r* ,",d ,os ,4s ,"t". 6s ".t+C r' C .rd' ,.s' ,C" ""r-
,*s^s,$,
t*.
-"ff'
2 Source: Regulatory Research Associates.
I
2
J
4
5
6
7
8
9
l0
l1
t2
13
t4
l5
t6
t7
l8
t9
a.
A.
9.30 percent. This indicates that Ms. Carlock's and Mr. Gorman's ROE
recommendations are lower than the vast majority of returns allowed by
Commissions during the past five years. The median authorized ROE during this
period for gas distribution companies was 9.73 percent, and there have been 4l
decisions (or 34.1 percent) with explicit ROE awards of 9.90 percent or higher,
which corroborates the reasonableness of my recommended 9.90 percent cost of
coilrmon equity.
Do you have other general concerns with Staf?s and Mr. Goman's analyses
and recommendations.
Neither Staff nor Mr. Gorman appear to have taken into consideration the fact that
interest rates are rising and are expected to continue to increase in20l7, as the
Federal Reserve normalizes monetary policy after aprolonged period of holding
interest rates artificially low following the financial crisis and Great Recession.
At its December 2016 Federal Open Market Committee ("FOMC") meeting, the
Federal Reserve not only raised the target federal funds rate by 25 basis points as
expected, but it also announced the Committee members' expectations that the
federal funds rate will be increased an additional TS basis points in2017 based on
the current economic outlook for employment and inflation.3 According to Blue
Chip Financial Forecasts, 89 percent of those surveyed after the FOMC's
FOMC, "Economic Projections of Federal Reserve Board members and Federal Reserve Bank
presidents," December 74,2016, Figure 2.
Gaske, Reb. 5
Intermountain Gas Company
3
I
2
J
4
5
6
7
8
9
December meeting expect the Federal Reserve will raise short-term interest rates
againat either the March or June meeting.a In response to the question about how
much they expect the Federal Reserve will raise interest rates in 2017,53 percent
ofthose surveyed expect an increase of50 basis points, 29 percent expect an
increase of75 basis points, and 13 percent expect an increase of 100 basis points.s
Since the filing of my Direct Testimony in August 2016, yields on 30-year
Treasury bonds have increased approximately 90 basis points (from 2.23 percent
on August 12,2016 to 3.1I percent in mid-January 2017). In a period of
anomalous financial market conditions due to the Federal Reserye's extraordinary
interventions, the results of an ROE analysis based on recent historical data (such
as dividend yields in the DCF model or the risk-free rate in the CAPM) need to be
interpreted carefully. For example, in two recent decisions the FERC expressed
concem that Federal Reserve actions may have artificially reduced current
dividend yields and the results of the DCF model.6 Expectations for higher
interest rates indicate that capital costs for public utilities will be higher on a
going-forward basis than in recent years.
l0
1l
12
13
t4
15
t6
17 II. RELATIVE RISK OF INTERMOUNTAIN'S IDAHO GAS OPERATIONS
Blue Chip Financial Forecasts, Vol. 36, Issue No. 1, January 1,2017.
rbid.
Opinion No. 531,147 FERC tl6l,23 a QOID; affd in Opinion No. 531-B, 150 FERC 'll6l,165 (March
3,2015); and Opinion No. 551, 156 FERC, n61,234 (Sept. 28, 2016),para.120-122.
Gaske, Reb. 6
Intermountain Gas Company
4
5
6
I
2
J
4
5
6
7
8
9
a.
A.
Do you agree with Ms. Carlock's conclusion concerning the risks of
Intermountain relative to the proxy companies?
No. Ms. Carlock does not provide a risk analysis and appears to ignore the
greater business and financial risks of Intermountain.
For example, Ms. Carlock ignores the circumstances of the proxy companies
when she recommends that approval of the proposed Fixed Cost Collection
Mechanism ("FCCM") should be accompanied by a corresponding reduction in
the authorized return on common equlty for Intermountain Gas. As shown in
Exhibit No. 5, Schedule 7 of my Direct Testimony, 66.7 percent of the operating
companies held by the companies in the proxy group have rate design
mechanisms that reduce volumetric risks and 66.7 percent also have monthly
customer charges for residential customers that are higher than the $10.00 being
proposed by Intermountain Gas. Thus, the typical proxy company currently has
mechanisms that reduce their rate design risk. However, Ms. Carlock's
recoflrmendation to reduce Intermountain's allowed return on equity by 25 basis
points if the FCCM is approved indicates that she incorrectly believes the proxy
companies do not already have mechanisms to reduce their rate design risks.
As noted on page 38, lines 5-9 of my Direct Testimony, my analysis assumed that
the proposed FCCM and customer charge would be approved, and that such
approval would tend to eliminate differences in that particular risk element. Thus,
a slight upward adjustment in the allowed rate of return on equity would be
Gaske, Reb. 7
Intermountain Gas Company
l0
11
t2
13
t4
l5
t6
t7
18
t9
2t
20
I
)
J
4
5
6
7
8
9
a.
A.
required if these proposals are not approved, but no downward adjustment would
be appropriate ifthey are not approved.
In addition, her recommended rate of return is insufficient relative to current risk
premium and capital asset pricing model estimates of the cost of capital. As a
result, Ms. Carlock's recommended rate of retum does not adequately reflect the
greater risks of Intermountain.
Please explain your disagreement with Mr. Gorman's assessment of the
Company's business risk
At one point in his testimony Mr. Gorman concludes that"the proxy group is less
rislE, butreasonably comparable in investment risk"7 to Intermountain Gas.
Later, however, he incorrectly claims that the criteria I used to select the proxy
group ensues that there are no differences in risk.8 Although I hied to select
proxy companies that were as similar as possible to Intermountain there were
some significant risk differences that remain.
As explained in my Direct Testimony, the typical proxy company has a more
diversified economy in its service territory and is between 9 and 22 times larger
than Intermountain Gas' Idaho jurisdictional gas distribution operations.e That is
an unavoidable fact because there are no comparably-sized, publicly-traded
companies with analysts' consensus growth rate estimates. The higher rate of
rbid.
Direct Testimony of Michael P. Gorman, pageT2,lines l-7.
Direct Testimony of J. Stephen Gaske, at 3 l.
Gaske, Reb. 8
Intermountain Gas Company
10
ll
t2
13
t4
15
t6
t7
18
t9
7
8
9
1
2
J
4
return required by smaller utility operations has been demonstrated empirically.l0
Moody's Investors Service has described how it considers the diversity of utility
operations as a risk. Specifically, in "Rating Methodology for Regulated Electric
and Gas Utilities" Moody's stated:
We also consider the diversity of utility operations (e.g., regulated
electric, gas, woter, steam) when there are material operations in
more than one area. Economic diversity is typically afunction of the
population, size and breadth of the tenitory and the businesses that
drive its GDP and employment. For the size of the teruitory, we
typically consider the number of customers and the volumes of
generation and/or throughput. For breodth, we consider the number
of sizeable metropolitan areos served, the economic diversity and
vitality in those metropolitan areas, and any concentration in a
particular area or industry.tl
Much of Intermountain Gas' Idaho service territory is characterized by the small
size and small town lack of diversity described by Moody's. Moody's rating
methodology confirms that companies with those attributes have elevated risk,
which suggests that an allowed retum above the return required for the typical
proxy company is appropriate in this proceeding.
On pages 22-32 of his testimony, Mr. Gorman provides general information on
the utility industry, including authorized returns, capital spending trends,
credit rating agency commentary, and utility stock price per{ormance. Do you
have any comments on this section?
Michael Annin, Equity and the Small-StockEJpcl, Public Utilities Fortnightly, October 15, 1995.
Moody's, "Rating Methodolory: Regulated Electric and Gas Utilities," December 23,2013,p. 19.
Gaske, Reb. 9
Intermountain Gas Company
5
6
7
8
9
10
1l
t2
l3
t4
l5
t6
t7
18
t9
20
2t
22
23
a
l0
ll
lA.Much of the industry perspective that Mr. Gorman provides in this section of his
testimony pertains to electric utilities, not gas distribution companies. As such,
this evidence is not particularly relevant to the determination of a fair return for
Intermountain Gas because the business and operating risks for electric utilities
are not the same as those for gas distibution companies.
With regard to his analysis of gas utilities in general, Mr. Gorman's claim that gas
utility stocks have been more stable than the general stock market since 200412 is
debatable. The basis for his claim is the graph in Figure 3, page 31 of his
testimony. However, the standard deviation of gas utility retums shown on Mr.
Gorman's Figure 3 is 17.5 percent, while the standard deviation of the S&P 500
retums is only 15.8 percent. This wider spread in the probable returns suggests
that Mr. Gorman is overestimating the relative stability of gas utility retums.
I agree with Mr. Gorman that capital spending forecasts for the natural gas
industry are considerably higher than the historical average since 2005.13 As Mr.
Gorman states, "this capital investment is exceeding internal sources of funds to
the gas utilities, requiring them to seek external capital to fund higher capital
investment." Intermountain Gas is also engaged in a large capital spending
program over the next few years in order to comply with federal pipeline safety
requirements and to replace aging infrastructure. Given this need for ongoing
Direct Testimony of Michael P. Gorman, page 30, line 3l to page 31, line 4.
Ibid, at 25.
Gaske, Reb. l0
Intermountain Gas Company
2
J
4
5
6
7
8
9
10
1l
t2
13
t4
15
t6
t7
t8
19
t2
l3
1
2
J
4
5
6
7
8
9
10
l1
t2
13
t4
15
t6
t7
18
t9
20
access to capital, it is very important that the authorized return on common equity
for Intermountain Gas be set at a level that allows the Company to compete for
capital on reasonable terms with comparable risk utilities.
a. What is your conclusion regarding the risk analyses of Staffand Mr. Gorman?
A. The cost of common equity recommendations of both Staffand Mr. Gorman fail
to adequately reflect the greater business and financial risks of Intermountain's
gas distribution operations in comparison to the risks of the proxy companies.
The Company has greater business risks, and Intermountain also has above-
average financial risks due to its proposed common equity ratio being lower than
the proxy group median.la Intermountain therefore has a cost of capital that is
above the average or median for the proxy companies and none of the other
parties' rate of return recommendations sufficiently reflect this fact.
III. DCF ANALYSES OF MR. GORMAN AND MR. ROGERS
a. Mr. Gorman's DCF Analyses
a. Please summarize Mr. Gorman's DCF analyses.
A. Mr. Gorman constructed a Constant Growth DCF model using analyst growth
rates, a Constant Growth DCF model using sustainable growth rates, and a Multi-
Stage Growth DCF model. These models produce a range of average ROE
estimates from 7 .79 percent to 9.69 percent. Mr. Gorman derives a range of DCF
retums from 8.99 percent (based on the median results of his constant growth
Gaske, Reb. I I
Intermountain Gas Company
l4 Ibid, at 38
1
2
J
4
5
6
7
8
9
10
11
t2
l3
t4
t5
t6
l7
18
t9
DCF analysis using analysts' earnings growth rates) to 9.69 percent (based on the
average results of his constant growth DCF analysis using sustainable growth
rates), and he concludes that the DCF studies support a return on equity of 9.40
percent. ls
a. Do you agreewith Mr. Gorman's conclusion regarding his DCF returms?
A. No, I do not. Mr. Gorman states that "his DCF studies support a return on equity
of 9.40 percent for the proxy companies," which is the same as the median return
of my Basic DCF analysis which relies on analysts' earnings growth estimates.
However, Mr. Gorman does not include the required flotation cost adjustment, nor
does he make an appropriate adjustment for the gteater risk of Intermountain Gas'
Idaho natural gas distribution operations relative to the average proxy company.
In fact, Mr. Gorman's DCF based on analysts' estimates shows a range of 6.84 -
13.95 percent, and his DCF based on sustainable growth indicates a range of 7.12
- 12.73 percent. In addition, when Mr. Gorman's "Multi-Stage" DCF analysis is
re-calculated using "Sustainable" growth as the second stage, the resulting
analysis indicates a range of 7.14 - 12.22 percent, with a median of 9.74 percent.
Consequently, my estimate of 9.90 percent is well within the range of Mr.
Gorman's reasonable DCF results.
a. What growth rate estimates does Mr. Gorman use in his three DCF models?
Gaske, Reb. 12
Intermountain Gas Company
15 Direct Testimony of Michael P. Gorman, at 52.
2
3
4
5
6
7
8
9
1A.
11
t2
l3
t4
15
t6
t7
18 a.
t9
20 A.
2t
For his Constant Growth DCF model, Mr. Gorman uses a simple average of three
different consensus estimates of earnings growth (Zacks, Reuters, and SNL
estimated growth rates), whereas I use those reported by Zacks and Yahoo!
Finance, which are both based on consensus forecasts. Mr. Gorman's Sustainable
Growth DCF model uses a growth rate based on Value Line's three-to-five year
projections of eamings, dividends, earned retum on book equity, and projected
book value growth from stock issuances. The results of these inputs to his
analysis are similar to mine. Both Mr. Gorman's analysts' earnings growth
estimates of 6.24 percent and his sustainable growth estimate of 6.55 percent are
somewhat higher than my corresponding growth rate estimates.
Mr. Gorman's Multi-Stage DCF model uses growth rates for each proxy company
that are a form of weighted average of the analysts' eamings growth forecasts for
each company (in years l-5) and the nominal GDP growth forecast (in years l1-
200). Mr. Gorman gives significant weight to his long-term growth rate, which is
based on U.S. projected nominal GDP growttr, by assuming that each proxy
company's growth rate will converge to the projected growth rate for U.S. GDP of
4.25 percentwithin l0 years. As discussed below, this is inappropriate.
Are analysts' growth rates generally a superior measure of long-term investor
expectations?
Yes. Although analysts' longest-term growth forecasts are typically expressed as
five-year forecasts, these forecasts generally represent growth rate expectations for a
longer period of time than the five-years expressed in the forecast. There is a large
Gaske, Reb. 13
Intermountain Gas Company
l0
22
I
2
J
4
5
6
7
8
9
10
1t
t2
13
t4
15
t6
t7
a.
A
amount of literature that suggests analystso growttr rate forecasts are a superior
measure of the long-term growth rate expectations that are reflected in stock prices.
For example, Vander Weide and Carleton found that analysts' eamings growth rate
forecasts have a very highly significant relationship with stock prices.r6 This
indicates that the analysts' earnings growth estimates are an accurate estimator of
long-term growth rate expectations implicit in stock prices, even though the
analysts' eamings growth estimates are putatively five-year estimates. Similarly,
Marston, Harris and Crawford examined publicly available data from 1982-1985
and found that plausible meastres of risk are more closely related to expected
retums derived from a constant growth DCF model than to those derived from multi-
stage growth models.lT
Did Mr. Gorman provide any assessment of the growth rates used in his
Constant Growth DCF model to growth estimates by reference to other
benchmarks?
Yes. Mr. Gorman notes that the average eamings growth estimate of 6.24 percent
for his proxy group in his Constant Growth DCF model is "higher than my long-
term sustainable growth rute of 4.25yo,;tt8 and he observes that "the median DCF
Vander Weide, J.H. and Carleton, W.T., "Investor Growth Expectations: Analysts vs. History," 7fte
Journal of Portfolio Management, Spring 1988, pp. 78-82.
F. Marston, R. Harris, and P. Crawford, "Risk and Return in Equity Markets: Evidence Using Financial
Analysts' Forecasts," n Handbook of Security Analysts' Forecasting and Asset Allocation, J. Guerard
and M. Gultekin (eds.), Greenwich, CT, JAI Press; as described in R. Harris and F. Marston, "Estimating
Shareholder Nsk Premia Using Analysts' Growth Forecasts," Financial Management, Summer 1992,
p.64.
Direct Testimony of Michael P. Gorman, at 43.
Gaske, Reb. 14
Intermountain Gas Company
t6
t7
l8
I
2
J
4
5
6
7
8
9
result for the proxy group more accurately reflects the central tendency of the
group"le due to the outlier growth rate for South Jersey Industries. He states that
"a long-term sustainable growth rate for a utility stock cannot exceed the growth
rate of the economy in which it sells its goods and services"20 and therefore the
long-term GDP growth rute of 4.25 percent is the maximum logical growth rate.
However, it is important to note that the GDP growth rate is an average for all
activities in the economy. At any given point in time, some companies or
industries grow faster than the economy, while other companies or industries are
growing slower or declining. Thus, it is not unusual for the growth rates of some
companies or industries to be below or above the average GDP growth rate for
significant periods of time. In addition, the use of GDP growth rates in Mr.
Gorman's Multi-Stage DCF analysis is flawed in that it assumes that over the
long-term, all companies in the proxy group converge to the same growth rate.
That is why it is important to place primary reliance upon company-specific
growth rate information in order to distinguish between sectors and companies
with declining, or below average gtowth, and those that are expected to comprise
the above-average growth sectors and companies.
Do you agreewith Mr. Gorman's use of projected nominal GDP growth rates
as the second-stage sustainable growth rate in his analysis?
rbid.
rbid.
Gaske, Reb. 15
Intermountain Gas Company
10
11
t2
l3
t4
15
t6
t7
18 a.
t9
l9
20
2
3
4
5
6
7
8
9
lA.
1l a.
t2
13 A.
t4
l5
t6
t7
18
No. On page 1 of his Exhibit No. 307, Mr. Gorman calculates the long-run
"sustainable" growth rate to be 6.55 percent - not 4.25 percent. Mr. Gorman
states that"a sustainable long-term earnings retention ratio will help gauge
whether analysts current three-to-five-year growth rate projections can be
sustained over an indefinite period of time."2l Because Mr. Gorman's sustainable
growth rate, 6.55 percent, exceeds the analysts' growth rate of 6.24 percent,
investors reasonably can expect the analysts' growth rates to be sustained over an
indefinite period of time. Thus, investors would not expect the proxy companies'
average growth rates to decline to the forecasted growth rate in U.S. GDP within
any time period that is materially significant for the DCF analysis.
Is there a more reasonable way to calculate Mr. Gorman's Multi-Stage DCF
model?
Yes. Mr. Gorman's use of U.S. GDP growth in calculating a Multi-Stage DCF
yields an implausibly low median return of 7.57 percent for the proxy companies.
As shown on attached Exhibit No. 33, Schedule 1, when one uses Mr. Gorman's
"Sustainable" growth rates as the second stage of the DCF model, the result is a
range of 7.14 - 12.22 percent, and a much more plausible and reasonable median
of 9.74 percent.
Gaske, Reb. 16
Intermountain Gas Company
10
2t Ibid, at 45.
I
2
J
4
5
6
7
8
9
10
11
t2
13
t4
l5
t6
t7
t8
t9
b. Staff DCF Analyses
a. Please summarize Mr. Rogerst testimony regarding the use of analystst
earnings growth rates in the DCF model.
A. Mr. Rogers states that "growth rate utilized by Mr. Gaske in the Basic DCF is
simply the growth estimates for each of the companies in the proxy group from
analysts at both Zacks Investment Research and Thomson First Call. The values
from both analysts are averaged together to determine the growth rate for each of
the proxy companies.tt22 \fu. Rogers expresses concern with the Basic DCF
analysis because he appears to believe that "unless, by chance, both analysts have
correctly predicted the actual growth rate, then the model will inevitably be based
on incorrect estimates and incomplete information."23 Finally, he concludes that
"the estimates from both analysts vary by up to 20Yo," which leads him to
conclude that "the model is inherently flawed and will be based to some degree on
incorrect estimates."24
a. What is your response?
A. As a preliminary matter, I want to clarifr that the earnings growth estimates from
Zacks and Thomson First Call are consensus estimates of industry analysts that
provide eamings estimates for each proxy group company; they are not the
estimate of a single analyst employed by Zacks or First Call, as Mr. Rogers
Direct Testimony of Mark Rogers, at 9.
Ibid, at 10.
rbid.
Gaske, Reb. 17
Intermountain Gas Company
22
23
24
I
2
J
4
5
6
7
8
9
l0
ll
t2
t3
t4
l5
t6
t7
l8
t9
appears to believe. More importantly, Mr. Rogers misunderstands the
significance of growth rate estimates. The earnings growth estimates for the
proxy group companies represent the information on which investors are basing
their decision to buy or sell the common equity of each company, and at what
prices. The relevant issue is not whether the earnings growth rate estimate is
ultimately correct or incorrect; rather, the key issue is whether the growth rates
reflect investors' expectations at the time that they buy or sell the stocks. The
consensus growth rate estimates are a good estimate of investors' growth rate
expectations when they make buy/sell decisions. I do not agree with Mr. Rogers
that the DCF model is flawed simply because there is a range of growth rate
estimates. lndeed, this diversity of opinion is one reason why some investors sell
the stocks and others buy the stocks. It is the consensus of those diverse opinions
that leads to an equilibrium in the market prices at which purchases/sales occur.
Please describe the reasons why Mr. Rogers argues the Commission should rely
on the Blended DCF analysis rather than the Basic DCF analysis.
Mr. Rogers argues that the Blended DCF analysis is more reliable and less
subjective because it combines earnings growth rates from analysts with retention
growth rates.2s Mr. Rogers also contends that "retention rates are the primary
driver of dividend growth and book value per share, which are the two primary
Gaske, Reb. 18
Intermountain Gas Company
a.
A.
25 Direct Testimony of Mark Rogers, at 11.
I
2
J
4
5
6
7
8
9
l0
ll
t2
13
t4
l5
t6
t7
l8
a.
A.
variables for attracting common equity investors."26 Mr. Rogers also notes that
Intermountain Gas has not filed a general rate case in decades, and argues that
"using a sustainable long-term growth rate more accurately reflects the reality of
Intermountain Gas in that the new rate of return may also be in effect for many
years."27
Do you agreewith Mr. Rogers' reliance on a Blended DCF analysis?
No, I do not. As discussed previously in my Rebuttal Testimony, academic
research has shown that earnings growth expectations are the most important
determinant of stock prices, not dividend growth or book value growth. The
retention growth rate changes based on management decisions to conserve cash,
manage the dividend payout ratio, or invest in capital projects that support growth
or enhance reliability and safety. In addition, analysts' earnings growth rates have
been shown to be the most reliable indicator of future dividend growth, and the
estimate on which investors rely when setting stock prices. Finally, the timing of
rate case filings has nothing to do with the appropriate growth rate in the DCF
model.
Mr. Rogers presents a regression analysis in which he finds that there is a near
perfect linear trend in dividend growth for MDU Resources, making future
Ibid, at 16.
Ibid, at 14.
Gaske, Reb. 19
Intermountain Gas Company
a.
26
27
I
2
J
4
5
6
7
8
9
A.
predictions on dividend payments for MDU Resources quite accurate.2s Please
comment on this analysis.
Mr. Rogers' regression analysis has an R-squared of 0.99, which causes him to
conclude that "[r]ather than relying on the opinion of analysts, using this type of
data from the proxy group of companies more accurately measures the proxy
group's future growth and the comparable return for Intermountain Gas."2e First,
MDU Resources is not a member of the proxy group for Intermountain Gas
because MDU Resources is a diversified company with a minor portion of its
business engaged in natural gas distribution operations. Second, Mr. Roger's
regression analysis incorrectly suggests that the risk of a company can be defined
by a linear regression of past dividend payments. Investment risk does not simply
involve the trend of past results but, instead, involves the uncertainty associated
with unexpected and unpredictable events in the future. Third, the regression
analysis of dividends presented by Mr. Rogers is not even particularly relevant for
investors. lnstead, earnings and stock prices are both far more relevant. As
shown on attached Exhibit No. 33, Schedule 2, a regression analysis of MDU
Resources' past earnings per share has an R-square of 0.04 which, if one were to
accept Mr. Rogers' methodology, suggests that MDU Resources' earnings are
entirely unpredictable and that the investment is highly risky. However,
Direct Testimony of Mark Rogers, at 13-14.
rbid.
Gaske, Reb. 20
Intermountain Gas Company
10
ll
t2
l3
t4
l5
t6
t7
l8
t9
28
29
I
2
J
reasonable forecasts of the future involve more than mere analysis of past trends,
and a regression of past trends does not establish either future expectations or a
level of predictability.
4 IV. CAPITAL ASSET PRICING MODEL ANALYSIS
sQ.Please describe your disagreement with Mr. Goman's use of the Capital Asset
Pricing Model to estimate the cost of common equity capital for Intemountain
Gas.
My primary disagreement with Mr. Gorman's CAPM analysis is his method of
estimating the market risk premium used in the analysis. Research studies provide
empirical support for the proposition that equity risk premia generally increase as
interest rates decrease, and vice versa. For example, as shown in the Risk
Premium analysis in Exhibit No. 5, Schedule 5 to my Direct Testimony, there is
an inverse relationship between the natural gas utility equity risk premia and
interest rates. Despite ttris fact, Mr. Gorman uses historical average market risk
premiums.
For example, Mr. Gorman calculated two altemative estimates for the risk premium.
He calls one of his estimates a "forward-looking" estimate, but that number is really
derived from the average historical real retum on common stocks from 1926-2015,
adjusted for projected inflation. From those historical returns, he subtracts the
current projected bond yield ftom Blue Chip Financial Forecasts to get a risk
premium estimate. Although two of the three elements in his "forward-looking" risk
premium are forwardJooking, the essential core element of the calculation is an
Gaske, Reb. 2l
Intermountain Gas Company
A.
6
7
8
9
10
11
t2
l3
t4
15
t6
t7
18
t9
2t
20
22
1
2
J
4
5
6
7
8
9
10
11
t2
l3
t4
l5
t6
t7
l8
t9
historical ayeragethat does not reflect current market conditions. Mr. Gorman's
other risk premium, which he calls an historical risk premium, is calculated
somewhat differently, but it is obviously based on historical averages rather than
current forward-looking data. 3 o
There generally is a sfrong inverse relationship between required risk premiums and
bond yields, and we are currently in a period with exceptionally low bond yields and
above-average risk premiums. In addition, in my Direct Testimony I calculated a
tue forwardJooking market risk premium using the S&P 500 companies, which is
considerably higher than historical averages. Thus, Mr. Gorman's historical average
data assumes inappropriately low market risk premiums for current market
conditions.
a. Does Mr. Goman's CAPM analysis produce plausible results?
A. No. As shown in ExhibitNo. 317,Mr. Gorman's CAPM analysis based on the
long-term historical average risk premium produces an implausibly low return
estimate of 7.86 percent; a return this low has never been awarded to a gas
distribution company in either the last five years or the last 30 years.3l
a. Did you also develop a CAPM result in your Direct Testimony?
A. Yes. As shown in Table 2 onpage29 of my Direct Testimony, if one were to use
the CAPM as a benchmark of a reasonable retum, the most reasonable current
30 Direct Testimony of Michael P. Gorman, p. 61.
Source: Regulatory Research Associates.
Gaske, Reb. 22
Intermountain Gas Company
3l
I
2
J
4
5
6
7
8
9
l0
1l
t2
13
t4
15
t6
t7
estimate would be 9.7 percent for the typical proxy company. This estimate based
on forward-looking data is consistent with the DCF estimates and far more
plausible than Mr. Gorman's much lower CAPM estimate that is based on
unadjusted historical data.
V. FLOTATION COST ADJUSTMENT
a. What are Mr. Gorman's concerns with your estimate of flotation costs?
A. Mr. Gorman asserts that the flotation cost adjustment for Intermountain Gas "is
not based on known and measurable costs for IGC".32 In addition, Mr. Gorman
believes I should have identified Intermountain Gas' actual and verifiable
flotation costs that are properly allocated to regulated operations, show the time
period over which these costs were incurred, and show how they have been
treated for ratemaking purposes in the past.33
a. How do you respond to these concerns?
A. Mr. Gorman mis-states the purpose of my flotation cost adjustment. He claims
that the "adjustment is intended to recover the cost a utility incurred" in the past
and he opposes it because it is not "based on IGC's actual and verifiable flotation
expenses."3a As I explained in my Direct Testimony:
A more important purpose of a flotation cost adjustment is to
establish a return that is sfficient to enable a company to attroct
capital onreasonoble terms. Thisfundamental requirement of afair
rate of return is analogous to the well-understood basic principle
Direct Testimony of Michael P. Gorman, at 69.
Ibid, at71.
Ibid, at70-71.
Gaske, Reb. 23
Intermountain Gas Company
l8
t9
20
2T
32
33
34
1
2
J
4
5
6
7
8
9
that afirm, or an individual, should maintain a good credit rating
even when they do not expect to be borrowing money in the near
future. Regardless ofwhether a company can confidently predict its
need to issue new common stock several years in advance, it should
be in a position to do so on reosonable terms at all times without
dilution ofthe bookvalue of the existing investors' common equity.3s
The primary purpose of the flotation cost adjustment is to be consistent with the
capital attraction standard which requires that the return be sufficient to enable the
company to raise capital on reasonable terms on a forwardJooking basis. In this
regard, it is similar to an insurance premium. A company is not required to show
that it has had accidents or catastrophes in the past in order to include an
insurance premium in its cost of service. Instead, the point of the insurance
premium is to ensure that the company can pay for future costs that may or may
not ever materialize. Mr. Gorman's suggestion that flotation costs can only be
recovered after the fact misses the entire point of the capital attraction standard.
Moreover, I am not aware of any ratemaking or regulatory accounting convention
that provides for the amortization and recovery of past flotation costs associated
with issuing common equity.
Please summarize your position regarding a flotation cost adjustment as it
relates to Mr. Gorman's testimony and your revised results.
For the reasons explained in my Direct Testimony, I continue to believe that a
flotation cost adjustment is reasonable in this case. However, to address part of
Gaske, Reb. 24
Intermountain Gas Company
10
11
t2
13
t4
l5
t6
t7
l8
te a.
2t A.
20
22
3s Direct Testimony of J. Stephen Gaske, page 17,lines 8-17
I
2
J
4
5
6
7
8
9
Mr. Gorman's concem, I calculated the actual flotation costs incurred by MDU
Resources Group as a result of its three most recent public offerings in 1998,
2002, arlLd2004. The average flotation cost for these three issuances (shown in
Exhibit No. 33, Schedule 3) was 3.6 percent, which is consistent with my flotation
cost adjustment of 4.0 percent.
a. What arguments does Mr. Rogers provide in opposition to your calculation of
A.
the flotation cost adjustment?
While Mr. Rogers agrees that recovery of flotation costs is appropriate for
Intermountain Gas,36 he takes issue with the method I have used to apply the
flotation cost adjustment to the return estimate for each proxy group company. In
particular, Mr. Rogers argues that the flotation cost adjustment should only apply
to the dividend yield component of the retum, not the entire retum estimate.3T
Is there support in academic literature for your approach, which multiplies the
entire retum by a specified factor to adjust for flotation costs?
Yes. Myron Gordon, who is credited with developing the constant growth DCF
model for estimating rate of retum, has stated that a regulatory agency should set
the allowed rate of return greater than the investor return requirement so as to
allow the firm to issue stock at a price that will yield net proceeds equal to book
value. Professor Gordon advocates the following adjustment:
10
11
t2
13 a.
t4
15 A.
t6
t7
18
t9
36
31
Direct Testimony of Mark Rogers, at 17.
Ibid, at 19-20.
Gaske, Reb. 25
Intermountain Gas Company
I
2
J
4
5
6
The agency need only estimate the proportion that the proceeds per
share on an issue bear to the price of the stock and adjust the allowed
rate of return so that the price per share is the indicated ratio of the
bookvalue per share. If the proceeds on an issue are 9I percent of
marlret price, the agency should maintain market price at about 110
percent of book value.38
In order to meet this requirement, the flotation cost adjustrnent must be applied to
the entire rate of retum. The flotation cost adjusftnent that I have proposed attempts
to meet the same standard.
IO \rI. RISK PREMITIM ANALYSIS
t2 A.
Please summarize Mr. Gorman's bond yield plus equity risk premium analysis.
In addition to his CAPM analysis, Mr. Gorman includes two additional Risk
Premium approaches to estimate Intermountain Gas' cost of equity.
Mr. Gorman's first approach calculates the annual risk premium for each year
from 1986 through September 20l6by taking the difference between regulatory
commission-authorized equity returns and long-term Treasury bond yields.3e
From thatdata, Mr. Gorman selected the 1987-1991 S-year average risk premium
of 4.17 percent as the low end of his range, and the 2012-2016 5-year average risk
premium of 6.68 percent as the high end. He then used a weighted average of
these two numbers to derive his risk premium of estimate of 6.10 percent.ao
However, if one wanted to know the most recently required risk premium, it
13
t4
15
t6
t7
l8
t9
2t
Myron J. Gordon, The Cost ofCapital to a Public Utility,Michigan State University, l974,pp 165-166.
Exhibit No. 312.
.25 x4.l7oh + .75 x 6.680/o:6.100/o
Gaske, Reb. 26
Intermountain Gas Company
7
8
9
11 a.
20
3E
39
40
1
2
J
4
5
6
7
8
9
10
11
t2
13
14 a.
l5
r6
t7 A.
18
r9
would be more accurate to simply use the 2012-2016 average of 6.68 percent.
When added to the U.S. Treasury bond yield of 3.4 percent used by Mr. Gorman,
that approach indicates a utility return requirement of 10.08 percent.al
Mr. Gorman's second approach calculates the average risk premium for the period
1986 through September 2016 as the difference between the average authorized
equity retums for natural gas distribution companies and the concurrent A-rated
utility bond yield. From that data Mr. Gorman again used the 1987-1991 5-year
average of 2.80 percent, and the 2012-2016 5-year average of 5.51 percent to
produce a weighted average of 4.8 percent.a2 Again, if one wants to know the most
recently required risk premium, it would make more sense to simply use the 2012-
2016averageof5.51 percent. Whenaddedtotheutilitybondyield of 4.33 percent
used by Mr. Gorman, that approach indicates a current utility return requirement of
9.84 percent.a3
What would have been the overall result if Mr. Gorman had limited his gas
utility risk premium calculations to the most recent five-year period instead
blending in another live-year period that is 25 years out of date?
At page 58, lines 20-21and page 64, Table 10 of his testimony, Mr. Gorman
indicates that the mid-point of his two gas utility risk premium is 9.3 percent,
which also is his recommended rate of return for Intermountain. However, if he
Direct Testimony of Michael P. Gorman, at 58, Iine 14.
0.25 x2.80o/o + 0.75 x 5.51o/o:4.80%.
Direct Testimony of Michael P. Gorman, at 55, line 18.
Gaske, Reb. 27
Intermountain Gas Company
4l
42
43
I
2
J
had employed the more reasonable approach of using only the most recent five-
years of risk premiums, the mid-point of his two risk premium analyses would
have been 9.96 percent:
Utility Risk Premi wn 2012-2Ol 644
Plus: Current Bond Yieldas
Required Gas Utility ROE
6.68%
3.40o/o
r0.08%9.96Vo
5.51%
4.33%
9.84%
4
5
6
7
8
9
l0
1l
12
13
a
A.
In other words, if Mr. Gorman had not inappropriately mixed 1988-1991 data into
his analysis, his gas utility risk premium analysis would have indicated a required
rate ofreturn of 10.0 percent.
Mr. Gorman suggests that your large and small company Risk Premium
analysis is flawed because it is based on the broad market for common stocks
and does not reflect the below market risk of Intermountain Gas and utility
operations in general.a6 How do you respond?
As discussed in my Direct Testimony, the purpose of my small company
historical Risk Premium analysis is to serve as a benchmark to assess the
Michael P. Gorman, Exhibit No. 312, Col.4, lines 3l and 34; and Exhibit No. 313, Col. 4, lines 31 and
34.
Direct Testimony of Michael P. Gorman, page 58, lines 14 and 18.
Direct Testimony of Michael P. Gorman, at 74.
Gaske, Reb. 28
Intermountain Gas Company
M
45
46
Table I
Mr. Gorman's Risk Premium Analysis Using Current Data
ROE based on
T-Bond Yield
Mid-
Point
ROE based on
Utility Bond
Yield
I
)
J
4
5
6
7
8
9
l0
11
t2
13
t4
15
t6
t7
18
t9
reasonableness of my DCF analysis and to place in context the Company's
requested ROE of 9.90 percent.aT The small company risk adder serves as a
useful indicator of the cost of capital for Intermountain Gas because a gas
distribution utility must offer potential retums that allow it to compete for equity
capital with other investments of comparable risk. I indicated that gas distribution
companies generally have lower risks than the average of all small publicly-traded
companies. However, the significant average risk premiums earned by small
companies are informative, and provide some relevant context for the authorized
return for lntermountain Gas' Idaho gas distribution operations. Therefore, I
believe this information is relevant for purposes of demonstrating the
reasonableness of my recommended rate of retum. I have not used my small
company Risk Premium analysis to establish the recommended cost of common
equity capital for the Company, but only as a general benchmark to corroborate
the reasonableness of my DCF results.
In addition to your historical benchmark risk premiums, did you calculate a
current risk premium specific to natural gas utilities?
Yes.
Do you agree with Mr. Gorman that gas utility risk premiums are not inversely
related to bond yields?
Gaske, Reb. 29
Intermountain Gas Company
a.
A.
a.
47 Direct Testimony of J. Stephen Gaske, at 39.
1A.No, I do not. According to Mr. Gorman, academic research does not support the
"simplistic inverse relationship between equity risk premiums and interest rates"48
that are contained in my Risk Premium analysis. Mr. Gorman argues that
"researchers have found that the relationship changes over time and is influenced
by changes in perception of the risk of bond investments relative to equity
investments, and not simply changes in interest rates."4e However, the Risk
Premium analysis in ExhibitNo. 5, Schedule 5 of my Direct Testimony
demonstrates that there is in fact an inverse relationship between natural gas
utility equity risk premia and interest rates.
Using data similar to the analysis in Mr. Gorman's Exhibit Nos. 312 and 313, my
regression produced the following relationship:
Intercept + Coefficient x Bond Yield: Utility Risk Premium
0.0845 - 0.5632 x Bond Yield = Utility Risk Premium
The regression statistics indicate that this equation is statistically significant and
the R-square reveals that more than79 percent of the variation in the risk
premium is explained by the bond yield. The negative coeffrcient in the above
equation demonstrates the inverse relationship between bond yields and the
natural gas utility risk premium. For every change of 100 basis points in the bond
yield, the natural gas utility risk premium changes by approximately 56 basis
Direct Testimony of Michael P. Gorman, at 75.
rbid.
Gaske, Reb. 30
Intermountain Gas Company
2
J
4
5
6
7
8
9
l0
11
t2
l3
t4
15
t6
t7
18
t9
48
49
I
2
points in the opposite direction. Thus, Mr. Gorman's observations clearly do not
apply to natural gas utilities.
VII. MARKET DCF ANALYSIS
a. How do you respond to Mr. Gorman's concer:ns regarding your Market DCF
Analysis?
A. Mr. Gorman has concems with the analysts' projected growth rates for the S&P
500 companies that I use in my Market DCF analysis because those growth rate
are above the average projected growth rate in the U.S. economy of 4.25.s0
However, Mr. Gorman estimates that the long-term "sustainable" growth rate of
the proxy companies is 6.55 percent.5l Thus, there is no reason to think that
investors' growth expectations for specific companies is limited to the projected
growth in the economy.
Moreover, my current Market DCF rate of retum estimate indicated by analysts'
$owth rate projections is 12.1 percent, which is very close to the 12.0 percent
long-term average return eamed by large-company common stocks during the
period 1926-2016.52 Thus, a cunent market DCF rate of retum estimate that is
virtually identical to the average return achieved during the past 90 years is
clearly sustainable in the long run. Like my Risk Premium analysis, the purpose
of my Market DCF analysis is to serve as a benchmark to assess the
Direct Testimony of Michael P. Gorman, at 77.
Gorman Exhibit No. 307 , page l, line 8, col. I I .
See Direct Testimony of Michael P. Gorman, page 61, lne22.
Gaske, Reb. 31
Intermountain Gas Company
J
4
5
6
7
8
9
10
l1
t2
13
t4
l5
t6
t7
l8
19
50
5l
52
1
2
aJ
4
5
6
7
8
9
reasonableness of my DCF analysis and provide context for my recommended
ROE of 9.90 percent and to estimate a current market risk premium for my
CAPM analysis.s3
As noted earlier, analysts' eamings growth rate forecasts are a superior measure of
the long-term growth rate expectations that are reflected in stock prices. My
approach to conducting a Market DCF is virtually identical to one adopted by the
Federal Regulatory Energy Commission ("FERC') in a recent order. In response to
arguments similarto those proffered by Mr. Gorman inthis proceeding, the FERC
concluded:
We are also unpersuaded that the growth rate projection in the
NETOs' CAPM study was skewed by the NETOs' reliance on analysts'
projections of non-utility companies' medium-term earnings growth,
or that the studyfailed to consider that those analysts' estimates reflect
unsustainable short-term stock repurchase programs and are not
long-term projections. As explained above, the NETOs based their
growth rate input on datafrom IBES, which the Commission has found
to be a reliable source of such data. Thus, the time periods used for
the growth rate projections in the NETOs'CAPM study are the time
periods over which IBES forecasts earnings growth. Petitioners'
arguments against the time period on which the NETOs' CAPM
analysis is based are, in ffict, arguments that IBES data ore
insuficient in a CAPM study.sa
Thus, the FERC did not agree with the argument that analysts' projections for the
S&P 500 are unsustainable and not reliable for estimating the cost of capital for a
broad-based market index.
Direct Testimony of J. Stephen Gaske, at 39.
150 FERC'!J61,165, Docket Nos. ELll-66-001 Opinion No. 531-B, para. ll2.
Gaske, Reb. 32
Intermountain Gas Company
10il
t2
l3
t4
l5
l6
t7
l8
19
20
2t
22
23
24
25
53
54
I Q. On page 78 of his testimony, Mr. Gorman argues that the S&P 500 companies I
2 use in my Market DCX'analysis have risk characteristics that are significantly
3 different than the risks encountered by Intermountain Gas and its parent
4 company. What is your response?
5 A. I agree that those companies have different risks, which is why my recommended
6 rate of return for Intermountain Gas' Idatro gas distribution operations of 9.9
7 percent is significantly less than the 12.1 percent DCF rate of return estimated for
8 the market as a whole.ss Moreover, as shown earlier, if one were to use a CAPM
9 beta to adjust for the differences in risk, the result is an indicated rate of return of
l0 9.7 percent for the average proxy company.56
I I YI[. CAPITAL STRUCTURE
12 a. At pages 3436 of his testimony Mr. Goman recommends reducing
13 Intermountain Gas'ratemaking capital strucfure from 50.0 percent common
14 equity to 48.0 percent Is this change appropriate?
15 A. No. Mr. Gorman provides two flawed reasons for his recommendation. First, he
16 argues that "for the period ending December 31,2015" Intermountain's common
17 equity ratio was approximately 48 percent.sT However, Mr. Gorman ignores the
l8 fact that Intermountain's more recent quarterly common equity ratio at the time of
19 the rate filing was approximately 52 percent (i.e., 51.85%). As discussed in the
Schedule 6, Direct Testimony Exhibit No 5.
Direct Testimony of J. Stephen Gaske, at 29.
Direct Testimony of Michael P. Gorman, p. 34, lines 14-17
Gaske, Reb. 33
Intermountain Gas Company
55
56
57
I
2
3
4
5
6
Prepared Direct Testimony of Intermountain witness Mr. Mark Chiles,
Intermountain attempts to maintain a target common equity ratio of 50.0 percent
and that, although the common equrty ratio varies from time to time, it has
generally been maintained slightly above 50.0 percent in recent years. Table C2
on page 3 of Mr. Chiles' testimony shows the following common equity ratio
history:
Table2
Intermountain Gas Company
Common Equiry Ratio
t2t3U20t3 54.27%
t2l3U20t4 s2.40%
t2t3U20ts 47.95%
613012016 51.85%
Mr. Gorman's use of the lowest number in recent years, December 2015, which
was not even the most recent value, is not indicative of a current or expected
common equity ratio for Intermountain.
Second, Mr. Gorman argues that the "proxy group has an average common equity
ratio of 48.0% (including short-term debt)..." However, he also observes that the
proxy companies have an average common equity ratio of ". .. 53.6% (excluding
short-term debt) ..."s8 It is appropriate to include only long-term debt in the
capital structure because that matches the long-term nature of the rate base.
Gaske, Reb. 34
Intermountain Gas Company
7
8
9
t0
11
t2
l3
t4
58 Direct Testimony of Michael P. Gorman, p. 39, lines 16-18.
I Intermountain's filed equity ratio of 50.0 percent is comfortably within the range
of equity ratios of the proxy companies, and as Ms. Carlock testifies "is
reasonable based on the analysis ofhistorical, current and projected capital
structures for Intermountain Gas and the proxy group."5e As shown in my Direct
Testimony Exhibit No. 5, Schedule 8, the proxy company coillmon equity ratios
are in a range between 47.5 percent and 58.9 percent with a median equity ratio of
54.33 percent. Six of the seven proxy companies have common equity ratios
greater than the 50.00 percent level filed by Intermountain Gas. Thus, the
Company's common equity ratio is neither unusual nor extreme and should be
adopted by the Commission.
What effect does the capital strucfure have on the costs of doing business?
Most large companies are financed using a mix of debt and equity capital.
Including a reasonable amount of debt in the capital structure can provide a low-
cost source of frrnds because the common equity holders shield lenders from a
portion of the risks of the company. However, the requirement to pay a fixed
level of interest and repay principal as scheduled, causes the possibility of
bankruptcy or other financial distress to increase as the firm takes on more debt.
Financial 'oleverage" provided by fixed debt payments also tends to translate
relatively small fluctuations in a company's operating income into much larger
variations in the net income available to common stockholders. When the
Gaske, Reb. 35
Intermountain Gas Company
2
3
4
5
6
7
8
9
10
ll a.
t2 A.
t3
t4
l5
t6
t7
l8
l9
20
59 Direct Testimony of Terri Carloch at 8.
1
2
J
4
sQ.
6
7
8A.
9
l0
1l
t2 a.
l3
t4 A.
15
t6
t7
t8
t9
20
2t
22
proportion of debt is increased beyond some level, both the lenders and the
stockholders require greater rates of refurn on their investments to compensate for
the greater risks involved. In financial theory, there is an optimal range of equity
ratios that minimizes the overall cost of capital of a company.
Is it common for commissions to adjust the ratemaking capital structure when
the capital structure is nomal in comparison with companies that have similar
risks?
No. Because there are numerous factors that go into establishing a company's
capital structure, the common approach is for regulators to recognize that, unless
the capital structure is extreme, the appropriate capital structure is a matter for
management discretion and judgment.
What factors are important for determining the appropriate capital structure
for a company?
The amount of debt that is economical for a firm depends on its business risks and
the perceived probability that it could experience unexpected difficulties that
would render it unable to meet its debt obligations. Although firms in the same
industry generally tend to have similar business risks, there is often a general,
very broad, range of equity ratios associated with companies in particular
industries. Firms in the same industry have different capital structures for many
reasons. For example, within a given industry, there may be wide differences in
the vintages of capital and operating strategies of individual companies. Another
important factor is the quality of a firm's earnings in terms of cash flow and
Gaske, Reb. 36
Intermountain Gas Company
I
2
J
4
5
6
7
8
9
l0
l1
t2
13
t4
l5
t6
t7
l8
t9
20
2t
22
continuing operations. When all factors are considered the managers of a
company are usually in the best position to evaluate the prospective risks and
operating needs of their company and determine the most appropriate capital
structure.
a. In addition to individual differences in business risks, are there other important
factors that can determine the appropriate capital structure for a company?
A. Yes. Another important factor is the transaction cost of raising new capital. In
order to borrow funds from outside sources, a company typically pays issuance
costs that are close to one percent of the amount borrowed. In contrast, raising
new cofllmon equity funds from outside sources generally involves flotation costs
that are 3-5 percent of the amount of capital raised. In addition, on a percentage
basis, flotation and issuance costs generally are proportionately lower for larger
issues. Consequently, there often is a "pecking order," whereby firms attempt to
raise as much new capital as possible from intemally-generated retained earnings
and issue debt only when intemal funds are not sufficient to finance attractive
projects and maintain the desired dividend levels. The higher flotation costs
associated with raising equity capital from extemal sources means that, up to a
point, it is less expensive to issue debt for as much external financing as possible
before tuming to the external equity markets.
Different companies also have different patterns of needs for financing. A
company might take on large amounts of debt to finance new projects, but then
pay down its debt and increase its equity ratio over time after the project is in
Gaske, Reb. 37
Intermountain Gas Company
I
2
J
4
5
6
7
8
9
10
ll
t2
13
t4
l5
l6
t7
l8
t9
20
2t
22
a.
A.
service. When a company's debt ratio is high, its financial flexibility is restricted.
This means that its ability to undertake additional projects is limited, and it may
not be able to refinance its debt or raise new capital if adverse circumstances
arise.
Thus, when one considers financing costs and the often uneven pattem of capital
investments, there may be times when achieving the target capital structure may
not be as desirable as minimizing the issuance costs that the firm incurs as it
operates on a dynamic basis. A well-managed company might reasonably
maintain a relatively high equity ratio for extended periods of time and then
undertake alarge amount of additional debt to finance a new project. The
important point is that wide differences in capital structures exist within any given
industry from time to time and a determination of the "appropriate" capital
structure for a particular company should not be made in a vacuum which ignores
that company's unique history, business needs and circumstances.
At pages 2l and35 of his testimony, Mr. Gorman cites a settlement involving
Cascade Natural Gas Company as support for his capital structure and return
on equity recommendations. Does a settlement provide a valid precedent?
No. It is well established that settlements cannot be relied upon to support a
position regarding any particular element of the settlement because there are
likely to be multiple tradeoffs and considerations involved in reaching a
settlement. In fact, section 13 on page 9 of the Cascade settlement specifically
states:
Gaske, Reb. 38
Intermountain Gas Company
I
2
3
4
5
"No Stipulating Party shall be deemed to have agreed that any
provision of this Stipulation is appropriate for resolving issues in
any other proceeding ... "
Thus, it is inappropriate for Mr. Gorman to cite that settlement as a precedent in
an Intermountain proceeding or any other proceeding.
IX.SUMMARY
What does your analysis of Mr. Gorman's cost of capital testimony indicate?
As discussed earlier, Mr. Gorman's recommendation to reduce the common
equity ratio from 50 percent to 48 percent ignores Intermountain's demonstrated
adherence to a 50-50 target capital structure. Moreover, Intermountain Gas'
proposed equity ratio of 50.0 percent is below the median equity ratio for
companies in the proxy group and is toward the lower end of the range of equity
ratios. As such, the proposed equity ratio of 50.0 percent is reasonable for
ratemaking pu{poses, but is also a source of above-average financial risk.
Mr. Gorman and I both agree that Intermountain is riskier than the proxy
company group, however, his recommended rate of retum fails to adequately
reflect Intermountain's greater risk.60
In reviewing his cost of common equity analyses, I identified several flaws that,
when corrected, indicate that my proposed return on equity is reasonable. First,
his analysis of gas utility risk premiums mixed current data with data that is 25
years out of date. When only the current data is used his gas utility risk premium
60 Direct Testimony of Michael P. Gorman, page 39,line 14-15
Gaske, Reb. 39
Intermountain Gas Company
6
7
8
9
a.
A.
10
11
t2
13
t4
15
t6
t7
18
t9
2T
20
I
2
J
4
5
6
7
8
9
l0
11
t2
l3
t4
l5
t6
t7
18
t9
20
2l
22
a.
A.
analysis indicates a current required return on equity of 10.0 percent for natural
gas utility companies. Second, the core factor in Mr. Gorman's CAPM analysis is
based on average historical retums from 1926-2015. When a proper and
reasonable current market risk premium is used, his CAPM analysis would
suggest a median rate of return of 9.7 percent for the proxy companies. Third,
his multi-stage DCF analysis assumes that investors' expect the growth rate of all
proxy companies to decline to the average growth rate in the economy within a
relatively short time period. However, it is far more reasonable to expect that
after an initial time period, the analysts' projected growth rates will at least equal
the eamings retention growth rates, which Mr. Gorman calls "sustainable"
growth. When sustainable growth is used in his multi-stage DCF instead of U.S.
GDP growth, the median proxy company ROE is 9.74 percent. Considering
these corrections and his other analyses in the context of the gxeater risk of
Intermountain, Mr. Gorman's analyses confirm the reasonableness of my 9.9
percent recommended return on common equity.
What does your analysis of Staffs cost of capital evidence indicate?
Staffand I both agree that Intermountain's proposed common equity ratio is
reasonable. However, Staff s cost of common equity analysis is based solely on
DCF analyses and fails to consider other estimation methods which indicate that
the cost of common equity capital for the proxy companies currently is greater
than the DCF analyses would suggest. Moreover, Staff s analysis fails to
adequately consider Intermountain's greater risks relative to those of the proxy
Gaske, Reb. 40
Intermountain Gas Company
1 companies. For example, two of Staff s seven proxy companies have Basic DCF
rates of return significantly above 10.0 percent, which places my 9.9 percent
recommended return on common equity clearly within azone of reasonableness.
Please summarize the conclusions of your Rebuttal Testimony.
The refurns on common equity recommended by Staff and Mr. Gorman are
inadequate to meet the tests of a reasonable rate of return because they do not
consider the relative business and financial risks of Intermountain Gas compared
with the proxy group companies, and because they do not take into account the
expectations for higher interest rates in the near future.
Does this conclude your Prepared Rebuttal Testimony?
Yes.
Gaske, Reb. 4l
Intermountain Gas Company
2
J
4
5
6
7
8
9
10
ll
a.
A.
a.
A.