INT Gelhaus Direct.pdf

I
I
I
I
I
I
I
I
I 1
I
2
3
I 4
I
5
6
I 7
I
8
9
I 10
I
11
12
I 13
I
14
15
I 16
I
17
18
I
I
BEFORE THE IDAHO PUBLIC UTILITIES COMMISSION
INTERMOUNTAIN GAS COMPANY )
)
CASE NO. U-1034-88
Prepared Di rect Testimony
of
LARRY F. GELHAUS
Q. Please state your name and business address.
A. My name isLa rry F. Gel haus, and my busi ness address is One Market
Plaza, san Francisco, California.
Q. What is your professi on?
A. I am a certified public accountant and a manager in the
international firm of Arthur Andersen & Co., independent public
accountants.
Q. What is the nature of the work performed by Arthur An dersen& Co.?
A. Arthur Andersen & Co. is an international independent public
accounti ng fi rm wi th 140 offi ces located in 42 countri es throughout
the world, of which 64 are in the United States. We audit all types
of businesses, both regulated ~nd non-regulated; and, in the public
utility field, we audit approximately one-third of the electric and
gas compani es in theUni ted States, a substanti al porti on of the
independent telephone industry and most of the major gas
transmission companies. In addition to audits, we perform many
other services for utility clients such as assisting in the
preparati on of rate case data, install i ng property records systems,
I
I
I
1
I 2
I 3
4
I 5
I 6
7
I 8
I 9
10
I 11
I 12
13
I 14
I 15
I
16
17
I 18
I
19
20
I 21
I 22
23
I 24
I 25
26
I
I
-2-
performi ng depreci ati on studi es and ass i sti ng in securi ti es
regi strati ons.
Q. Mr. Gelhaus, what is your educational and professional background?
A. I graduated from Stanford University in 1966 with a Bachelor of Arts
degree in economics. I then attended the Stanford University
Graduate School of Business, where I received a Master of Business
Administration degree in 1968, with finance as my field of
concentration. I taught various courses in business administration
as an Assi stant Professor of Management at Eastern Washi ngton State
College until 1971, when I joined Arthur Andersen & Co. I became a
manager in the Regul ated Industri es Di vi s i on of our audi t practi ce
in 1976, and have specialized in both audit and consulting work for
a wide range of utility companies since joining the firm. My audit
experi ence i ncl udes servi ng as manager on several uti 1 ity company
audi t engagements on the West Coast, and as an instructor in our
fi rm-wi de regul ated industry trai ni ng program. My uti 1 i ty company
consul ti ng experi ence i ncl udes property record system revi ews and
installations, review of a cost study methodology for AT&Tt review
of the rates department for a medium-sized utility, and rate. case
work. The rate case work includes supervising working capital lead-
lag studies and preparing testimony on proper working capital
allowances. It also includes supervising work to substantiate other
rate base items, including plant in service, and preparing testimony
re 1 ated to such work. In addi ti on, I was i nvo 1 ved in the
development of a presentation on the working capital component of
rate base, whi ch descri bed procedures for conducti ng a 1 ead-l ag
I
I
I
I
1
2
I 3
I
4
5
I 6
I
7
8
I 9
I
10
11
I 12
I 13
14
I 15
I 16
17
I 18
I 19
20
I 21
I 22
23
I 24
I 25
26
I
I
-3-
study, and participated in the initial presentation before a
committee of the Edison Electric Institute. I am a certified public
accountant and a member of both the Cali forni a CPA Soci ety and the
American Institute of Certified Public Accountants.
Q. What is your firm1s relationship with Intermountain Gas Company?
A. We have served as independent audi tors of the Company for many
years.
Q. Woul d you please describe the subject matter of your testimony in
thi s proceedi ng?
A. Yes. We were asked by the Company to assist it in performing a
1 ead-l ag study to determine the amount of i nvestor-provi ded worki ng
capital which should be considered in determining rate base in this
proceeding. I will describe both the results of this study and the
study itself which was performed under my supervision.
Q. Please define the working capital concept as it relates to a
regulated utility company.
A. The working capital allowance, in the sense that it applies to a
regulated utility company, can be described as the economic input of
capital in excess of that used to finance net utility plant in order
to operate the utility business. This capital is necessary (l) to
cover the time period, or IIlagll, from the time that the Company
incurs costs of providing service to the time that it is paid for
that service by its customers, and (2) to cover any deferred cost,
exclusive of net utility plant, which must be financed by investors
pendi ng recovery of such cost from customers. The items fi nanced by
investors i ncl ude accounts recei vabl e, i ncl udi ng recei vabl~s for
I
I
I
1
I 2
I 3
4
I 5
I 6
7
I 8
I 9
10
I 11
I 12
13
I 14
I 15
16
I 17
I 18
I
19
20
I 21
I
22
23
I 24
I 25
26
I
I
-4-
servi ces rendered but not yet bi 11 ed, i nventori es of materi al s,
supplies and fuels that must be maintained for purposes of
permi tti ng effi ci ent operati ons, and current and past expendi tures
which will be collected in the future. In addition, any
compensating cash balances which must be maintained should be
included in the working capital allowance if the investors are not
compensated through rate of return or otherwi se.
Q. Is thi s di fferent from the defi ni ti on of working capi ta 1 as it is
used in the ordi nary busi ness or accounti ng sense?
A. Yes. The IIworking capital al1owancell, as the term is used in the
utility industry, must be differentiated from the term IIworking
capitalll, as it is used in the ordinary business or accounting sense
of current assets less current liabilities. IIWorking capital
II as
applied to commercial enterprises and as used by financial
institutions and others in evaluating the financial standing of
commerci a 1 enterpri ses means current assets 1 ess current 1 i abi 1 i ti es
whi ch may be expressed ei ther as a quanti ty or as a rati o. The
concept is used there for purposes of evaluating the liquidity of a
commercial business at a point in time.
Further, although the cost of financing the excess of
current assets over current liabilities must be recovered in the
aggregate sales prices if the enterprise is to remain viable, this
cost is not ordinarily used as a direct determinant of sales prices
for goods and servi ces of a commerci a 1 enterpri see
On the other hand, since a utility's sale price is set b~
the Commission rather than by competition, the cost of the capital
I
I
I
1
I 2
I 3
4
I 5
I 6
7
I 8
I 9
10
I 11
I 12
13
I 14
I 15
I
16
17
I 18
I
19
20
I 21
I
22
23
I 24
I
25
26
I 27
I
-5-
needed to finance the utility's working capital must be included in
the revenue requi rement if the investor is to be compensated for all
the capital he has devoted to the business. As I stated, the
working capital allowance as used in utility ratemaking is the
investors' capital devoted to unrecovered costs other than net
utility plant, and the cost of financing this capital to be included
in the revenue requirement is computed by applying the rate of
return to the computed working capi ta 1 allowance.
Whi 1 e total working capi ta 1 is of concern to all
commerci al enterpri ses, the uti 1 ity ratemaki ng process must focus
its attenti on di rectly on the measurement of the amount of i nvestor-
financed working capital. It is this objective -- measuring
investor-financed working capital -- which makes the utility working
capital computation more complex than the mechanical computation
generally associated with defining the working capital of commercial
enterpri ses.
Q. Is your definition of utility working capital the same as cash
working capital?
A. It is not the same if, as is typically the case, cash working
capital means a limited concept which fails to consider the
investment requi red in capi tal cost items whi ch may not
contemporaneously resul tin cash transacti ons . Eventually, however,
all the utility's revenue requirements involve transactions which
relate to cash -- the cash investment by equity and debt holders,
the cash expenditures for utility plant as well as for current
operating expenses, interest and dividends, and the collection of
cash from the customers.
I
I
I
1
I 2
I 3
4
I 5
I 6
7
I 8
I 9
10
I 11
I U
I
13
14
I 15
I
16
17
I 18
I
19
20
I 21
I 22
23
I 24
I 25
26
I
I
-6-
In simple terms, the investors must supply the capital to
finance the receivables from customers from the date the service is
rendered to the date of payment. Additional investment is needed to
fi nance inventories and other costs that are to be coll ected out of
future revenues. Further, if any compensati ng bank balance
requi rements must be mai ntai ned, the amount of these balances must
be added to the working capital allowance if the cost to finance
these balances i s not refl ected in rate of return.
The investor capital needed to finance receivables,
inventories, deferrals and compensating bank balances can be offset
to the extent that the Company can properly del ay payments for
labor, goods, servi ces and certai n taxes beyond the date that such
labor, goods, and taxes are utilized in providing service to the
customer.
Q. Could you please explain how a utility company's working capital
allowance can be determi ned?
A. There are several methods that are used. No method will produce an
allowance that is precisely correct as a measure of the company's
working capital. The purpose of the determination should be to
arrive at an amount that is reasonable, does not contain obvious
defects and whi ch is not so time-consumi ng to compute as to offset
the benefits of arriving at what may be perceived as a more accurate
determination.
A number of juri sdi cti ons i ncl udi ng the Company and
Commi ssi on staff in previ ous proceedi ngs have used an arbi trary
formul a (often 45 days of operati ons and mai ntenance expenses wi th
I
I
I
1
I 2
I 3
4
I 5
I 6
7
I 8
I 9
I
10
11
I 12
I
13
14
I 15
I 16
17
I 18
I 19
20
I 21
I 22
23
I 24
I 25
26
I
I
-7 -
or without certain offsets) to determine the working capital
allowance. This method has the benefit of ease of computation, but
it is not tailored to the specific experience of individual
companies and does not consider all working capital components.
Many have recognized these deficiencies and are using, or are
currently consi deri ng usi ng, III ead-l agll studi es to determi ne working
capital requirements (this includes the Federal Energy Regulatory
Commission, long a proponent of the formula method, which has
recently proposed to incorporate 1 ead-l ag methodology in determi ni ng
work i ng capi tal allowances).
A third approach is commonly referred to as the IIbalance
sheetll method. Thi s approach has 1 ess acceptance because it cannot
be appl ied in a conceptually sound manner without a great deal of
additional analysis and study.
Q. Would you explain the formula method and its application?
A. The formul a method seeks to measure the funds whi ch must be invested
in the uti 1 i ty company between the time that operati on and
mai ntenance expenses are incurred and the repl eni shment of these
funds is recei ved from customers. In practi ce, thi s formul a
generally involves the inclusion in rate base of one-eighth of the
annual revenue requi rements, after pro forma adjustments, for
operati on and mai ntenance expenses. Addi ti ona 1 amounts are added to
the rate base for materi a 1 s, supp 1 i es and gas i nventori es. The one-
ei ghth of the year approximates 45 days. The 45 days is an
arbitrary figure, intended to represent the net period of time
between when payments are made to employees and vendors for servi ce
I
I
I
I
1
2
I 3
I
4
5
I 6
I
7
S
I 9
I 10
11
I 12
I 13
14
I 15
I 16
17
I is
I 19
20
I 21
I 22
23
I 24
I 25
26
I
I
-s-
provi ded to customers and payments for that servi ce are recei ved
from customers. The formul a method is rel ati vely simpl e and can be
readi ly adjusted for the effect of pro forma adj ustments. However,
it is not a tai lored approach -- that is, it is not based upon
speci fi c ana lyses of a company' s experi ence and does not cons i der
the del ay in the coll ecti on from customers of costs of servi ce other
than operation and maintenance expenses (e.g., depreciation, current
and deferred income taxes, and capital costs).
Q. You menti oned that offsets are sometimes app 1 i ed to the resul ts
obtai ned from the formul a method. Woul d you elaborate?
A. Yes. If it is thought, because a particular prepaid or accrued
expense i tern has an unusual payment pattern, that the answer given
by the formula method significantly misrepresents reality, an
adjustment will be made for the particular item.
Q. Were offsets employed in the Company's last general rate proceeding
before the Commi ss ion?
A. Yes. The Commission accepted the staff recommendation to reduce the
cash allowance obtai ned from the formul a by both average accrued
franchi se taxes and average accrued property taxes, on the basi s
that the Company coll ects these taxes before they are pai d to the
government.
Q. Was this offsetting technique proper?
A. No. Thi s technique assumes that the Company has the use of the
amounts rel ati ng to franchi se taxes and property taxes from the date
that the cost is incurred (i .e., the date that service is rendered
to the customer) unti 1 the payments for such i terns are made. These
I
I
I 1
I 2
I 3
4
I 5
I 6
7
I 8
I 9
10
I 11
I 12
13
I 14
I 15
16
I 17
I 18
19
I 20
I 21
22
I 23
I 24
25
I 26
I
I
-9-
amounts woul d be avai 1 abl e only if there were no 1 ag in revenue
coll ecti on, whi ch woul d mean that the Company recei ved payment for
servi ce on the same date that the servi ce was rendered. Thi sis, of
course, not the case.
Q. Are there other flaws in the practice of applying offsets to the
formul a?
A. Yes. The selection of offsets can become an exercise whereby either
side (Company or staff) chooses to use only those items which
further its view or position. This merely adds further distortion
to what is already an arbitrary result obtained from the formula.
For example, it could be argued that to the extent the Company has
prepai d or deferred costs such as rate case expenses, whi ch the
Company has already paid, but which will not be recovered in
revenues for several years, an offset is justified which adds an
amount to working capi ta 1 .
Q. You referred to the II bal ance sheetll method of determi ni ng the
working capital allowance. Would you explain that method and its
advantages and di sadvantages?
A. The balance sheet method can be descri bed generally as a co~putati on
where the balances per books of current assets, prepayments, and
deferred charges, less noninterest-bearing current liabilities, are
totaled. This approach superficially appears to be simple and to
yi el d a resul t of equi val ent accuracy to those obtai ned by other
methods. However, there are several significant problems, both
conceptual and practical, which arise when one attempts to utilize
thi s method to compute the working capi ta 1 allowance. .
I
I
I
1
I 2
I 3
4
I 5
I 6
7
I 8
I 9
10
I 11
I 12
13
I 14
I 15
I
16
17
I 18
I
19
20
I 21
I
22
23
I 24
I 25
26
I
I
-10-
First, by its nature, this method is historically based --
it does not consi der the effects of the pro forma adjustments
refl ected in revenues and expenses of the test year. In most cases,
this by itself significantly limits the usefulness of this method.
Second, the balance sheet method generally relies upon
reported end-of-period (or end-of-month) balances in the various
balance sheet accounts. If either average of month-end bal ances or
the end of the test period balances are used, there is no reflection
of the significant fluctuations in the balances which regularly
occur wi thi n each month. Thi s shortcomi ng coul d presumably be
offset by deriving and computing daily balances for each balance
sheet account. However, this refinement, even if it could be
adopted, woul d s i gni fi cantly increase the amount of effort and
related cost involved and void the apparent benefit of simplicity.
Third, the amounts reported monthly on a company's balance
sheet generally include balances related to non-rate base, non-
recurring or non-utility activities. Identification of the balances
related to such activities could be, in many cases, a formidable
task. An example would be accounts payable and accruals related to
constructi on work in process excl uded from the rate base.
In summary, the perceived simplicity of the balance sheet
method is obtai ned only at the cost of accuracy. The si gni fi cant
practical problems, which arise when one attempts to obtain
reasonable accuracy, remove entirely any aspect of simplicity.
Q. Pl ease descri be the 1 ead-l ag method.
A. The lead-lag method involves a study of the time lag between the
I
I
I
1
I 2
I 3
4
I 5
I 6
7
I 8
I 9
10
I 11
I 12
13
I 14
I 15
16
I 17
I 18
19
I 20
I 21
I
22
23
I 24
I 25
26
I
I
-11-
date when the customer recei ves servi ce from the company and the
date when the customer pays for such service, and the lead times
between the date that employees and suppl i ers render servicé to the
company and the date that the company is requi red to pay for such
service.
Once these 1 ags and 1 eads are determi ned, they are
applied, as appropriate, to each revenue and expense item in the
revenue requi rement. A major advantage of thi s approach over others
is that it is based upon specific analyses of the experience of a
company and is appl i ed to the speci fic accounts of the company.
Q. Does the 1 ead-l ag method correct the defi ci enci es of the formul a and
offsets approach?
A. If properly applied, it does. It provides an analysis specifically
tailored to the accounts and experience of a particular company,
which should yield the most accurate result. By including all
components of the revenue requi rement, wei ght can be gi ven to the
lag in recovery of capital costs, which is proper, and any need for
offsets is eliminated, thereby eliminating the possibility of the
mi suse or improper appl i cati on of such offsets whi ch is so often the
case.
Q. Does the 1 ead-l ag method correct the defi ci enci es of the balance
sheet method whi ch you i denti fi ed?
A. If properly applied, it does. In the first place, all allowed pro
forma adjustments are factored in, and a better match occurs because
only the revenue and cost of servi ce uti 1 i zed in the rate case are
used to measure investor capi tal requi rement. Si nce an annual i zed
I
I
I
1
I 2
I 3
4
I 5
I 6
7
I 8
I 9
10
I 11
I 12
13
I 14
I 15
16
I 17
I 18
I
19
20
I 21
I
22
23
I 24
I 25
26
I
I
-12-
revenue requi rement is used, there is no problem of month-end
cutoffs; and since only jurisdictional cost of service amounts are
util ized, non-util ity activities are effectively excl uded. .
Q. You mention lIif properly appliedll as a qualifier with respect to the
1 ead-l ag method. Coul d you expl ai n?
A. Yes. The results of lead-lag studies are oftentimes limited in
their application to operation and maintenance expenses. When the
lead-lag study results are used in this way, working capital
requi rements are understated because no wei ght is gi ven to the 1 ag
in recovery of capital costs -- depreciation and property taxes on
plant, the common equity return and related income taxes which
shoul d be earned, the preferred stock cost and the interest cost.
These capi ta 1 costs may consti tute a s i gni fi cant porti on of the
tota 1 revenue requi rement.
Q. Which method did you use in your study to determine the amount of
investor-owned working capital which should be included in rate base
to determi ne an adequate revenue requi rement?
A. We used a lead-lag study method to determine that portion of the
working capital allowance which relates to the investor capital
requi red to fi nance the net 1 ag in recovery of all of the cost of
providing service to the Company's customers. The results reflect
those net (lead) lag days specifically related to the Company, due
to its customer payment practi ces, employee and vendor payment
practi ces and other appropri ate economi c real i ties of its operati ng
environment. By considering all components of the Company's revenue
requirement, including return on invested capital, the results are
I
I
I
1
I 2
I 3
4
I 5
I 6
7
I 8
I 9
10
I 11
I 12
I
13
14
I 15
I 16
17
I 18
I
19
20
I 21
I 22
23
I 24
I 25
26
I
I
-13-
not affected by arbitrarily including or excluding certain costs.
We call thi s 1 ead-l ag study method the IIRevenue Requi rementll 1 ead-
1 ag method. Uti 1 i zati on of thi s approach provi des a consi stent
benchmark for the Company and the Commi ssi on, both in thi s
proceeding and for the future.
Q. Pl ease summari ze the resul ts of your working capi ta 1 study.
A. The total working capital component of rate base is shown on
Exhibit 102, Schedule 2, Page 1. The total working capital
component on thi s Exhi bi tis spl i t between (1) the estimated amount
of the capital needed to finance the net (lead) lag in the recovery
of expenses and costs of provi di ng uti 1 i ty servi ce (Line 9) and (2)
capital invested in assets other than net utility plant required to
permit efficient operation (Lines 7 and 8).
Q. How was the amount of capital required to finance the net (lead) lag
in recovery of expenses and costs of providing utility service as
shown on Line 9 determi ned?
A. We performed a lead-lag study, which is summarized on Exhibit 104.
Thi s study was performed under my supervi si on and di recti on. As
shown on this Exhibit, applying the days determined by our study
(Column (c)) to the test period revenues, expenses and costs
i ndi cates that the number of net 1 ag days in recovery of the
expenses and costs is zero. Thi s number of days, 0% when converted
to a percent of a year, is appl ied to the adjusted test year revenue
requirement of $158,809,555 to determine the amount of working
capital the investors must provide to cover all costs of service not
yet coll ected from customers. In thi s case, the amount is zero
I
I
I
1
I 2
I 3
4
I 5
I 6
7
I 8
I 9
10
I 11
I 12
I
13
14
I 15
I 16
17
I 18
I 19
20
I 21
I 22
23
I 24
I 25
26
I
I
-14-
si nce on the average, revenues are coll ected at approximately the
same time as expenses and costs are pai d.
Q. Please explain the dollar amounts in Column (b) of Exhibit 104.
A. These are the 12-month test period amounts of each basic category of
revenues, expenses and costs, after both normal i zing and pro forma
adjustments. The derivation of the subtotals in Column (b) is shown
on Exhibit 102, Schedule 1, Col umn H. These subtotal s were then
separated by source -- for example, of the $12,129,311 of Operations
and Mai ntenance Expense, $6,121,346 represents 1 abor, $2,979,504
accounts payable, etc.
Q. Pl ease exp 1 a in how the days in Co 1 umn (c) were computed.
A. We performed our study on the hi stori ca 1 12-month peri od ended
September 30, 1979, which coincided with the Company's fiscal year
and whi ch represented a recent and representati ve 12-month peri ode
We obtained from the Company's books and records the revenue and
cost data for thi s peri od segregated into the same categori es as in
Column (a) of Exhibit 104. We then performed analyses of the
amounts booked for thi s 12-month peri od, and, as a resul t of such
analyses, were able to determine the applicable lags in revenue
collection and leads in obtaining goods and services prior to
payment. The days shown in Column (c) represent these resul ts.
Q. Referring to Exhibit 104, why were cyclical and non-cyclical
revenues analyzed separately?
A. Because bi 11 i ng practi ces di ffer for customers in each category
which affect the computation of lag days. Non-cyclical, customers
are bi 11 ed at the end of the month for service rendered for that
-15-
month, based on month-end meter readi ngs. For cycl i ca 1 customers,
meter readi ngs and bi 11 i ngs occur throughout the month (dependi ng
upon the specific cycle), and there is a lag of several days for
bi 11 processi ng between the meter read date and the bi 11 i ng date.
Q. Please explain how the 38.09-day collection lag in cyclical revenue
shown on Line 2, Column (c) was determined.
A. Our objecti ve was to determi ne the 1 ag from the date of the
rendering of service to the customer to the date that the payment
for that servi ce was recei ved by the Company. For thi s purpose, it
was assumed that service was rendered ratably during the period
between meter readi ngs so that the mi dpoi nt of the meter readi ng
period coul d be used as a proxy for the date that service was
rendered duri ng the peri ode
The average lag of 38.09 days in the collection of
cyclical billings was determined by dividing average daily cyclical
accounts receivable by average daily cyclical revenues. To this
number were added (l) the average number of days from the mi dpoi nt
of the meter reading period to the meter reading date, and (2) the
average number of days from the meter readi ng date to the bi 11 i ng
date. The total is the average 38.09-day lag shown for cyclical
revenues.
Q. Were the non-cyclical lag days computed similarly?
A. Yes.
Q. Were non-utility revenues, expenses and costs excluded from your
study?
A. Yes.
'J
I
I
I
1
2
I 3
I
4
5
I 6
I 7
8
I 9
I 10
11
I 12
I 13
14
I 15
I 16
17
I 18
I 19
20
I 21
I 22
23
I 24
I 25
26
I
I
-16-
Q. How was the 1 ead time for purchased gas cost computed?
A. It was computed by ascertai ni ng for all purchased gas costs booked
during the 12 months ended September 30, 1979, the applicable bills
(invoices) and delivery periods, together with the dates such bills
~ere pai d by the Company. The wei ghted average number of days from
the mi dpoi nt of the del i very peri ods covered by the bi 11 s to the
bill payment dates was 35.00.
Q. How were the accounts payable and materi a 1 sand supp 1 i es 1 eads of
15.67 days on Lines 7 and 8 determined?
A. To have conducted an analysi s of 100% of the voucher transactions
woul d have requi red revi ewi ng thousands of transacti ons. For thi s
reason, we selected a random sample of vouchers for ana lys is. We
included in our analysis only vouchers charged to operations and
mai ntenance expense accounts and materi al sand suppl i es. We
computed the number of days lead from the date a good was received,
or the mi dpoi nt of the peri od duri ng whi ch the Company recei ved a
servi ce bei ng pai d for, to the date the Company pai d each i nvoi ce.
This analysis indicated that the average lead for voucher payments
was 15.67 days.
Q. How did you determine the 14.82-day lead shown on Line 9 with
respect to 1 abor costs?
A. The Company has two types of payroll -- biweekly and monthly. We
determi ned the 1 ead from the mi dpoi nt of the pay peri od to the date
the employees are pai d for both types of payroll, for each pay
period during our study year. In addition, we calculated the
average lead time for the deposit of tax withholdings (income and
I
I
I
I
1
2
I 3
I
4
5
I 6
I 7
8
I 9
I 10
11
I 12
I 13
14
I 15
I 16
17
I 18
I 19
20
I 21
I 22
23
I 24
I 25
26
I
I
-17 -
social security) and other withholdings. The average leads were
then wei ghted by the rel ated doll ar amounts, and the resul tis the
14.82-day computed 1 ead shown on Line 9.
Q. What is the credit shown in Column (b) for C&K?
A. These are refunds to customers resul ti ng from termi nati on of the C&K
joint venture exploration program.
Q. How was the number of lead days for C&K determined?
A. It is assigned the same number of days as overall operating revenues
in order to negate the effect of C&K in the study. Thi sis proper
because, on the books, the other si de of the C&K entry to expense is
to revenues, wi th no effect on net income.
Q. What does the 1I0ther" category consi st of?
A. The most significant items included in the "Otherll category are
amortization of prepaid insurance and regulatory commission expense,
pension expense, automobile expense and provision for uncollectible
accounts.
Q. How was the net 61.01 1 ag day amount for 1I0ther" determi ned?
A. Each of the significant items was analyzed separately to determine
the time 1 ag between payment for and consumpti on of the servi ce
i nvol ved. We then computed a wei ghted average of all the items
analyzed, which was 61.01 net lag days.
Q. How were the lead days applicable to the general taxes shown on
Lines 14 through 17 in Column (c) determined?
A. For property and franchi se taxes, whi ch are assessed on a cal endar-
year basis, we analyzed all payments made pertaining to calendar
1978 and computed the wei ghted average number of days from the
I
I
I
I
1
2
I 3
I
4
5
I 6
I 7
8
I 9
I 10
11
I 12
I 13
14
I 15
I 16
17
I 18
I 19
20
I 21
I 22
23
I 24
I 25
26
I
I
-18-
midpoint of calendar 1978 to the payment dates for each of these
taxes. We performed a similar analysis for unemployment taxes and
the employer's portion of FICA, except that the period analyzed was
the 12 months ended September. 30, 1979.
Q. Please describe the method used to calculate lead days of 91.50 on
Line 20 applicable to current Federal income taxes.
A. The computed lead of 91.50 days was calculated based on statutory
requi rements of the Internal Revenue Code -- namely, a payment of
20% of estimated total Federal income taxes on January 15, March 15,
June 15, and september 15. The remaining 20% is payable 10% on
December 15 and 10% on March 15 of the following year.
Q. Pl ease descri be how the 284.00 1 ead day fi gure for current state
income taxes was determi ned.-
A. It was determined by computing the number of days from the midpoint
of fi sca 1 year 1979 to the payment of the estimated 1 i abi 1 ity for
this time period.
Q. Please explain your rationale for including the amounts of interest
on debt and preferred stock dividends in this calculation and how
the lead days assigned to these items were determined.
A. These amounts are i ncl uded in the total revenue requi rement whi ch
will be collected from the customers in rates. The collection of
such revenues in rates is subject to the same 35.45-day lag as are
all other expense and cost items that make up the total revenue
requi rement. Al though these costs accrue dai ly, the cash payments
to debt hol ders and preferred stock-hol ders are made pursuant to the
contractua 1 agreements between these investors and the Company.
I
I
I
1
I 2
I 3
4
I 5
I 6
I
7
8
I 9
I
10
11
I 12
I
13
14
I 15
I 16
17
I 18
I 19
20
I 21
I 22
23
I 24
I 25
26
I
I
- 19-
Since the payment schedule does not require daily payments
consi stent wi th the incurrence of the costs, but instead provi des
for some delay in payment, the payment schedule does produce an
economic benefit. While we believe that this economic benefit
belongs to the investors who have suppl i ed the Company's capi tal, we
have i ncl uded the 1 ead times resul ti ng from the payment schedul e for
both interest and preferred stock dividends in our study. By
reducing the working capital requirement, this has the effect of
passi ng the economi c benefi t to the customers.
The 15-day 1 ead shown on Line 27 for short-term interest
expense is based on the Company's policy of making interest payments
on short-term debt at the end of each month. The 90-day lead shown
for long-term interest expense on Line 28 is based on semiannual
interest payment due dates on 1 ong- term debt and the Company IS
policy of paying such interest the day before the due date.
The 45-day 1 ead shown on Line 29 for preferred stock
dividends is based on the Company's policy of paying such dividends
on the day following the three-month period to which the dividends
rel ate.
Q. Zero lead days are shown for certain cost amounts on Exhibit 104,
including depreciation (Line 25), provisions for deferred income
taxes and investment tax credit-net (Li nes 22 and 23), and cost of
common equi ty (Line 30). Why have such amounts been i ncl uded in the
study wi th no 1 ead days?
A. As is the case for other expenses and costs, these amounts are part
of the total revenue requirement and are subject to the 35.45-day
I
I
I 1
I 2
I 3
4
I 5
I 6
7
I 8
I 9
10
I 11
I 12
13
I 14
I 15
16
I 17
I 18
19
I 20
I 21
I
22
23
I 24
I
25
26
I
I
-20-
lag in collection described earlier. The investor should be
compensated on a dai ly basi s for these costs, and, to the extent
that he is not, he has an investment in working capital related to
these items on which he should expect a return.
Q. Even though there is a 35.45-day lag in the collection of revenues,
is there not an offsetti ng 1 ead that shoul d be considered?
A. For reasons which I will explain, we do not believe that the
computat i on of any 1 ead is appropri ate for any of these cost items.
Wi th respect to depreci ati on, rate base is reduced by the
accumul ated depreciati on. No adjustment to thi s rate base reduction
is made to reflect the fact that the provision for depreciation is
credited to this account as the service is rendered to the customer,
but the depreci ati on wi 11 not be co 11 ected from the customer unti 1,
on the average, 35.45 days later. Therefore, there is no offsetting
lead for the provision for depreciation.
Wi th respect to the provi si on for deferred income taxes,
even though rate base has not been reduced by the accumul ated
portion of such provision, the accumulated portion has been included
in the cost of capital computation at a zero effective cost rate
(Exhibit 102, Schedule 6, Page 1, Line 5), which has the same effect
as a rate base reduction. No adjustment is made to the accumulated
deferred income taxes i ncl uded in the cost of capi tal computati on to
reflect the fact that the provi si on for deferred income taxes is
credi ted to thi s account as servi ce is rendered to the customer, but
will not be collected until, on the average, 35.45 days later.
I
I
I 1
I 2
I 3
4
I 5
I 6
7
I 8
I 9
10
I 11
I 12
13
I 14
I 15
16
I 17
I 18
19
I 20
I 21
I
22
23
I 24
I
25
26
I
I
-21-
Therefore, there is no offsetti ng 1 ead for the provi si on for
deferred income taxes.
With respect to the investment tax credit, Congress stated
in 1971 that the benefits of the credit are to be shared between the
customers and the investors. The Company has made the tax e 1 ecti on
whi ch provi des that the credi t may be used to reduce revenue
requi rements over the book service 1 i fe of the property, but the
unamorti zed amounts may not be used to reduce rate base. The effect
is that the investment credi t, once uti 1 i zed to reduce taxes
otherwi se payable, becomes a source of funds on whi ch the uti 1 i ty is
entitled to earn its overall rate of return. For this reason, there
are no lead days associated with the investment tax credit which may
appropriately be passed on to the customer.
Similarly, with respect to the cost of common equity, the
investors are entitled to their return on capital employed ~ the
servi ce is rendered to customers. Therefore, there is no 1 ead, as
there is when vendors and taxing au thori ties delay requ i red payment
days beyond when servi ce is rendered.
Q. Is there a lead that should be reflected in the working capital
study because dividends to common shareholders are only paid
quarterly?
A. No. The common shareholders are entitled to the return on their
equity investment as service is rendered. They, in theory, can take
out their earnings daily as dividends or can reinvest the entire
amount dai ly. To the extent that they do the 1 atter, they have made
an additional investment on which they are also entitled to a return
I
I
I
1
I 2
I 3
4
I 5
I 6
7
I 8
I 9
10
I 11
I 12
I
13
14
I 15
I
16
17
I 18
I
19
20
I 21
I 22
23
I 24
I 25
I
I
-22-
the next day. To the extent that the common equity investors
receive their reinvested earnings as quarterly cash dividends, the
common shareholders are no longer enti tl ed to a return on sûch
rei nvested earni ngs si nce they no longer exi st. For accounti ng
purposes, retai ned earni ngs are reduced and common di vi dends payabl e
are increased as common dividends are declared. However, for
purposes of computing the common shareholders' invested capital,
such divi dends payabl e are effectively a part of common sharehol der
investment until paid out in cash.
Q. Pl ease summari ze the resul ts of your 1 ead-l ag study.
A. When the days determined from our study in Column (c) are applied,
i tern by i tern, to the test-peri od amounts shown in Column (b), the
resul tis a net 1 ag in recovery of all expenses and costs of zero
days, as shown on Line 32 in Column (c). This means that during the
test year, the customer had provided working capital equivalent to
zero days of total revenue, or $0 (0 + 365 X $158,809,555). Because
of the zero net lag, the cash requirement is also zero.
Q. Why have additional items of working capital been included as shown
on Exhibit 102, Schedule 2, Page 1, Lines 7 and 8 to arrive at the
total worki ng capi tal component of rate base?
A. Our lead-lag study determined the net (lead) lag in recovery of
expenses and costs. In making thi s determi nati on, we have treated
inventory i terns as consumed when purchased, when in fact the Company
is required to maintain a substantial investment in materials and
supplies and liquified natural gas (LNG) inventory to promote
I
I
I
1
I 2
I 3
4
I 5
I 6
7
I 8
I 9
10
I
I
I
I
I
I
I
I
I
I
I
I
I
-23-
effi ci ent servi ceo Therefore, it is necessary to i ncl ude thi s
investment in working capi ta 1.
Q. Pl ease descri be how the investment in materi a 1 sand supp 1 i es and LNG
inventories of $679,400 and $2,712,933, respectively, shown on
Exhibit 102, SChedule 2, Page 1, Lines 7 and 8 were computed.
t
A. These amounts represent 12-month averages of book and projected
ba 1 ances of mater; a 1 sand supp 1 i es and LNG i nventori es dur; ng the
test year, as detailed on Exhibit 102, SChedule 2, Pages 5 and 6.
Q. Does this complete your direct testimony ;n this proceeding?
A. Yes.