20060821INT to Staff 1-8.pdf

EXECUTIVE OFFICES
INTERMOUNTAIN GAS COMPANY RECEIVED
555 SOUTH COLE ROAD. P.O. BOX 7608 . BOISE IDAHO 83707 . [phone] . FAX: zOn~o
AUG 2\ PM~:' 8
IDAHO PU81.\C
UTILITIES COMMISSIO,~
August 21 , 2006
Jean Jewell
Idaho Public Utilities Commission
472 W Washington St.
O. Box 83720
Boise, ID 83720-0074
Re: First Production Request of the Commission Staff
Case No. INT-06-
Dear Ms. Jewell:
Pursuant to the above referenced Case Number, attached are the original and seven copies
of Intermountain Gas Company s response to the First Production Request of the
Commission Staff.
If you have any questions regarding the attached, please contact me at 377-6168.
SinCerelY
~MC r--
~::~
;. I
Gas Supply and Regulatory Affairs
Record Holder: Mike McGrath, Director, Gas Supply and Regulatory Affairs, Intermountain Gas
Company, 555 S Cole Rd, Boise, Idaho, 83709, [phone]
Request No.1: Economic Forecast.In the IRP the Company uses the Idaho economic forecast
published by John S. Church, dated May 2005. Did Mr. Church publish a later or updated
forecast? If so , was that the subsequent work used in preparation of the IRP?
Response: Mr. Church, as a matter of business, regularly publishes updated forecasts
reflecting the then current state of the economy and economic outlook. Intermountain
Gas Company s filed IRP embedded the most recent economic forecast available at the
time from Idaho Economics which is to say the May 2005 Idaho Economics forecast.
Preparer: Byron Defenbach, Manager, Market Products and Services
Witness: Mike McGrath
Request No.2: Linear Regression Modeling. The demand forecast model used by the Company
relies on the single variable of design heating-degree-days to predict natural gas usage. There is
a statement indicating that other variables have been considered. a) What are those other
variables and how do they affect the output of the model? b) The Durbin-Watson statistic
suggests that better specification of the model is possible. Has the Company followed up on that
indication? If so, what were the results?
The statement in the Production Request that
, "
The demand forecast model used by the
Company relies on the single variable of design heating-degree-days to predict natural
gas usage" is not correct. A closer inspection of the equations included in the IRP filing
would reveal that the peak day regression equations for November, December and
January all contain both a heating degree day variable and a weekend variable. The
weekend variable helps to capture, or model, the differing consumption behavior between
the week days and weekends.
When building regression models, the Company regularly tests a variety of economic
variables ' correlation to natural gas consumption. These variables include , but are not
limited to: CPI, GDP , Prime Rate, Mortgage Rate, Idaho Housing Starts, Idaho Per
Capita Personal Income, and usage trends. Statistical tests on these variables have shown
thus far that they are not statistically significant in explaining changes in usage per
customer per degree day.
The Company continually looks for new ways to improve the linear regression models.
An area the Company is investigating as a future enhancement is the inclusion of an
autoregressive variable which would help to address any serial correlation that the
Durbin-Watson statistic suggests may be present in the current models.
Preparer: Lori Blattner, Sr. Financial Analyst
Witness: Mike McGrath
Request No.3: Natural Gas Pricing . The price curve used for the natural gas price forecast
(Exhibit 4, Appendix A, Chart 4-2) is not cited. Please provide the source for this chart. Did the
RESPONSE OF INTERMOUNTAIN GAS
TO FIRST PRODUCTION RESPONSE
- 1 -AUGUST 21 , 2006
Company consider and review other price forecasts for this application? If so, what were they?
Were different price forecasts considered for application to the differing economic scenarios?
Response: The source for the natural gas pricing data was IGI Resources Inc, a BP
Energy Company. This data incorporated the daily NYMEX market close plus applicable
basin differentials to determine the projected monthly price.
The Company utilized additional price forecasts in the IRP model. One such additional
forecast was from the November 16 2005 market close. The impact of using the differing
price forecasts was shown not to materially alter the optimization results.
Intermountain did not utilize non-NYMEX driven price forecasts which would have been
required had the Company attempted to incorporate or correlate natural gas price
variation with differing Idaho demand or growth scenarios. Had such data been available
and included, its use could have potentially resulted in nine different optimization
scenarios (low customer growth, base case customer growth, and high customer growth
as a base for each of the pricing scenarios) not adding any measurable improvement in
the Company s ability to plan for needed capital improvements to meet projected peak
demands.
Preparer: David Swenson, Sr. Financial Analyst
Witness: Mike McGrath
Request No.4: Sensitivity Analysis. Please identify and describe any sensitivity analysis
performed, and please provide the results of the sensitivity analysis. In your response, include a
full description and range of inputs that were varied for the sensitivity analysis.
Response: The models were run using three differing demand levels (High Growth, Base
Case and Low Growth scenarios) and a wide range of resource options as highlighted in
the appendix of the filed IRP.
The High Growth, Base Case and Low Growth scenarios coupled with a wide range of
resource options represent the sensitivity analysis boundaries that the Company believes
the actual load duration curves will fall within. The filed IRP provides a detailed
description of the results of the sensitivity analysis generated by examining the impacts
of these growth and resource outcomes on the Company s load duration curve forecasts.
As well, the Company utilized additional price forecasts in the IRP model. One such
additional forecast was from the November 16, 2005 market close. The impact of using
the differing price forecasts was shown not to materially alter the optimization results
Preparer: David Swenson, Sr. Financial Analyst
Witness: Mike McGrath
Request No.5: Has the Company considered addressing the effects of price elasticity on the
IRP process? If so, what were the results? Ifnot, why not?
RESPONSE OF INTERMOUNTAIN GAS
TO FIRST PRODUCTION RESPONSE
- 2 -AUGUST 21 , 2006
Response: The Company has considered addressing the effects of price elasticity, or the
correlation between consumption and burner-tip prices, on the IRP process. There are
several reasons, however, that the Company believes it is not appropriate to include the
effects of price elasticity on a Peak Day (Load Duration Curve) analysis.
One of the primary goals of the IRP process is to ensure that the Company is able to
provide safe, reliable natural gas service to our customers on a severe cold, or "design
temperature day or, over a period of extreme cold days. The winter of 1990 represented
what Southern Idaho did, and can, experience for record cold temperatures. Since that
time, Intermountain s service area has experienced record customer growth, volatile
natural gas prices and warmer than normal winter temperatures. During this high-growth
warm winter temperature period, Intermountain s customers were better able to respond
to price increases and conserve on natural gas usage by adjusting their thermostats
downward, and/or by employing more permanent conservation measures. Throughout
the recent period of volatile natural gas prices, the Company has continued to promote
the wise and efficient use of natural gas by actively encouraging its customers to
conserve on their natural gas usage through media campaigns, information on the website
and bill stuffers focusing on conservation tips. The Company credits these efforts with a
measurable decline in weather adjusted baseload usage. However, thermal efficiencies
tend to deteriorate with record cold temperatures, and even at lower thermostat settings
our customers ' ability to conserve is diminished. If Southern Idaho were to experience
record cold temperatures again as it last did back in 1990, many furnaces would run
continuously just to keep up, even if customers set their thermostats at a lower
temperature setting. The Company is aware that other LDC's in the Northwest have also
observed that while baseload consumption does appear to have varying degrees of
correlation with natural gas prices, design weather consumption has not proven to be
significantly correlated to natural gas prices. Including lower usage due to price elasticity
during design weather, therefore, has the potential to jeopardize or severely
underestimate our usage requirements during record cold temperatures.
Additionally, preliminary research indicates that for a given heating degree day, peak
usage per customer on the Sun Valley and Idaho Falls laterals is higher than the average
peak usage per customer for the rest of Intermountain s customers. As the IRP document
explained, the Company has installed metering equipment on both laterals so the
Company will be able to better model, or predict, the usage per customer for those two
specific areas of our service territory. Unfortunately, the Company does not yet have
enough data to make statistically significant calculations regarding usage per customer
for the Idaho Falls and Sun Valley laterals, but as more data is collected, we look forward
to having enough data to make those calculations in a future IRP. Using a price elasticity
adjusted usage per customer for the Idaho Falls and Sun Valley laterals would potentially
cause the Company to underestimate usage requirements for record cold days on these
two very sensitive areas of our system.
Also, the Company, as previously mentioned, includes a low growth scenario in the IRP
filing. Adding a price elasticity component on top of this low growth scenario implies an
ability to sophisticate the modeling process to include the effects of price on an already
low usage per customer forecast. Also as mentioned above in the Company s response to
RESPONSE OF INTERMOUNTAIN GAS
TO FIRST PRODUCTION RESPONSE
- 3 -AUGUST 21 , 2006
Request No., the inclusion of price would add six additional scenarios for consideration
a sophistication of the proverbial brick.
The Company will continue to model and correlate the relationship between record cold
temperatures and consumption as more representative historical data becomes available
while at the same time ensuring the safe, reliable delivery of natural gas on the coldest
design" degree day.
Preparer: Lori Blattner, Sr. Financial Analyst
Witness: Mike McGrath
Request No.6: In the section entitled "Resource Optimization " a "Fill" resource is defined as a
generic resource used to eliminate a deficit. Does the Company have specific resource types
such as storage, spot, or term contract that it will use for each requirement? If so, please identify
those resources.
Response: Studies such as the IRP are an invaluable piece of the Company s long-term
planning to identify predicted deficits compared against a list of potential resource
solutions. In specific, the model provides the Company with direction as to locale, term
seasonality and other requirements and suggests, given its selection of potential options
an efficient combination of resources to consider when meeting future needs.
It is important to note that the results of the optimization represent a snapshot in time and
the best available resources for consideration at that time. As well, the Company
continually seeks the best method for managing resource deficits before they arise. So as
anticipated deficits become more imminent, IRP model solutions may indeed lose some
or all of their practical relevance before implementation, and updated solutions may be
called for.
While the shorter-term gas supplies are treated as an economic commodity and are
generally assumed to be filled by "spot" supplies, the other "Fill" resources used in the
model such as interstate capacity, storage and, to some extent, longer-term gas supplies
are not. These types ofresources often become available unexpectedly, at a moments
notice, and are generally market driven in terms of volume, timing and term (i.e. they
rarely perfectly match the Company s needs). So, absent publicly announced
infrastructure expansion (which none were known when the IRP was developed), the
Company must watch the market for appropriate open market opportunities. The
Company is therefore vigilant in watching for, and evaluating the appropriateness of
such market opportunities if and when they do arise. But the nature of those resources
does not allow the Company to know of specific contracts, parties or other details
beforehand. However, knowing what resource is needed, as well as where and when
such resources are available allows the Company to timely address their appropriateness
and attempt to negotiate favorable terms and conditions where applicable.
See response No.7 for more specific information on distribution capacity resources.
Preparer: David Swenson, Sr. Financial Analyst
Witness: Mike McGrath
RESPONSE OF INTERMOUNTAIN GAS
TO FIRST PRODUCTION RESPONSE
- 4-AUGUST 21, 2006
Request No.7: What are the specific, planned actions that will eliminate the peak day
distribution deficits identified in the Resource Optimization section for the Sun Valley, Idaho
Falls and Canyon County service laterals?
Response: In response to design weather peak distribution deficits, the Company s IRP
model selects the most optimal solution based on a best-cost type evaluation. However
as stated in Response No., at one level these stark numbers simply provide the
Company notice of the level and timing of deficits at a particular point in time based on
the model inputs. While the model determines the best mix of resources to solve deficits
the Company is continually evaluating new technology and/or applications of this new
technology to find the best available resource portfolio in this changing energy
marketplace. The specific planned actions that the Optimization Model determined were
the best resources for the Company to employ to eliminate distribution deficits are as
follows:
1- Canyon County
a. 2007 - the relatively small deficit of 3 740 therms is eliminated by utilizing
mobile LNG. From a real-world standpoint, it is possible that a deficit this
small occurring on only one day may be also resolved from short-term
operational efficiencies.
b. 2008-11 - distribution main looping and pressure upgrades that add 170 000
therms of additional delivery capacity resolves all deficits thru 2011.
2- Idaho Falls Lateral-
a. 2007 - the 30 450 therm deficit is filled by using mobile LNG
b. 2008 -09 - the deficits are almost entirely filled by adding 80 000 therms of
capacity to the distribution facilities via the Phase V looping project.
c. 2010-2011 these deficits are entirely filled by utilizing the Phase XI looping
project that adds another 90 000 therms of capacity.
3- Sun Valley Lateral -
a. 2009 - the 370 therm deficit is filled via mobile LNG.
b. 2010-2011 - the deficits for these years are entirely filled via a looping project
that increases capacity by 40 000 therms.
Notwithstanding the above, subsequent to completing and filing the 2007-11 IRP with the
Commission, the Company has further refined its plans to eliminate the projected
distribution deficits in a more cost efficient manner. Interestingly enough, some of the
early research conducted for this IRP led to discovering some technology enhancements
that will allow the Company to forestall some very expensive pipeline upgrades in two
areas of the system. These refinements are explained below:
Canyon County -
a. 2007-08 - further study of historical pressures from Northwest Pipeline and
the ability to use the Company s system linepack for "temporary" storage
indicate that in the near-term, the peak day deficits will lag prior estimates and
the deficits will occur a later year and at a lesser magnitude. Consequently, the
looping project that was to occur in 2007 can now be delayed until 2008.
RESPONSE OF INTERMOUNTAIN GAS
TO FIRST PRODUCTION RESPONSE
- 5 -AUGUST 21 , 2006
b. 2008-11 - the looping project now planned for 2008 will continue to eliminate
all remaining deficits projected thru 2011.
Idaho Falls Lateral-
a. 2007-08 - updated analysis forecast that, due to higher Northwest Pipeline
pressure and temporary use of lateral "line pack", any deficits would occur
later and would be smaller than earlier projected. Additionally, the Company
has continued to research the mobile LNG technology and found that it can be
employed on a larger scale over a longer period of time than was assumed in
the IRP. Consequently, the looping projects previously projected for 2007 and
2009 may now be delayed until after 2011 saving the Company approximately
$7.5 million.
3. Sun Valley Lateral -
a. Through the use of mobile LNG, the pipeline looping that was previously
projected to occur in 2007 is now likely to be postponed until after 2011
saving the Company over $2.5 million.
Preparer: David Swenson, Sr. Financial Analyst
Witness: Mike McGrath
Request No.8: Commission Order 27098 changed the requirement for addressing efficiency
measures to require only "a general explanation with each IRP filing of whether or nor(t) there
are cost effective DSM opportunities." What efficiency measures did the Company review and
consider in the 2006 IRP process and what were the results of those reviews?
Response: All ofIntermountain Gas Company s Integrated Resource Plans that were
filed subsequent to the above referenced Order Number, including the 2006 IRP, included
a listing of "cost effective DSM opportunities." These measures are more fully explained
in the Company s filed IRP as contained in the section "The Efficient and Direct Use of
Natural Gas.
As previously noted, the Company actively promotes the wise and efficient use of natural
gas. The aforementioned media campaigns, conservation information and education on
the Company s web site, and conservation minded bill stuffers all focus on cost effective
energy efficiency measures. The Company credits these efforts with a measurable
decline in weather adjusted baseload consumption.
Preparer: Byron Defenbach, Manager, Market Products and Services
Witness: Mike McGrath
RESPONSE OF INTERMOUNTAIN GAS
TO FIRST PRODUCTION RESPONSE
- 6 -AUGUST 21 , 2006