TransCanada PipeLines Limited Reply Evidence of Paul R. Carpenter Application for Approval of Mainline of The Brattle Group, Inc. 2013 – 2030 Settlement RH-001-2014
Reply Evidence of Paul R. Carpenter of The Brattle Group, Inc.
August 22, 2014
NATIONAL ENERGY BOARD
IN THE MATTER OF the National Energy Board Act and the Regulations made thereunder;
AND IN THE MATTER OF an Application by TransCanada PipeLines Limited, pursuant to Part I and IV of the National Energy Board Act for approval of a negotiated settlement of 2013- 2020 Mainline tolls and tariff terms;
AND IN THE MATTER OF a decision of the National Energy Board dated March 2013, pursuant to Hearing Order RH-003- 2011, and National Energy Board Order TG-002-2013. [RH-001- 2014]
TRANSCANADA PIPELINES LIMITED
______
APPLICATION FOR APPROVAL OF TOLLING SETTLEMENT
REPLY EVIDENCE OF PAUL R. CARPENTER THE BRATTLE GROUP, INC. ______
August 22, 2014
To: The Secretary National Energy Board 517 10th Avenue S.W. Calgary, Alberta T2R 0A8
WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
TABLE OF CONTENTS
I. Qualifications and Summary...... 1
II. “Economic Withholding” and Market Power ...... 8
III. Secondary Market Competition and Downstream Prices ...... 16
IV. Dr. Cicchetti’s Regressions ...... 29
V. Pricing Discretion and Upstream Netbacks ...... 42
VI. Pricing Discretion, Toll Stability and Transparency ...... 46
VII. Pricing Discretion and Business Risk ...... 48
i WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 I. QUALIFICATIONS AND SUMMARY
2 Q1. Please state your name, address and position. 3 A1. My name is Paul R. Carpenter. I am a Principal and Chairman of The Brattle Group, 4 an economic and management consulting firm with offices in Cambridge, 5 Massachusetts, Washington D.C., New York City, San Francisco, London, Rome and 6 Madrid. My office is located at 44 Brattle Street, Cambridge, Massachusetts 02138.
7 Q2. Will you briefly describe your educational background and professional 8 qualifications?
9 A2. Yes. I am an economist specializing in the fields of industrial organization, antitrust, 10 finance, and energy and regulatory economics. I received a Ph.D. in Applied 11 Economics and an M.S. in Management from the Massachusetts Institute of 12 Technology, and a B.A. in Economics from Stanford University. I have been 13 involved in research and consulting on the economics and regulation of the natural 14 gas, oil and electric utility industries in North America and abroad for thirty years. I 15 frequently have testified before federal, state and Canadian regulatory commissions, 16 in federal court and before the U.S. Congress, on issues of pricing, competition and 17 regulatory policy in these industries. Outside of North America, I have advised 18 governments and regulatory bodies on the structure of their natural gas markets and 19 the pricing of gas transmission services. These assignments have included testimony 20 before the U.K. Monopolies and Mergers Commission and the Australian 21 Competition Tribunal, and advice to the European Commission and the governments 22 of, and regulators in, Greece, Ireland, the Netherlands, New Zealand and Australia.
23 For at least 30 years I have been extensively involved in the evaluation of the 24 economics and regulation of the natural gas industry in North America. In Canada, I 25 have advised pipeline and distribution companies and have previously testified before 26 the National Energy Board on matters relating to pipeline competition and capacity 27 expansion, including the Alliance Pipeline Ltd. certification proceeding, and on the
1 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 subject of business risk in the context of cost of capital. I gave evidence on business 2 risk in the RH-003-2011 proceeding at the request of TransCanada.
3 Q3. Could you elaborate on your experience and qualifications in the areas of 4 competition policy and antitrust as applied to regulated utilities including 5 natural gas pipelines?
6 A3. Yes. From the earliest days in my career to the present I have been involved in 7 matters related to the application of competition policy and antitrust to natural gas 8 pipelines and regulated utilities. For example, I was the principal economic expert 9 witness in two of the seminal antitrust lawsuits brought against natural gas pipelines 10 in the early years of U.S. pipeline regulatory reform: State of Illinois v. Panhandle 11 Eastern Pipeline Co. (Fed. Ct. for C.D. Illinois), and City of Chanute, et al. v. 12 Williams Natural Gas (Fed. Ct. for Kansas). In the area of mergers and acquisitions, 13 I have been the economic expert witness in several proceedings where the merger of 14 regulated gas assets and electric utilities raised issues of vertical market power, 15 including the merger that created Sempra Energy and the proposed merger between 16 Exelon and Public Service Gas & Electric Co. I also have experience in the 17 evaluation of claims of market manipulation. For example, in 2008 I was the 18 economist retained by the enforcement staff of the Federal Energy Regulatory 19 Commission (“FERC”) to evaluate and testify as to its claims of market manipulation 20 in gas commodity and derivatives trading by an energy transportation and trading 21 company. Further details of my educational and professional background, as well as a 22 listing of my publications, are provided in my curriculum vitae, which is appended to 23 this testimony as Attachment A.
24 Q4. What assignment were you given in this proceeding? 25 A4. I have been asked by TransCanada PipeLines Limited (TransCanada) to review the 26 intervenors’ written evidence in this proceeding bearing on the issue of 27 TransCanada’s pricing discretion for short term firm and interruptible services (STFT 28 and IT), including the evidence of the Canadian Association of Petroleum Producers 29 (CAPP) and its witness Dr. Ren Orans, and the evidence of Centra Manitoba (Centra)
2 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 and its witness Dr. Charles Cicchetti. I have been asked to respond to the opinions 2 expressed in this evidence, that TransCanada has been using its pricing discretion in 3 ways which are detrimental to economic efficiency and price formation in the markets 4 served by the TransCanada Mainline. Based on these claims, the intervenor evidence 5 recommends that the Board limit the pricing discretion granted to TransCanada in the 6 Board’s RH-003-2011 Decision.
7 Q5. Dr. Carpenter, how would you characterize the intervenor written evidence on 8 pricing discretion to which you are responding?
9 A5. The intervenor evidence on pricing discretion is based on a series of assertions, each 10 of which deserves close scrutiny. These assertions include:
11 1. A claim that pricing discretion equates to the “economic withholding” of short 12 term services by TransCanada that is economically inefficient. In other 13 words, that TransCanada is exercising market power by virtue of its pricing 14 discretion. (CAPP/Orans)
15 2. Assertions that TransCanada’s use of its pricing discretion has affected 16 downstream commodity prices at key trading hubs (Centra/Cicchetti and 17 CAPP/Orans).
18 3. A claim that the reduction in netbacks to NIT observed during the first few 19 months after the implementation of the Board’s RH-003-2011 Decision Model 20 was due to TCPL’s pricing discretion and that such effects could happen again 21 in the future (Orans).
22 4. The position that pricing discretion should be limited because it creates 23 volatility and toll instability for producers and that it lacks transparency 24 (Orans).
25 5. The view that pricing discretion was permitted by the Board in RH-003-2011 26 solely to mitigate the business risk associated with the Decision Model, and 27 that Board approval of the terms of the Settlement here eliminates that 28 business risk (CAPP/Orans).
3 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 2 I conclude that these intervenor assertions are either false or unproven, and therefore 3 that their proposals that the Board limit TransCanada’s short term service pricing 4 discretion are unwarranted.
5 Q6. What are your principal conclusions regarding each of these premises? 6 A6. My conclusions are as follows:
7 1. I disagree with the opinions expressed by CAPP and Dr. Orans’ that 8 TransCanada’s use of its pricing discretion constitutes “economic withholding” 9 of capacity that is economically inefficient.
10 First, the concept of “economic withholding” that originated in the regulation of 11 wholesale electricity markets has no application in this case without a finding that 12 TransCanada has market power with respect to short term services (STFT and IT). I 13 demonstrate using the tools typically employed by economists in competition law 14 settings that TransCanada does not possess market power in the relevant product 15 market (short term transportation services including “bundled” sales of delivered gas) 16 and in the relevant geographic market (at least as large as the area served by the major 17 hubs and other delivery points on the Mainline.) I show that TransCanada’s short 18 term services face significant competition in the relevant market from holders of 19 pipeline and storage capacity on other pipelines into the Dawn Hub, and holders of 20 FT capacity on the Mainline who use their diversion and alternative receipt point 21 rights to effectively compete with the Mainline’s discretionary services. A further 22 constraint on any alleged market power with respect to short term services is the 23 recourse that shippers have under NEB regulation to contract for FT services at 24 regulated tolls. None of the intervenor witnesses provide evidence that establishes 25 TransCanada’s market power in these relevant markets.
26 Second, the economic efficiency concept that CAPP and Dr. Orans are propounding 27 in their assertions regarding “economic withholding” is entirely short term in nature, 28 such that the only consideration is whether short-term pipeline throughput is 29 maximized. They completely ignore the long-run and dynamic efficiency benefits of
4 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 encouraging the long-term contracting of pipeline capacity. Pipeline regulation in 2 Canada is a “contract carriage” regime. It is well-recognized that policies which help 3 ensure that shippers meet their needs for pipeline capacity with long-term contracts, 4 help minimize the business risk faced by pipeline investors, thus reducing the cost of 5 capital and enabling economically efficient (long term) investment decisions and 6 lower tolls for shippers. CAPP’s recommendation to limit TransCanada’s pricing 7 discretion flies in the face of such policies.
8 9 2. The regression analyses provided by Dr. Cicchetti in his written evidence, that 10 purport to confirm his hypothesis that TransCanada’s use of its pricing discretion 11 (IT bid floor prices) has affected downstream commodity prices, is poorly 12 conceived and executed and its results are spurious and unreliable.
13 First, Dr. Cicchetti’s hypothesis is unjustified and unsupported at the outset. Given 14 the volume of Mainline IT flows at the major delivery points, and particularly the 15 Dawn Hub, compared to the volume of transactions that determines prices at those 16 locations, one would not expect TransCanada’s IT bid floor prices to have affected 17 downstream prices. In my opinion, the better hypothesis to test would have been the 18 reverse, that TransCanada’s IT bid floor prices responded to market conditions as 19 reflected in market prices. None of Dr. Cicchetti’s regressions permit one to reject 20 that hypothesis.
21 Second, all of Dr. Cicchetti’s regressions suffer from a statistical malady known as 22 “autocorrelation.” When I correct for this problem in his regression specification, I 23 discover that the corrected regressions are very weak, with low explanatory power 24 indicating that variables have likely been omitted from the analysis. Moreover, the 25 corrected regressions for his “counterfactual” Group C locations, such as California 26 that are geographically remote from the Mainline, have the economically implausible 27 result that Mainline IT bid floor prices affect daily spot prices in those locations. I 28 conclude that Dr. Cicchetti’s regressions are producing spurious results and are 29 entirely unreliable.
5 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1
2 3. The claim by Dr. Orans that TransCanada’s use of its pricing discretion led to 3 lower netbacks to NIT during the first three months after the implementation of 4 the RH-003-2011 Decision Model and that this should be a concern to WCSB 5 producers in the future is unproven. Nowhere in his evidence does Dr. Orans 6 discuss whether there have been previous episodes of lower netbacks, or whether 7 other potential transitory effects of the implementation of the Decision Model or 8 other market factors unrelated to its implementation, may have had an impact on 9 price spreads during this period.
10 The netback effects discussed by Dr. Orans are entirely spot-market related. When 11 one examines the forward market price spreads at the time one does not see any 12 evidence of a step change in the market’s expectations regarding netbacks caused by 13 pricing discretion. This is important because WCSB producers have the ability to 14 mitigate any risks associated with spot netback effects by hedging using forward 15 markets. Moreover, producers can ensure that they are “connected” to downstream 16 markets by contracting for FT capacity on the Mainline or other pipelines such as 17 Alliance/Vector, just as many shale-gas producers in the U.S are currently contracting 18 for new pipeline capacity to guarantee market access. WCSB producers have the 19 advantage that the pipeline is already in the ground and that capacity is available.
20
21 4. In my opinion, the regulatory principle of “toll stability” is really a concern 22 about uncertainty and variability in regulated FT tolls over time and not short- 23 term services, the value of which is derivative of inherently volatile short-term 24 market basis-differentials. I do not think it appropriate, nor is it efficient, for 25 pipeline regulators to attempt to “stabilize” the market value of short term 26 services, even if that were possible without regulating the entire secondary 27 market.
28 Similarly, CAPP’s proposal to require TransCanada to provide advance authorization 29 of monthly bid floors and to increase its short term data posting requirements simply
6 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 serves to limit TransCanada’s ability to compete effectively in the secondary market, 2 and it serves only to benefit other, unregulated secondary market participants. I see 3 no efficiency-enhancing reason to require TransCanada to adhere to such 4 requirements when other competitors have no such obligation.
5
6 5. I disagree with the view held by CAPP and Dr. Orans that TransCanada’s pricing 7 discretion should be limited because the Settlement terms imply a reduction in 8 business risk.
9 First, the mitigation of business risk was not the only reason that the Board 10 sanctioned increased pricing discretion in RH-003-2011. Pricing discretion was also 11 increased in order to encourage shippers to contract for FT service and to generate 12 revenues for the Mainline in order to keep tolls lower in the future and recover its 13 costs of service. Indeed, in addition to approving increased pricing discretion in the 14 Decision, the Board approved an ROE of 11.5% on 40% equity thickness. The 15 Settlement involves an agreed reduction of ROE from 11.5% to 10.1% which reflects 16 the fact that business risk has been reduced under the Settlement. Second, the 17 retention of the pricing discretion approved by the Board was a term of the Settlement 18 which the parties agreed in combination with other elements of the package, so 19 CAPP’s and Dr. Orans’ position must be that the allocation of risks negotiated in the 20 settlement is somehow unbalanced.
21 Q7. How is the remainder of your written evidence organized? 22 A7. I begin in Section II by discussing the concepts of “economic withholding” and 23 market power as they relate to the issues before the Board in this proceeding. Then I 24 turn to my analysis of the competition for TransCanada’s short term services in the 25 secondary market including diversions and the formation of prices in downstream 26 markets in Section III. In Section IV I evaluate Dr. Cicchetti’s regression analyses 27 regarding the impact of TransCanada’s IT bid floor prices on downstream prices. I 28 address Dr. Orans’ claims regarding the impact of pricing discretion on netbacks at 29 NIT in Section V. In Sections VI I address the issues of pricing discretion and toll
7 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 stability and transparency, and I conclude in Section VII with a discussion of the 2 pricing discretion and business risk issue.
3 II. “ECONOMIC WITHHOLDING” AND MARKET POWER
4 Q8. Dr. Orans, at page 11, lines 17-18 of his written evidence, says that 5 “TransCanada has used its newly authorized pricing flexibility for short term 6 services to engage in economic withholding of short term pipeline capacity, to 7 the detriment of market participants.” How do you interpret this accusation?
8 A8. Dr. Orans’ appears to be taking the extreme position that any time a pipeline is 9 offering short term services on a given path at a price above the short run value of 10 transportation on that path (as measured by the path’s spot market basis differential – 11 the difference between the spot commodity price at the delivery location and receipt 12 point), that it is engaging in “economic withholding” and that such a pricing policy is 13 economically inefficient and harmful to market participants. He reveals the essence 14 of his position at page 27, line 17 through page 28, line 1 of his evidence:
15 Efficient use of a pipeline occurs when short-term prices are set at 16 levels to maximize the efficient flow of gas from low to high cost 17 markets; a pipeline thus encourages economically efficient flows if it 18 offers interruptible transportation service at a cost (sic) between its 19 variable cost and the value of transportation in a competitive market. 20 Whenever short-term prices for transmission services are set at levels 21 that restrict efficient flows, the pipe is performing a form of economic 22 withholding. 23 I disagree with Dr. Orans’ apparent view that this is the economically meaningful 24 way to evaluate TransCanada’s use of its pricing discretion in this case.
25 Q9. On what basis do you disagree? 26 A9. First, the economic efficiency concept Dr. Orans is propounding is entirely short-term 27 in nature, as is clear from the citation to his evidence above. Taken at face value, the 28 implication of this concept is that even on days where the spot value of transportation 29 is below the regulated toll for FT service (on a 100 percent load factor basis), the 30 pipeline must discount its IT service below the full cost-recovery toll or be accused of
8 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 “economic withholding.” If this were the standard, then most gas pipelines in North 2 America would be engaged in “economic withholding” on many days of the year 3 when they elect not to discount IT service. U.S. pipeline regulation, in particular, 4 does not require any pipeline to discount its IT service below the maximum (FT) 5 tariff rate.
6 Second, there are very good long-term efficiency reasons not to impose such a short 7 run policy formulation on pipelines. For example, given the high fixed cost, low 8 variable cost economics of pipeline investments, it is well-recognized that 9 encouraging the long-term contracting of pipeline capacity helps manage the business 10 risks faced by pipeline investors, thus enabling economically efficient (long term) 11 investment decisions. In this context, long-run efficiency would be compromised if 12 existing pipelines that have suffered contract non-renewals and less than full 13 utilization were forced to price short term services so as to maximize short-term 14 throughput.
15 Q10. Does Dr. Orans appear to recognize the policy merits of using discretionary 16 service pricing to encourage FT contacting in his written evidence?
17 A10. Yes, he does, at page 47, lines 4-6:
18 To the extent a shipper’s needs are firm, I agree that discretionary 19 services should be priced in a way that to (sic) encourages that shipper 20 not to meet his needs with STFT and IT services.
21 Thus, in this passage Dr. Orans acknowledges that there may be policy reasons to use 22 “economic withholding of short term capacity,” to pursue long-term contracting 23 objectives. But this view does not seem to be consistently held in his evidence. 24 Later, on the last page of his evidence (p. 67, lines 1-3), Dr. Orans returns to this 25 concept in support of the CAPP recommendations regarding pricing discretion, and 26 he puts it somewhat differently:
27 …TransCanada has proven its intent to use discretionary 28 pricing flexibility as a blunt instrument to force shippers to purchase 29 firm contracts at the expense of other ratemaking objectives…
30 Dr. Orans does not explain or reconcile this apparent contradiction in his evidence.
9 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 Q11. How does “economic withholding,” as Dr. Orans appears to use the concept, 2 relate to the “physical withholding” of pipeline capacity?
3 A11. As near as I can tell Dr. Orans never testifies that TransCanada has physically 4 withheld short-term capacity from the market, but he suggests in several places in his 5 evidence either that TransCanada’s economic withholding has allowed “capacity on 6 the pipeline to remain empty when gas could otherwise be transported to the benefit 7 of all parties,”1 or that “when this (IT) service is not available—either due to physical 8 constraints or economic withholding—gas hub prices diverge….”2 Note that in the 9 last reference Dr. Orans appears to equate the effects of economic withholding to the 10 effects of physical constraints. Nowhere in his presentation does Dr. Orans present 11 any evidence that TransCanada’s alleged economic withholding has led to physical 12 constraints on the pipeline, and I am not aware of any evidence which would support 13 that position.
14 Q12. Where does the concept of “economic withholding” as distinct from “physical 15 withholding” originate?
16 A12. My understanding is that the concept of economic withholding originated in the 17 context of wholesale electricity markets. In the context of electricity, where the 18 wholesale market is now frequently supplied by unregulated power generators, it was 19 deemed important to understand whether particular generators were in a position to 20 raise the wholesale price by either “physically” withholding capacity from the market 21 (say, by shutting down for unnecessary maintenance) or by “economically” 22 withholding capacity (e.g., by bidding a price so high that a more expensive generator 23 in the supply “stack” would be called on to run instead.) In either case, if the effect of 24 the withholding would be to remove enough capacity from the wholesale market to 25 cause the price to rise, then it might be said that the generator was in a position to 26 exercise market power. Of course, in real life electricity markets, whether or not the 27 economic (or physical) withholding of capacity by a particular generator would cause
1 Written evidence of Ren Orans, p.38, lines 13-15. 2 Ibid., p.48, lines 12-14.
10 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 prices to rise would be a function of how “steep” the supply curve of power 2 generation is, and how close the generator is to being the marginal source of supply 3 such that if withdrawn its capacity would affect the market-clearing price for power.
4 Q13. What is the relevance of this electricity market background to the concept of 5 “economic withholding?”
6 A13. I raise it here because it helps explain that economic withholding is only 7 economically inefficient when the firm that is doing the withholding possesses and is 8 exercising market power. That is the context, and assumption being made, when it is 9 said that the economic withholding of electric generation capacity is considered 10 harmful or inefficient in wholesale generation markets.
11 More generally, if a firm does not possess market power over the product or service 12 involved, then it is a “price taker” and a decision to not offer its product (e.g., at a 13 discount to the prevailing market price) is of no consequence to the market or to 14 market efficiency. Firms in workably competitive markets refrain from making 15 products or services available every day. For example, the owner of a high-rise 16 condominium may refrain from making short-term apartment leases available at a 17 discount in order to retain the option to lease the apartments at higher rents for longer 18 terms. Perhaps Dr. Orans would say that the developer is not maximizing the short 19 term occupancy of the building and therefore “economically withholding” capacity 20 from the market, but it is of no consequence to the relevant real estate market, and it 21 is perfectly rational behavior to be expected in competitive markets.
22 Q14. Has the concept of “economic withholding” been raised in the context of 23 Canadian wholesale electricity markets?
24 A14. Yes, the Alberta Market Surveillance Administrator (MSA) discusses the concept of 25 economic withholding in its Enforcement Guidelines.3 In these Guidelines, the MSA 26 employs a competition law approach to assessing the efficiency effects of wholesale
3 Offer Behaviour Enforcement Guidelines For Alberta’s Wholesale Electricity Market, Alberta Market Surveillance Administrator, January 14, 2011.
11 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 market offer behavior. In the context of “unilateral effects,” (i.e., single firm behavior 2 as opposed to collusion) the MSA distinguishes between conduct it terms “extraction” 3 and “extension.” Extraction conduct “is aimed at capturing surplus (profits) that a 4 market participant has created independent of the conduct’s effect on rivals,” while 5 extension is “conduct that increase surplus (profits) by weakening or eliminating the 6 competitive constraints imposed by rivals.”4 This is consistent with the competition 7 law approach to market power, which I describe below, where it is only the creation, 8 preservation or enhancement of market power that is problematic. The MSA goes on 9 to say:
10 Conduct of the first kind [extraction] is considered competitive and 11 consequently would not result in enforcement action from the MSA. 12 Conduct of the second kind [extension] poses a concern and is likely to 13 be subject to investigation and potential enforcement action. 14 In relation to offer behaviour this means market participants are free to 15 pursue individually profit maximizing behaviour that does not impact 16 on rivals conduct. This would include strategies typically characterized 17 as economic withholding, which the MSA defines as: 18 “economic withholding” means offering available supply at a 19 sufficiently high price in excess of the supplier’s marginal costs and 20 opportunity costs so that it is not called on to run and where, as a 21 result, the pool price is raised. Such a strategy is only profitable for a 22 firm that benefits from the higher price in the market.[citation omitted] 23 Similarly, market participants are free to offer below marginal cost and 24 opportunity cost such that they receive dispatch and lower pool price. 25 The MSA accepts that both this and economic withholding are in 26 different circumstances rational profit maximizing behaviour…. 27 In a workably competitive market the use of both strategies is 28 disciplined by the actions of competitors such that there is no 29 expectation that a market participant can exert significant control over 30 market outcomes.5
31 The MSA, in other words, treats “economic withholding” as presumptively pro- 32 competitive, profit-maximizing behaviour, even in wholesale electricity markets 33 where such withholding may have the effect of raising the wholesale price, as long as 34 the behaviour does not impede or prevent a competitive response.
4 Ibid, p.9. 5 Ibid., pp. 9-10.
12 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 In this case, there is not only the competitive response that the secondary market 2 provides to TransCanada’s discretionary services, but there is a regulated recourse toll 3 for FT service available to shippers seeking an alternative to relying on discretionary 4 services to meet firm requirements. Shippers who choose to contract for FT service, 5 and who then have transportation services to sell on the secondary market, become 6 additional direct competitors to Mainline discretionary services. As I describe later, 7 and as detailed in the company’s Additional Written and Reply evidence, this 8 competitive response has been experienced since the implementation of RH-003- 9 2011.6
10 Q15. You have suggested that “economic withholding” is only potentially an issue if 11 the firm in question has market power. What is market power?
12 A15. “Market power” has traditionally been defined by economists, legal scholars and by 13 North American courts as the power to raise price above the competitive level 14 profitably, for a non-transitory period of time, and not lose so many sales in the 15 process that the price increase must be rescinded.7 There are three key aspects to this 16 definition. The first is that to have market power a firm must have the power to raise 17 price above the competitive level profitably. The second condition is that the firm 18 must be able to raise price for a non-transitory period of time, i.e., it must be durable 19 or persistent. The third aspect of this definition has to do with the loss of sales either 20 through substitution or competitive entry – market power cannot be sustained in the 21 face of substitution away from the firm’s services or in the face of entry or potential 22 entry. Sometimes market power is referred to as the power to raise price profitably 23 and preclude entry. My understanding is that under U.S. and Canadian competition 24 law, neither the possession of market power, nor the exercise of market power to 25 charge prices above the competitive level is unlawful in itself. What I understand to
6 See also Response of CAPP to MAS IR-06. 7 William M. Landes and Richard A. Posner, “Market Power in Antitrust Cases,” Harvard Law Review, Vol. 94, No. 5, March 1981. See also, Canadian Competition Bureau, Abuse of Dominance Guidelines, September 2012, Sec 2.3, which states that “…in a general sense, market power is the ability of a single firm or group of firms to profitably raise prices above the competitive level, or other elements of competition such as quality, choice, service or innovation below the competitive level, for a significant period of time.”
13 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 be unlawful is conduct designed to create, preserve or enhance market power where 2 the conduct causes harm to competition in the relevant market.8 Harm to competition 3 means harm to the market as a whole, and not harm to a single or even a few 4 competitors. My understanding is that conduct designed to create, preserve or 5 enhance market power is lawful if there is a legitimate business justification for doing 6 so, particularly if the conduct has the effect of enhancing efficiency either in the short 7 run or in the long run.9
8 Q16. How does one typically assess whether a firm has market power? 9 A16. Screening tools have been developed over the years to assist in determining whether a 10 firm possesses market power. Typically this involves the definition of relevant 11 product and geographic markets for the product or service in question, and then the 12 identification of substitutes for, and alternative suppliers of, the product in question. 13 Armed with this information one can then assess the level of concentration in, and the 14 particular firm’s position in that market and market share. Because all firms in the 15 real world have some degree of market power, and the textbook example of “perfect 16 competition” is never observed, the inference one draws from concentration and 17 market share statistics is necessarily judgmental and various quantitative guidelines 18 have been developed to identify when a market may be more or less at risk for 19 anticompetitive conduct.
20 Q17. Have any of the intervenor witnesses assessed whether TransCanada has market 21 power with respect to its discretionary services in any relevant market?
22 A17. No, they have not, and it is a serious deficiency in their evidence. Their presentations 23 rely not on an analysis of the competition faced by TransCanada’s discretionary 24 services and substitutes for those services, but instead they rely indirectly on various 25 statistical analyses of downstream and upstream commodity prices, by which they 26 attempt to draw the inference that TransCanada’s discretionary pricing has caused 27 prices to increase in downstream markets, or decrease in upstream markets, during
8 Canadian Competition Bureau, Abuse of Dominance Guidelines, September 2012, Subsection 4. 9 Ibid., Subsection 3.2.
14 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 various time periods since the implementation of RH-003-2011. None of these 2 statistical analyses are sufficient to provide any evidence as to whether TransCanada 3 possesses or has exercised market power with respect to discretionary services. I will 4 evaluate these statistical analyses in detail later in my reply evidence and will explain 5 how Centra and Dr. Cicchetti, in particular, have confused the observation of 6 correlation in prices with causation.
7 Q18. Have any of the intervenor witnesses demonstrated that TransCanada’s use of 8 its pricing discretion has harmed competition in the secondary market, or that it 9 lacks a legitimate business justification?
10 A18. No they have not. While Dr. Orans speculates at pages 32 and 33 of his evidence that 11 the depth of the secondary market may not always constrain TransCanada’s pricing 12 behavior in the future, he acknowledges that the increase in FT contracting secured by 13 TransCanada when the RH-003-2011 Decision Model was implemented actually 14 increased the secondary market competition for Mainline discretionary services.10 15 Dr. Orans’ unfounded concerns about the uncertainty of the secondary market is not a 16 showing that TransCanada’s use of pricing discretion has driven competitors from the 17 market so as to create, enhance or preserve market power. Finally, there has been no 18 showing in any of the intervenor evidence that TransCanada’s use of its pricing 19 discretion lacks a legitimate business justification, including its use to encourage 20 increased FT contracting, to generate revenues for Mainline cost recovery and to 21 lower future Mainline tolls. All of these were reasons the Board gave in Decision 22 RH-003-2011 justifying the increased pricing discretion granted to TransCanada.
23 Q19. Have you evaluated whether TransCanada has market power with respect to the 24 provision of discretionary services on the Mainline? If so, what have you 25 concluded?
26 A19. Yes I have, and in my opinion TransCanada does not possess market power with 27 respect to the provision of IT or STFT services on the Mainline. As a result, there is
10 Orans Written Evidence, page 33, line22 – page 34, line 8.
15 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 no reason for the Board to limit in any way the pricing discretion that it authorized in 2 Decision RH-003-2011. Nothing has changed since the implementation of RH-003- 3 2011 to indicate that TransCanada faces any less competition for Mainline 4 discretionary services than the Board previously recognized as constraining 5 TransCanada’s use of such discretion. Moreover, the option to contract with the 6 Mainline at rate-regulated FT tolls is still available as a recourse to shippers who 7 desire firm service as an alternative to Mainline STFT and IT services.11 Over and 8 above the competition faced by TransCanada’s discretionary services in the 9 secondary market, the ability of shippers to contract at a rate regulated toll for FT 10 service acts as a further constraint on any market power it is alleged the Mainline 11 might possess in respect of discretionary services.
12 III. SECONDARY MARKET COMPETITION AND DOWNSTREAM PRICES
13 Q20. What is meant by the use of the terms “primary market” and “secondary 14 market” in the context of the regulation of gas pipelines?
15 A20. When parties and regulators refer to the “primary market,” they are usually referring 16 to the market for firm transportation service sold by the pipeline company, typically 17 under long terms contracts. The term “secondary market” then usually refers to the 18 market in which primary pipeline capacity is “released” or resold by shippers on a 19 shorter-term basis (either as pipeline capacity itself or “bundled” as delivered gas 20 supply using their pipeline capacity). The pipeline’s own primary short-term firm 21 and interruptible capacity that is uncontracted and available for discretionary services 22 then competes directly in the secondary market as an alternative to the use of the 23 capacity held by others. By encouraging the development of secondary markets,
11 When FT capacity is not available because the path on the pipeline is fully subscribed, shippers seeking firm capacity can participate in the pipeline’s open season process to contract for new or expanded capacity. Note that when FT capacity is fully subscribed, the pipeline itself will tend to have less capacity available itself to sell as discretionary services, and more third-party capacity is potentially available for sale in the secondary market in competition with the pipeline’s services. See also the Decision of the Ontario Energy Board in EB-2005-0551 quoted in Answer 23 below.
16 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 regulators introduce competition among and between the holders of primary capacity 2 and the pipeline itself (if the pipeline has capacity available for sale).
3 Q21. How do you define the “product market” in this case? 4 A21. The product market relevant to Mainline discretionary services includes the provision 5 of short term transportation services on pipelines and substitutes for that service, 6 including the pipeline and storage capacity that is bundled with the gas commodity 7 for delivery in the secondary market.
8 Q22. How do you define the “geographic market” in this case? 9 A22. The geographic market in this case is at least as large as the area served by the major 10 hubs (particularly Dawn, Ontario) and other delivery points on the Mainline. I say “at 11 least as large” because we know that there is competition among pipelines and gas 12 suppliers to serve significant geographic areas that extend well beyond the immediate 13 proximity of the interconnected hubs.
14 Q23. Have any Provincial regulators defined the geographic market as it relates to the 15 secondary market in which the Mainline’s discretionary services compete?
16 A23. Yes. In 2006, after an extensive evidentiary proceeding, the Ontario Energy Board 17 decided to forbear from regulation of Union Gas’ storage and transportation services 18 to non-franchise customers in Ontario. It defined the geographic market in this way:
19 The Board concludes that the geographic market extends beyond 20 Ontario, even though there is a lack of uncontracted firm pipeline 21 capacity. The Board is satisfied that there are reasonable alternative 22 means for storage customers in Ontario to access a broad market area. 23 This can be done through the secondary markets or through 24 participating in open seasons for new firm capacity. The Board is also 25 satisfied that there is access to suitable substitutes for Ontario storage 26 available in the broader market because there is direct evidence that 27 the alternatives are considered and are being used.12
12 Ontario Energy Board, Decision With Reasons, EB-2005-0551, 7 November 2006, p.37. (emphasis added)
17 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 While my reason for raising this decision has to do with the OEB’s view as to the 2 boundaries of the geographic market, note that in this case the product market 3 involved storage services and not just short term pipeline transportation services. 4 Storage services obviously involve withdrawals during periods of peak demand, and 5 the OEB found the geographic market to be broad and competitive even if 6 uncontracted firm pipeline capacity was not available. This is because the OEB 7 viewed the participation of storage customers in open seasons for new pipeline 8 capacity as a “reasonable alternative means” to access the broader geographic market, 9 as is evident from the quoted passage above.
10 Q24. How concentrated is the market for short term services in the geographic 11 market defined as the Dawn Hub?
12 A24. It is very unconcentrated. The risk of any one holder of pipeline or storage 13 withdrawal capacity serving the Dawn Hub exercising market power, including 14 TransCanada with respect to Mainline discretionary services, is extremely low. I 15 have evaluated the concentration of pipeline capacity holders to the Dawn area and 16 storage withdrawal capacity holders at Dawn using the Herfindahl-Hirschman Index 17 (HHI).13 I calculate an HHI of 434, which indicates an unconcentrated market and 18 little to no risk that market power could be exercised by transportation and storage 19 withdrawal capacity holders that serve the Dawn market. My analysis reflects an 20 unconcentrated commodity market at Dawn that includes the potential sellers who 21 hold either pipeline capacity to Dawn or storage withdrawal capacity at Dawn. 22 23 In performing this evaluation, I have used information regarding current shipper 24 capacity holdings (as of July 1, 2014) on Vector Pipeline (to St. Clair) and Panhandle
13 The HHI is a preferred measure of concentration by competition law enforcement agencies in the U.S. and Canada. It is defined as the sum of the squared market shares of the competitors in the market. Based on their experience, the Agencies generally classify markets into three types: • Unconcentrated Markets: HHI below 1500 • Moderately Concentrated Markets: HHI between 1500 and 2500 • Highly Concentrated Markets: HHI above 2500 See U.S. Department of Justice & Federal Trade Commission, Horizontal Merger Guidelines, Sec. 5.2, (2010).
18 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 Eastern (to Ojibway). This information is available from the “Index of Customer” 2 filings U.S. pipelines make to the U.S. FERC. I have also included current long-term 3 shipper capacity holdings on the Mainline from St. Clair to Union SWDA.14 I have 4 also assumed that TransCanada is effectively the holder of an additional 304,000 5 Dth/d of unsubscribed firm pipeline capacity to Dawn, capacity that it can sell on a 6 short-term or long-term basis. I account for storage withdrawal capacity holdings at 7 Dawn using Union Gas’ Index of Customers for ex-franchise storage services on its 8 system.
9 My calculation of the resulting HHI is included as Attachment B. As shown in 10 Attachment B, TransCanada’s share of transportation capacity to Dawn, and storage 11 withdrawal capacity at Dawn, is only 6.0%.
12 Q25. What about the Emerson and Iroquois delivery points on the Mainline? Is it 13 possible to develop measures of concentration for those points as you have done 14 for Dawn?
15 A25. Unlike the Dawn hub, delivery points like Emerson and Iroquois are really points at 16 which gas is “handed-off” for downstream delivery to the major markets of the 17 Midwest U.S. (in the case of Emerson) and New York/New England (in the case of 18 Iroquois). While gas commodity transactions occur at these points, the nature of the 19 competition is different than at the Dawn hub. In the case of Emerson and Iroquois, 20 the competition for TransCanada’s discretionary services primarily derives from the 21 ability of FT capacity holders on the Mainline to sell transportation capacity (by itself 22 or bundled with gas for delivery) in the secondary market.15 Since much of the 23 competition at these points to TransCanada’s short term services may come from FT 24 diversions, it is not straight-forward to construct an HHI at these points based on 25 capacity holdings at those locations. Such a measure, for example, would fail to
14 It is my understanding that the deliveries on Great Lakes to St. Clair is transported to Dawn (Union SWDA) via the TransCanada Mainline segment between St. Clair and Union SWDA. Hence, I do not include the pipeline capacity holdings on Great Lakes into St. Clair to avoid double counting. 15 If flow reverses at these points, which I understand happens at Emerson and could happen in the future at Iroquois, then capacity holdings on Great Lakes and the Iroquois Pipeline would also become relevant.
19 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 capture the potential for diversions. In the case of these points, I look primarily to 2 direct evidence of competition for short term services in the volume flow data at these 3 points since the implementation of RH-003-2011, as discussed below.
4 Q26. Are there potentially separate geographic markets served by the Mainline that 5 are associated with more isolated Mainline delivery points than the Dawn hub, 6 Emerson or Iroquois, for example?
7 A26. Potentially, yes, but the intervenor evidence seems to be focused on the major 8 delivery points, such as Dawn, Iroquois and Emerson.16 I’m not aware of any 9 suggestion that TransCanada has been using its pricing discretion to “economically 10 withhold” capacity from smaller delivery points, or that customers who have no 11 alternative but to transact at such points should be relying on STFT or IT service to 12 meet firm year-round requirements. At such points, customers continue to have the 13 ability to contract for FT capacity at regulated tolls. Delivery points on the Mainline 14 are also tied together competitively through the use of diversions and alternative 15 receipt points that permit holders of FT capacity at primary points to compete in the 16 secondary market by diverting gas to these other points. In any case, the principal 17 issue in this case as framed by the intervenors appears to be the effect of pricing 18 discretion on major downstream commodity markets (in the case of Centra’s 19 evidence), and resulting netbacks at NIT (in the case of CAPP).
20 Q27. Are there other quantitative measures one can use to evaluate the 21 competitiveness of the secondary market in the Dawn Hub area and at other 22 major Mainline delivery points?
23 A27. Yes, one can look to various measures of transaction volume, participation and 24 liquidity to assist in gauging the extent of competition in the area of the Dawn hub 25 and other Mainline delivery points. For example, one measure of liquidity is the 26 number and volume of transactions in the market. While total quantities are not 27 publically available, Platts (the trade publication organization that collects and
16 Dr. Cicchetti uses Emerson as a “proxy” for the Centram MDA, for example, because he says there are too few transactions at Centram for it to be a meaningful pricing location.
20 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 publishes the various hub pricing data relied on by market participants) does publish 2 the number and volume of fixed price transactions that go into its index price every 3 day at each hub. Figure 1 below shows the transaction volume at Dawn for every day 4 from the beginning of 2011 to the present. The average daily volume over this period 5 was 970 MMcf/day. On this measure Dawn was the third largest hub in North 6 American behind only NIT/AECO and the Chicago hub.17 Figure 2 shows the 7 number of transactions each day that contributed to the Platts index. On average over 8 this period there were 127 fixed price transactions per day at Dawn that made up the 9 Platt’s index (again 3rd in rank among North American hubs). It is important to 10 recognize that these data involve only fixed price transactions. Many more 11 transactions would be entered into at the hub which would be indexed to the 12 published price. There appears to be very little change in these transaction figures for 13 Dawn post-July 2013, another indication that TransCanada’s pricing discretion has 14 done nothing to affect the very high liquidity of this important market hub connected 15 to the Mainline. Over the same time period (from July 1, 2013 through June 30, 16 2014), the volume of TransCanada STFT and IT transactions were 75 MMcf/day, and 17 69 MMcf/day respectively, showing how small TransCanada’s discretionary service 18 activity has been relative to the size and liquidity of the Dawn hub.
19 Figures 3 and 4 depict the same Platts quantity and number of fixed price transactions 20 data for Emerson, while Figures 5 and 6 report transaction volumes and numbers for 21 deliveries to Iroquois. While obviously not as liquid as the Dawn hub, these locations 22 have sufficient volumes and number of fixed price transactions for Platts to compute 23 a price index at these locations. As I will show later, at these two locations 24 TransCanada faces significant competition from FT capacity holders exercising their 25 diversion and alternate receipt point rights, which are part of the secondary market. 26 This was particularly true at Iroquois during the winter of 2013/2014 when diversions
17 Interestingly, on this measure Henry Hub has fallen to eleventh place among hubs in North America. (also eleventh on the basis of number of transactions) Some analysts have suggested that this may be due to the growing importance of Marcellus shale relative to Gulf Coast gas supplies. See “Supply Shifts Challenge Henry Hub as Benchmark,” Platts Gas Daily, August 5, 2013.
21 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 (which have scheduling priority over Mainline IT services) effectively crowded out 2 IT service.
Figure 1 Transaction Volume at Dawn Hub Jan 1, 2011 – July 25, 2014
Figure 2 Number of Transactions at Dawn Hub Jan 1, 2011 – July 25, 2014
22 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
Figure 3 Transaction Volume at Emerson Jan 1, 2011 – July 25, 2014
Figure 4 Number of Transactions at Emerson Jan 1, 2011 – July 25, 2014
23 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
Figure 5 Transaction Volume at Iroquois Jan 1, 2011 – July 25, 2014
Figure 6 Number of Transactions at Iroquois Jan 1, 2011 – July 25, 2014
1 In Figure 7 I have plotted the pipeline flow volumes delivered into the Dawn Hub by 2 month from January 2011 to the present on the TransCanada, Vector and Union
24 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 pipelines (the last of which also includes receipts from the MichCon, Bluewater and 2 Panhandle Eastern pipeline systems). Transactions using the capacity on all of these 3 pipelines are a source of liquidity for the Dawn Hub. From July 2013 through mid- 4 August 2014, total flows into Dawn on these three pipelines averaged 2.46 Bcf/day. 5 Of this total, the Mainline’s share (including capacity held by FT shippers) was 0.75 6 Bcf/day, or 30.4 percent. TransCanada’s combined STFT and IT volumes averaged 7 0.14 Bcf/day into Dawn over the post-July 2013 period,18 representing only 5.8% of 8 the total.
Figure 7 Average Daily Deliveries to Dawn Hub by Pipeline and Month Jan 1, 2011 - Aug 15, 2014
9 Q28. Are there other measures of secondary market competition that relate to 10 discretionary service volumes on the Mainline?
11 A28. Yes. Based on the Mainline’s experience since the July 2013 implementation of 12 pricing discretion, one can observe how FT shippers’ use of diversions has been a 13 significant and effective competitor to Mainline IT services, particularly at points like
18 Based on authorized nominations from the period July 1, 2013 – June 30, 2014, produced by TransCanada in response to NEB Information Request 2.13, Attachment 1.
25 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 Iroquois. Figures 8, 9, and 10 below illustrate this point in the case of flows to the 2 Dawn Hub, Emerson, and Iroquois, respectively. For all three delivery points, 3 TransCanada’s Mainline IT services played a significant role (especially to Emerson) 4 before the implementation of Decision RH-003-2011. Since the Decision went into 5 effect, however, the exercise of diversion rights has increased dramatically for 6 nominations to these locations, effectively competing with and even “crowding out” 7 IT nominations. This is especially true for deliveries to Iroquois, where FT Diversion 8 nominations have effectively replaced IT since July 2013.
Figure 8 Mainline FT Diversion and IT Nominations to Dawn Hub January 2013 – June 2014
700 July 1, 2013
600
500
400 IT 300 FT Diversions
200
100 Authorized Nominations (MMcf/d)
0
Source: TransCanada response to NEB Information Request 2.13, Attachment 1.
26 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
Figure 9 Mainline FT Diversion and IT Nominations to Emerson January 2013 – June 2014
1,200 July 1, 2013
1,000
800
600 IT
400
FT Diversions 200 Authorized Nominations (MMcf/d)
0
Source: TransCanada response to NEB Information Request 2.13, Attachment 1. Figure 10 Mainline FT Diversion and IT Nominations to Iroquois January 2013 – June 2014
350 July 1, 2013
300
250 IT
200 FT Diversions
150
100
50 Authorized Nominations (MMcf/d)
0
Source: TransCanada response to NEB Information Request 2.13, Attachment 1.
1 Q29. Can one infer anything about the competition for Mainline discretionary 2 services by considering TransCanada’s bid floor pricing experience since July 3 2013?
27 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 A29. Yes, and the Company’s evidence and responses to IR’s provide further examples of 2 how competition in the secondary market served to constrain its discretionary pricing 3 over the past year. For example, in its response to NEB IR 1.22 (a), TransCanada 4 indicates that the increased contracting for FT on the Mainline since the 5 implementation of RH-003-2011 contributed to a substantial increase in the use of 6 diversions by shippers in competition with discretionary services. Diversions 7 increased from 11% of system demand in the year prior to July 2013 to 21% of 8 system demand from July 1, 2013 to March 31, 2014. During this same period, 9 “shippers contracted above the minimum bid floor approximately 49% of the time for 10 STFT service and 18% of the time for IT service.”
11 Q30. Dr. Orans claims that he has observed instances during the severe winter of 12 2013/2014 when TransCanada’s bid floors for IT were set above the spot price 13 basis differential yet there was a portion of the Mainline that remained 14 unutilized. How do you respond to this?
15 A30. Yes, I have reviewed Dr. Orans discussion of this experience. In my opinion Dr. 16 Orans fails to appreciate the kinds of day-to-day frictions that occur in markets of this 17 sort (and perhaps unlike the electricity markets where most of his experience lies). 18 Such frictions can become more acute when prices become volatile due to weather 19 and potential capacity constraints on parts of the network (including constraints on 20 other pipelines and with respect to storage capacity). One cannot expect that all 21 segments of any pipeline network will be running full when constraints begin to occur 22 on some parts of the network, or are anticipated to occur. With respect to 23 TransCanada’s IT bid floors, because it establishes these floors in advance of market 24 outcomes, it must attempt to anticipate where the market is going to move on any 25 given day. In most trading markets, even the most liquid markets, one observes a 26 phenomenon known as the “bid-ask spread” where bids to sell differ significantly 27 from offers to buy, particularly when markets are volatile. One would expect offer 28 and bid prices posted in advance to differ necessarily from where the market 29 ultimately lands in respect of an “average daily price.” Other market participants, 30 such as firm capacity holders, also have to anticipate how much capacity they plan to
28 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 nominate -- flows which can change at the last minute depending on what happens to 2 demand. Transactional friction alone may prevent all of that capacity from being 3 taken up before the close of trading and delivery. Thus, one should expect to see 4 instances where TransCanada’s IT bid floors ended up exceeding the average basis 5 differential after the fact, and that the pipeline was also not fully utilized, particularly 6 when the market was exhibiting large day-to-day variations in demand and prices, as 7 it did last winter.
8 Q31. What do you conclude from all of this evidence regarding the competition faced 9 by Mainline discretionary services in downstream geographic market(s)?
10 A31. I conclude that TransCanada does not possess market power with respect to the 11 provision of Mainline discretionary services, and that it was not in a position to 12 exercise market power in commodity markets served by the Mainline by virtue of its 13 STFT and IT pricing discretion. 14
15 IV. DR. CICCHETTI’S REGRESSIONS
16 Q32. What conclusions does Dr. Cicchetti reach in his written evidence regarding 17 TransCanada’s pricing discretion?
18 A32. Dr. Cicchetti reaches two conclusions regarding TransCanada’s Mainline pricing 19 discretion: a) that “increased short term bid floor prices has resulted in non- 20 competitive commodity prices and anomalous pricing outcomes”; and b) that “effects 21 on consumers are significant enough to consider introducing constraints on the degree 22 of pricing discretion authorized.”19 When Dr. Cicchetti refers to “non-competitive 23 commodity prices” and “anomalous pricing outcomes” he is referring to the 24 commodity price of natural gas at various downstream locations either directly 25 connected to the Mainline, such as Dawn, or “indirectly interconnected” locations 26 such as Chicago.
27 Q33. What does Dr. Cicchetti say is the basis for these conclusions?
19 Written Evidence of Dr. Charles Cicchetti, p.26, lines 5-11.
29 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 A33. Dr. Cicchetti says that he relies on a series of regression analyses, which he says 2 confirm his “hypothesis that higher IT bid floor prices on the TCPL Mainline would 3 cause commodity prices to increase at interconnected hubs.”20
4 Q34. What is the basis for Dr. Cicchetti’s “hypothesis” that higher IT bid floor prices 5 caused increased commodity prices at interconnected hubs?
6 A34. It is not quite clear from his written evidence what the basis is for this hypothesis, 7 other than Dr. Cicchetti’s observation that there were some “highly unusual and 8 seemingly non-competitive commodity price movements at hubs connected to TCPL 9 this past winter,”21 and that this was the first winter during which TransCanada was 10 authorized to use its pricing discretion for IT services.
11 Q35. Do you believe that this is a reasonable hypothesis to make? 12 A35. No, not at all, and using the Dawn Hub as an example, the share of transaction 13 volume at Dawn last winter represented by Mainline IT volume is miniscule as 14 compared to the total volume of competing transactions that form commodity prices 15 at Dawn on any given day. On the basis of this fact alone (as well as the other 16 statistics regarding Dawn liquidity and competitiveness I have described above), the 17 likelihood that one should expect Mainline IT bid floor prices to affect the 18 commodity price of gas at Dawn is near zero. The better hypothesis would be the 19 reverse, that downstream commodity prices are affecting the IT bid floors established 20 by TransCanada. In other words, that TransCanada’s IT pricing behavior was 21 following the market, not determining the market. Indeed, this is exactly what the 22 Board expected TransCanada to do when it authorized the existing pricing discretion:
23 We recognize that giving TransCanada the flexibility to increase and 24 decrease bid floors may give it the opportunity to charge very high 25 tolls in certain markets and at certain times, for example, during 26 significant weather events. We are of the view, however, that it is 27 important to provide TransCanada with the necessary tools to capture
20 Ibid., p.35, lines 4-6. 21 Ibid, p.28, lines 28-30.
30 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 market opportunities, if and when they arise, and to recover costs 2 associated with its system from those who use it.22
3 This passage indicates that it was the Board’s hypothesis that TransCanada would set 4 its bid floors to capture market opportunities, such as occasioned by cold weather 5 events. There is nothing in Dr. Cicchetti’s casual observations of price movements 6 last winter to suggest that TransCanada has been “creating” market opportunities as 7 opposed to “capturing” them, as the Board expected it to do.
8 Dr. Cicchetti’s further extension of his hypothesis to the Chicago hub and other 9 markets “indirectly interconnected” to the Mainline has even less rationale.
10 Q36. Does Dr. Cicchetti explore whether there might be explanations other than 11 TransCanada’s pricing discretion for why commodity prices in northern and 12 eastern markets exhibited “anomalous price surges” last winter?
13 A36. No. In his evidence Dr. Cicchetti does not explore such alternative explanations to 14 any great extent and he certainly does not do so in his regression analyses reported in 15 his written evidence that I review below.23 It is well established that the price spikes 16 that were experienced in northern and eastern markets last winter were due to a 17 combination of factors, including weather events and hence demand increases, 18 pipeline capacity constraints in parts of the region, and low and potentially depleted 19 storage inventories. These alternative explanations of what happened in the market 20 last winter are also described in some detail by Union Gas, Enbridge Gas Distribution 21 and Gaz Metro in their response to NEB IR 1.2. It has come to my attention that Dr. 22 Cicchetti has very recently provided the results of additional regressions, in which he 23 purports to have incorporated a “winter index” variable. I will access this newly 24 provided analysis in due course. However, the conclusions in Dr. Cicchetti’s evidence
22 Decision RH-003-2011, p.126. 23 For example, in response to MAS IR’s 1.31 and 1.45 Dr. Cicchetti acknowledges that he has not considered the impact of pipeline bottlenecks in his analysis beyond the correlation of prices with the Henry Hub included in his regressions. See also the response of Centra to MAS IR 1.40. I understand that just prior to the completion of this reply evidence, Dr. Cicchetti has submitted some additional regression results that attempt to account for weather effects. I have not yet had the opportunity to review these additional regressions.
31 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 were based on the regression analyses cited therein, regressions which I believe 2 produced spurious and unreliable results.
3 Q37. Have you seen any other authoritative summaries of what happened in gas 4 markets last winter?
5 A37. In a June 2014 report on natural gas storage challenges, the U.S. Department of 6 Energy, Energy Information Administration (EIA) addressed the role of both acute 7 and sustained colder-than-normal weather events last winter in straining the 8 supply/demand balance for natural gas, driving storage levels to their lowest since 9 2003 and leading to price spikes, including record high prices in the Northeast and 10 Mid-Atlantic.24
11 The report details the confluence of events that led demand to consistently outpace 12 supply over the five month period from November 2013 through March 2014, as 13 illustrated in Figure 11.
14 Figure 11
15 The EIA emphasizes that “every major region of the contiguous United States 16 experienced waves of significantly colder-than-normal temperatures,” and explains
24 U.S. Energy Information Administration, Issues and Trends: Natural Gas. Record winter withdrawals create summer storage challenges, June 12, 2014.
32 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 how the extreme cold contributed to 9% growth in U.S. natural gas consumption 2 compared to the previous winter. The five months from November 2013 through 3 March 2014 all set same month U.S natural gas consumption records, and included…
4 • The top six overall natural gas consumption days going back to 5 January 2005. 6 • Six of the top ten residential/commercial consumption days dating 7 back to January 2005. 8 • The top nine industrial consumption days going back to January 2005.
9 The EIA concludes that the weather-driven demand increases were the primary cause 10 of tightening in U.S. natural gas markets and the attendant price spikes. They further 11 explain that while the higher prices created incentives for dry gas production to 12 increase 3% from 2012-13 winter season levels, the 0.3 Bcf/d increase in supply fell 13 far short of the 7.5 Bcf/d consumption increase.25 Furthermore, they conclude that the 14 weather played a primary role in the increased volatility of prices.
15 “Prices responded to sharp daily demand increases and continued 16 drawdowns from storage by increasing significantly in the northeastern 17 United States, as occurred [in the winter of 2012-2013]. However 18 these price spikes were not confined to the Northeast; they also 19 occurred at trading hubs serving consumers in the central and western 20 United States.”
21 It is clear that the U.S. Energy Information Administration attributes the recent winter 22 price spikes to a tighter supply/demand balance caused by sustained extreme cold 23 weather across the North American continent, and to the attendant record natural gas 24 consumption in the face of constraints on storage and pipeline delivery capacity.26 25 The EIA’s findings confirm that Dr. Cicchetti’s hypothesis that TransCanada’s 26 pricing discretion caused the price spikes last winter is baseless. There is no mention 27 in the report of the use by TransCanada of its pricing discretion in the provision of 28 short-term services on the Mainline.
25 Ibid. 26 In contrast to the EIA’s analysis, see Centra’s response to MAS IR 1.40 c)-f).
33 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 Q38. How does Dr. Cicchetti test his hypothesis that TransCanada’s IT bid floor 2 prices caused the price spikes?
3 A38. Dr. Cicchetti constructs a series of regression analyses for the purpose of testing his 4 hypotheses that TransCanada’s IT bid floor prices affected commodity prices at 5 various pricing locations. He selects 14 pricing locations and divides them into three 6 groups: Group A includes pricing locations that are directly connected to the Mainline 7 (Centram MDA,27 Emerson, Dawn and Iroquois); Group B includes pricing locations 8 “indirectly connected” to the Mainline (Niagara, Chicago, Dominion South and Mich- 9 Con); and Group C includes pricing locations where Dr. Cicchetti’s hypothesis is that 10 there should be no effect of TransCanada’s bid floor prices (AECO/NIT28, PG&E 11 City Gate, San Juan, Permian, SoCal City Gate, and Leidy).
12 Q39. Do you have any concerns about Dr. Cicchetti’s selection of these locations, or 13 the groups he created?
14 A39. It is not clear to me from Dr. Cicchetti’s presentation whether he selected these three 15 groups before he ran his regressions, or after examining his results. But since he says 16 he is testing the hypotheses that Groups A and B would be affected by TransCanada’s 17 IT bid floors, while Group C (as a counterfactual) would not be affected, I assume 18 that he chose the groupings in advance – which would be the only correct scientific 19 method to employ. It is also unclear to me where Dr. Cicchetti draws the line with 20 respect to “indirectly connected” locations. There is no part of the major interstate or 21 inter-provincial pipeline system in North America that is not “indirectly 22 interconnected” in some fashion. I am also somewhat concerned that he left out, 23 without explanation, some key pricing locations in the Northeast US (such as 24 Algonquin and Transco Zone 6 NY) which were affected by significant price spikes 25 last winter, in part due to pipeline capacity constraints into the Northeast, and which 26 had little or nothing to do with the Mainline. As a result, his “counterfactual” Group
27 Dr. Cicchetti says that he uses the Emerson pricing location as a proxy for the Centram MDA due to the lack of reported market prices Centram MDA. 28 Note that by including AECO/NIT in Group C, Dr. Cicchetti holds a view apparently contrary to that of Dr. Orans with respect to the impact of TransCanada’s discretionary pricing on upstream netbacks.
34 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 C includes only points, with the possible exception of Leidy, PA (located in the heart 2 of the Marcellus production area), which were not affected to anywhere near the same 3 extent by last winter’s extreme weather in the Northern U.S.
4 Q40. What regression specification does Dr. Cicchetti employ and what does he claim 5 about his results?
6 A40. Dr. Cicchetti uses as his independent variable (i.e., the variable to be “explained”) the 7 natural log of the reported daily spot price at the pricing location he is testing. The 8 dependent variables (or “explanatory variables”) differ across his regressions, but in 9 all cases he employs the natural log of the Henry Hub spot price as one explanatory 10 variable, and a “winter dummy” variable that attempts to differentiate winter days 11 from other days.29 Then, depending on which locations he is testing, he uses the 12 natural log of the TransCanada IT bid floor prices at Emerson, Centram, Dawn or 13 Iroquois as an additional explanatory variable. If, on the basis of his regressions, the 14 coefficient on the IT bid floor prices is positive and statistically significant, then Dr. 15 Cicchetti says that at those locations TransCanada’s IT bid floor prices are affecting 16 downstream prices. Based on these specifications, he claims that his explanatory 17 variables explain about 80 percent of the variation in prices, and that the results 18 permit him to reject the hypothesis that TransCanada’s IT bid floor prices did not 19 affect prices at the Group A and B locations. Further, he concludes that based on the 20 results for Group C, he can reject the hypothesis that TransCanada’s IT bid floor 21 prices at Emerson affected prices at the hubs “where TCPL does not deliver natural 22 gas.”30
23 Q41. Have you been able to replicate Dr. Cicchetti’s regression results using his 24 underlying data?
25 A41. I have replicated Dr. Cicchetti’s results using the data he provided in TCPL/Centra IR 26 1.11, attachment (b), which appears to contain the calculated values for the variables
29 Dr. Cicchetti also includes an “Emerson event” dummy in certain of his regressions to test whether certain days when the Emerson IT bid floor price was unusually high were of significance. 30 Cicchetti Written Evidence at p. 36, line 16.
35 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 he defined as inputs to his regressions. However, at this juncture, and without access 2 to Dr. Cicchetti’s electronic workpapers, I cannot verify that these values are 3 consistent with or correctly calculated from the raw hub price and IT bid floor data he 4 provided in response to TCPL/Centra IR 1.8. Nevertheless, I believe that the results I 5 have obtained using Dr. Cicchetti’s calculated regression inputs are adequate for the 6 purposes of my evaluation below. I report my replication of Dr. Cicchetti’s results 7 for his Groups A, B and C Regressions in the first Panels of the following Tables.
Table 1 Replication and Correction of Cicchetti Regressions Group A: Hub Commodity Prices “Directly Connected” to TCPL’s Market Areas Panel 1. Cicchetti Regressions Under Original Specifications (Natural Logs)
Cicchetti Regression No.>> 1 2 3 4 5 6 24 Dependent Variable>> Emerson Emerson Dawn Iroquois Dawn Iroquois MichCon
Henry Hub 1.405*** 1.625*** 1.460*** 1.533*** 1.738*** 1.703*** 1.377*** (0.107) (0.0865) (0.110) (0.115) (0.0859) (0.105) (0.110) Emerson Bid Floor 0.186*** 0.222*** 0.113*** (0.0200) (0.0199) (0.0243) Centram Bid Floor 0.213*** 0.190*** (0.0196) (0.0202) Dawn Bid Floor 0.239*** (0.0215) Iroquois Bid Floor 0.199*** (0.0228) Winter Dummy 0.0823** 0.0718*** 0.0157 0.236*** 0.0386 0.275*** 0.0524 (0.0324) (0.0272) (0.0340) (0.0361) (0.0270) (0.0331) (0.0335) Emerson Event Dummy 0.283*** 0.220*** 0.331*** (0.0642) (0.0638) (0.0783) Constant -0.601*** -0.847*** -0.853*** -0.881*** -0.939*** -0.848*** -0.534***
Observations 316 316 316 316 316 316 316 R-squared 0.746 0.821 0.745 0.767 0.832 0.792 0.703 DW 0.385 0.502 0.306 0.461 0.492 0.582 0.436
Panel 2. Cicchetti Regressions Under Corrected Specifications (First Differences of Natural Logs)
Cicchetti Regression No.>> 1 2 3 4 5 6 24 Dependent Variable>> Emerson Emerson Dawn Iroquois Dawn Iroquois MichCon
Henry Hub 0.892*** 0.733*** 0.908*** 0.365** 0.825*** 0.297* 1.161*** (0.158) (0.151) (0.148) (0.168) (0.142) (0.170) (0.183) Emerson Bid Floor 0.165*** 0.161*** 0.0599** (0.0218) (0.0206) (0.0246) Centram Bid Floor 0.0940*** 0.0987*** (0.0192) (0.0223) Dawn Bid Floor 0.135*** (0.0207) Iroquois Bid Floor 0.0197 (0.0214) Winter Dummy 0.00120 0.000893 0.00139 0.00190 -0.00119 0.0106 0.000454 (0.0160) (0.0161) (0.0148) (0.0170) (0.0152) (0.0182) (0.0185) Emerson Event Dummy 0.00944 0.0197 -0.0409 (0.0292) (0.0275) (0.0328) Constant -0.00110 -0.00130 -0.00116 -0.000784 -0.00123 -0.000904 -0.000602
Observations 315 315 315 315 315 315 315 R-squared 0.177 0.272 0.247 0.020 0.310 0.036 0.186 DW 2.298 2.280 2.341 2.181 2.366 2.193 2.492
Notes: Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1
36 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
Table 2 Replication and Correction of Cicchetti Regressions Group B: Hub Commodity Prices “Indirectly Connected” to TCPL’s Market Areas Panel 1. Cicchetti Regressions Under Original Specifications (Natural Logs)
Cicchetti Regression No.>> 7 8 22 9 10 11 23 Dependent Variable>> Niagara Dom. South MichCon Niagara Chicago Dom. South MichCon
Henry Hub 1.361*** 0.954*** 1.406*** 1.646*** 1.398*** 0.997*** 1.594*** (0.106) (0.0275) (0.113) (0.0903) (0.0935) (0.0250) (0.0929) Emerson Bid Floor 0.196*** 0.0952*** 0.0292*** 0.180*** (0.0209) (0.0216) (0.00578) (0.0215) Dawn Bid Floor 0.205*** 0.0417*** 0.190*** (0.0208) (0.00536) (0.0220) Winter Dummy 0.0313 -0.0196** 0.0269 0.0536* 0.0812*** -0.0140* 0.0416 (0.0329) (0.00849) (0.0348) (0.0284) (0.0294) (0.00787) (0.0292) Emerson Event Dummy 0.125* 0.338*** 0.0566*** 0.210*** (0.0671) (0.0694) (0.0186) (0.0690) Constant -0.710*** -0.101*** -0.758*** -0.846*** -0.529*** -0.107*** -0.784***
Observations 316 316 316 316 316 316 316 R-squared 0.722 0.900 0.692 0.784 0.737 0.910 0.773 DW 0.380 0.388 0.400 0.550 0.601 0.430 0.526
Panel 2. Cicchetti Regressions Under Corrected Specifications (First Differences of Natural Logs)
Cicchetti Regression No.>> 7 8 22 9 10 11 23 Dependent Variable>> Niagara Dom. South MichCon Niagara Chicago Dom. South MichCon
Henry Hub 0.304** 1.043*** 1.102*** 0.228* 0.758*** 1.032*** 1.020*** (0.134) (0.0430) (0.183) (0.133) (0.194) (0.0430) (0.180) Emerson Bid Floor 0.0696*** 0.156*** 0.0149** 0.162*** (0.0192) (0.0280) (0.00621) (0.0260) Dawn Bid Floor 0.0243 0.0106* 0.130*** (0.0188) (0.00601) (0.0256) Winter Dummy 0.00240 -0.000721 0.000604 0.00738 0.00946 -0.00138 0.00342 (0.0134) (0.00430) (0.0183) (0.0142) (0.0207) (0.00459) (0.0192) Emerson Event Dummy -0.0215 -0.0392 0.00400 -0.00733 (0.0256) (0.0374) (0.00830) (0.0347) Constant -0.000884 -2.22e-05 -0.000695 -0.00102 -0.000465 -3.46e-05 -0.000786
Observations 315 315 315 315 315 315 315 R-squared 0.027 0.672 0.202 0.062 0.165 0.677 0.242 DW 2.429 1.906 2.478 2.427 2.323 1.907 2.520
Notes: Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1
37 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
Table 3 Replication and Correction of Cicchetti Regressions Group C: Hub Commodity Prices Not Connected to TCPL’s Market Areas Panel 1. Cicchetti Regressions Under Original Specifications (Natural Logs)
Cicchetti Regression No.>> 12 13 14 15 16 17 18 Dependent Variable>> AECO NIT PG&E Gate El Paso, San Juan El Paso, Permian SoCal City Gate Leidy Leidy
Henry Hub 1.435*** 0.991*** 1.042*** 1.045*** 0.912*** 0.882*** 1.019*** (0.0653) (0.0338) (0.0346) (0.0341) (0.0255) (0.116) (0.121) Emerson Bid Floor -0.0446*** -0.00897 -0.0112 -0.0145* -0.00438 -0.0615** (0.0151) (0.00782) (0.00800) (0.00788) (0.00588) (0.0268) Dawn Bid Floor -0.0363 (0.0236) Winter Dummy 0.124*** -0.0148 0.0153 0.0131 0.0116 0.153*** 0.158*** (0.0205) (0.0106) (0.0109) (0.0107) (0.00801) (0.0365) (0.0375) Emerson Event Dummy 0.121** 0.0454* 0.0677*** 0.0536** 0.0581*** 0.119 (0.0485) (0.0251) (0.0257) (0.0253) (0.0189) (0.0862) Constant -0.858*** 0.0964** -0.0892* -0.0914* 0.176*** -0.252 -0.398***
Observations 316 316 316 316 316 316 316 R-squared 0.783 0.825 0.847 0.847 0.888 0.362 0.356 DW 0.276 1.042 0.929 0.880 0.722 0.316 0.310
Panel 2. Cicchetti Regressions Under Corrected Specifications (First Differences of Natural Logs)
Cicchetti Regression No.>> 12 13 14 15 16 17 18 Dependent Variable>> AECO NIT PG&E Gate El Paso, San Juan El Paso, Permian SoCal City Gate Leidy Leidy
Henry Hub 0.811*** 1.075*** 1.296*** 1.395*** 0.783*** 1.249*** 1.257*** (0.0853) (0.0908) (0.0869) (0.0816) (0.0591) (0.174) (0.173) Emerson Bid Floor 0.0593*** 0.0764*** 0.0765*** 0.0624*** 0.0397*** 0.00803 (0.0123) (0.0131) (0.0126) (0.0118) (0.00855) (0.0251) Dawn Bid Floor 0.0169 (0.0242) Winter Dummy 0.00385 0.00151 0.000374 0.000157 0.00106 -0.00976 0.000535 (0.00910) (0.00970) (0.00928) (0.00872) (0.00631) (0.0185) (0.0173) Emerson Event Dummy -0.0168 -0.00581 -0.000119 -0.000904 -0.00235 0.0521 (0.0165) (0.0175) (0.0168) (0.0158) (0.0114) (0.0335) Constant 0.000369 1.40e-05 -0.000414 -0.000332 -0.000143 -0.00249 -0.00252
Observations 315 315 315 315 315 315 315 R-squared 0.317 0.420 0.514 0.558 0.440 0.165 0.157 DW 1.976 2.443 2.440 2.239 2.309 2.026 2.025
Notes: Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1
1 Q42. What have you discovered about Dr. Cicchetti’s regressions during your 2 attempts to replicate his results?
3 A42. I have discovered that all of his regressions suffer from a very serious problem known 4 as “autocorrelation.”31 This statistical problem, when serious, can produce entirely 5 spurious regression results.32 In order for regressions of the type estimated33 by Dr. 6 Cicchetti to be valid, there are a series of statistical assumptions that must be met.
31 Also known as “serial correlation.” 32 See for example, C.W.J. Granger and P. Newbold, “Spurious Regressions in Econometrics,” Journal of Econometrics, 2 (1974) 111-120. 33 Using the Ordinary Least Squares (OLS) method of estimation.
38 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 One assumption is that the residuals, or “errors” produced by the equation (the 2 portion of the daily spot prices each day that is “unexplained” by the dependent 3 variables) must be random, like “white noise.” If these errors are correlated in some 4 way, then the regression coefficient estimates are likely biased and the significance 5 tests will produce exaggerated results. Furthermore, the R-squared statistic (the 6 percentage of the variation in prices that is “explained” by the equation) will be 7 biased significantly upwards. Autocorrelation arises as a problem frequently when 8 one is working with time series price data, particularly daily prices. This is because 9 today’s daily spot price tends to be in part a function of yesterday’s price, and so on.
10 Q43. How do you know that Dr. Cicchetti’s regressions exhibit an autocorrelation 11 problem?
12 A43. There is a diagnostic test statistic that every regression computer package reports, 13 called the Durbin-Watson (D-W). The D-W ranges from 0.0 to 4.0. The closer the 14 D-W is to 0.0, the more likely is the existence of severe positive autocorrelation. A 15 D-W closer to 2.0 would indicate that the problem is less likely to be prevalent. A D- 16 W closer to 4.0 would indicate a severe negative autocorrelation problem. Dr. 17 Cicchetti did not report the D-W statistic along with the other regression outputs in 18 his Attachment 1 to his evidence, but I have reported it along with the results in 19 Tables 1, 2, and 3. As shown, all of Dr. Cicchetti’s regressions in Groups A and B 20 have D-W statistics between 0.3 and 0.6, indicating that autocorrelation is likely 21 present and severe in all of these results. The D-W statistics for his Group C 22 regressions are between 0.2 and 1.1, and also exhibit the problem.
23 Q44. Are there ways of correcting for the autocorrelation problem in regression 24 analysis?
25 A44. Yes there are. One approach is to specify one’s data in terms of first differences, not 26 absolute levels. In this case, that would be interpreted using Dr. Cicchetti’s variables 27 as testing the hypothesis that daily changes in spot prices are a function of changes in 28 Henry Hub prices and changes in TransCanada bid floor prices.
39 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 Q45. Have you undertaken such a correction to Dr. Cicchetti’s regressions, and if so, 2 what are your results?
3 A45. Yes, I have, and the results are reported in the second panels of Tables 1, 2, and 3 for 4 Groups A, B and C respectively. The first result to notice is that the D-W statistic is 5 now at or somewhat above the 2.0 level, so the positive autocorrelation problem is no 6 longer likely to be present. The second result to notice is the R-squared statistic is 7 now significantly lower than that reported in Dr. Cicchetti’s original results. Instead 8 of explaining approximately 80 percent of the variation in prices, these results 9 indicate that the corrected equations explain typically less than 25 percent of the 10 variation in the changes in prices, and in some cases there is little-to-no explanatory 11 power. For example, the first corrected regression shown in Panel 2 of Table 1 with 12 respect to the Iroquois pricing location (i.e., “Cicchetti Regression #4”) shows an R- 13 squared statistic of 0.02, indicating that the equation explains only 2 percent of the 14 variation in Iroquois daily price movements. The only explanatory variable in that 15 regression with a significant coefficient is the change in the Henry Hub price. 16 TransCanada’s IT bid floor price at Iroquois has no significant explanatory power.
17 Q46. What do these corrected results show in general for Dr. Cicchetti’s Groups A, 18 B?
19 A46. For most of the corrected regressions in Groups A and B, the change in the Henry 20 Hub price coefficient is positive and statistically significant, as one would expect. 21 Also, the change in the TransCanada IT bid floor prices at the various locations tend 22 to have positive coefficients and they are statistically significant, which we should 23 expect to see if TransCanada’s bid floor prices are attempting to track changes in the 24 factors that affect market prices. But the explanatory power of the equations are very 25 low, as described above, indicating that there likely exist important variables that are 26 omitted from Dr. Cicchetti’s equations and that should be included. If TransCanada’s 27 bid floor prices are correlated with the market conditions that TransCanada is reacting 28 to in determining its bid floors, then in the absence of variables which attempt to 29 capture prevailing market conditions, the bid floor prices may simply be acting as a 30 proxy for market conditions in these equations.
40 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 Q47. What about the corrected results for Dr. Cicchetti’s Group C? 2 A47. The corrected results for Group C are reported in Panel 2 of Table 3. Recall that this 3 was Dr. Cicchetti’s “counterfactual” group, where his original results caused him to 4 reject his hypothesis “with extreme statistical significance” that TransCanada’s IT bid 5 floor prices affected spot prices at the distant locations in Group C. The corrected 6 version of his regressions changes that result entirely. The corrected results now 7 indicate that changes in the TransCanada Emerson IT bid floor price appears to have 8 a positive, and statistically significant effect on changes in prices in California (at 9 PG&E City Gate and SoCal City Gate), at El Paso Permian and San Juan, and at 10 AECO/NIT (i.e., higher Emerson bid floors actually raise the NIT price—contrary to 11 Dr. Orans’ netback effect). These results make no economic sense.
12 Q48. What do you conclude from your review and correction of Dr. Cicchetti’s 13 regression results?
14 A48. I conclude that Dr. Cicchetti’s regression analysis is poorly conceived and executed 15 and that its results are unreliable. First, the hypothesis that he attempts to test is 16 poorly framed and it has the causation reversed. A more plausible hypothesis is that 17 TransCanada’s bid floor prices are reacting to changes in market prices, not causing 18 changes in market prices. Second, Dr. Cicchetti makes no attempt to account for 19 other explanations for why gas commodity prices in northern and eastern markets 20 were unprecedented last winter, and his regression specification does not control for 21 these other factors. The high explanatory power of his regressions as indicated in the 22 R-squared statistic is inflated due to a rampant autocorrelation problem. An attempt to 23 correct the problems with his analysis demonstrates that his regressions are likely 24 producing spurious results, and that his data do not support his counterfactual. For all 25 of these reasons I would recommend that the Board give no weight to any the 26 regression results provided by Dr. Cicchetti, particularly since when they are 27 “corrected” as I have discussed, they fail to provide any meaningful results.
41 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 V. PRICING DISCRETION AND UPSTREAM NETBACKS
2 Q49. To this point you have discussed the claims that TransCanada’s pricing 3 discretion has affected downstream prices. Do any of the intervenor witnesses 4 allege that pricing discretion has affected upstream prices?
5 A49. Yes, and this appears to be the principal concern of Dr. Orans in his evidence for 6 CAPP. Dr. Cicchetti for Centra has not claimed, nor has he presented any evidence to 7 suggest, that pricing discretion has affected upstream gas prices since the 8 implementation of RH-003-2011.
9 Q50. What is the nature of Dr. Orans evidence with respect to the upstream effects of 10 TransCanada’s short term pricing discretion?
11 A50. Dr. Orans looks at the relationship between TransCanada’s bid floors for STFT and 12 IT services, price spreads and flows on the Mainline during three time periods: July 13 2013 – September 2013 (Period 1), October 2013 – December 2013 (Period 2), and 14 January 2014 – March 2014 (Period 3). Dr. Orans observes that during his Period 1, 15 increasing bid floors for discretionary services corresponded to an increase in the 16 (spot market) price spread between Empress and Emerson “that ultimately led to a 17 substantial decrease in netback prices in Alberta and the disconnection of the NIT hub 18 from North American trading hubs.”34
19 Q51. Does the netback effect Dr. Orans observes in his Period 1 extend into his 20 Periods 2 and 3?
21 A51. No. Dr. Orans observes that in Period 2 the price “spreads between Empress and 22 Emerson decreased dramatically—presumably due to the increased depth of the 23 secondary market that resulted from increased FT-NR contracting, as these two 24 events are well correlated.”35 During Period 3 Dr. Orans observes that extreme cold 25 weather contributed to “extremely high and volatile gas prices in eastern markets,” 26 and that “TransCanada’s bid floors for IT service during this period track relatively
34 Written Evidence of Dr. Ren Orans, p.21, line 11 – p.22, line 2. 35 Ibid., p.22, lines 6-9.
42 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 closely to the value of transportation…”36 Thus, the netback effects he describes 2 were strictly a Period 1 phenomenon according to his own evidence. Based on these 3 comments, I would think Dr. Orans would support my opinion that Dr. Cicchetti has 4 the causation reversed in his regressions, i.e., that downstream market prices 5 determined TransCanada’s bid floors and not the opposite.
6 Q52. Have there been periods prior to the implementation of RH-003-2011 when price 7 spreads between NIT and eastern markets widened?
8 A52. Yes. TransCanada’s response to NEB IR 1.24 identifies several such events over the 9 last ten years, prior to the implementation of pricing discretion for Mainline short 10 term services.
11 Q53. You mentioned that in describing his Period 1 netback effect, Dr. Orans focusses 12 his observations on the Empress to Emerson spot price spread. Do you have any 13 concerns about that focus?
14 A53. Yes, I do. First, Emerson is not a very liquid delivery point on the TransCanada 15 Mainline, and hence the reported spot commodity price at Emerson can be subject to 16 a myriad of influencing factors. For example, in addition to the implementation of 17 pricing discretion on July 1, 2013, one of the other effects of Decision RH-003-2011 18 was the elimination of RAM credits, which can be thought of as “free” transportation 19 service when utilized. My understanding is that the elimination of RAM credits had a 20 disproportionate impact on flows at Emerson during the implementation of RH-003- 21 2011. Nowhere in his evidence does Dr. Orans discuss the other potential transitory 22 effects of the implementation of RH-003-2011 that could have had an impact on price 23 spreads during his Period 1. Instead he leads the reader to infer that the effects are 24 entirely due to pricing discretion.
25 More importantly, I think Dr. Orans’ emphasis on spot price spreads is misplaced. 26 Spot price spreads are simply the difference between daily spot commodity price 27 realizations at two locations. They are not forward-looking in any way. In my
36 Ibid., p.22, lines 13-16.
43 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 opinion it is likely to be more revealing to evaluate whether forward basis spreads 2 were affected during this period, and I would focus on more liquid markets such as 3 Dawn and not Emerson. The forward spread can be viewed as the gas trading 4 market’s expectation as to the movement of future price spreads. If pricing discretion 5 was expected to have a sustained effect on netback prices, then one would expect that 6 the implementation of pricing discretion would have been accompanied by a sharp 7 and sustained increase in the forward spread. For example, in response to NEB IR 8 2.15, TCPL presented a graph of the NIT to Dawn forward price strip (November 9 2014 – October 2015) observed over time from June 2012 through May 2014. I have 10 reproduced this graph below. The lower line on the graph represents the NIT to 11 Dawn forward spread, and while it does vary slightly over time, there is no indication 12 of a sustained increase in the forward spread that corresponds to Dr. Orans’ Period 1.
Figure 12
13 Q54. What does Dr. Orans say about whether the netback effect he observed in his 14 Period 1 would likely occur in the future?
15 A54. Dr. Orans extrapolates from his Period 1 observation of spot price spreads to 16 conclude that “TransCanada’s unlimited flexibility,…effectively allows it to
44 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 disconnect the NIT hub from North American markets without notice.”37 I do not 2 believe that his evidence provides any basis for this claim. More importantly, he 3 provides no explanation as to why it would be in TransCanada’s economic interest to 4 disconnect the NIT hub from North American markets. Given TransCanada’s and the 5 Mainline’s dependence on the WCSB to provide future competitive supplies for its 6 long haul services, I would think TransCanada would have every economic incentive 7 to maintain the “connectivity” of NIT to eastern markets.
8 Q55. If Dr. Orans were correct about this, wouldn’t producers in the WCSB 9 necessarily be harmed because their access to Eastern markets would be at times 10 restricted, or because netbacks would be more volatile?
11 A55. If what Dr. Orans suggests was correct, WCSB producers would only be harmed if 12 they chose not to mitigate or eliminate such risks. For example, WCSB producers 13 could guarantee their “connection” to eastern markets by contracting for firm capacity 14 on the Mainline, on other pipelines such as Alliance/Vector, or they could hedge the 15 basis differential in the forward financial market. If volatility is a concern with 16 respect to spot markets at NIT, producers could hedge their production forward, 17 thereby laying off the volatility risk on counterparties willing to bear it. Dr. Oran 18 testifies that “[t]hrough discussions with CAPP members, I understand that producers 19 place a large intrinsic value on access to large, liquid, and connected trading hubs.”38 20 While I do not disagree with this, what Dr. Orans does not mention in this passage is 21 that the producers who place a large intrinsic value on the connection that the 22 Mainline provides to eastern markets are not paying for that optionality if they do not 23 hold capacity on the Mainline. I would observe that across North America, as new 24 shale gas production has come on line, many producers are finding it necessary to 25 commit to long-term contracts underpinning large investments in new pipeline 26 infrastructure to ensure their access to liquid downstream markets. Unlike many of 27 these producers, a WCSB producer currently has the advantage of an existing pipeline
37 Ibid., p.35, lines 7-9. 38 Ibid., p.35, lines 10-12.
45 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 in the ground with FT capacity availability that can provide guaranteed access to 2 eastern markets if the producer is willing to pay the regulated FT toll for that service. 3
4 VI. PRICING DISCRETION, TOLL STABILITY AND TRANSPARENCY
5 Q56. Do you share the concern expressed by Dr. Orans at pages 35 and 36 of his 6 evidence that “continuing to allow TransCanada to exercise unlimited pricing 7 flexibility going forward presents producers with uncertainty in their access to 8 liquid gas markets”?
9 A56. No, I do not share this concern. First, spot basis differentials along major pipeline 10 corridors have been volatile in North America in varying degrees for as long as basis 11 differentials have been in existence. This is because spot basis differentials are just a 12 reflection of the volatility of the spot commodity markets at each end of the pipe. 13 This kind of volatility has not been a barrier to investment in new production, in part 14 because producers have been able to hedge their positions either by contracting in the 15 forward market or by contracting for firm pipeline capacity to liquid destinations, if 16 they want to lay off the volatility risk. Dr. Orans fails to mention these avenues for 17 risk mitigation and why they are not available to WCSB producers.
18 Q57. Would the CAPP proposal to limit TransCanada’s discretionary service pricing 19 flexibility reduce such volatility as Dr. Orans suggests?
20 A57. I don’t believe so, and Dr. Orans has not explained how it would occur. Again, spot 21 basis volatility is a function of the spot commodity price volatility at each end of the 22 pipe. Limiting TransCanada’s pricing discretion does nothing to limit the ability of 23 other competitors in the secondary market from transacting at volatile spot prices. 24 Again, spot basis volatility is a characteristic of most pipeline paths in North 25 America, and on paths where the pipeline itself does not have significant pricing 26 discretion over short term services.
46 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 Q58. Does the existing pricing discretion violate the regulatory objective of toll 2 stability or lead to “inefficient and disruptive tolls” as CAPP suggests in it 3 evidence at page 4, lines 18-19?
4 A58. Not in my opinion. To me the concept of “toll stability” is really a concern about 5 uncertainty and variability in regulated FT tolls over time. I do not think it 6 appropriate, nor is it efficient, for pipeline regulators to attempt to “stabilize” the 7 market value of short-term transportation services, even if that were possible without 8 regulating the entire secondary market. CAPP’s proposal to limit TransCanada’s 9 discretion is, in practical effect, a proposal to regulate one competitor in the 10 secondary market while the rest of the participants remain unregulated. All this serves 11 to do is limit TransCanada’s ability to capture some portion of the economic rent 12 available on the pipeline. Perhaps that is the intent of CAPP’s proposal, but if so, 13 there is nothing economically efficient about it.
14 With respect to whether the existing pricing discretion leads to “inefficient and 15 disruptive tolls,” I’ve discussed above the inappropriateness of focusing on short-run 16 efficiency alone at the expense of long run efficiency. I’m not sure what CAPP means 17 by the phrase “disruptive tolls,”39 but none of the statistical evidence which has been 18 presented by the intervenor witnesses has established that the pricing discretion 19 granted by the Board in RH-003-2011 causes “toll disruption” or has disrupted the 20 secondary market in any way.
21 Q59. Does your view as to the competitive effects of CAPP’s proposal extend to its 22 recommendation requiring TransCanada to provide advance notification of 23 monthly bid floors and to increase its short term data posting requirements?
24 A59. Yes it does. In my opinion these proposals simply serve to limit TransCanada’s 25 ability to effectively compete in the secondary market to the benefit of other, 26 unregulated secondary market participants. I see no efficiency-enhancing reason to
39 In response to TransCanada IR 1.2 (b), CAPP defined “disruptive tolls” as “tolls that disrupt the functioning of the natural gas market including the integrated nature of the North American natural gas market.” In other words, disruptive tolls are tolls that disrupt the market.
47 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 require TransCanada to adhere to such posting requirements when other competitors 2 have no such requirements. Dr. Orans appears to take the view that TransCanada 3 should be treated differently in this respect on the grounds that it is a regulated natural 4 monopoly that earns an allowed rate of return (see CAPP response to TransCanada IR 5 1.15 (b).) But the “natural monopoly” feature that Dr. Orans is referring to has to do 6 with regulated FT tolls, which still exist under the terms of the Settlement as a 7 recourse. Dr. Orans has not established, nor do I believe he can establish, that 8 TransCanada is a “natural monopoly” or has market power with respect to the 9 discretionary services at issue here. The Board need not hamstring TransCanada as 10 an effective competitor in the provision of discretionary services in the secondary 11 market, nor should it.
12 VII. PRICING DISCRETION AND BUSINESS RISK
13 Q60. At page 44, lines 11-12 of his evidence, Dr. Orans responds to a question about 14 the terms of the Settlement by saying that “the material reduction to the risk 15 faced by TransCanada’s shareholders [in the Settlement] necessitates a more 16 conservative framework for pricing flexibility.” Do you agree?
17 A60. No. Retaining the existing pricing discretion for STFT and IT services is a feature of 18 the Settlement. Dr. Orans is essentially suggesting that the allocation of risks in the 19 Settlement is somehow unbalanced, and that pricing discretion should be limited in 20 order to rebalance those risks. He provides no evidence or basis for that proposition. 21 Furthermore, he goes on to say (at page 44, lines 12-17):
22 Unlimited pricing flexibility was provided to TransCanada as an 23 important instrument with which to manage the incremental risks due 24 to multi-year fixed tolls set below the pipeline’s direct cost of service 25 and the associated potential for disallowed costs. Since these 26 incremental risks vanish under the proposed settlement, the need for 27 unlimited pricing flexibility disappears.
28 As TransCanada’s business risk expert for cost of capital in the RH-003-2011 29 proceeding, in my opinion Dr. Orans has mischaracterized the effects of the Decision 30 Model, and the proposed Settlement, on TransCanada’s business risk. First, pricing 31 discretion was authorized by the Board in RH-003-2011 for reasons that go well
48 WRITTEN REPLY EVIDENCE OF PAUL R. CARPENTER
1 beyond risk mitigation.40 Specifically, pricing discretion was viewed by the Board as 2 a means to encourage shippers to sign FT contracts instead of relying on short-term 3 services to meet year-round needs, and as an important revenue generation tool to 4 permit TransCanada to recover costs and maintain lower and more stable FT tolls in 5 the future.41 Second, there is no sense in which under the terms of the Settlement, 6 what Dr. Orans refers to as “incremental risks due to multi-year fixed tolls” somehow 7 “vanish”, and there is nothing in the Settlement as I understand it which eliminates 8 the potential for disallowed costs as described by the Board in Decision RH-003- 9 2011.
10 Q61. Do you acknowledge that the Settlement would have the effect of reducing the 11 Mainline’s business risk relative to the Decision Model in RH-003-2011?
12 A61. Yes, and in particular it helps reduce the competitive risks faced by the Mainline (see 13 TransCanada’s Additional Written Evidence, pages 22-23, and its response to NEB 14 IR 1.18). But the terms of the Settlement account for that risk reduction by 15 TransCanada’s agreement to reduce its allowed return from 11.5% on 40% equity in 16 RH-001-2011 to 10.1% on 40% equity under the Settlement. This is not just a “tacit 17 acknowledgement” as Dr. Orans refers to it on page 43, line 7 of his evidence, but it 18 is an explicit one made by the negotiating parties.
19 Q62. Does this complete your written reply evidence? 20 A62. Yes, it does.
40 See TransCanada’s Additional Written Evidence, pp.16-17. 41 Reasons for Decision, RH-003-2011, pp. 2-3.
49 Attachment A to the Written Reply Evidence of Paul R. Carpenter Page 1 of 15 PAUL R. CARPENTER Principal
Cambridge, MA +1.617.864.7900 [email protected]
Dr. Paul Carpenter holds a Ph.D. in applied economics and an M.S. in management from the Massachusetts Institute of Technology, and a B.A. in economics from Stanford University. He specializes in the economics of the natural gas, oil and electric utility industries. Dr. Carpenter was a co- founder of Incentives Research, Inc. in 1983. Prior to that he was employed by the NASA/Caltech Jet Propulsion Laboratory and Putnam, Hayes & Bartlett, and he was a post-doctoral fellow at the MIT Center for Energy Policy Research. He is currently a Principal and Chairman of The Brattle Group.
AREAS OF EXPERTISE
Energy economics Regulation Corporate planning Pricing Policy Antitrust
EXPERIENCE
Natural Gas and Electric Utility Industries
Consulting and testimony on nearly all of the economic and regulatory issues surrounding the transition of the natural gas and electric power industries from strict regulation to greater competition. These issues have included stranded investments and contracts, design and pricing of unbundled and ancillary services, evaluation of supply, demand and price forecasting models, the competitive effects of pipeline expansions and performance-based ratemaking. He has consulted on the regulatory and competitive structures of the gas and electric power industries in the U.S., Canada, the United Kingdom, continental Europe, Australia and New Zealand.
Valuation and Damages
Expert testimony before courts, tribunals and in arbitrations concerning asset valuation and damages associated with breach of contract, bankruptcy and commercial disputes. Experience includes expert testimony in U.S. federal and state courts, the British High Court of Justice, the Australian Competition Tribunal and various arbitration and mediation panels in Australia, Canada, New Zealand and the U.S.
Antitrust
Expert testimony in several of the seminal cases involving the alleged denial of access to regulated facilities; analysis of relevant market and market power issues, business justification defenses, and damages.
1
Attachment A to the Written Reply Evidence of Paul R. Carpenter Page 2 of 15 Paul R. Carpenter
Regulation
Studies and consultation on alternative ratemaking methodologies for oil and gas pipelines, on “bypass” of regulated facilities before the U.S. Congress; advice and testimony before several state utility commissions and the National Energy Board of Canada on new facility certification policy.
Finance
Research on business and financial risks in the regulated industries and testimony on risk, cost of capital, and asset valuation for network industries, airports and seaports in the U.S., Canada., Australia and New Zealand.
PROFESSIONAL AFFILIATIONS
American Economic Association
ACADEMIC HONORS AND FELLOWSHIPS
Stewart Fellowship, 1983 MIT Fellowships, 1981, 1982, 1983 Brooks Master’s Thesis Prize (Runner-up), MIT, 1978
PUBLICATIONS
“Pipeline Regulatory Issues Arising From Oil and Natural Gas Production Growth in North America” with Matthew O’Loughlin and Steve Levine, The Energy Law Advisor, Volume 8, No. 1, February 2014.
“A Framework for Analyzing Market Manipulation.” Review of Law and Economics, September 2012 (with Shaun D. Ledgerwood)
“Shale Gas and Pipeline Risk,” Public Utilities Fortnightly, April 2012 (with Steven H. Levine, A. Lawrence Kolbe and Bente Villadsen)
“Options for Reforming the Building Blocks Framework.” Report to the Australian Energy Market Commission, 16 December 2009, (with Toby Brown).
“The Advent of U.S. Gas Demand Destruction and Its Likely Consequences for the Pricing of Future European Gas Supplies,” (with Carlos Lapuerta and Morten Frisch), 16 March 2005.
2
Attachment A to the Written Reply Evidence of Paul R. Carpenter Page 3 of 15 Paul R. Carpenter
“REx Incentives: Performance Based Ratemaking (PBR) Choices that Reflect Firms’ Performance Expectations,” (with Johannes P. Pfeifenberger and Paul C. Liu), The Electricity Journal, November 2001.
“Asset Valuation and the Pricing of Monopoly Infrastructure Services: A Discussion Paper,” (with Carlos Lapuerta) 28 July 2000.
“Competition in Gas Pipeline Markets: International Precedent for Regulatory Coverage Decisions,” Report to the National Competition Council of Australia (with Judy Chang), June 2000.
“Methodologies for Establishing National and Cross-Border Systems of Pricing of Access to the Gas System in Europe,” Report to the European Commission (with Carlos Lapuerta and Boaz Moselle), February 2000.
“A Critique of Light-handed Regulation: The Case of British Gas,” (with Carlos Lapuerta), Northwestern Journal of International Law & Business, Volume 19, No. 3, Spring 1999.
“Separate Marketing of Natural Gas by Joint Venture Producers in Australia,” (with Jurgen Weiss), prepared for Optima Energy, Australia, submitted to the Upstream Issues Working Group, Australian and New Zealand Minerals and Energy Council, 26 September 1998.
“Likely Trends in Canadian Natural Gas Imports,” (with Matthew P. O’Loughlin and Gao-Wen Shao), Natural Gas, Volume 14, No. 8, March 1998.
“Pipeline Pricing to Encourage Efficient Capacity Additions,” (with Frank C. Graves and Matthew P. O’Loughlin), prepared for Columbia Gas Transmission Corporation and Columbia Gulf Transmission Company, February 1998.
“The Outlook for Imported Natural Gas,” (with Matthew P. O’Loughlin and Gao-Wen Shao), prepared for The INGAA Foundation, Inc., July 1997.
“Basic and Enhanced Services for Recourse and Negotiated Rates in the Natural Gas Pipeline Industry,” (with Frank C. Graves, Carlos Lapuerta, and Matthew P. O’Loughlin) May 29, 1996, prepared for Columbia Gas Transmission Corporation, Columbia Gulf Transmission Company.
“Estimating the Social Costs of PUHCA Regulation,” (with Frank C. Graves) submitted on behalf of Central and South West Corp. to the U.S. Securities and Exchange Commission in its Request for Comments on the Modernization of Regulation of Public Utility Holding Companies, File No. S7-32-94, February 6, 1995.
“Review of the Model Developer’s Report, Natural Gas Transmission And Distribution Model (NGTDM) Of The National Energy Modeling System,” December 1994, prepared for U.S. Department of Energy, Energy Information Administration and Oak Ridge National Laboratory under Subcontract No. 80X- SL220V.
3
Attachment A to the Written Reply Evidence of Paul R. Carpenter Page 4 of 15 Paul R. Carpenter
“Pricing of Electricity Network Services to Preserve Network Security and Quality of Frequency Under Transmission Access,” (with Frank C. Graves, Marija Ilic, and Asef Zobian) response to the Federal Energy Regulatory Commission’s Request for Comments in its Notice of Technical Conference Docket No. RM93-19-000, November 1993.
“Creating a Secondary Market in Natural Gas Pipeline Capacity Rights Under FERC Order No. 636,” (with Frank C. Graves) draft December 1992, Incentives Research, Inc.
“Review of the Component Design Report, Natural Gas Annual Flow Module, National Energy Modeling System,” August 1992, prepared for the U.S. Department of Energy, Energy Information Administration.
“Unbundling, Pricing, and Comparability of Service on Natural Gas Pipeline Networks,” (with Frank C. Graves) November 1991, prepared for the Interstate Natural Gas Association of America. “Review of the Gas Analysis Modeling System (GAMS): Final Report of Findings and Recommendations,” August 1991, prepared for the U.S. Dept. of Energy, Energy Information Administration.
“Estimating the Cost of Switching Rights on Natural Gas Pipelines,” (with F.C. Graves and J.A. Read) The Energy Journal, October 1989.
“Demand-Charge GICs Differ from Deficiency-Charge GICs,” (with F.C. Graves) Natural Gas, Vol. 6, No. 1, August 1989.
“What Price Unbundling?” (with F.C. Graves) Natural Gas, Vol. 5 No. 10, May 1989.
Book Review of Drawing the Line on Natural Gas Regulation: The Harvard Study on the Future of Natural Gas, Joseph Kalt and Frank Schuller, eds., in The Energy Journal, April 1988.
“Adapting to Change in Natural Gas Markets,” (with Henry D. Jacoby and Arthur W. Wright) in Energy, Markets and Regulation: What Have We Learned?, Cambridge: MIT Press, 1987.
Evaluation of the Commercial Potential in Earth and Ocean Observation Missions from the Space Station Polar Platform, Prepared by Incentives Research for the NASA Jet Propulsion Laboratory under Contract No. 957324, May 1986.
An Economic Comparison of Alternative Methods of Regulating Oil Pipelines, (with Gerald A. Taylor) Prepared by Incentives Research for the U.S. Department of Energy, Office of Competition, July 1985.
“The Natural Gas Policy Drama: A Tragedy in Three Acts,” (with Arthur W. Wright) MIT Center for Energy Policy Research Working Paper No. 84-012WP, October 1984.
Oil Pipeline Rates and Profitability under Williams Opinion 154 , (with Gerald A. Taylor), Prepared by Incentives Research for the U.S. Department of Energy, Office of Competition, September 1984.
4
Attachment A to the Written Reply Evidence of Paul R. Carpenter Page 5 of 15 Paul R. Carpenter
Natural Gas Pipelines After Field Price Decontrol: A Study of Risk, Return and Regulation, Ph.D. Dissertation, Massachusetts Institute of Technology, March 1984. Published as a Report to the U.S. Department of Energy, Office of Oil and Gas Policy, MIT Center for Energy Policy Research Technical Report No. 84-004.
The Competitive Origins and Economic Benefits of Kern River Gas Transmission, Prepared by Incentives Research, Inc., for Kern River Gas Transmission Company, February 1994.
“Field Price Decontrol of Natural Gas, Pipeline Risk and Regulatory Policy,” in Government and Energy Policy, Richard L. Itteilag, ed., Washington D.C., June 1983.
“Risk Allocation and Institutional Arrangements in Natural Gas,” (with Arthur W. Wright) invited paper presented to the American Economic Association Meetings, San Francisco, December 1983.
“Vertical Market Arrangements, Risk-shifting and Natural Gas Pipeline Regulation,” Sloan School of Management Working Paper No. 1369-82, September 1982 (Revised April 1983).
Natural Gas Pipeline Regulation After Field Price Decontrol (with Dr. Henry Jacoby and Arthur W. Wright), prepared for U.S. Department of Energy, Office of Oil and Gas Policy, MIT Energy Lab Report No. 83-013, March 1983.
Book Review of An Economic Analysis of World Energy Problems, by Richard L. Gordon, Sloan Management Review, Spring 1982.
“Perspectives on the Government Role in New Technology Development and Diffusion,” (with Drew Bottaro) MIT Energy Lab Report No. 81-041, November 1981.
International Plan for Photovoltaic Power Systems (co-author), Solar Energy Research Institute with the Jet Propulsion Laboratory Prepared for the U.S. Department of Energy, August 1979.
Federal Policies for the Widespread Use of Photovoltaic Power Systems (contributor), Jet Propulsion Laboratory Report to the U.S. Congress DOE/CS-0114, March 24, 1980.
“An Economic Analysis of Residential, Grid-connected Solar Photovoltaic Power Systems,” (with Gerald A. Taylor) MIT Energy Laboratory Technical Report No. 78-007, May 1978.
SPEECHES/PRESENTATIONS
“Russia, Ukraine, and Trade,” Georgetown Center for Business and Public Policy, Washington D.C., April 23, 2014.
“The Uncertain Future For ANS LNG Exports,” Energy in Alaska Conference, Anchorage, Alaska, December 3, 2012.
5
Attachment A to the Written Reply Evidence of Paul R. Carpenter Page 6 of 15 Paul R. Carpenter
“The Uncertain Future for Alaska North Slope Gas in the Lower-48,” Energy in Alaska Conference, Anchorage, Alaska, December 2, 2011.
“Economic Perspectives on Gas Trading Markets and Regulation,” Canada/U.S. Energy Transactions Conference, Vancouver B.C., August 24, 2010.
“Incentive Regulation – Design: Key Plan Components,” Alberta Utility Commission Workshop on Performance Based Regulation, Edmonton, Alberta, May 26, 2010.
“LNG Access Policy and California,” California Resources Agency Workshop on LNG, June 1, 2005.
Opening Remarks at the Eighth Central and Eastern European Power Industry Forum (CEEPIF 2001), Budapest, March 29, 2001.
“CPUC v. El Paso Merchant Energy, et al., FERC Docket No. RP00-241-000,” ABA Forum, Washington, DC, September 6, 2001.
“Overseas Experience B Lessons for Australian Gas and Power Markets from California and Europe,” 2001 Gas Industry Forum, The Australian Gas Association, Melbourne, Victoria, Australia, June 26, 2001.
“Liberalizing Energy Markets: Lessons from California’s Crisis,” 20th Annual Conference on US-Turkish Relations, Washington, DC, March 27, 2001.
“Opening Remarks from the Chair: Rates, Regulations and Operational Realities in the Capacity Market of the Future,” AIC conference on “Gas Pipeline Capacity ‘97,” Houston, Texas June 17, 1997.
“Lessons from North America for the British Gas TransCo Pricing Regime,” prepared for AIC conference on: Gas Transportation and Transmission Pricing, London, England, October 17, 1996.
“GICs and the Pricing of Gas Supply Reliability,” California Energy Commission Conference on Emerging Competition in California Gas Markets, San Diego, Ca. November 9, 1990.
“The New Effects of Regulation and Natural Gas Field Markets: Spot Markets, Contracting and Reliability,” American Economic Association Annual Meeting, New York City, December 29, 1988.
“Appropriate Regulation in the Local Marketplace,” Interregional Natural Gas Symposium, Center for Public Policy, University of Houston, November 30, 1988.
“Market Forces, Antitrust, and the Future of Regulation of the Gas Industry,” Symposium on the Future of Natural Gas Regulation, American Bar Association, Washington D.C., April 21, 1988.
“Valuation of Standby Tariffs for Natural Gas Pipelines,” Workshop on New Methods for Project and Contract Evaluation, MIT Center for Energy Policy Research, Cambridge, March 3, 1988.
6
Attachment A to the Written Reply Evidence of Paul R. Carpenter Page 7 of 15 Paul R. Carpenter
“Long-term Structure of the Natural Gas Industry,” National Association of Regulatory Utility Commissioners Meeting, Washington D.C., March 1, 1988.
“How the U.S. Gas Market Works or Doesn’t Work,” Ontario Ministry of Energy Symposium on Understanding the United States Natural Gas Market, Toronto, March 18, 1986.
“The New U.S. Natural Gas Policy: Implications for the Pipeline Industry,” Conference on Mergers and Acquisitions in the Gas Pipeline Industry, Executive Enterprises, Houston, February 26-27, 1986.
Various lectures and seminars on U.S. natural gas industry and regulation for graduate energy economics courses at Massachusetts Institute of Technology, 1984-96.
Panelist in University of Colorado Law School workshop on state regulations of natural gas production, June 1985. (Transcript published in University of Colorado Law Review.) “Oil Pipeline Rates after the Williams 154 Decision,” Executive Enterprises, Conference on Oil Pipeline Ratemaking, Houston, June 19-20, 1984.
“Issues in the Regulation of Natural Gas Pipelines,” California Public Utilities Commission Hearings on Natural Gas, San Francisco, May 21, 1984.
“The Natural Gas Pipelines in Transition: Evidence From Capital Markets,” Pittsburgh Conference on Modeling and Simulation, Pittsburgh, April 20, 1984.
“Financial Aspects of Gas Pipeline Regulation,” Pittsburgh Conference on Modeling and Simulation, Pittsburgh, April 19-20, 1984.
“Natural Gas Pipelines After Field Price Decontrol,” Presentations before Conferences of the International Association of Energy Economists, Washington D.C., June 1983, and Denver, November 1982.
“Spot Markets for Natural Gas,” MIT Center for Energy Policy Research Semi-annual Associates Conference, March 1983.
“Pricing Solar Energy Using a System of Planning and Assessment Models,” Presentations to the XXIV International Conference, The Institute of Management Science, Honolulu, June 20, 1979.
TESTIMONY
Antitrust/Federal Court/Arbitration
In the matter of an Arbitration between the Northwest Shelf Joint Venture and Verve Energy, Perth Western Australia, July 2013.
In the District Court in and for Tulsa County, the State of Oklahoma, Bettina M. Whyte, as Trustee for the SemGroup Litigation Trust v. PriceWaterhouseCoopers, LLP, April 2013.
7
Attachment A to the Written Reply Evidence of Paul R. Carpenter Page 8 of 15 Paul R. Carpenter
In the matter of an Arbitration between the Gippsland Basin Joint Venture and Australia Gas Light, Melbourne, Victoria, March 2013.
In the British High Court of Justice, Queen’s Bench Division, Commercial Court, GB Gas Holdings Limited v. Accenture, June 2011 (expert report).
In the matter of an Arbitration between Woodside Energy Ltd and Alinta Sales Pty Ltd, Perth, Western Australia, July 2009, September 2009.
In the Arbitration between Niska Gas Storage US, LLC and Alenco Inc., 2007.
In the Arbitration between the Southwest Queensland Producers and Xstrata, Ltd., Brisbane, Australia, 2006.
In the Superior Court of the State of California, County of San Diego, Natural Gas Anti-trust Cases I, II, III, & IV, February 2006, May 2006, June 2006 (declarations).
In the United States District Court for the State of California, County of Los Angeles, Central District, TXU Energy Services Company v. American Remedial Technologies, March 2003, April 2003.
In the United States District Court for the Northern District of Alabama, Northeastern Division, The City of Huntsville d/b/a Huntsville Utilities v. Proliance Energy, LLC, February 2003, June 2003, February 2005.
In the Arbitration between Wellington International Airport Ltd., and Air New Zealand and Qantas Airways Ltd., August 2002.
In the United States District Court for the Eastern District of Virginia, Alexandria Division, Hess Energy Inc. v. Lightning Oil Company, Ltd., July 2002.
In the United States District Court for the District of Colorado, The Farm Credit Bank of Wichita, formerly known as The Federal Land Bank of Wichita, et al., v. Atlantic Richfield Company, April 2001.
In the United States Bankruptcy Court for the District of Delaware, KCS Energy, Inc., et al., Debtors: Chapter 11, November 2000.
Mediation between Methanex LTD, et al and Westgate Port, New Zealand, May 2000.
In the matter of the Arbitration between American Central Gas Company v. Union Pacific Resources and Duke Energy Fuels, et al., July 2000.
In the United States District Court for the Western District of Missouri, Riverside Pipeline Company, L.P., et al., v. Panhandle Eastern Pipeline Company, September 1998.
In the United States District Court, District of Columbia, United States of America, Dept. of Justice v. Enova Corporation, August 1998.
8
Attachment A to the Written Reply Evidence of Paul R. Carpenter Page 9 of 15 Paul R. Carpenter
In the matter of the Arbitration between Western Power Corp. and Woodside Petroleum Corp., et al., Perth, Western Australia, May-July 1998.
In the United States District Court for the District of Montana, Butte Division, Paladin Associates, Inc. v. Montana Power Company, November- December 1997.
In the United States District Court for the District of Colorado, Atlantic Richfield Co. v. Darwin H. Smallwood, Sr., et al., July 1997.
In the Australian Competition Tribunal, Review of the Trade Practices Act Authorisations for the AGL Cooper Basin Natural Gas Supply Arrangements, on behalf of the Australian Competition and Consumer Commission, February 1997.
In the Southwest Queensland Gas Price Review Arbitration, Adelaide, South Australia, May 1996.
In the matter of the Arbitration between Amerada Hess Corp. v. Pacific Gas & Electric Co., May 1995.
In re Columbia Gas Transmission Corp., Claims Quantification Proceeding in the U.S. Bankruptcy Court for the District of Delaware, Before the Claims Mediator, July and November 1993.
Deposition Testimony in Fina Oil & Gas v. Northwest Pipeline Corp. and Williams Gas Supply (New Mexico) 1992.
Testimony by Affidavit in James River Corp. v. Northwest Pipeline Corp. (Fed. Ct. for Oregon) 1989. Deposition and Testimony by Affidavit in Merrion Oil and Gas Col, et al., v. Northwest Pipeline Corp. (Fed. Ct. for New Mexico) 1989.
Deposition Testimony in Martin Exploration Management Co., et al. v. Panhandle Eastern Pipeline Co. (Fed. Ct. for Colorado) 1988 and 1992.
Trial Testimony in City of Chanute, et al. v. Williams Natural Gas (Fed. Ct. for Kansas) 1988.
Deposition Testimony in Sinclair Oil Co. v. Northwest Pipeline Co. (Fed. Ct. for Wyoming) 1987.
Deposition and Trial Testimony in State of Illinois v. Panhandle Eastern Pipeline Co. (Fed. Ct. for C.D. Ill) 1984-87.
Economic/Regulatory Testimony
Before the National Energy Board of Canada, TransCanada Pipeline Ltd. Application for Approval of Tariff Amendments, Docket RH-1-2013, August 2013.
Before the National Energy Board of Canada, Application of Chevron Canada Ltd. For a Priority Destination Designation on the TransMountain Pipeline, Docket MH-002-2012, December 2012.
9
Attachment A to the Written Reply Evidence of Paul R. Carpenter Page 10 of 15 Paul R. Carpenter
Before the National Energy Board of Canada, TransCanada PipeLines Limited, NOVA Gas Transmission Ltd., Foothills Pipe Lines Ltd. - Business and Services Restructuring Proposal and Mainline Tolls for 2012-2013, Docket RH-003-2011, September 2011, May 2012.
Before the Alberta Utilities Commission, Rate Regulation Initiative, Proceeding 566, ATCO Gas and ATCO Electric Performance Based Regulation Applications, July 2011, April 2012.
Before the Federal Energy Regulatory Commission and the Regulatory Commission of Alaska, BP Pipelines (Alaska) Inc., FERC Docket No. IS09-348, RCA Docket P-08-9, June 2011.
Before the Illinois Commerce Commission, Northern Illinois Gas Company d/b/a/ Nicor Gas Company, Reconciliation of Revenues Collected under Gas Adjustment Charges with Actual Costs Prudently Incurred, Docket No. 01-0705, May 2011.
Before the Regulatory Commission of Alaska, In the Matter of the Petition by Aurora Energy, LLC to Exempt Steam Heat Rates from Economic Regulation, Docket No. U-10-158, December 2010, January 2011.
Before the Federal Energy Regulatory Commission, Kern River Gas Transmission, Docket No. RP04-274, June 2010, September 2010.
Before the Michigan Public Service Commission, SEMCO Energy Gas Company, Case No. U-16169, June 2010.
Before the Régie de L’Énergie, Société en Commandite Gaz Métro Cause Tarifaire 2010, Docket No. R- 3690-2009, May 2009.
Before the Ontario Energy Board, The Cost of Capital in Current Economic and Financial Market Conditions, Docket No. EB-2009-0084, April 2009 (report).
Before the Alberta Utilities Commission, In The Matter Of Alberta Utilities Commission 2009 Generic Cost of Capital Hearing, Application No. 1578571, November 2008.
Before the Federal Energy Regulatory Commission, Energy Transfer Partners, LP, Energy Transfer Company, ETC Marketing, Ltd., Houston Pipeline Company, Docket No. IN06-3-003, September 2008, May 2009.
Before the Regulatory Commission of Alaska, In the Matter of the Tariff Revision, Designated as TA167- 4, Regarding a Proposed Gas Sales Agreement Between ENSTAR Natural Gas Company and ConocoPhillips Alaska, Inc. and a Proposed Gas Sales Agreement Between ENSTAR and Marathon Oil Company, Docket No. U-08-58, May 2008, July 2008.
Before the California Public Utility Commission, Application of Pacific Gas & Electric Co. for Authorization to Enter Into Long-Term Natural Gas Transportation Arrangements with Ruby Pipeline, Docket No. A.07-12-021, May 2008, June 2008.
10
Attachment A to the Written Reply Evidence of Paul R. Carpenter Page 11 of 15 Paul R. Carpenter
Before the National Energy Board of Canada, In the Matter of Trans Québec and Maritimes Pipeline Inc., Docket RH-1-2008, December 2007, September 2008, October 2008.
Before the Ontario Energy Board, Multi-year Incentive Rate Regulation for Natural Gas Utilities, Docket EB-2007-0606/0615, August 2007, September 2007, November 2007, December 2007.
Before the Régie de L’Énergie, Société en Commandite Gaz Métro Cause Tarifaire 2008, Docket No. R- 3630-2007, May 2007, August 2007.
Before the Ontario Energy Board, Application by Enbridge Gas Distribution Inc. for an Order or Orders Approving or Fixing Just and Reasonable Rates and Other Charges for the Sale, Distribution, Transmission and Storage of Gas Commencing January 1, 2007, Docket No. EB-2006-0034, August 2006, February 2007.
Before the California Public Utilities Commission, In the Matter of the Application of San Diego Gas & Electric Company (U 902 G) and Southern California Gas Company (U 904 G) for Authority to Integrate Their Gas Transmission Rates, Establish Firm Access Rights, and Provide Off-System Gas Transportation Services, Docket No. A. 04-12-004, July 2006.
Before the Federal Energy Regulatory Commission, Gas Transmission Northwest Corporation, Docket No. RP06-407, June 2006, October 2006 (affidavits).
Before the Regulatory Commission of Alaska, in the matter of the Gas Sales Agreement Between ENSTAR Natural Gas Company, A Division of SEMCO Energy Inc. And Marathon Oil Company filed as TA139-4, Docket No. U-06-2, March 2006, May 2006.
Before the Ontario Energy Board, Application by Union Gas Limited for an Order or Orders Approving or Fixing Just and Reasonable Rates and Other Charges for the Sale, Distribution, Transmission and Storage of Gas Commencing January 1, 2007, Docket No. EB-2005-0520, January 2006.
Before the New Jersey Board of Public Utilities, in the matter of the Joint Petition of Public Service Electric and Gas Company and Exelon Corporation For Approval of a Change in Control of Public Service Electric and Gas Company, and Related Authorizations, Docket No. EM05020106, November 2005, December 2005, January 2006, March 2006.
Before the Pennsylvania Public Utility Commission, Application for Approval of the Merger of Public Service Enterprise Group and Exelon Corporation, Docket No. A-110550F0160, June 2005, August 2005, September 2005.
Before the National Energy Board of Canada, in the matter of TransCanada Pipelines LTD., RH-2-2004 Phase II, Cost of Capital, January 2005.
Before the California Public Utilities Commission, Order Instituting Investigation into the Gas Market Activities of Southern California Gas Company, San Diego Gas and Electric, Southwest Gas, Pacific Gas and Electric, and Southern California Edison and their Impact on the Gas Price Spike Experience at the California Border from March 2000 through May 2001 on behalf of Southern California Edison, Docket
11
Attachment A to the Written Reply Evidence of Paul R. Carpenter Page 12 of 15 Paul R. Carpenter
No. I. 02-11-040, December 2003, May 2004, June 2004.
Before the Alberta Energy and Utilities Board in the matter of Alberta Energy and Utilities Board Generic Cost of Capital Hearing on behalf of Nova Gas Transmission LTD, Proceeding No. 1271597, November 2003.
Before the Federal Energy Regulatory Commission (FERC), California Public Utilities Commission v. El Paso Natural Gas Company, El Paso Merchant Energy-Gas, L.P., and El Paso Merchant Energy Company on behalf of Southern California Edison, Docket No. RP00-241-000, May 2001, February 2002.
Before the National Energy Board of Canada, in the matter of TransCanada Pipelines, Ltd. Fair Return Application, March 2002.
Before the California Public Utilities Commission, Application of Wild Goose Storage Inc. to Amend its Certificate of Public Convenience and Necessity to Expand and Construct Facilities For Gas Storage Operation, Docket No. A. 01-06-029, November 2001.
Before the California Public Utilities Commission, Application of Southern California Gas Company Regarding Year Six (1999-2000) Under Its Experimental Gas Cost Incentive Mechanism and Related Gas Supply Matters, Application No. 00-06-023, (On behalf of Southern California Edison Company), November 2001.
Before the U.S. Congress, House of Representatives, Subcommittee on Energy Policy, Natural Resources and Regulatory Affairs, Hearings on California Natural Gas Market, October 2001.
Before the New Zealand Commerce Commission, Inquiry into Airfield Activities at Auckland, Wellington and Christchurch International Airports, July 2000, August 2001.
Before the National Energy Board of Canada in the matter of the National Energy Board Act and the Regulations made thereunder; and in the matter of an Application by TransCanada PipeLines Limited for orders pursuant to Part I and Part IV of the National Energy Board Act, June 2001.
Before the California Assembly, Subcommittee on Energy Oversight, Hearings into the Causes of the Natural Gas Price Increases During the California Energy Crisis, April 2001.
Before the California Public Utilities Commission, CPN Pipeline Co. & CPN Gas Marketing Co. v. Pacific Gas & Electric, Case No. C00-09-021, October 2000.
Before the California Public Utilities Commission in the matter of Southern California Gas Co. for Authority to Implement a Rate for Peaking Service, Application No. 00-06-032, (On behalf of Kern River Gas Transmission and Questar Southern Trails Pipeline Co.), September 2000.
Before the Federal Energy Regulatory Commission (FERC), California Public Utilities Commission v. El Paso Natural Gas Company, El Paso Merchant Energy-Gas, L.P., and El Paso Merchant Energy Company, Docket No. RP00-241-000, August 2000.
Kern River Gas Transmission, Federal Energy Regulatory Commission (FERC) Docket No.
12
Attachment A to the Written Reply Evidence of Paul R. Carpenter Page 13 of 15 Paul R. Carpenter
RP99-274-003, August 2000.
Before the California Public Utilities Commission, Rulemaking on the Commission’s Own Motion to Assess and Revise the Regulatory Structure Governing California’s Natural Gas Industry, California Natural Gas Market Conditions Report, Docket No. R.98-01-011, on behalf of Southern California Edison, July 1998.
Before the National Energy Board of Canada, Application of Alliance Pipeline Ltd., Hearing Order GH- 3-97, December 1997, April 1998.
Before the California Public Utilities Commission, Pacific Enterprises, Enova Corporation, et al. Merger Proceedings, Docket A.96-10-038, on behalf of Southern California Edison, August 1997.
In the Superior Court of the State of California for the County of Los Angeles, Pacific Pipeline System Inc. v. City of Los Angeles, on behalf of Pacific Pipeline System Inc., January 1997.
Before the U.K. Monopolies and Mergers Commission, British Gas Transportation and Storage Price Control Review, on behalf of Enron Capital and Trade Resources Limited, January 1997.
Northern Border Pipeline Company, Federal Energy Regulatory Commission (FERC) Docket No. RP96- 45-000, July 1996.
Wisconsin Electric Power Co., Northern States Power Co. Merger Proceedings. FERC Docket No. EC 95-16-000, on behalf of Madison Gas & Electric Co., Wisconsin Citizens Utility Board and the Wisconsin Electric Cooperative Association, May 1996.
Before the California Public Utilities Commission, Application of PG&E for Amortization of Interstate Transition Cost Surcharge, Application 94-06-044, on behalf of El Paso Natural Gas, December 1995.
Tennessee Gas Pipeline Company, FERC Docket No. RP95-112-000, on behalf of JMC Power Projects, September 1995.
Before the National Energy Board of Canada, Drawdown of Balance of Deferred Income Taxes Proceeding, RH-1-95, on behalf of Foothills Pipe Lines Ltd., September 1995.
Pacific Gas Transmission, FERC Docket No. RP94-149-000, on behalf of El Paso Natural Gas, May 1995.
Before the California Public Utilities Commission, Application of Pacific Pipeline System, Inc., A.91-10- 013, on behalf of PPSI, April 1995.
Before the National Energy Board of Canada, Multipipeline Cost of Capital Proceeding, RH-2-94, on behalf of Foothills Pipe Lines Ltd., November 1994.
Before the California Public Utilities Commission, Pacific Gas & Electric 1992 Operations Reasonableness Review, Application 93-04-011, on behalf of El Paso Natural Gas, November 1994.
Before the National Energy Board of Canada, Foothills Pipe Lines (Alta.) Ltd., Wild Horse Pipeline
13
Attachment A to the Written Reply Evidence of Paul R. Carpenter Page 14 of 15 Paul R. Carpenter
Project, Order No. GH-4-94, October 1994.
Iroquois Gas Transmission System, L.P., FERC Docket No. RP94-72-000, on behalf of Masspower and Selkirk Cogen Partners, September 1994.
Tennessee Gas Pipeline Co., FERC Docket No. RP91-203-000, on behalf of JMC Power Projects and New England Power Company, February, May 1994.
Before the California Public Utilities Commission, on the Application of Pacific Gas & Electric Company to Establish Interim Rates for the PG&E Expansion Project, July 1993.
Before the Florida Public Service Commission, Petition of Florida Power Corporation for Order Authorizing A Return on Equity for Florida Power’s Investment in the SunShine Intrastate and the SunShine Interstate Pipelines, FPSC Docket No. 930281-EI, June 4, 1993.
Before the Florida Public Service Commission, Application for Determination of Need for an Intrastate Natural Gas Pipeline by SunShine Pipeline Partners, FPSC Docket No. 920807-GP, April-May 1993.
Northwest Pipeline Corp., et. al., FERC Docket No. IN90-1-001, February 1993. City of Long Beach, Calif., vs. Unocal California Pipeline Co., before the California Public Utilities Commission, Case No. 91-12-028, February 1993.
Alberta Energy Resources Conservation Board, on Applications of NOVA Corporation of Canada to Construct Facilities, January 1993.
Before the California Public Utilities Commission, on the Application of Pacific Gas & Electric Co. to guarantee certain financing arrangements of Pacific Gas Transmission Co. not to exceed $751 million, 1992.
Mississippi River Transmission Co., FERC Docket No. RP93-4-000, October 1992, September 1993.
Unocal California Pipeline Co., FERC Docket No. IS92-18-000, August 1992.
Before the California Public Utilities Commission, in the Rulemaking into natural gas procurement and system reliability issues, R.88-08-018, June 1992.
Alberta Energy Resources Conservation Board, Altamont & PGT Pipeline Projects, Proceeding 911586, March 1992.
Before the California Utilities Commission, on the Application of Southern California Gas Company for approval of capital investment in facilities to permit interconnection with the Kern River/Mojave pipeline, A.90-11-035, May 1992.
Northern Natural Gas, FERC Docket No. RP92-1-000, October 1991.
Florida Gas Transmission, FERC Docket No. RP91-1-187-000 and CP91-2448-000, July 1991.
14
Attachment A to the Written Reply Evidence of Paul R. Carpenter Page 15 of 15 Paul R. Carpenter
Tarpon Transmission, FERC Docket No. RP84-82-004, January 1991.
Before the California Public Utilities Commission, on the Application of Pacific Gas & Electric Co. to Expand its Natural Gas Pipeline System, A.89-04-033, May 1990 and October 1991.
CNG Transmission, FERC Docket No. RP88-211, March 1990.
Panhandle Eastern Pipeline, FERC Docket No. RP88-262, March 1990.
Mississippi River Transmission, FERC Docket No. RP89-249, October 1989, September 1990.
Tennessee Gas Pipeline, FERC Docket No. CP89-470, June 1989.
Empire State Pipeline, Case No. 88-T-132 before the New York Public Service Commission, May 1989.
Before the U.S. Congress, House of Representatives, Committee on Energy and Commerce, Subcommittee on Energy and Power, Hearings on “Bypass” Legislation, May 1988.
Tennessee Gas Pipeline, FERC Docket No. RP86-119, 1986-87.
Mojave Pipeline Co., FERC Docket No. CP85-437, 1987-88.
Consolidated Gas Transmission Corp., FERC Docket No. RP88-10, 1988.
Panhandle Eastern, FERC Docket No. RP85-194, 1985.
On behalf of the Natural Gas Supply Association in FERC Rulemaking Docket No. RM85-1, 1985-86.
On behalf of the Panhandle Eastern Pipeline Co. in FERC Rulemaking Docket No. RM85-1, 1985.
15
Attachment B to the Written Reply Evidence of Paul R. Carpenter Page 1 of 1
HHI Based on Storage Withdrawal Capacity at Dawn And Capacity Holdings on Pipelines into Dawn Hub
Storage Withdrawal Capacity Holdings on Total Share Square of Capacity at Dawn Pipelines into Dawn (MMBtu/d) (%) % Share (MMBtu/d) Hub (MMBtu/d)
Tenaska Gas Storage LLC 478,337 50,904 529,241 10.47 109.53 DTE Energy Trading Inc. 0 346,358 346,358 6.85 46.91 Enbridge Gas Distribution Inc 162,817 175,000 337,817 6.68 44.63 Shell Energy North America US 260,000 73,500 333,501 6.60 43.49 TransCanada PipeLines Ltd. 0 304,063 304,063 6.01 36.16 BP Canada Energy Markt. Corp. 6,000 250,001 256,000 5.06 25.63 Gaz Metro Limited Partnership 200,459 0 200,459 3.96 15.71 Cargill Inc. 154,000 30,000 184,001 3.64 13.24 J. Aron & Co. 84,001 79,050 163,051 3.22 10.40 Suncor Energy Marketing Inc. 154,358 5,962 160,319 3.17 10.05 Westcoast Energy Enterprises 0 160,000 160,000 3.16 10.01 Castleton Commodities 84,000 64,001 148,002 2.93 8.57 Powerex Corp. 144,374 0 144,374 2.86 8.15 Union Gas Ltd. 0 127,000 127,000 2.51 6.31 Goreway Station Partnership by its managing partner Goreway Power Station Holdings ULC 121,321 0 121,321 2.40 5.76 Bluewater Gas Storage LLC 80,000 38,334 118,334 2.34 5.48 United Energy Trading Canada, ULC 56,000 60,000 116,000 2.29 5.26 PT Petrochina International 44,910 58,006 102,917 2.04 4.14 ONEOK Energy Services Co L.P. 0 100,000 100,000 1.98 3.91 Tidal Energy Marketing (US) 0 85,000 85,000 1.68 2.83 York Energy Centre LP 83,080 0 83,080 1.64 2.70 EDF Trading North America LLC 58,500 19,904 78,404 1.55 2.40 Koch Canada Energy Services, LP 75,001 0 75,001 1.48 2.20 ConocoPhillips Co. 0 65,000 65,000 1.29 1.65 Yankee Gas Services Co. 0 60,200 60,200 1.19 1.42 DTE Gas Co. 0 50,000 50,000 0.99 0.98 Exelon Generation Company, LLC 47,999 0 47,999 0.95 0.90 Repsol Energy North America 24,000 20,000 44,000 0.87 0.76 Thorold CoGen L.P. by its General Partner Northland Power Thorold Cogen GP Inc. 41,704 0 41,704 0.82 0.68 Greenfield Energy Centre LP 40,000 0 40,000 0.79 0.63 Portlands Energy Centre L.P., by its General Partner, Portlands Energy Centre Inc. 37,913 0 37,913 0.75 0.56 TransCanada Power 33,363 0 33,363 0.66 0.44 AltaGas Ltd. 31,847 0 31,847 0.63 0.40 St. Clair Power, L.P. 26,899 0 26,899 0.53 0.28 Bay State Gas Company 0 26,645 26,645 0.53 0.28 Energy America LLC 0 25,000 25,000 0.49 0.24 Sequent Energy Mgmt LP 0 25,000 25,000 0.49 0.24 Barclays Canadian Commodities Limited 24,000 0 24,000 0.47 0.23 NJR Energy Services Company 24,000 0 24,000 0.47 0.23 Freepoint Commodities LLC 3,000 20,000 23,000 0.45 0.21 Emera Energy Services Inc 0 20,000 20,000 0.40 0.16 Merrill Lynch Commodities Inc. 0 20,000 20,000 0.40 0.16 Southern Connecticut Gas Co. 0 18,300 18,300 0.36 0.13 Twin Eagle Resource Management Canada, LLC 0 15,000 15,000 0.30 0.09 Brooklyn Union Gas Co. 0 12,500 12,500 0.25 0.06 KeySpan Gas East Corp. 0 12,500 12,500 0.25 0.06 Noble Americas Gas & Power Corp. 0 10,000 10,000 0.20 0.04 St. Lawrence Gas Company, Inc. 9,905 0 9,905 0.20 0.04 CT Natural Gas Corp. 0 9,700 9,700 0.19 0.04 Northland Power Inc. 8,281 0 8,281 0.16 0.03 Northern Utilities Inc. 0 6,070 6,070 0.12 0.01 MIECO INC. 6,000 0 6,000 0.12 0.01 1425445 Ontario Limited o/a Utilities Kingston 5,118 0 5,118 0.10 0.01 Ag Energy Co‐operative Ltd. 2,400 0 2,400 0.05 0.00 Energy Source Natural Gas Inc. 240 0 240 0.00 0.00
Total 2,613,824 2,443,000 5,056,824 100.00 434.42
Sources: Storage withdrawal capacity at Dawn from Index of Customers for storage customers on Union Gas website. Capacity holdings on pipelines to Dawn Hub from Index of Customers for each pipeline from SNL, and from TransCanada response to ANE 1.3a. Notes: Pipeline capacity holdings includes Panhandle Eastern, Vector, and TransCanda pipelines. Storage withrawal capacity excludes in‐franchise Dawn storage withdrawal capacity of Union Gas. HHI calculated as the sum of the square of each capacity holder's share of total capacity. Figures have been converted from GJ/d to MMBtu/d.