1016-00004 · Dunmore Energy Holdings · Expert Engagement

whitfield-analysis-summary-memo.docx DOCUMENT Finance & Banking

15 / 35 ← → ✕

PRIVILEGED AND CONFIDENTIAL — ATTORNEY WORK PRODUCT

MEMORANDUM

Date: November 15, 2023

To: Catherine Bloom, Responsible Partner, Calderwood & Harkness LLP Hannah Brinkley, Lead Associate, Calderwood & Harkness LLP

From: Patrick Gallagher, Working Associate, Calderwood & Harkness LLP

Re: Summary of Expert Economic Analysis — Dr. Leonard Whitfield, Whitfield Economic Advisors LLC; FERC Docket No. IN23-7-000

C&H Matter Number: 1016-00004

C&H Billing Code: 71420.00001

Client: Dunmore Energy Holdings Ltd. / Dunmore Power Trading LLC

I. Purpose and Scope of Memorandum

The purpose of this memorandum is to summarize the expert economic analysis prepared by Dr. Leonard Whitfield, Principal Expert, Whitfield Economic Advisors LLC ("WEA"), 1700 Pennsylvania Avenue NW, Suite 550, Washington, D.C. 20006, for the benefit of the Calderwood & Harkness LLP ("C&H") case team handling the above-referenced enforcement matter before the Federal Energy Regulatory Commission ("FERC" or the "Commission").

Dr. Whitfield's expert report (the "Whitfield Report") was prepared to rebut the market impact and unjust profits calculations set forth by FERC's Office of Enforcement ("OE") in its Preliminary Findings letter dated October 2, 2023 (the "Preliminary Findings Letter"). The Preliminary Findings Letter contains detailed quantitative analyses prepared by Dr. Samuel Breck, Senior Economist, FERC Division of Analytics & Surveillance, which purport to establish both the existence and magnitude of market harm allegedly caused by the trading activities of Dunmore Power Trading LLC ("DPT"). Dr. Whitfield's Report addresses each of Dr. Breck's principal quantitative conclusions and identifies significant methodological deficiencies that, in Dr. Whitfield's opinion, result in a substantial overstatement of both market impact and alleged unjust profits.

WEA was retained pursuant to an engagement letter dated July 10, 2023, executed between Calderwood & Harkness LLP and Dr. Whitfield. The engagement provides for compensation at an hourly rate of $475 per hour for Dr. Whitfield's personal time, with staff economist rates ranging from $225 to $375 per hour depending on seniority. The engagement was structured through counsel to preserve the protection of the attorney work product doctrine, as Dr. Whitfield's analysis and conclusions were developed at the direction of C&H attorneys in anticipation of contested proceedings before FERC and, potentially, in federal court.

The Whitfield Report is scheduled to be submitted to FERC OE on November 20, 2023, as an attachment to our supplemental submission in FERC Docket No. IN23-7-000. This memorandum summarizes Dr. Whitfield's principal findings and conclusions for internal review and strategic planning purposes in advance of that formal submission. The memorandum is intended to facilitate discussion among the case team regarding the implications of Dr. Whitfield's analysis for our settlement strategy and for the framing of Dunmore's defenses in any future proceedings.

Nothing in this memorandum is intended to constitute legal advice to the client independent of the broader representation, and the contents should not be shared outside of the C&H case team without the express approval of the Responsible Partner.

II. Background: FERC Staff's Allegations and Quantification

A. Overview of Staff's Theory

FERC Staff alleges that Dunmore Power Trading LLC ("DPT"), the wholly owned power marketing subsidiary of Dunmore Energy Holdings Ltd. ("DEH"), engaged in a course of market manipulation in the Midcontinent Independent System Operator, Inc. ("MISO") Day-Ahead and Real-Time energy markets in violation of Section 222 of the Federal Power Act, 16 U.S.C. § 824v, and the Commission's Anti-Manipulation Rule, 18 C.F.R. § 1c.2. The investigation was initiated following a referral report dated April 3, 2023, prepared by Velthorn Market Analytics, Inc. ("Velthorn"), the MISO Independent Market Monitor, under the supervision of Dr. Patricia Engel, President of Velthorn.

The Relevant Period identified by FERC Staff spans from January 3, 2022 through March 17, 2023 — a period of approximately fourteen and one-half months. The pricing nodes at issue are the Indiana Hub and the Minnesota Hub, both located within the MISO footprint. These two pricing nodes are key commercial reference points within the MISO market and are the settlement locations for a substantial volume of financial transmission rights, virtual transactions, and bilateral power contracts.

Staff's theory of the case encompasses three interrelated components of an alleged manipulative scheme:

1. Uneconomic Day-Ahead Bidding. FERC Staff alleges that Kyle Renfro, Senior Trader on DPT's MISO Day-Ahead Desk, submitted Day-Ahead supply offers for Dunmore's gas-fired generation units at prices significantly below marginal cost. According to Staff, these offers ranged from negative $15 per megawatt-hour to negative $40 per megawatt-hour at the Indiana Hub. Staff contends that these below-cost offers were designed to create artificial congestion on transmission pathways between the Indiana Hub and the Minnesota Hub, and that DPT held Financial Transmission Rights ("FTRs") that would benefit financially from the resulting congestion differentials. In Staff's view, the negative offers lacked any legitimate economic rationale and were submitted solely to benefit DPT's FTR positions.

2. Cross-Product Manipulation. FERC Staff further alleges that Anastasia Petrov, a Trader on DPT's MISO Real-Time Desk, executed Real-Time virtual positions — specifically, Increment Offers ("INCs") and Decrement Bids ("DECs") — at the Minnesota Hub that were designed to amplify the congestion differential between the Indiana Hub and the Minnesota Hub. Staff characterizes these virtual transactions as complementary to the uneconomic Day-Ahead offers submitted by Mr. Renfro, asserting that the two sets of transactions operated in concert to maximize the spread between the two pricing nodes and thereby increase the value of DPT's FTR positions.

3. Physical-Financial Coordination. Finally, FERC Staff alleges that DPT coordinated the timing and volume of natural gas nominations through Dunmore Midstream Partners LP ("DMP") (EIN: 47-8831205), a sister entity within the Dunmore corporate family, to support the uneconomic generation offers described above and to mask DPT's true marginal costs. Specifically, Staff contends that DPT procured gas volumes through DMP at non-market rates or on terms that obscured the true economic cost of fuel delivery to DPT's generation units, thereby creating the appearance that the negative Day-Ahead offers were economically rational when, in Staff's view, they were not.

Staff's theory thus rests on the premise that the three components — uneconomic Day-Ahead bidding, cross-product virtual trading, and coordinated gas nominations — constituted a single, integrated scheme to manipulate congestion pricing within MISO for the purpose of enriching DPT's FTR portfolio at the expense of other market participants.

B. FERC Staff's Quantification

FERC Staff's Preliminary Findings Letter includes detailed quantitative estimates of both the volume of allegedly manipulative activity and the resulting financial impact. The key figures, as stated in the Preliminary Findings Letter and the accompanying analysis prepared by Dr. Breck, are as follows:

• Total Day-Ahead uneconomic generation offers: Staff calculated that DPT submitted approximately 2,840 gigawatt-hours ("GWh") of generation offers at prices below estimated marginal cost during the Relevant Period. This figure encompasses all generation offers at the Indiana Hub that were priced at or below negative $15 per megawatt-hour.

• Total FTR positions benefiting from alleged manipulation: Staff identified FTR positions held by DPT with a notional value of approximately $47.3 million. Staff asserts that these FTR positions were directionally aligned with the congestion patterns created by DPT's uneconomic Day-Ahead offers and therefore represent the financial instruments through which DPT captured the fruits of its allegedly manipulative conduct.

• Total virtual transactions at Minnesota Hub: Staff identified approximately 1,620 GWh of virtual transactions (INCs and DECs) executed by Ms. Petrov at the Minnesota Hub during the Relevant Period that Staff contends were designed to amplify the congestion differential.

• Estimated "unjust profits": Based on the analysis prepared by Dr. Breck, FERC Staff estimated that DPT realized "unjust profits" of $12.8 million attributable to the alleged scheme. This figure reflects the total incremental profits that, in Staff's estimation, DPT earned on its FTR positions and virtual transactions as a result of the congestion patterns purportedly created by the uneconomic Day-Ahead offers.

• Initial penalty demand: In addition to disgorgement of the $12.8 million in estimated unjust profits, Staff proposed an initial civil penalty of $18 million. This penalty figure reflects a base penalty of $9 million, which was doubled to account for the duration and scope of the alleged conduct under the FERC Penalty Guidelines.

• Total initial demand: Staff's combined initial demand thus totals $30.8 million, comprising the $18 million civil penalty and the $12.8 million disgorgement amount.

These figures represent Staff's opening position and are the principal target of Dr. Whitfield's rebuttal analysis, as summarized below.

III. Dr. Whitfield's Qualifications

Dr. Leonard Whitfield holds a Ph.D. in Economics from the Massachusetts Institute of Technology, with a specialization in energy markets and industrial organization. He is the Principal and founder of Whitfield Economic Advisors LLC, a boutique consulting firm focused on the economic analysis of wholesale electricity and natural gas markets. Dr. Whitfield has over twenty-five years of experience in the analysis of organized wholesale electricity markets, with particular expertise in the MISO and Southwest Power Pool ("SPP") regional transmission organization footprints.

Throughout his career, Dr. Whitfield has served as an expert consultant and testifying expert in numerous proceedings before FERC, including both enforcement proceedings under Section 222 of the Federal Power Act and rate proceedings under Sections 205 and 206. He has also provided expert testimony in related state regulatory proceedings and in federal court litigation arising from wholesale electricity market disputes.

Dr. Whitfield's areas of expertise include, but are not limited to, market impact analysis, congestion pricing in locational marginal price ("LMP") markets, the valuation and risk management of Financial Transmission Rights, the economics of virtual trading (including INCs and DECs), unit commitment and dispatch economics for thermal generation resources, and the analysis of bidding behavior in organized wholesale electricity markets. He has authored peer-reviewed articles on congestion pricing methodology and has presented at numerous industry conferences on wholesale market design and market manipulation detection.

Dr. Whitfield has been retained as an expert in contested proceedings before FERC on multiple prior occasions, and his analyses have been cited with approval by FERC Staff in at least two prior enforcement matters. His curriculum vitae, which is included as an appendix to the Whitfield Report, details over forty prior expert engagements in energy market disputes.

Based on his extensive experience and qualifications, Dr. Whitfield is well qualified to opine on the specific issues raised in this proceeding, including the economic rationale for DPT's Day-Ahead bidding behavior, the drivers of congestion pricing at the Indiana Hub, the market impact of DPT's trading activity, and the methodological adequacy of Staff's unjust profits calculations.

IV. Summary of Dr. Whitfield's Key Findings

Dr. Whitfield's Report contains four principal findings, each of which addresses a discrete element of FERC Staff's theory and quantification. The findings are summarized in turn below.

A. Finding 1: DPT's Negative Day-Ahead Offers Were Consistent with Portfolio Optimization and Contractual Minimum Generation Obligations

Dr. Whitfield conducted a comprehensive analysis of DPT's Day-Ahead offer data at the Indiana Hub across the entire Relevant Period, examining the pricing, volume, timing, and frequency of the offers that FERC Staff characterizes as "uneconomic." This analysis included a detailed review of DPT's internal bidding records, generation dispatch records, fuel procurement documentation, and contractual arrangements with DMP and third-party counterparties.

Dr. Whitfield's principal conclusion on this issue is that the negative offers submitted by Mr. Renfro — ranging from negative $15 per megawatt-hour to negative $40 per megawatt-hour — were consistent with a legitimate portfolio optimization strategy employed by DPT and did not constitute uneconomic or manipulative bidding.

Central to this conclusion is the fact that DPT's gas-fired generation units at the Indiana Hub are part of Dunmore's broader fleet of 2,650 megawatts of U.S. natural gas-fired generation capacity. A significant portion of these units were subject to contractual minimum generation commitments — specifically, minimum take-or-pay obligations under gas supply agreements with Dunmore Midstream Partners LP and minimum dispatch requirements under long-term power purchase agreements with certain MISO load-serving entities. These contractual commitments required certain units to remain online and generating at specified minimum output levels regardless of prevailing Day-Ahead clearing prices. Under these circumstances, negative offers reflected the economic reality of the opportunity costs associated with curtailing or shutting down generation in the face of binding contractual obligations.

Dr. Whitfield explains that when a generator faces binding minimum generation constraints — whether driven by contractual take-or-pay obligations, gas pipeline scheduling requirements, or physical operating limitations — it is entirely rational, and commonplace, for the generator to submit Day-Ahead offers at negative prices. By offering at negative prices, the generator ensures that it will be dispatched in the Day-Ahead market and thereby avoid the potentially greater costs associated with contract breach, gas pipeline imbalance penalties, or the physical and financial costs of cycling thermal units through start-up and shut-down sequences.

To contextualize DPT's bidding behavior, Dr. Whitfield identified comparable negative bidding patterns by other MISO market participants during the same Relevant Period. His analysis of publicly available MISO offer data revealed that multiple other generators at the Indiana Hub and at neighboring pricing nodes submitted Day-Ahead offers at negative prices during the Relevant Period, and that the frequency and magnitude of DPT's negative offers were not anomalous relative to the broader market. This finding is significant because it demonstrates that DPT's bidding behavior, rather than being indicative of a manipulative scheme, was consistent with the behavior of similarly situated market participants responding to comparable market conditions and contractual constraints.

Dr. Whitfield further concludes that FERC Staff's marginal cost calculation — which formed the basis for Staff's determination that DPT's offers were "below cost" — failed to account for several material cost components. Specifically, Staff's marginal cost estimate did not incorporate: (a) start-up and shut-down costs for cycling gas-fired units, which can be substantial for combined-cycle and combustion turbine units of the type operated by DPT; (b) minimum run-time constraints embedded in MISO's unit commitment protocols, which effectively commit generators to operate for a specified number of hours once started and thereby create sunk costs that should be reflected in incremental offer pricing; and (c) contractual obligations requiring minimum gas throughput on DMP pipelines, which imposed real economic costs on DPT that would be avoided only by continuing to generate and consume gas at the required minimum volumes.

Finally, Dr. Whitfield emphasizes the importance of evaluating DPT's Day-Ahead offers in the context of DPT's broader generation and trading portfolio, rather than in isolation. DPT managed a diversified portfolio comprising 4,280 megawatts of total U.S. net generation capacity, including 2,650 megawatts of natural gas-fired generation, 1,130 megawatts of wind generation, and 500 megawatts of solar generation. Within such a diversified portfolio, it is standard industry practice to optimize individual unit offers with reference to the overall portfolio economics, including the interaction between generation dispatch, transmission congestion, and hedging positions. Dr. Whitfield concludes that DPT's negative Day-Ahead offers, viewed in the context of this broader portfolio, reflected a rational and legitimate optimization strategy that is common among sophisticated portfolio managers in organized wholesale electricity markets.

B. Finding 2: Congestion Patterns at Indiana Hub Were Primarily Driven by Transmission Outage Events, Not DPT's Bidding

Dr. Whitfield performed a rigorous regression analysis of congestion pricing at the Indiana Hub during the Relevant Period, spanning from January 3, 2022 through March 17, 2023. The purpose of this analysis was to determine the extent to which the congestion patterns observed at the Indiana Hub — and upon which FERC Staff's theory of manipulation critically depends — were attributable to DPT's Day-Ahead bidding behavior versus other, exogenous market factors.

Dr. Whitfield's key finding on this issue is that transmission outage events were the dominant causal factor in the congestion patterns that FERC Staff attributed to DPT's bidding. Specifically, Dr. Whitfield identified a series of both planned and unplanned transmission outages on key MISO transmission facilities during the Relevant Period that independently created the congestion differentials Staff attributed to DPT. These outage events included planned maintenance on major 345 kV transmission lines serving the Indiana Hub area, as well as several forced outages on critical transmission facilities resulting from equipment failures and severe weather events.

Dr. Whitfield's correlation analysis demonstrated that the periods of highest congestion at the Indiana Hub coincided with identified transmission outage events, not with DPT's bidding patterns. In particular, the most significant congestion episodes during the Relevant Period — which accounted for a disproportionate share of the total FTR profits Staff attributes to DPT — occurred during intervals in which one or more major transmission facilities in the Indiana Hub area were out of service. During these same intervals, DPT's Day-Ahead offers were not materially different in pricing or volume from DPT's offers during non-outage periods, suggesting that the congestion was driven by the transmission system topology, not by DPT's offer behavior.

When Dr. Whitfield controlled for transmission outage events in his regression model — that is, when the model included variables accounting for the duration, location, and capacity impact of known transmission outages — the incremental congestion attributable to DPT's Day-Ahead offers was statistically insignificant in the majority of hourly intervals during the Relevant Period. This result held across multiple model specifications and robustness checks, and was consistent regardless of whether Dr. Whitfield used actual Day-Ahead LMP data or shadow price decomposition data from MISO's market clearing engine.

This finding directly and substantially undermines FERC Staff's theory — and the analysis prepared by Dr. Breck — that DPT's uneconomic offers "created" the congestion from which DPT's FTR positions profited. If the observed congestion patterns were primarily driven by transmission outage events that had nothing to do with DPT's bidding, then the causal chain upon which Staff's theory of manipulation depends is broken. The FTR profits that Staff attributes to manipulative conduct were, in large part, simply the result of DPT holding hedging positions that happened to be in the money during periods of outage-driven congestion — a result that is entirely consistent with the legitimate hedging purpose for which FTRs are designed.

C. Finding 3: Actual Market Impact Estimated at No More Than $2.1 Million (Versus FERC Staff's $12.8 Million)

In addition to his analysis of congestion causation, Dr. Whitfield conducted an independent "but-for" market impact analysis to estimate the actual profits, if any, that were attributable to DPT's bidding behavior during the Relevant Period. This analysis represents the core quantitative rebuttal to Dr. Breck's $12.8 million unjust profits estimate.

Dr. Whitfield's methodology involved comparing actual MISO market outcomes — including Day-Ahead and Real-Time LMPs, congestion components, and FTR settlement values at the Indiana Hub and Minnesota Hub — to a counterfactual scenario in which DPT's Day-Ahead offers at the Indiana Hub were priced at estimated marginal cost rather than at the negative prices actually submitted. In constructing this counterfactual, Dr. Whitfield controlled for transmission outages and other exogenous market conditions (including weather, load levels, fuel prices, and the bidding behavior of other market participants) to isolate the incremental effect, if any, of DPT's negative offers on congestion pricing and, by extension, on DPT's financial positions.

The result of this analysis is that the actual market impact attributable to DPT's bidding behavior is estimated at no more than $2.1 million over the entire Relevant Period. This figure represents the incremental change in DPT's FTR profits and virtual trading profits that would result from repricing DPT's Day-Ahead offers at estimated marginal cost, after controlling for transmission outage effects and other exogenous factors.

This $2.1 million estimate stands in stark contrast to FERC Staff's estimate of $12.8 million in unjust profits, representing a reduction of approximately 84 percent. Dr. Whitfield attributes this substantial discrepancy to four key differences between his methodology and Staff's:

First, Staff's analysis did not adequately control for the effects of transmission outage events on congestion pricing. As discussed in Finding 2, transmission outages were the primary driver of congestion at the Indiana Hub during the Relevant Period. By failing to control for outage events, Staff's model attributed outage-driven congestion — and the associated FTR profits — to DPT's bidding behavior, resulting in a significant upward bias in the estimated market impact.

Second, Staff attributed all FTR profits during the Relevant Period to the alleged scheme, even though the FTR positions would have generated positive returns independent of DPT's Day-Ahead bidding due to the outage-driven congestion that prevailed during much of the Relevant Period. Dr. Whitfield's counterfactual analysis demonstrates that DPT's FTR portfolio would have been profitable during the Relevant Period regardless of whether DPT had submitted negative Day-Ahead offers, because the congestion that drove FTR profitability was overwhelmingly attributable to transmission outage events rather than to DPT's bidding.

Third, Staff's marginal cost calculation used a simplified heat rate approach — multiplying a generic heat rate by a daily natural gas price index — that did not account for DPT's actual delivered gas costs through DMP infrastructure, start-up and shut-down costs for thermal units, or the minimum generation economics described in Finding 1. By understating DPT's true marginal costs, Staff's analysis overstated the degree to which DPT's offers were "uneconomic," thereby inflating the volume of offers treated as manipulative and the associated market impact.

Fourth, Staff aggregated profits across both Indiana Hub FTR positions and Minnesota Hub virtual positions, creating a cumulative overcount. Dr. Whitfield's analysis demonstrates that a portion of the virtual trading profits at the Minnesota Hub were attributable to market conditions unrelated to congestion at the Indiana Hub and therefore should not be included in the unjust profits calculation even under Staff's own theory.

Dr. Whitfield notes that his $2.1 million estimate represents a conservative upper bound and may overstate the actual market impact attributable to DPT's bidding behavior. The estimate does not fully isolate the effect of seasonal weather patterns and load variability on congestion pricing, and it does not account for the possibility that, in the counterfactual scenario, other market participants might have submitted similarly priced offers that would have produced comparable congestion outcomes. To the extent these additional factors were incorporated, the estimated market impact could be even lower than $2.1 million.

D. Finding 4: FTR Positions Were Acquired Months Before the Relevant Period as Routine Hedging

Dr. Whitfield reviewed the acquisition timeline and underlying rationale for DPT's FTR portfolio, with particular attention to the FTR positions identified by FERC Staff as instruments through which DPT allegedly captured the profits of its manipulative scheme.

Dr. Whitfield's conclusion is that the FTR positions identified by FERC Staff — with a notional value of approximately $47.3 million — were acquired months before the start of the Relevant Period (January 3, 2022) as part of DPT's routine hedging program. Specifically, the bulk of the FTR portfolio in question was acquired through MISO's annual FTR auction process, with the relevant auction clearing dates falling in the third and fourth quarters of 2021 — well in advance of the January 2022 start of the Relevant Period. Additional FTR positions were acquired through MISO's monthly FTR auctions, but these, too, were acquired on dates preceding the allegedly manipulative bidding activity by periods ranging from several weeks to several months.

Dr. Whitfield compared the acquisition dates, volumes, and pricing of DPT's FTR positions during this period against DPT's historical hedging patterns in prior years. He found that DPT's FTR acquisitions for the 2022 and 2023 planning years were consistent with the size, direction, and pricing of FTR positions DPT had acquired in each of the five preceding planning years. There was no discernible increase in FTR acquisition volume or shift in portfolio direction that would suggest DPT was building FTR positions in anticipation of, or in coordination with, the allegedly manipulative Day-Ahead bidding.

DPT's FTR portfolio was managed by the trading desk under the supervision of Derek Sung, Senior Vice President of Power Marketing & Trading, who oversaw a team of fourteen traders. The FTR positions were acquired and managed under DPT's standard risk management framework, which included board-approved risk limits, daily mark-to-market reporting, and periodic portfolio reviews by the risk management committee. Dr. Whitfield found no evidence that the FTR acquisition process deviated from these established protocols during the Relevant Period.

The temporal separation between FTR acquisition and the alleged manipulative bidding is significant because it undermines FERC Staff's inference of manipulative intent. Under Section 222 of the Federal Power Act and 18 C.F.R. § 1c.2, a finding of manipulation requires evidence of scienter — i.e., that the respondent acted with the intent to defraud or manipulate. If DPT acquired its FTR positions months before the allegedly manipulative bidding began, the inference that DPT's Day-Ahead offers were submitted for the purpose of benefiting those FTR positions is substantially weakened. The more natural inference, consistent with Dr. Whitfield's findings, is that DPT acquired FTR positions as part of its routine hedging activities and subsequently submitted Day-Ahead offers for legitimate portfolio optimization reasons unrelated to the FTR positions.

Dr. Whitfield also noted that other MISO market participants with generation portfolios comparable in size and composition to DPT's held FTR positions of similar magnitude and direction during the same period. This observation further supports the conclusion that DPT's FTR portfolio reflected standard industry hedging practice rather than a building block of a manipulative scheme.

V. Assessment of FERC Staff's Methodology and Weaknesses

Dr. Whitfield's Report dedicates a substantial section to identifying specific methodological weaknesses in the analysis prepared by Dr. Samuel Breck, Senior Economist, FERC Division of Analytics & Surveillance. These weaknesses, which Dr. Whitfield addresses both individually and in the aggregate, are summarized below.

1. Failure to Control for Transmission Outages. Staff's regression model, which purported to isolate the effect of DPT's bidding on congestion pricing at the Indiana Hub, did not include a variable for known transmission outages. This omission constitutes a textbook example of omitted variable bias — a well-recognized econometric error that arises when a regression model fails to account for a factor that is correlated with both the independent variable of interest (DPT's bidding) and the dependent variable (congestion pricing). Because transmission outages are a major determinant of congestion pricing and occurred with significant frequency during the Relevant Period, the omission of an outage variable from Staff's model resulted in the inflation of the estimated impact of DPT's bidding. In effect, Staff's model attributed to DPT's bidding the congestion effects that were actually caused by transmission outages, leading to a substantial overstatement of market impact.

2. Overinclusive Profit Calculation. Staff's unjust profits estimate of $12.8 million included all FTR profits and all virtual trading profits at the Minnesota Hub during the Relevant Period, rather than isolating only the incremental profits attributable to the allegedly manipulative bids. This approach is inconsistent with the established legal and economic standard for disgorgement, which requires that only the profits causally attributable to the wrongful conduct be disgorged. By including profits that DPT would have earned regardless of its Day-Ahead bidding behavior — including FTR profits driven by outage-related congestion and virtual trading profits attributable to factors unrelated to Indiana Hub congestion — Staff's calculation significantly overstated the appropriate disgorgement amount.

3. Simplified Marginal Cost Estimation. Staff's determination that DPT's Day-Ahead offers were "below marginal cost" was based on a simplified cost methodology in which Staff multiplied a single heat rate by a daily natural gas price index to arrive at an estimated marginal cost for each of DPT's generation units. This approach did not account for DPT's actual delivered gas costs, which include pipeline transportation charges and scheduling costs associated with gas delivery through DMP infrastructure. Nor did Staff's methodology account for start-up and shut-down costs, which can be substantial for gas-fired combined-cycle and combustion turbine units and which are properly included in the economic evaluation of generation offers. Additionally, Staff's marginal cost estimate did not reflect DPT's contractual minimum generation obligations, which — as discussed in Finding 1 — imposed real economic costs that justified negative offer pricing. The result of Staff's simplified methodology was an understatement of DPT's true marginal costs and a corresponding overstatement of the degree to which DPT's offers were "uneconomic."

4. Absence of a Rigorous Counterfactual. Staff's "but-for" analysis assumed that, in the absence of DPT's below-cost offers, no other market participant would have submitted similarly priced offers at the Indiana Hub. Dr. Whitfield found this assumption to be unsupported given the observable competitive bidding patterns at the Indiana Hub during the Relevant Period. As noted in Finding 1, multiple other market participants submitted negative Day-Ahead offers at the Indiana Hub and at neighboring pricing nodes during the Relevant Period, and there is no basis in the record for assuming that these other participants would not have continued to do so in the counterfactual scenario. By failing to account for the likely bidding behavior of other market participants in the but-for scenario, Staff's analysis overstated the incremental impact of DPT's offers on congestion pricing.

Dr. Whitfield concludes that these methodological weaknesses, individually and collectively, resulted in a significant overstatement of both the market impact of DPT's bidding activity and the alleged unjust profits attributable to DPT. Dr. Whitfield's independent analysis, which corrects for these deficiencies, yields an estimated market impact of no more than $2.1 million — approximately 16 percent of Staff's estimate.

VI. Implications for Settlement Strategy

Dr. Whitfield's analysis provides a strong foundation for challenging FERC Staff's Preliminary Findings and, in particular, for contesting the $12.8 million disgorgement claim as speculative and unquantifiable with sufficient certainty to support an enforcement action or penalty assessment.

The $2.1 million market impact estimate established by Dr. Whitfield provides a defensible upper bound that should serve as the anchor for Dunmore's negotiating position in any settlement discussions with FERC OE. If the Commission were to accept Dr. Whitfield's analysis — or even to accept that Staff's methodology is subject to the weaknesses Dr. Whitfield identifies — the total demand would be substantially reduced from the $30.8 million initially proposed by Staff. Even if Staff were to reject Dr. Whitfield's market impact figure entirely, the identified methodological deficiencies in Dr. Breck's analysis create significant litigation risk for Staff and should incentivize a more reasonable settlement posture.

The portfolio optimization defense articulated in Finding 1 and the FTR acquisition timing evidence presented in Finding 4 directly support the argument that DPT's conduct lacked the manipulative intent required to sustain a violation under 18 C.F.R. § 1c.2 and Section 222 of the Federal Power Act. The Commission has consistently held that a violation of the Anti-Manipulation Rule requires proof of scienter, and Dr. Whitfield's findings provide substantial economic evidence that DPT's trading activities were motivated by legitimate business considerations — portfolio optimization, contractual compliance, and routine hedging — rather than by an intent to manipulate market prices.

Dr. Whitfield's findings are consistent with and reinforce the legal and factual arguments presented in C&H's Response to Preliminary Findings submitted on November 6, 2023. In particular, the transmission outage analysis presented in Finding 2 complements the legal arguments regarding causation that were central to the November 6 submission, and the $2.1 million market impact estimate provides quantitative support for the position that Staff's disgorgement demand is grossly excessive.

The transmission outage analysis is particularly effective as a defense tool because it provides an independent, data-driven, and empirically verifiable explanation for the congestion patterns that FERC Staff attributed entirely to DPT's bidding. This analysis shifts the narrative from one in which DPT's bidding "created" artificial congestion to one in which congestion was primarily driven by infrastructure constraints — a framing that is both more consistent with the data and substantially more favorable to Dunmore's position.

Recommendation. The case team recommends deploying Dr. Whitfield's findings aggressively in the upcoming settlement negotiations to reduce both the penalty and disgorgement components of Staff's demand. The submission of the Whitfield Report on November 20, 2023, should be accompanied by a cover letter emphasizing the key findings and requesting a settlement conference at which Dr. Whitfield can present his analysis directly to OE staff. In the interim, the case team should prepare Dr. Whitfield for potential questions regarding his methodology, data sources, and conclusions, and should coordinate with Dr. Whitfield on any supplemental analyses that may be needed to address Staff's anticipated counterarguments.

VII. Status of Expert Fees

WEA's engagement with C&H provides for compensation at an hourly rate of $475 per hour for Dr. Whitfield's personal time, with staff economist rates as provided in the engagement letter dated July 10, 2023. Through the expected date of matter closure, total WEA fees are estimated at approximately $312,500.

Fees accrued to date (as of the date of this memorandum, November 15, 2023) reflect work through the preparation and finalization of the expert report, including data collection and analysis, regression modeling, preparation of the counterfactual market impact study, drafting and revision of the expert report, and coordination with C&H attorneys. The expert report is scheduled for submission to FERC OE on November 20, 2023.

Additional fees may accrue beyond the current estimate if Dr. Whitfield is needed for settlement conference support, preparation of supplemental analyses in response to Staff inquiries, or participation in any formal hearing proceedings. The case team will monitor fee accruals and provide updated estimates to the client as appropriate.

VIII. Attachments and Cross-References

The following documents are referenced in this memorandum and should be consulted for additional detail:

1. Expert Report of Dr. Leonard Whitfield, Whitfield Economic Advisors LLC — Final report to be submitted to FERC OE on November 20, 2023, as an attachment to C&H's supplemental submission in FERC Docket No. IN23-7-000. The report contains Dr. Whitfield's full analysis, supporting data, regression results, and appendices, including his curriculum vitae.

2. FERC Staff Preliminary Findings Letter, dated October 2, 2023 — Letter from FERC OE to C&H setting forth Staff's preliminary findings, including the factual allegations, legal theories, quantification of unjust profits, and initial penalty demand. This letter is the document to which Dr. Whitfield's report is principally responsive.

3. C&H Response to Preliminary Findings, submitted November 6, 2023 — C&H's written response to the Preliminary Findings Letter, setting forth Dunmore's legal and factual defenses. Dr. Whitfield's expert report supplements and reinforces the arguments presented in this submission.

4. WEA Engagement Letter, dated July 10, 2023 — Engagement letter between Calderwood & Harkness LLP and Dr. Leonard Whitfield, Whitfield Economic Advisors LLC, establishing the terms of Dr. Whitfield's retention, including the hourly rate, scope of engagement, and confidentiality obligations.

5. Velthorn Market Analytics, Inc. Referral Report, dated April 3, 2023 — Report prepared by Velthorn Market Analytics, Inc. (Dr. Patricia Engel, President), in its capacity as the MISO Independent Market Monitor, referring certain trading patterns by DPT to FERC OE for investigation. This referral initiated the investigation that resulted in FERC Docket No. IN23-7-000.

PRIVILEGED AND CONFIDENTIAL — ATTORNEY WORK PRODUCT — PREPARED IN ANTICIPATION OF LITIGATION

Calderwood & Harkness LLP 1001 Fannin Street, Suite 4200 Houston, TX 77002

PRIVILEGED AND CONFIDENTIAL — ATTORNEY WORK PRODUCT
MEMORANDUM
Date: November 15, 2023
To: Catherine Bloom, Responsible Partner, Calderwood & Harkness LLP Hannah Brinkley, Lead Associate, Calderwood & Harkness LLP
From: Patrick Gallagher, Working Associate, Calderwood & Harkness LLP
Re: Summary of Expert Economic Analysis — Dr. Leonard Whitfield, Whitfield Economic Advisors LLC; FERC Docket No. IN23-7-000
C&H Matter Number: 1016-00004
C&H Billing Code: 71420.00001
Client: Dunmore Energy Holdings Ltd. / Dunmore Power Trading LLC
I. Purpose and Scope of Memorandum
The purpose of this memorandum is to summarize the expert economic analysis prepared by Dr. Leonard Whitfield, Principal Expert, Whitfield Economic Advisors LLC ("WEA"), 1700 Pennsylvania Avenue NW, Suite 550, Washington, D.C. 20006, for the benefit of the Calderwood & Harkness LLP ("C&H") case team handling the above-referenced enforcement matter before the Federal Energy Regulatory Commission ("FERC" or the "Commission").
Dr. Whitfield's expert report (the "Whitfield Report") was prepared to rebut the market impact and unjust profits calculations set forth by FERC's Office of Enforcement ("OE") in its Preliminary Findings letter dated October 2, 2023 (the "Preliminary Findings Letter"). The Preliminary Findings Letter contains detailed quantitative analyses prepared by Dr. Samuel Breck, Senior Economist, FERC Division of Analytics & Surveillance, which purport to establish both the existence and magnitude of market harm allegedly caused by the trading activities of Dunmore Power Trading LLC ("DPT"). Dr. Whitfield's Report addresses each of Dr. Breck's principal quantitative conclusions and identifies significant methodological deficiencies that, in Dr. Whitfield's opinion, result in a substantial overstatement of both market impact and alleged unjust profits.
WEA was retained pursuant to an engagement letter dated July 10, 2023, executed between Calderwood & Harkness LLP and Dr. Whitfield. The engagement provides for compensation at an hourly rate of $475 per hour for Dr. Whitfield's personal time, with staff economist rates ranging from $225 to $375 per hour depending on seniority. The engagement was structured through counsel to preserve the protection of the attorney work product doctrine, as Dr. Whitfield's analysis and conclusions were developed at the direction of C&H attorneys in anticipation of contested proceedings before FERC and, potentially, in federal court.
The Whitfield Report is scheduled to be submitted to FERC OE on November 20, 2023, as an attachment to our supplemental submission in FERC Docket No. IN23-7-000. This memorandum summarizes Dr. Whitfield's principal findings and conclusions for internal review and strategic planning purposes in advance of that formal submission. The memorandum is intended to facilitate discussion among the case team regarding the implications of Dr. Whitfield's analysis for our settlement strategy and for the framing of Dunmore's defenses in any future proceedings.
Nothing in this memorandum is intended to constitute legal advice to the client independent of the broader representation, and the contents should not be shared outside of the C&H case team without the express approval of the Responsible Partner.
II. Background: FERC Staff's Allegations and Quantification
A. Overview of Staff's Theory
FERC Staff alleges that Dunmore Power Trading LLC ("DPT"), the wholly owned power marketing subsidiary of Dunmore Energy Holdings Ltd. ("DEH"), engaged in a course of market manipulation in the Midcontinent Independent System Operator, Inc. ("MISO") Day-Ahead and Real-Time energy markets in violation of Section 222 of the Federal Power Act, 16 U.S.C. § 824v, and the Commission's Anti-Manipulation Rule, 18 C.F.R. § 1c.2. The investigation was initiated following a referral report dated April 3, 2023, prepared by Velthorn Market Analytics, Inc. ("Velthorn"), the MISO Independent Market Monitor, under the supervision of Dr. Patricia Engel, President of Velthorn.
The Relevant Period identified by FERC Staff spans from January 3, 2022 through March 17, 2023 — a period of approximately fourteen and one-half months. The pricing nodes at issue are the Indiana Hub and the Minnesota Hub, both located within the MISO footprint. These two pricing nodes are key commercial reference points within the MISO market and are the settlement locations for a substantial volume of financial transmission rights, virtual transactions, and bilateral power contracts.
Staff's theory of the case encompasses three interrelated components of an alleged manipulative scheme:
1.  Uneconomic Day-Ahead Bidding. FERC Staff alleges that Kyle Renfro, Senior Trader on DPT's MISO Day-Ahead Desk, submitted Day-Ahead supply offers for Dunmore's gas-fired generation units at prices significantly below marginal cost. According to Staff, these offers ranged from negative $15 per megawatt-hour to negative $40 per megawatt-hour at the Indiana Hub. Staff contends that these below-cost offers were designed to create artificial congestion on transmission pathways between the Indiana Hub and the Minnesota Hub, and that DPT held Financial Transmission Rights ("FTRs") that would benefit financially from the resulting congestion differentials. In Staff's view, the negative offers lacked any legitimate economic rationale and were submitted solely to benefit DPT's FTR positions.
2.  Cross-Product Manipulation. FERC Staff further alleges that Anastasia Petrov, a Trader on DPT's MISO Real-Time Desk, executed Real-Time virtual positions — specifically, Increment Offers ("INCs") and Decrement Bids ("DECs") — at the Minnesota Hub that were designed to amplify the congestion differential between the Indiana Hub and the Minnesota Hub. Staff characterizes these virtual transactions as complementary to the uneconomic Day-Ahead offers submitted by Mr. Renfro, asserting that the two sets of transactions operated in concert to maximize the spread between the two pricing nodes and thereby increase the value of DPT's FTR positions.
3.  Physical-Financial Coordination. Finally, FERC Staff alleges that DPT coordinated the timing and volume of natural gas nominations through Dunmore Midstream Partners LP ("DMP") (EIN: 47-8831205), a sister entity within the Dunmore corporate family, to support the uneconomic generation offers described above and to mask DPT's true marginal costs. Specifically, Staff contends that DPT procured gas volumes through DMP at non-market rates or on terms that obscured the true economic cost of fuel delivery to DPT's generation units, thereby creating the appearance that the negative Day-Ahead offers were economically rational when, in Staff's view, they were not.
Staff's theory thus rests on the premise that the three components — uneconomic Day-Ahead bidding, cross-product virtual trading, and coordinated gas nominations — constituted a single, integrated scheme to manipulate congestion pricing within MISO for the purpose of enriching DPT's FTR portfolio at the expense of other market participants.
B. FERC Staff's Quantification
FERC Staff's Preliminary Findings Letter includes detailed quantitative estimates of both the volume of allegedly manipulative activity and the resulting financial impact. The key figures, as stated in the Preliminary Findings Letter and the accompanying analysis prepared by Dr. Breck, are as follows:
•  Total Day-Ahead uneconomic generation offers: Staff calculated that DPT submitted approximately 2,840 gigawatt-hours ("GWh") of generation offers at prices below estimated marginal cost during the Relevant Period. This figure encompasses all generation offers at the Indiana Hub that were priced at or below negative $15 per megawatt-hour.
•  Total FTR positions benefiting from alleged manipulation: Staff identified FTR positions held by DPT with a notional value of approximately $47.3 million. Staff asserts that these FTR positions were directionally aligned with the congestion patterns created by DPT's uneconomic Day-Ahead offers and therefore represent the financial instruments through which DPT captured the fruits of its allegedly manipulative conduct.
•  Total virtual transactions at Minnesota Hub: Staff identified approximately 1,620 GWh of virtual transactions (INCs and DECs) executed by Ms. Petrov at the Minnesota Hub during the Relevant Period that Staff contends were designed to amplify the congestion differential.
•  Estimated "unjust profits": Based on the analysis prepared by Dr. Breck, FERC Staff estimated that DPT realized "unjust profits" of $12.8 million attributable to the alleged scheme. This figure reflects the total incremental profits that, in Staff's estimation, DPT earned on its FTR positions and virtual transactions as a result of the congestion patterns purportedly created by the uneconomic Day-Ahead offers.
•  Initial penalty demand: In addition to disgorgement of the $12.8 million in estimated unjust profits, Staff proposed an initial civil penalty of $18 million. This penalty figure reflects a base penalty of $9 million, which was doubled to account for the duration and scope of the alleged conduct under the FERC Penalty Guidelines.
•  Total initial demand: Staff's combined initial demand thus totals $30.8 million, comprising the $18 million civil penalty and the $12.8 million disgorgement amount.
These figures represent Staff's opening position and are the principal target of Dr. Whitfield's rebuttal analysis, as summarized below.
III. Dr. Whitfield's Qualifications
Dr. Leonard Whitfield holds a Ph.D. in Economics from the Massachusetts Institute of Technology, with a specialization in energy markets and industrial organization. He is the Principal and founder of Whitfield Economic Advisors LLC, a boutique consulting firm focused on the economic analysis of wholesale electricity and natural gas markets. Dr. Whitfield has over twenty-five years of experience in the analysis of organized wholesale electricity markets, with particular expertise in the MISO and Southwest Power Pool ("SPP") regional transmission organization footprints.
Throughout his career, Dr. Whitfield has served as an expert consultant and testifying expert in numerous proceedings before FERC, including both enforcement proceedings under Section 222 of the Federal Power Act and rate proceedings under Sections 205 and 206. He has also provided expert testimony in related state regulatory proceedings and in federal court litigation arising from wholesale electricity market disputes.
Dr. Whitfield's areas of expertise include, but are not limited to, market impact analysis, congestion pricing in locational marginal price ("LMP") markets, the valuation and risk management of Financial Transmission Rights, the economics of virtual trading (including INCs and DECs), unit commitment and dispatch economics for thermal generation resources, and the analysis of bidding behavior in organized wholesale electricity markets. He has authored peer-reviewed articles on congestion pricing methodology and has presented at numerous industry conferences on wholesale market design and market manipulation detection.
Dr. Whitfield has been retained as an expert in contested proceedings before FERC on multiple prior occasions, and his analyses have been cited with approval by FERC Staff in at least two prior enforcement matters. His curriculum vitae, which is included as an appendix to the Whitfield Report, details over forty prior expert engagements in energy market disputes.
Based on his extensive experience and qualifications, Dr. Whitfield is well qualified to opine on the specific issues raised in this proceeding, including the economic rationale for DPT's Day-Ahead bidding behavior, the drivers of congestion pricing at the Indiana Hub, the market impact of DPT's trading activity, and the methodological adequacy of Staff's unjust profits calculations.
IV. Summary of Dr. Whitfield's Key Findings
Dr. Whitfield's Report contains four principal findings, each of which addresses a discrete element of FERC Staff's theory and quantification. The findings are summarized in turn below.
A. Finding 1: DPT's Negative Day-Ahead Offers Were Consistent with Portfolio Optimization and Contractual Minimum Generation Obligations
Dr. Whitfield conducted a comprehensive analysis of DPT's Day-Ahead offer data at the Indiana Hub across the entire Relevant Period, examining the pricing, volume, timing, and frequency of the offers that FERC Staff characterizes as "uneconomic." This analysis included a detailed review of DPT's internal bidding records, generation dispatch records, fuel procurement documentation, and contractual arrangements with DMP and third-party counterparties.
Dr. Whitfield's principal conclusion on this issue is that the negative offers submitted by Mr. Renfro — ranging from negative $15 per megawatt-hour to negative $40 per megawatt-hour — were consistent with a legitimate portfolio optimization strategy employed by DPT and did not constitute uneconomic or manipulative bidding.
Central to this conclusion is the fact that DPT's gas-fired generation units at the Indiana Hub are part of Dunmore's broader fleet of 2,650 megawatts of U.S. natural gas-fired generation capacity. A significant portion of these units were subject to contractual minimum generation commitments — specifically, minimum take-or-pay obligations under gas supply agreements with Dunmore Midstream Partners LP and minimum dispatch requirements under long-term power purchase agreements with certain MISO load-serving entities. These contractual commitments required certain units to remain online and generating at specified minimum output levels regardless of prevailing Day-Ahead clearing prices. Under these circumstances, negative offers reflected the economic reality of the opportunity costs associated with curtailing or shutting down generation in the face of binding contractual obligations.
Dr. Whitfield explains that when a generator faces binding minimum generation constraints — whether driven by contractual take-or-pay obligations, gas pipeline scheduling requirements, or physical operating limitations — it is entirely rational, and commonplace, for the generator to submit Day-Ahead offers at negative prices. By offering at negative prices, the generator ensures that it will be dispatched in the Day-Ahead market and thereby avoid the potentially greater costs associated with contract breach, gas pipeline imbalance penalties, or the physical and financial costs of cycling thermal units through start-up and shut-down sequences.
To contextualize DPT's bidding behavior, Dr. Whitfield identified comparable negative bidding patterns by other MISO market participants during the same Relevant Period. His analysis of publicly available MISO offer data revealed that multiple other generators at the Indiana Hub and at neighboring pricing nodes submitted Day-Ahead offers at negative prices during the Relevant Period, and that the frequency and magnitude of DPT's negative offers were not anomalous relative to the broader market. This finding is significant because it demonstrates that DPT's bidding behavior, rather than being indicative of a manipulative scheme, was consistent with the behavior of similarly situated market participants responding to comparable market conditions and contractual constraints.
Dr. Whitfield further concludes that FERC Staff's marginal cost calculation — which formed the basis for Staff's determination that DPT's offers were "below cost" — failed to account for several material cost components. Specifically, Staff's marginal cost estimate did not incorporate: (a) start-up and shut-down costs for cycling gas-fired units, which can be substantial for combined-cycle and combustion turbine units of the type operated by DPT; (b) minimum run-time constraints embedded in MISO's unit commitment protocols, which effectively commit generators to operate for a specified number of hours once started and thereby create sunk costs that should be reflected in incremental offer pricing; and (c) contractual obligations requiring minimum gas throughput on DMP pipelines, which imposed real economic costs on DPT that would be avoided only by continuing to generate and consume gas at the required minimum volumes.
Finally, Dr. Whitfield emphasizes the importance of evaluating DPT's Day-Ahead offers in the context of DPT's broader generation and trading portfolio, rather than in isolation. DPT managed a diversified portfolio comprising 4,280 megawatts of total U.S. net generation capacity, including 2,650 megawatts of natural gas-fired generation, 1,130 megawatts of wind generation, and 500 megawatts of solar generation. Within such a diversified portfolio, it is standard industry practice to optimize individual unit offers with reference to the overall portfolio economics, including the interaction between generation dispatch, transmission congestion, and hedging positions. Dr. Whitfield concludes that DPT's negative Day-Ahead offers, viewed in the context of this broader portfolio, reflected a rational and legitimate optimization strategy that is common among sophisticated portfolio managers in organized wholesale electricity markets.
B. Finding 2: Congestion Patterns at Indiana Hub Were Primarily Driven by Transmission Outage Events, Not DPT's Bidding
Dr. Whitfield performed a rigorous regression analysis of congestion pricing at the Indiana Hub during the Relevant Period, spanning from January 3, 2022 through March 17, 2023. The purpose of this analysis was to determine the extent to which the congestion patterns observed at the Indiana Hub — and upon which FERC Staff's theory of manipulation critically depends — were attributable to DPT's Day-Ahead bidding behavior versus other, exogenous market factors.
Dr. Whitfield's key finding on this issue is that transmission outage events were the dominant causal factor in the congestion patterns that FERC Staff attributed to DPT's bidding. Specifically, Dr. Whitfield identified a series of both planned and unplanned transmission outages on key MISO transmission facilities during the Relevant Period that independently created the congestion differentials Staff attributed to DPT. These outage events included planned maintenance on major 345 kV transmission lines serving the Indiana Hub area, as well as several forced outages on critical transmission facilities resulting from equipment failures and severe weather events.
Dr. Whitfield's correlation analysis demonstrated that the periods of highest congestion at the Indiana Hub coincided with identified transmission outage events, not with DPT's bidding patterns. In particular, the most significant congestion episodes during the Relevant Period — which accounted for a disproportionate share of the total FTR profits Staff attributes to DPT — occurred during intervals in which one or more major transmission facilities in the Indiana Hub area were out of service. During these same intervals, DPT's Day-Ahead offers were not materially different in pricing or volume from DPT's offers during non-outage periods, suggesting that the congestion was driven by the transmission system topology, not by DPT's offer behavior.
When Dr. Whitfield controlled for transmission outage events in his regression model — that is, when the model included variables accounting for the duration, location, and capacity impact of known transmission outages — the incremental congestion attributable to DPT's Day-Ahead offers was statistically insignificant in the majority of hourly intervals during the Relevant Period. This result held across multiple model specifications and robustness checks, and was consistent regardless of whether Dr. Whitfield used actual Day-Ahead LMP data or shadow price decomposition data from MISO's market clearing engine.
This finding directly and substantially undermines FERC Staff's theory — and the analysis prepared by Dr. Breck — that DPT's uneconomic offers "created" the congestion from which DPT's FTR positions profited. If the observed congestion patterns were primarily driven by transmission outage events that had nothing to do with DPT's bidding, then the causal chain upon which Staff's theory of manipulation depends is broken. The FTR profits that Staff attributes to manipulative conduct were, in large part, simply the result of DPT holding hedging positions that happened to be in the money during periods of outage-driven congestion — a result that is entirely consistent with the legitimate hedging purpose for which FTRs are designed.
C. Finding 3: Actual Market Impact Estimated at No More Than $2.1 Million (Versus FERC Staff's $12.8 Million)
In addition to his analysis of congestion causation, Dr. Whitfield conducted an independent "but-for" market impact analysis to estimate the actual profits, if any, that were attributable to DPT's bidding behavior during the Relevant Period. This analysis represents the core quantitative rebuttal to Dr. Breck's $12.8 million unjust profits estimate.
Dr. Whitfield's methodology involved comparing actual MISO market outcomes — including Day-Ahead and Real-Time LMPs, congestion components, and FTR settlement values at the Indiana Hub and Minnesota Hub — to a counterfactual scenario in which DPT's Day-Ahead offers at the Indiana Hub were priced at estimated marginal cost rather than at the negative prices actually submitted. In constructing this counterfactual, Dr. Whitfield controlled for transmission outages and other exogenous market conditions (including weather, load levels, fuel prices, and the bidding behavior of other market participants) to isolate the incremental effect, if any, of DPT's negative offers on congestion pricing and, by extension, on DPT's financial positions.
The result of this analysis is that the actual market impact attributable to DPT's bidding behavior is estimated at no more than $2.1 million over the entire Relevant Period. This figure represents the incremental change in DPT's FTR profits and virtual trading profits that would result from repricing DPT's Day-Ahead offers at estimated marginal cost, after controlling for transmission outage effects and other exogenous factors.
This $2.1 million estimate stands in stark contrast to FERC Staff's estimate of $12.8 million in unjust profits, representing a reduction of approximately 84 percent. Dr. Whitfield attributes this substantial discrepancy to four key differences between his methodology and Staff's:
First, Staff's analysis did not adequately control for the effects of transmission outage events on congestion pricing. As discussed in Finding 2, transmission outages were the primary driver of congestion at the Indiana Hub during the Relevant Period. By failing to control for outage events, Staff's model attributed outage-driven congestion — and the associated FTR profits — to DPT's bidding behavior, resulting in a significant upward bias in the estimated market impact.
Second, Staff attributed all FTR profits during the Relevant Period to the alleged scheme, even though the FTR positions would have generated positive returns independent of DPT's Day-Ahead bidding due to the outage-driven congestion that prevailed during much of the Relevant Period. Dr. Whitfield's counterfactual analysis demonstrates that DPT's FTR portfolio would have been profitable during the Relevant Period regardless of whether DPT had submitted negative Day-Ahead offers, because the congestion that drove FTR profitability was overwhelmingly attributable to transmission outage events rather than to DPT's bidding.
Third, Staff's marginal cost calculation used a simplified heat rate approach — multiplying a generic heat rate by a daily natural gas price index — that did not account for DPT's actual delivered gas costs through DMP infrastructure, start-up and shut-down costs for thermal units, or the minimum generation economics described in Finding 1. By understating DPT's true marginal costs, Staff's analysis overstated the degree to which DPT's offers were "uneconomic," thereby inflating the volume of offers treated as manipulative and the associated market impact.
Fourth, Staff aggregated profits across both Indiana Hub FTR positions and Minnesota Hub virtual positions, creating a cumulative overcount. Dr. Whitfield's analysis demonstrates that a portion of the virtual trading profits at the Minnesota Hub were attributable to market conditions unrelated to congestion at the Indiana Hub and therefore should not be included in the unjust profits calculation even under Staff's own theory.
Dr. Whitfield notes that his $2.1 million estimate represents a conservative upper bound and may overstate the actual market impact attributable to DPT's bidding behavior. The estimate does not fully isolate the effect of seasonal weather patterns and load variability on congestion pricing, and it does not account for the possibility that, in the counterfactual scenario, other market participants might have submitted similarly priced offers that would have produced comparable congestion outcomes. To the extent these additional factors were incorporated, the estimated market impact could be even lower than $2.1 million.
D. Finding 4: FTR Positions Were Acquired Months Before the Relevant Period as Routine Hedging
Dr. Whitfield reviewed the acquisition timeline and underlying rationale for DPT's FTR portfolio, with particular attention to the FTR positions identified by FERC Staff as instruments through which DPT allegedly captured the profits of its manipulative scheme.
Dr. Whitfield's conclusion is that the FTR positions identified by FERC Staff — with a notional value of approximately $47.3 million — were acquired months before the start of the Relevant Period (January 3, 2022) as part of DPT's routine hedging program. Specifically, the bulk of the FTR portfolio in question was acquired through MISO's annual FTR auction process, with the relevant auction clearing dates falling in the third and fourth quarters of 2021 — well in advance of the January 2022 start of the Relevant Period. Additional FTR positions were acquired through MISO's monthly FTR auctions, but these, too, were acquired on dates preceding the allegedly manipulative bidding activity by periods ranging from several weeks to several months.
Dr. Whitfield compared the acquisition dates, volumes, and pricing of DPT's FTR positions during this period against DPT's historical hedging patterns in prior years. He found that DPT's FTR acquisitions for the 2022 and 2023 planning years were consistent with the size, direction, and pricing of FTR positions DPT had acquired in each of the five preceding planning years. There was no discernible increase in FTR acquisition volume or shift in portfolio direction that would suggest DPT was building FTR positions in anticipation of, or in coordination with, the allegedly manipulative Day-Ahead bidding.
DPT's FTR portfolio was managed by the trading desk under the supervision of Derek Sung, Senior Vice President of Power Marketing & Trading, who oversaw a team of fourteen traders. The FTR positions were acquired and managed under DPT's standard risk management framework, which included board-approved risk limits, daily mark-to-market reporting, and periodic portfolio reviews by the risk management committee. Dr. Whitfield found no evidence that the FTR acquisition process deviated from these established protocols during the Relevant Period.
The temporal separation between FTR acquisition and the alleged manipulative bidding is significant because it undermines FERC Staff's inference of manipulative intent. Under Section 222 of the Federal Power Act and 18 C.F.R. § 1c.2, a finding of manipulation requires evidence of scienter — i.e., that the respondent acted with the intent to defraud or manipulate. If DPT acquired its FTR positions months before the allegedly manipulative bidding began, the inference that DPT's Day-Ahead offers were submitted for the purpose of benefiting those FTR positions is substantially weakened. The more natural inference, consistent with Dr. Whitfield's findings, is that DPT acquired FTR positions as part of its routine hedging activities and subsequently submitted Day-Ahead offers for legitimate portfolio optimization reasons unrelated to the FTR positions.
Dr. Whitfield also noted that other MISO market participants with generation portfolios comparable in size and composition to DPT's held FTR positions of similar magnitude and direction during the same period. This observation further supports the conclusion that DPT's FTR portfolio reflected standard industry hedging practice rather than a building block of a manipulative scheme.
V. Assessment of FERC Staff's Methodology and Weaknesses
Dr. Whitfield's Report dedicates a substantial section to identifying specific methodological weaknesses in the analysis prepared by Dr. Samuel Breck, Senior Economist, FERC Division of Analytics & Surveillance. These weaknesses, which Dr. Whitfield addresses both individually and in the aggregate, are summarized below.
1.  Failure to Control for Transmission Outages. Staff's regression model, which purported to isolate the effect of DPT's bidding on congestion pricing at the Indiana Hub, did not include a variable for known transmission outages. This omission constitutes a textbook example of omitted variable bias — a well-recognized econometric error that arises when a regression model fails to account for a factor that is correlated with both the independent variable of interest (DPT's bidding) and the dependent variable (congestion pricing). Because transmission outages are a major determinant of congestion pricing and occurred with significant frequency during the Relevant Period, the omission of an outage variable from Staff's model resulted in the inflation of the estimated impact of DPT's bidding. In effect, Staff's model attributed to DPT's bidding the congestion effects that were actually caused by transmission outages, leading to a substantial overstatement of market impact.
2.  Overinclusive Profit Calculation. Staff's unjust profits estimate of $12.8 million included all FTR profits and all virtual trading profits at the Minnesota Hub during the Relevant Period, rather than isolating only the incremental profits attributable to the allegedly manipulative bids. This approach is inconsistent with the established legal and economic standard for disgorgement, which requires that only the profits causally attributable to the wrongful conduct be disgorged. By including profits that DPT would have earned regardless of its Day-Ahead bidding behavior — including FTR profits driven by outage-related congestion and virtual trading profits attributable to factors unrelated to Indiana Hub congestion — Staff's calculation significantly overstated the appropriate disgorgement amount.
3.  Simplified Marginal Cost Estimation. Staff's determination that DPT's Day-Ahead offers were "below marginal cost" was based on a simplified cost methodology in which Staff multiplied a single heat rate by a daily natural gas price index to arrive at an estimated marginal cost for each of DPT's generation units. This approach did not account for DPT's actual delivered gas costs, which include pipeline transportation charges and scheduling costs associated with gas delivery through DMP infrastructure. Nor did Staff's methodology account for start-up and shut-down costs, which can be substantial for gas-fired combined-cycle and combustion turbine units and which are properly included in the economic evaluation of generation offers. Additionally, Staff's marginal cost estimate did not reflect DPT's contractual minimum generation obligations, which — as discussed in Finding 1 — imposed real economic costs that justified negative offer pricing. The result of Staff's simplified methodology was an understatement of DPT's true marginal costs and a corresponding overstatement of the degree to which DPT's offers were "uneconomic."
4.  Absence of a Rigorous Counterfactual. Staff's "but-for" analysis assumed that, in the absence of DPT's below-cost offers, no other market participant would have submitted similarly priced offers at the Indiana Hub. Dr. Whitfield found this assumption to be unsupported given the observable competitive bidding patterns at the Indiana Hub during the Relevant Period. As noted in Finding 1, multiple other market participants submitted negative Day-Ahead offers at the Indiana Hub and at neighboring pricing nodes during the Relevant Period, and there is no basis in the record for assuming that these other participants would not have continued to do so in the counterfactual scenario. By failing to account for the likely bidding behavior of other market participants in the but-for scenario, Staff's analysis overstated the incremental impact of DPT's offers on congestion pricing.
Dr. Whitfield concludes that these methodological weaknesses, individually and collectively, resulted in a significant overstatement of both the market impact of DPT's bidding activity and the alleged unjust profits attributable to DPT. Dr. Whitfield's independent analysis, which corrects for these deficiencies, yields an estimated market impact of no more than $2.1 million — approximately 16 percent of Staff's estimate.
VI. Implications for Settlement Strategy
Dr. Whitfield's analysis provides a strong foundation for challenging FERC Staff's Preliminary Findings and, in particular, for contesting the $12.8 million disgorgement claim as speculative and unquantifiable with sufficient certainty to support an enforcement action or penalty assessment.
The $2.1 million market impact estimate established by Dr. Whitfield provides a defensible upper bound that should serve as the anchor for Dunmore's negotiating position in any settlement discussions with FERC OE. If the Commission were to accept Dr. Whitfield's analysis — or even to accept that Staff's methodology is subject to the weaknesses Dr. Whitfield identifies — the total demand would be substantially reduced from the $30.8 million initially proposed by Staff. Even if Staff were to reject Dr. Whitfield's market impact figure entirely, the identified methodological deficiencies in Dr. Breck's analysis create significant litigation risk for Staff and should incentivize a more reasonable settlement posture.
The portfolio optimization defense articulated in Finding 1 and the FTR acquisition timing evidence presented in Finding 4 directly support the argument that DPT's conduct lacked the manipulative intent required to sustain a violation under 18 C.F.R. § 1c.2 and Section 222 of the Federal Power Act. The Commission has consistently held that a violation of the Anti-Manipulation Rule requires proof of scienter, and Dr. Whitfield's findings provide substantial economic evidence that DPT's trading activities were motivated by legitimate business considerations — portfolio optimization, contractual compliance, and routine hedging — rather than by an intent to manipulate market prices.
Dr. Whitfield's findings are consistent with and reinforce the legal and factual arguments presented in C&H's Response to Preliminary Findings submitted on November 6, 2023. In particular, the transmission outage analysis presented in Finding 2 complements the legal arguments regarding causation that were central to the November 6 submission, and the $2.1 million market impact estimate provides quantitative support for the position that Staff's disgorgement demand is grossly excessive.
The transmission outage analysis is particularly effective as a defense tool because it provides an independent, data-driven, and empirically verifiable explanation for the congestion patterns that FERC Staff attributed entirely to DPT's bidding. This analysis shifts the narrative from one in which DPT's bidding "created" artificial congestion to one in which congestion was primarily driven by infrastructure constraints — a framing that is both more consistent with the data and substantially more favorable to Dunmore's position.
Recommendation. The case team recommends deploying Dr. Whitfield's findings aggressively in the upcoming settlement negotiations to reduce both the penalty and disgorgement components of Staff's demand. The submission of the Whitfield Report on November 20, 2023, should be accompanied by a cover letter emphasizing the key findings and requesting a settlement conference at which Dr. Whitfield can present his analysis directly to OE staff. In the interim, the case team should prepare Dr. Whitfield for potential questions regarding his methodology, data sources, and conclusions, and should coordinate with Dr. Whitfield on any supplemental analyses that may be needed to address Staff's anticipated counterarguments.
VII. Status of Expert Fees
WEA's engagement with C&H provides for compensation at an hourly rate of $475 per hour for Dr. Whitfield's personal time, with staff economist rates as provided in the engagement letter dated July 10, 2023. Through the expected date of matter closure, total WEA fees are estimated at approximately $312,500.
Fees accrued to date (as of the date of this memorandum, November 15, 2023) reflect work through the preparation and finalization of the expert report, including data collection and analysis, regression modeling, preparation of the counterfactual market impact study, drafting and revision of the expert report, and coordination with C&H attorneys. The expert report is scheduled for submission to FERC OE on November 20, 2023.
Additional fees may accrue beyond the current estimate if Dr. Whitfield is needed for settlement conference support, preparation of supplemental analyses in response to Staff inquiries, or participation in any formal hearing proceedings. The case team will monitor fee accruals and provide updated estimates to the client as appropriate.
VIII. Attachments and Cross-References
The following documents are referenced in this memorandum and should be consulted for additional detail:
1.  Expert Report of Dr. Leonard Whitfield, Whitfield Economic Advisors LLC — Final report to be submitted to FERC OE on November 20, 2023, as an attachment to C&H's supplemental submission in FERC Docket No. IN23-7-000. The report contains Dr. Whitfield's full analysis, supporting data, regression results, and appendices, including his curriculum vitae.
2.  FERC Staff Preliminary Findings Letter, dated October 2, 2023 — Letter from FERC OE to C&H setting forth Staff's preliminary findings, including the factual allegations, legal theories, quantification of unjust profits, and initial penalty demand. This letter is the document to which Dr. Whitfield's report is principally responsive.
3.  C&H Response to Preliminary Findings, submitted November 6, 2023 — C&H's written response to the Preliminary Findings Letter, setting forth Dunmore's legal and factual defenses. Dr. Whitfield's expert report supplements and reinforces the arguments presented in this submission.
4.  WEA Engagement Letter, dated July 10, 2023 — Engagement letter between Calderwood & Harkness LLP and Dr. Leonard Whitfield, Whitfield Economic Advisors LLC, establishing the terms of Dr. Whitfield's retention, including the hourly rate, scope of engagement, and confidentiality obligations.
5.  Velthorn Market Analytics, Inc. Referral Report, dated April 3, 2023 — Report prepared by Velthorn Market Analytics, Inc. (Dr. Patricia Engel, President), in its capacity as the MISO Independent Market Monitor, referring certain trading patterns by DPT to FERC OE for investigation. This referral initiated the investigation that resulted in FERC Docket No. IN23-7-000.
PRIVILEGED AND CONFIDENTIAL — ATTORNEY WORK PRODUCT — PREPARED IN ANTICIPATION OF LITIGATION
Calderwood & Harkness LLP 1001 Fannin Street, Suite 4200 Houston, TX 77002

PRIVILEGED AND CONFIDENTIAL — ATTORNEY WORK PRODUCT

MEMORANDUM

Record

Parties

Dates

Amounts