On Friday, June 17, 2016, the Federal Energy Regulatory Commission (the “Commission”) issued an Order Assessing Civil Penalties against ETRACOM LLC (“ETRACOM”) and ETRACOM’s founding member and majority owner, Michael Rosenberg (together, “Respondents”).1 The Commission held that the Respondents violated section 222 of the Federal Power Act and section 1c.2 of the Commission’s regulations (the “Anti-Manipulation Rule”). Specifically, the Commission found that between May 14, 2011 and May 31, 2011 (the “Manipulation Period”), ETRACOM and Rosenberg engaged in a “cross-commodity” scheme in which they submitted virtual supply offers at the New Melones intertie at the border of the California Independent System Operator (“CAISO”) wholesale electric market with the intent to lower power prices artificially at New Melones in order to increase the value of ETRACOM’s Congestion Revenue Rights (“CRRs”) positions that settled based upon power prices at that location. Respondents offered two primary defenses to Staff’s allegations, each of which the Commission rejected: (1) that CAISO was not a well-functioning market and (2) that ETRACOM based its trading activity on market fundamentals.

Consistent with Staff’s recommendation, the Commission assessed $2.4 million in civil penalties against ETRACOM and $100,000 in penalties against Rosenberg. Additionally, the Commission ordered ETRACOM to disgorge $315,072 in unjust profits.

The Products at Issue and the Commission’s Allegations

The alleged scheme in the ETRACOM Assessment Order involved CRRs and virtual transactions. CRRs are “financial instruments that settle at an amount equal to the difference in day-ahead congestion costs between two locations,” a source and a sink.2 The CRR holder receives a payment if the congestion in a given hour is in the same direction as the CRR position, and the holder incurs a charge if congestion occurs in the opposite direction. CRR values fluctuate based upon changes in locational marginal prices (“LMP”) at their source and sink. Virtual transactions can take the form of supply offers and demand bids. Either trade can alter the value of a CRR position because, although they settle financially, virtual supply and demand transactions are evaluated in CAISO’s day-ahead market process along with physical power supply and demand transactions. Virtual transactions can therefore affect day-ahead LMPs and, consequently, CRRs with a source or sink that has the same LMP. Such was the case, the Commission concluded, with ETRACOM’s trading.

The Commission observed a significant change in ETRACOM’s trading of virtual transactions and CRRs at the New Melones intertie in early May 2011. In April 2011, ETRACOM’s CRRs at New Melones profited from import congestion into CAISO.3 These positions netted ETRACOM approximately $195,000 in profits.4 During this same period, ETRACOM engaged in virtual transactions at 22 locations, but not at New Melones. ETRACOM acquired larger CRR positions that profited from import congestion into CAISO at New Melones beginning in early May.5 But then the economics of that position changed significantly. Although ETRACOM profited on its New Melones intertie CRR position from import congestion for the first week of May, the second week brought unexpected export congestion in most off-peak hours.6 The reverse in congestion resulted in a $23,624 loss for ETRACOM on its CRR position.

According to the Commission, Rosenberg then devised a manipulative scheme for ETRACOM to enter into uneconomic virtual trades to restore the profitability of the New Melones CRR positions by sending false supply signals into the market and driving down power prices in the day-ahead market at New Melones. The false supply signals were intended to restore the import congestion from which ETRACOM’s CRR positions would profit. The Commission concluded that the strategy worked, and in the hours that ETRACOM placed its virtual supply offers, no export congestion occurred.7 Consequently, Respondents allegedly expanded and continued this strategy for the rest of the Manipulation Period, eventually offering virtual supply for almost every hour.8 This scheme, the Commission concluded, resulted in ETRACOM earning over $690,122 in total revenue from its CRR positions.9

According to the Commission, several factors demonstrate that this pattern of trading was an intentional, manipulative scheme. First, the trading pattern itself evidenced Respondents’ scienter. The Commission pointed to the uneconomic nature of ETRACOM’s trades, specifically noting that ETRACOM’s virtual transactions lost money in 96% of the hours they cleared at New Melones.10 As indicated above, the Commission determined that ETRACOM’s virtual trading at New Melones differed from its trading at other locations where it also engaged in virtual trading.11 Second, notwithstanding ETRACOM’s argument that its internal communications evidenced no manipulative intent, the Commission believed these communications suggested that ETRACOM accepted its virtual trading losses because of the profitability of its CRR positions.12 Third, the Commission explained that the difference in virtual trading between the pre-manipulation period and the Manipulation Period demonstrated that ETRACOM executed an intentional, manipulative scheme.13

The Commission Rejected Each of Respondents’ Arguments

Respondents argued that their conduct was legitimate and not manipulative. While Respondents offered many arguments, we focus on the three primary categories below.

  1. CAISO was not a Well-Functioning Market

    Respondents argued that any seemingly uncompetitive trades resulted not from a manipulative scheme, but from CAISO’s flawed market design. The market suffered software pricing errors that sent incorrect price signals, caused unforeseeable outcomes, and created a dysfunctional and uncompetitive market.14 Asserting that they were unaware of these market design flaws and software errors, Respondents argued that their transactions were rational responses to what was not a “well-functioning market.”15

    Respondents asserted that they subsequently learned but did not know at the time that submitting any virtual trade of any size at New Melones could set the LMP and create congestion at that intertie – even where no physical power flowed and transmission capacity was not constrained. Despite CAISO’s alleged knowledge of this market flaw, also called “phantom congestion,” CAISO permitted market participants to purchase CRRs at New Melones until July 2011. Respondents argued that they were unaware of this phantom congestion and that ETRACOM’s small offers could not have set the LMP or created congestion if the market operated properly.

    Respondents further asserted that a “software pricing error” at New Melones caused the intertie price to be incorrectly set at $0, instead of the bid price, when the lowest-price virtual supply offer was positive and only virtual supply offers were submitted. Respondents and other market participants who were purportedly unaware of this error responded by submitting virtual supply offers at $0 or negative prices to increase their likelihood of clearing the market.

    Respondents urged the Commission to analyze their trading, with these design flaws in mind, from a “forward-looking chronological perspective” rather than to discern “fraud by hindsight.”16 Respondents asserted that they were unaware of the market design flaw and software errors and believed that the New Melones intertie was a well-functioning and competitive market during the Manipulation Period. Because they could not reasonably have known that their trading would cause the alleged market harm, they could not have intended the manipulation alleged.

    The Commission acknowledged that the CAISO market is not perfect, but found that Respondents failed to demonstrate that flawed market design and software errors were the reason for the virtual trading behavior in question. Moreover, the Commission stated that “market manipulation is not excused simply because there are market inefficiencies or even market dysfunction.”17 The Commission instead focused on its hindsight view of the pattern of trading and emphasized that Respondents continuously and “knowingly engaged in money-losing virtual trading” that benefitted their CRR positions regardless of whether CAISO’s market flaws and software errors created false price signals and incentives.18
  1. Trading Activity Based Upon Market Fundamentals and Expected Hydro Event

    Respondents also maintained that ETRACOM traded based upon sound economic principles and that the company’s trading strategy reflected genuinely-held beliefs about market fundamentals.19 Respondents pointed to contemporaneous documents purportedly indicating that they anticipated a large hydro event that, had it occurred (it did not), would have benefitted their virtual supply offers and vindicated their trading strategy.20 Rejecting this argument, the Commission concluded that no credible evidence supported Respondents’ assertion that their virtual trading reflected their expectation of a large hydro event. Instead, the Commission determined that “the only credible reason” for ETRACOM’s trading pattern was that Respondents intended for their unprofitable virtual trading to benefit their CRR positions.21
  1. Evidence of Manipulative Intent

    In addition to rejecting Respondents’ explanations for their trading activity, the Commission laid out several factors it considered indicative of manipulative intent. First, the timing and pattern of Respondents’ virtual transactions at New Melones during the Manipulation Period differed from Respondents’ virtual trading in the periods before and after. According to the Commission, Respondents’ virtual trades during the Manipulation Period were in a direction that tended to drive the New Melones price downward to the benefit of ETRACOM’s CRR position. Second, Respondents sustained consistent losses on their virtual supply transactions at New Melones during the Manipulation Period. Third, the extrinsic evidence—including communications and trading data—demonstrated that Respondents “closely tracked their performance at New Melones, were aware of their virtual trading losses, and were aware that their virtual trades impacted prices.”22 Consistent with its analysis in other penalty actions, the Commission determined in general that Respondents’ explanations of their trading and intent were simply “implausible and not credible.”23
  1. Procedural Arguments

    Respondents advanced several procedural arguments that the Commission also rejected.24 Respondents argued that FERC Staff failed timely to provide them with an unredacted copy of the CAISO Department of Market Monitoring (“DMM”) Referral when the case was first referred to the Commission in July 2011. Stating that they did not receive an unredacted copy of the DMM Referral until July 2014, Respondents argued that this delay impeded their ability to defend themselves.25 However, the Commission concluded that Respondents had ample time to review the record and suffered no prejudice because FERC Staff provided ETRACOM with the unredacted DMM Referral almost two years before Respondents submitted their answer to the Order to Show Cause.26

    Respondents also contended that FERC Staff ignored its duty to “search the files” of the DMM for exculpatory information.27 Rejecting this argument as well, the Commission held that FERC Staff’s duty to search for exculpatory information does not extend to entities separate from the Commission. Accordingly, FERC Staff had no duty to search the third party DMM’s files.28


The ETRACOM Assessment Order presents another in a series of enforcement actions in which, notwithstanding the absence of any evidence of specific intent to engage in fraudulent or manipulative conduct, the Commission nonetheless imposed millions of dollars in penalties based on an after-the-fact analysis of an alleged pattern of trading losses. Contrary evidence and arguments were dismissed as “not credible.” Affirmative evidence of market dysfunction was disregarded.