Benton County Wind Farm LLC v. Duke Energy Indiana, Inc.

In the United States Court of Appeals For the Seventh Circuit ____________________ No. 15-­‐‑2632 BENTON  COUNTY  WIND  FARM  LLC, Plaintiff-­‐‑Appellant, v. DUKE  ENERGY  INDIANA,  INC., Defendant-­‐‑Appellee. ____________________ Appeal from the United States District Court for the Southern District of Indiana, Indianapolis Division. No. 1:13-­‐‑cv-­‐‑01984-­‐‑SEB-­‐‑TAB — Sarah Evans Barker, Judge. ____________________ ARGUED FEBRUARY 26, 2016 — DECIDED DECEMBER 6, 2016 ____________________ Before POSNER, FLAUM, and EASTERBROOK, Circuit Judges. EASTERBROOK, Circuit Judge. In 2005 Duke Energy Indiana offered to buy 100 megawatts of renewable energy at a price high enough to enable potential sellers to finance the con-­‐‑ struction of wind turbines. As part of the deal Duke would acquire renewable-­‐‑energy credits that buyers or generators of wind energy can trade or sell to other utilities that lack wind generation. Benton County Wind Farm (Benton) ac-­‐‑ cepted Duke’s offer and built a 100-­‐‑megawatt facility that 2 No. 15-­‐‑2632 became operational in 2008. The contract between Duke and Benton requires Duke to pay Benton for all power delivered during the next 20 years. Duke does not have its own trans-­‐‑ mission lines in Benton County, and the contract requires Benton to deliver to lines owned by Northern Indiana Public Service Company (NIPSCO) or some other place designated by the regional transmission organization, the Midcontinent Independent System Operator (MISO). Electrical grids throughout North America are connected, and it is essential to ensure that none of the transmission lines becomes overloaded or fails to convey power to cus-­‐‑ tomers that are counting on it. The ten regional transmission organizations in North America develop technical standards for how smaller networks connect with each other. They also employ tools to monitor networks in order to prevent over-­‐‑ loads or imbalances, which can cause blackouts. Our opinion in MISO Transmission Owners v. FERC, 819 F.3d 329 (7th Cir. 2016), describes some of this regulatory and coordination function, and it includes a map showing MISO’s territory, which spans the middle of the continent from Manitoba through Louisiana—all or part of 15 states plus one prov-­‐‑ ince. It shares Indiana with PJM Interconnection, a regional transmission organization whose territory includes Chicago, New York City, and all or part of 13 states plus the District of Columbia. Only MISO’s decisions affect this case. Regional transmission organizations have concluded that the price system is the best tool to balance loads on the net-­‐‑ works. Potential buyers of energy bid for power to be deliv-­‐‑ ered over the network (this is done principally through utili-­‐‑ ties such as Duke and NIPSCO, which aggregate end-­‐‑users’ demands); potential sellers such as Duke (on behalf of Ben-­‐‑ No. 15-­‐‑2632 3 ton) also submit bids for sale, and the regional transmission organization accepts the bid that clears the market. When Benton’s wind farm started producing, the bidding was conducted once a day. Now it is conducted every five minutes—necessarily by computers. MISO uses a variant of a Vickrey auction to decide which bids are accepted at what price. Here’s a simple illustration. Buyer 1 bids $60 per meg-­‐‑ awatt-­‐‑hour (MWh) for 200 megawatts of power; Buyer 2 bids $40 for another 200; Buyer 3 bids $30 for a further 200. If the transmission grid in the area can carry 300 megawatts, then Buyer 1 gets 200 megawatts and Buyer 2 gets 100; Buyer 3 gets nothing. The bid price is set at $40 per MWh, which is what the marginal buyer is willing to pay; in a Vickrey auc-­‐‑ tion, all buyers and sellers receive the same price. (Treasury securities are sold using a similar system.) Meanwhile Seller 1 offers 100 megawatts at $20 per MWh, Seller 2 offers 100 megawatts at $30, Seller 3 100 megawatts at $40, and Seller 4 100 megawatts at $50. The market-­‐‑clearing price and quanti-­‐‑ ty are $40 for 300 megawatts. MISO accepts the bids from Sellers 1, 2, and 3, and all three receive $40 per megawatt-­‐‑ hour. For some kinds of suppliers, such as wind farms, the marginal cost of generating any unit of output is small, even though the capital cost of building wind turbines is high. Ra-­‐‑ ther than accept no sales, Seller 4 may cut its price to $10 per MWh. Then the prevailing offer would be $30 (enough to attract a total of 300 megawatts, the most the local grid can carry), and all three buyers would pay $30. Sellers 1, 2, and 3 may not take this lying down. They may cut their own bids. If all sellers bid only enough to cover their marginal costs, the price in such a market could fall to, say, $1 per MWh, 4 No. 15-­‐‑2632 and even at that price one of the four potential sellers would be unable to make a sale. This is roughly what has happened in central Indiana. When Benton started operating it was the only wind farm in the area, and NIPSCO’s facilities could carry its entire out-­‐‑ put. Duke purchased and paid for everything Benton could produce, and MISO cleared the transfers to the regional grid. But central Indiana has excellent conditions for generating power from wind, and by 2015, when the district court is-­‐‑ sued its opinion, aggregate capacity of local wind farms was not 100 megawatts but 1,745 megawatts. More wind farms are being built. The capacity of the local transmission grid has been exceeded. It is no longer possible for all of the local wind farms to generate power at the same time, because the grid cannot accept their full output. And because local gen-­‐‑ eration capacity substantially exceeds local transmission ca-­‐‑ pacity, the market-­‐‑clearing price in MISO’s auction has fall-­‐‑ en—indeed, the price sometimes is negative, and then would-­‐‑be producers must pay MISO to take the power off their hands, and buyers get free electricity. Prices near or be-­‐‑ low zero induce some producers to stop supplying electrici-­‐‑ ty and thus reduce output to what the grid can carry. Until the end of February 2013 MISO allowed wind farms to deliver to the grid no matter what other producers (coal, nuclear, solar, hydro, and so on) were doing, which meant that other classes of producers had to cut back. Sometimes the market price in this must-­‐‑carry-­‐‑wind-­‐‑power system fell below zero, which meant that wind generation alone had overtaxed the local grid. When that happened Duke paid a negative price, displacing other wind farms to ensure that Benton ran at capacity. So if the auction price was minus No. 15-­‐‑2632 5 $10/MWh, Duke would pay MISO that amount and pay Ben-­‐‑ ton for the power; it would receive nothing for this power (save the potential value of renewable-­‐‑energy credits) and charge the loss to its customers. Duke could recover some of the loss in its role as a buyer of power from MISO’s grid, be-­‐‑ cause even if the power on NIPSCO’s grid goes north (Duke’s operations are in southern Indiana), a lower price on NIPSCO’s network will depress prices on other grids, which will buy from NIPSCO and tell other sources to curtail their own output. But Duke believes that it loses more in its role as seller of Benton’s power than it gains in its role as buyer from MISO. On March 1, 2013, the rules changed to put wind farms constructed after 2005 on a par with other classes of produc-­‐‑ ers. Benton lost its status as a must-­‐‑run facility. Duke re-­‐‑ sponded to the new system by deciding to bid exactly $0, all the time, to put Benton’s power on the grid. When this bid is accepted, Duke gets the market-­‐‑clearing price (usually posi-­‐‑ tive but sometimes zero) and pays Benton the contract price (roughly $52 per MWh). But when the market-­‐‑clearing price in MISO’s auction falls below $0, and Duke’s bid therefore is rejected, MISO instructs Benton not to deliver any power. Once Benton generates power it must deliver it (otherwise it would fry its own equipment), so an order not to deliver power equates to an order not to generate power, and Ben-­‐‑ ton must stop its turbines from rotating. Under MISO’s new system, with Duke’s standing bid of $0/MWh, Benton has gone from delivering power 100% of the time the wind al-­‐‑ lowed to delivering (and being paid) only 59% of the time that the weather can drive its turbines at their capacity. 6 No. 15-­‐‑2632 In this litigation Duke takes the position that, when MISO tells Benton to stop delivering power, it does not owe Benton anything. Benton takes the position that Duke could put Benton’s power on the grid by making a lower bid (MISO accepts bids as low as negative $500 per MWh), thereby displacing other producers’ power, and that when Duke elects not to do this it owes liquidated damages under the contract. Sometimes for load-­‐‑balancing or other technical reasons MISO tells Benton to stop delivering power even when the market price exceeds zero and Duke’s bid nomi-­‐‑ nally has been accepted. Benton acknowledges that in this situation Duke need not pay damages. The district court sided with Duke, ruling that it need pay only for power delivered to the “Point of Metering” where it is measured and passes to the local grid; when MISO issues a stop order that quantity is zero. 2015 U.S. Dist. LEXIS 181563 (S.D. Ind. Oct. 9, 2015). The parties have a second contract that requires Duke to cooperate, reasonably, in marketing Benton’s power; the district judge found that bidding $0 is “reasonable” cooperation because it usually leads Duke to suffer an out-­‐‑of-­‐‑pocket loss, since the market price will be less than what Duke must pay Benton. Indeed, on this understanding Duke might be entitled to bid $52 in MISO’s auction and ensure that it makes a profit on reselling every megawatt-­‐‑hour that it buys from Benton. This is a contract dispute, so we must set out the contrac-­‐‑ tual clauses that matter. We have tried to be parsimonious; interested readers can find more details in the district court’s opinion. There are two contracts—the first requiring Duke to buy Benton’s power, the second requiring Duke to cooperate with Benton. The parties call the first the Renewable Wind No. 15-­‐‑2632 7 Energy Purchase Power Agreement or PPA; they call the se-­‐‑ cond the Joint Energy Sharing and Operating Agreement or JESOA. We discuss the second contract briefly at the end of this opinion. For now, we refer to the first contract as “the contract.” We have already mentioned one clause. The contract re-­‐‑ quires Duke to purchase Benton’s output, which it defines as “the entire electrical output of the Plant delivered to the Point of Metering” (emphasis added). A separate clause de-­‐‑ fines that point as where Benton connects with the local grid (either NIPSCO’s or another designated by MISO). Benton relies principally on §4.6(a) of the contract, a liquidated-­‐‑ damages clause captioned “Buyer’s Failure to Accept Deliv-­‐‑ ery of Electrical Output”: In the event that Buyer fails to accept delivery of all of the Elec-­‐‑ trical Output at the Point of Metering, whether due to Buyer’s failure to obtain Transmission Service (if applicable) or for any reason other than Seller’s failure to perform, an Emergency Condition, a Force Majeure Event that prevents such acceptance pursuant to Article 14 or the proper exercise by Buyer of its sus-­‐‑ pension rights pursuant to Section 15.2(a), then Buyer shall pay to Seller as liquidated damages an amount equal to the positive difference, if any, between (i)(x) the amount that would have been payable by Buyer to Seller hereunder if such Electrical Output had been accepted by Buyer plus (y) additional trans-­‐‑ mission charges, if any, reasonably incurred by Seller in deliver-­‐‑ ing the Electrical Output to such third party purchaser and (ii) the net amount, if any, that Seller using Commercially Reasona-­‐‑ ble Efforts, actually realizes through remarketing of such Electri-­‐‑ cal Output to Persons other than Buyer, provided that in the event Seller is unable to remarket such Electrical Output, then the net amount described in clause (ii) shall be $0 and the damages owed by Buyer shall also include the then-­‐‑current amount of the PTC (on a per MWh basis) on an After-­‐‑Tax Basis for each MWh of such Electrical Output that Seller was unable to remarket. The 8 No. 15-­‐‑2632 damages provided in this Section 4.6 shall be the sole and exclu-­‐‑ sive remedy of Seller for any failure of Buyer to accept delivery of Electrical Output that it is required to accept hereunder. One more long clause matters. It is §6.4, captioned “Trans-­‐‑ mission”: Buyer represents that it intends to deliver and sell all of the Elec-­‐‑ trical Output to [MISO] at the Point of Metering and does not in-­‐‑ tend to utilize any Transmission Services. If Buyer nevertheless utilizes Transmission Services for the Electrical Output during the Term or is required (due to a change in the applicable trans-­‐‑ mission rules) to use Transmission Services in order to accept de-­‐‑ liveries of the Electrical Output at the Point of Metering, then Buyer shall be responsible for arranging for all Transmission Services required to effectuate Buyer’s acceptance of delivery of and purchase of Electrical Output, including, without limitation, obtaining Transmission Service, in an amount of capacity equal to the Designated Nameplate Capacity Rating, and shall be re-­‐‑ sponsible for the payment of any charges related to such Trans-­‐‑ mission Services hereunder, including, without limitation, charges for transmission or wheeling services, ancillary services, imbalance, control area services, congestion charges, location marginal pricing, transaction charges and line losses. The Parties acknowledge that the purchase price of Electrical Output does not include charges for such Transmission Services, all of which shall be paid by Buyer. Finally, there is a definition of “transmission services” as: all transmission or wheeling services, scheduling services, im-­‐‑ balance services, OASIS, congestion and congestion manage-­‐‑ ment services, tagging services, dispatch services, ancillary ser-­‐‑ vices, control area services, and other transmission services nec-­‐‑ essary for Buyer to accept Electrical Output at the Point of Me-­‐‑ tering and transmit, and deliver Electrical Output from the Point of Metering, using the highest priority transmission service available. No. 15-­‐‑2632 9 Many other clauses and definitions potentially have some bearing, but we think that these few decide the case. The parties agree that Indiana law governs, but they do not rely on any principles unique to Indiana. The dominant principle is that courts follow contractual language unless ambiguity permits the use of parol evidence. The parties agree that this contract is clear (though not on what it means), and we too think it unnecessary to go beyond the document’s language. Benton tells us that §4.6(a) is a take-­‐‑or-­‐‑pay clause, requir-­‐‑ ing Duke to pay for energy whether taken or not. The district court was not persuaded, and neither are we, for then it would require Duke to pay Benton even if the reason for non-­‐‑delivery is an instruction that MISO issues independent of how much Duke bid in the auction and independent of how much transmission capacity is available. MISO might issue such an order if, for example, there is a decline in de-­‐‑ mand on the buyers’ side of the market or a technical fault in some other grid, which cannot accept as much power from NIPSCO’s lines. Yet Benton concedes that Duke need not pay when it re-­‐‑ ceives such a stop order. Duke says, without contradiction from Benton, that the market-­‐‑clearing price is positive 80% of the time and Duke’s $0 bid thus is accepted (just as a neg-­‐‑ ative $500/MWh bid would have been), but that MISO al-­‐‑ lowed Benton to generate power only 59% of the time; the difference between 80% and 59% must be attributable to MISO’s decisions rather than Duke’s bid. If Duke need not pay Benton for energy when MISO’s choices, alone, account for non-­‐‑generation, §4.6(a) can’t be a standard take-­‐‑or-­‐‑pay clause. Nor does it call itself a take-­‐‑or-­‐‑pay requirement; it calls itself a liquidated-­‐‑damages clause. 10 No. 15-­‐‑2632 But the opposite view—that if energy is not generated and so does not cross the Point of Metering, and never counts toward actual output, for any reason at all (including Duke’s entry of a standing $52/MWh bid), then Benton need not be paid—also is unfaithful to the contractual language. Section 4.6(a) makes it clear that some reasons for Duke’s failure to take energy excuse payment; and from the limited range of reasons that justify nonpayment it follows that oth-­‐‑ er reasons are inadequate and that payment remains due. The key to resolving the parties’ dispute lies toward the beginning of §4.6(a), which requires Duke to pay if it “fails to accept delivery of all of the Electrical Output at the Point of Metering, whether due to Buyer’s failure to obtain Trans-­‐‑ mission Service (if applicable) or for any reason other than … [a list].” This covers the sort of situation that prevailed after MISO changed its dispatch rules at the end of February 2013 and no longer deemed Benton a must-­‐‑carry generator. As of March 2013, Benton was being told to stop 41% of the time because transmission was unavailable at the price Duke was willing to offer—and could have been unavailable even if Duke had bid negative $500/MWh, if owners of the re-­‐‑ maining local wind farms had made the same negative bid. With insufficient transmission capacity, someone (or a lot of someones) had to stop delivering energy to NIPSCO’s facili-­‐‑ ties no matter what price Duke offered. But the contract provides what is to happen when the stoppage is “due to Buyer’s failure to obtain Transmission Services”. Duke is to pay for power not taken. Duke could build its own transmission lines or buy extra capacity from NIPSCO or some other firm. (Our opinion in MISO Trans-­‐‑ mission Owners describes the process by which MISO allo-­‐‑ No. 15-­‐‑2632 11 cates the rights to build new lines or augment existing ones.) If there is a market for transmission services, as there surely is in central Indiana where more and more wind power is becoming available, then there will be a supply of transmis-­‐‑ sion lines. It is only a matter of time until more capacity is built, whether by Duke or someone else. And §4.6(a) tells us that, until this happens, Duke must pay Benton. The risk of inadequate transmission was contemplated by the contract-­‐‑ ing parties and allocated to Duke. By accepting this risk, Duke enabled Benton to finance its project; otherwise poten-­‐‑ tial investors might have feared exactly the overcapacity sit-­‐‑ uation that has come to pass. Duke wanted Benton’s facili-­‐‑ ties to exist and called them into existence by promising to pay even if a shortfall of transmission services should lead to curtailment of deliveries. Duke resists that conclusion by pointing to the opening of §6.4, and some equivalent language elsewhere in the con-­‐‑ tract, which relate that Duke did not plan or want to operate transmission lines, contemplated immediately handing Ben-­‐‑ ton’s power to MISO at the Point of Metering, “and does not intend to utilize any Transmission Services.” That’s fine as a statement of Duke’s goal; maybe it believed that extra transmission capacity would be unnecessary or that NIPSCO would add to its own capacity as wind farms were built. But §6.4 does not say that Duke will never need to add transmis-­‐‑ sion capacity itself or that it is excused from paying Benton if it chooses not to. To the contrary, three parts of the contract strongly imply that Duke must do what is needed to make transmission ca-­‐‑ pacity available. One is the contract’s definition of “trans-­‐‑ mission services” to include “other transmission services 12 No. 15-­‐‑2632 necessary for Buyer to accept Electrical Output at the Point of Metering” (emphasis added). The second is in §4.6(a), which says that Duke must pay if the failure to deliver power is caused by “Buyer’s failure to obtain Transmission Service (if applicable)”. Now go back to the second sentence of §6.4 for the third, which tells us that “[i]f Buyer nevertheless utilizes Transmission Services for the Electrical Output during the Term or is required (due to a change in the applicable transmission rules) to use Transmission Services in order to accept deliveries of the Electrical Output at the Point of Me-­‐‑ tering” then Buyer (Duke) must pay the full cost. What would be the point of this clause, if Duke never has an obli-­‐‑ gation to obtain transmission service for the power Benton is able to generate? Sections 4.6(a) and 6.4 read together tell us that Duke must arrange for new transmission services if they prove to be necessary for Duke to accept all of Benton’s power after a “change in [MISO’s] applicable transmission rules”. The district court rejected this line of reasoning, 2015 U.S. Dist. LEXIS 181653 at *71–73, because MISO has not required Duke to add transmission capacity. In other words, the court understood the word “required” in §6.4 to mean “required by MISO” and the parenthetical clause “if applicable” in §4.6(a) to mean “if required under §6.4.” Yet §6.4 does not say “required by MISO”. It says “required (due to a change in the applicable transmission rules) … in order to accept de-­‐‑ liveries of the Electrical Output at the Point of Metering.” And “Electrical Output” is defined, as we have already quoted, as all of the power that Benton generates, not just the power that can coexist on NIPSCO’s lines with all other wind-­‐‑generated power in the area. MISO’s role in §6.4 is not to require Duke to build transmission capacity, but to No. 15-­‐‑2632 13 change the rules of dispatching power over whatever trans-­‐‑ mission capacity happens to exist. MISO did that; the upshot was that it no longer accepted all of Benton’s output; and the consequence under §4.6(a) and §6.4 is that Duke must either build (or arrange for) more transmission capacity or pay Benton the amount specified in §4.6(a). Potential buyers and sellers of electricity could and did foresee when negotiating this contract (and others like it) that electrical grids may be swamped by new sources of re-­‐‑ newable power, which usually is located far from the centers of demand. They needed to allocate the risk of that devel-­‐‑ opment, which predictably would compel MISO to alter its rules for which sources could put power on the grid. Allo-­‐‑ cating the risk to Benton would have made it hard, perhaps impossible, to finance the project’s construction, while leav-­‐‑ ing Duke and similar utilities no incentive to expand the re-­‐‑ gional grids as wind power became available. Allocating the risk to Duke facilitates both construction of renewable-­‐‑ energy sources and better incentives to match the size of the transmission grid to the capacity for local generation. We read this contract as allocating the risk to Duke, which means that Benton receives the compensation provided by §4.6(a) and Duke has the right incentives to build or buy ex-­‐‑ tra transmission capacity. Duke contended in the district court that MISO’s 2013 rules are an “Emergency Condition” for the purpose of §4.6(a) and prevent any recovery. It has not renewed that ar-­‐‑ gument on appeal, perhaps because it is hard to think of a long-­‐‑term set of rules for pricing and dispatching power as an “emergency.” We could imagine an argument that an un-­‐‑ anticipated change in MISO’s rules is enough of an “emer-­‐‑ 14 No. 15-­‐‑2632 gency” to give Duke time to build or acquire new transmis-­‐‑ sion capacity without needing to compensate Benton in the interim, but MISO announced the new rules years in ad-­‐‑ vance and phased them in slowly. Duke did not attempt to add transmission capacity in the time between the rules’ an-­‐‑ nouncement and their 2013 application to post-­‐‑2005 wind farms—and as far as we can tell it has not attempted to build or buy new transmission capacity in Benton County since then. This line of argument therefore is unavailable. We have so far not discussed the terms of the second con-­‐‑ tract, which the parties call JESOA. Because we have con-­‐‑ cluded that Benton prevails under the first contract, the se-­‐‑ cond would be important only if it entitles Benton to a larger recovery. The damages clauses of the two contracts differ (as do the clauses that determine each party’s responsibilities), so that it is possible in principle that Duke could be liable under one, the other, or both, and owe different damages under each. But we do not understand Duke to contend that its recovery under the second contract would exceed its re-­‐‑ covery under the first. Indeed, Benton’s briefs in this court mention the second contract only once, in passing, and make nothing of it substantively. We therefore think it unneces-­‐‑ sary to decide whether Duke is liable under the second con-­‐‑ tract and, if so, what damages that contract would provide. The judgment is reversed, and the case is remanded with instructions to determine the relief to which Benton is enti-­‐‑ tled. No. 15‐2632 15 POSNER, Circuit Judge, concurring. I agree with the deci‐ sion to reverse the judgment of the district court and remand for a calculation of damages. But I think the majority opin‐ ion’s analysis could be simplified, and in addition I disagree with the majority’s discussion of damages for the breach of the second contract. This is a diversity suit that presents issues of Indiana con‐ tract law. Benton County Wind Farm, the plaintiff and ap‐ pellant, operates a plant in northwestern Indiana that uses wind to push turbines that generate electricity, which it sells. In 2006, before construction of the wind farm had begun, Duke Energy, a large electrical company also in Indiana, signed a 20‐year contract with Benton in which Duke agreed either to pay a fixed price for the output of the wind‐ powered electrical plant that Benton was planning to build, or to refuse to accept the output and instead pay liquidated damages to compensate Benton for the loss of business. In 2007 Benton proposed to construct additional turbines, which would increase the wind farm’s capacity to generate electricity; and Duke and another buyer, called Vectren Power Supply (not a party to this case), agreed to split the purchase of the additional power. A second contract, this one between Benton on the one hand and both Duke and Vectren on the other, defined the amounts Vectren and Duke would each purchase, resolved certain issues arising from the fact that there would be two buyers for Benton’s output, and forbade Duke to take steps to reduce Benton’s output. The first contract is the “Renewable Wind Energy Project Purchase Power Agreement” (I’ll call it just the “Purchase Power Agreement”) and the second (discussed in the majori‐ ty opinion only for its relevance to liability) is the “Joint En‐ 16 No. 15‐2632 ergy Sharing and Operating Agreement.” I’ll discuss the two contracts in that order. MISO (Midcontinent Independent System Operator)— the Regional Transmission Organization that coordinates and to a considerable extent controls the transmission of electricity in a number of midwestern and southern states, including Indiana, and also in a chunk of Canada—buys en‐ ergy from producers like Benton. It had begun acquiring wind‐powered electricity on the basis of competitive offers instead of buying all the wind energy offered to it. Some‐ times producers of wind energy would even have to pay MISO to induce it to accept their energy, which they were willing to do because even if they lost money on the sale they’d get a valuable tax credit for producing renewable en‐ ergy. Duke would be an intermediary between Benton and MISO, buying from Benton and selling to MISO. It offered the energy it was buying from Benton to MISO at a price of $0/MWh (that is, at a zero price for each unit of energy equal to the amount of electricity that a megawatt of output would transmit to MISO over one hour). Obviously that offer price was not a market price, and MISO paid Duke (as it did the other suppliers of electricity to the transmission grid) the market price if it exceeded the offer price. If however the market price happened to be zero, Duke still would deliver the energy to MISO at the free offer price (i.e., $0/MWh), but it would never sell to MISO at a negative price, as that would mean that Duke was paying both Benton (for the en‐ ergy) and MISO (for accepting delivery of the energy). It might seem that a market price would never be nega‐ tive, but actually it could be because of an excessive supply No. 15‐2632 17 of wind energy and/or inadequate transmission capacity. And when it was negative, and Duke therefore wouldn’t pay MISO to take Benton’s energy, MISO would tell Benton to stop producing; otherwise the electricity could keep coming, even though MISO was surfeited with electricity, which is why it wouldn’t accept any more wind‐powered electricity unless paid to take it. The Purchase Power Agreement, the first of the two agreements between Benton and Duke at issue in this case, requires that “Buyer [Duke] shall accept and purchase from Seller [Benton] Electrical Output of the Plant,” and “Seller will not have the right to sell to third parties any of the Elec‐ trical Output” unless Duke has refused to accept it. The con‐ tract goes on to provide that “in the event that Buyer fails to accept delivery of all of the Electrical Output at the Point of Metering, whether due to Buyer’s failure to obtain Transmis‐ sion Service … or for any [other] reason” (with some excep‐ tions), Duke must pay Benton liquidated damages unless Benton can find some other company to buy the power that Duke is refusing to accept from it. The liquidated damages are to consist of the contract price for the power plus the production tax credit that Benton would have earned by producing the power. The credit is a tax break that the fed‐ eral government provides to producers of renewable energy sources, such as wind power, to encourage efficiency in the production and transmission of electricity. U.S. Department of Energy, “Renewable Electricity Production Tax Credit (PTC),” http://energy.gov/savings/renewable‐electricity‐prod uction‐tax‐credit‐ptc (visited December 5, 2016). Duke argues and the district court ruled that because the contract defines “Electrical Output” as “the entire electric 18 No. 15‐2632 energy output of the [Benton] Plant delivered to the Point of Metering,” Duke has no liability for refusing power not de‐ livered to that point. The district court’s ruling ignores, however, the fact—not contested by Duke, and surely known by senior staff in the electrical‐generation and trans‐ mission industry of Indiana and therefore implicit in any contract made by the electrical firms in that market—that it’s physically impossible for Duke to reject electricity that has reached the Point of Metering. Electricity dispatched by Ben‐ ton flows to the Point of Metering but doesn’t stop there, be‐ cause traveling as it does at upwards of half the speed of light it enters almost instantaneously into the transmission grid. It’s not that electricity can’t be stopped because of the speed at which it travels; every time one turns off an appli‐ ance that draws electricity the electrical flow to the appliance is stopped. But a flow of electricity can’t be stopped at the Point of Metering, because it’s merely the point at which electricity flowing from the Benton Wind Farm enters the grid (the Purchase Power Agreement refers to it as an “inter‐ connection point”). There is no switch at that point, which could be turned off to stop the flow of electricity. Once ener‐ gy is generated by Benton and transmitted to the Point of Metering, Duke has no way to prevent it from flowing into the grid. This is not to say that points of metering are unim‐ portant; they play an important role in billing and more gen‐ erally in managing the flow of electricity between electrical companies. See New York Independent System Operator, “Revenue Metering Requirements Manual” p. 1‐1 (August 2013), www.nyiso.com/public/webdocs/markets_operations/ No. 15‐2632 19 documents/Manuals_and_Guides/Manuals/Administrative/ rev_mtr_req_mnl.pdf (also visited on December 5, 2016). But a point of metering is not a wall or an on‐off switch. Article 8 of the Purchase Power Agreement is explicit that the equip‐ ment at the point of metering consists of meters (measuring devices), not on‐and‐off switches or shut‐off valves. Because there is no such equipment at the Point of Meter‐ ing, the only way Duke can refuse to receive Benton’s elec‐ tricity is to tell it not to send its output to (which also means beyond) the Point of Metering. Unless required to pay liqui‐ dated damages to Benton when it tells Benton not to send electricity to the Point of Metering, Duke would be avoiding all liability simply by telling Benton not to send electricity Duke’s way; the liquidated‐damages clause in the contract would thus be a nullity. Benton further appeals to a provision in the contract which states that “the Parties will reasonably cooperate with each other with respect to the bidding and scheduling with … the RTO [i.e., MISO] of the Electrical Output to be sold and delivered by Seller [Benton] and accepted and pur‐ chased by Buyer [Duke]. Buyer will be responsible for all such bidding and scheduling.” Reasonable cooperation would appear to require that Duke not block Benton from supplying power to MISO without compensating Benton in accordance with the liquidated‐damages provision. This in‐ terpretation is reinforced by section 6.3 of the contract, which provides that “nothing in Section 6.2 … shall require Seller [i.e., Benton] to take any action effecting … any reduc‐ tion in the Electrical Output.” By ordering and thus compel‐ ling Benton to reduce its delivery of energy to the Point of Metering, Duke could be thought to be violating section 6.3 20 No. 15‐2632 by requiring Benton to reduce its output, and therefore to be required to pay liquidated damages to compensate Benton for the loss of revenue resulting from the reduction in deliv‐ ery. Another provision in the Purchase Power Agreement states, however, that the “Seller [i.e., Benton] will not have the right to sell to third parties any of the Electrical Output” unless Duke “fails to accept delivery.” The clause we’ve itali‐ cized frees Benton to sell to other electrical companies if Duke refuses to buy from it, and if Benton sells to other companies at the same price that Duke would pay, it would not be entitled to liquidated damages, because it wouldn’t have suffered a loss (aside from extra transmission expenses, which the contract covers). Similarly, if Benton finds another buyer willing to buy its energy but only at a lower price than Duke is willing to pay, the liquidated damages owed by Duke to Benton will fall by the amount of revenue that Ben‐ ton is able to recoup from the new buyer. That the Purchase Power Agreement itself does not men‐ tion that electricity generated by Benton and fed into Duke’s transmission line does not stop at the Point of Metering, but continues unaltered into the transmission grid, is not fatal to Benton’s argument for liquidated damages. A court cannot decide a suit for breach of contract by ignoring facts critical to the alleged breach. Krieg v. Hieber, 802 N.E.2d 938, 944 (Ind. App. 2004). “This is upon the principle that the court may be placed, in regard to the surroundings and circum‐ stances, as nearly as possible in the position of the parties whose writings are to be interpreted.” Ransdel v. Moore, 53 N.E. 767, 769 (Ind. 1899). The district judge indicated aware‐ ness of the physics of transmission, how a wind turbine No. 15‐2632 21 works, and how MISO structures its bidding process. All these were uncontested facts essential to understanding the contracts at issue, facts of which the judges on this panel can take judicial notice. And though hardly necessary we can also appeal to the familiar analogy of the medieval law re‐ garding “blood letting” in the streets of the Italian city of Bo‐ logna—the law that, as famously explained in William Blackstone’s Commentaries on the Laws of England, vol. 2, p. 60 (1765), stated that “whoever drew blood in the streets should be punished with the utmost severity.” Blackstone asked whether the law should have been interpreted to make pun‐ ishable a surgeon “who opened the vein of a person that fell down in the street with a fit.” He thought not, saying that “the fairest and most rational method to interpret the will of the legislator, is by exploring his intentions at the time when the law was made, by signs the most natural and probable. And these signs are either the words, the context, the subject matter, the effects and consequence, or the spirit and reason of the law … . As to the effects and consequence, the rule is, where words bear either none, or a very absurd signification, if literally understood, we must a little deviate from the received sense of them” (emphases added). The law did not mention surgeons, but Blackstone thought it obvious that the legisla‐ tors, who must have known something about surgeons (ac‐ tually “barber surgeons”), had not intended the law to apply to them. It is likewise obvious that firms engaged in the pro‐ duction and transmission of electricity know that it doesn’t stop at a “Point of Metering,” as if it were water stopped by a dam. Another factor to be considered, however, is the duration of the contract—20 years. As pointed out in an amicus curiae brief submitted by the American Wind Energy Association 22 No. 15‐2632 “in Support of Neither Party,” wind energy entrepreneurs must make a large investment in creating wind farms, and having a predictable flow of revenue is important in ena‐ bling the entrepreneurs to attract the needed investment. Benton County Wind Farm will have lost that predictable flow if the district court’s decision is affirmed. Cutting the other way, however, is the pincers that Duke Energy has been placed in as a result of developments apparently not foreseen by the parties when they drafted the Purchase Power Agreement back in 2006—namely the sprouting of a number of other wind energy farms in Indiana where once Benton had been one of only a few. The electrical energy transmitted by the growing Indiana wind energy industry crowded the transmission grid and led to efforts by MISO to reduce the flow. The electricity that Duke buys from Benton is sold to MISO at the Point of Metering at what is called the Locational Marginal Price (LMP), which is based on energy costs, congestion costs, and line losses. The price is set uni‐ laterally by MISO rather than negotiated with Duke. As ad‐ ditional wind energy farms came on line, the congestion component of the LMP soared to the point at which sellers to MISO, such as Duke, had to pay MISO to take their electrici‐ ty; that is, the price to MISO had turned negative. That meant that for electricity bought from Benton and sold to MISO at the point of metering, Duke would be losing money because it would be paying both Benton for the electricity and MISO for accepting the electricity forwarded to it by Duke. Duke could avoid such a loss by bidding $0/MWh to MISO, so that upon receiving a negative‐price offer from MISO (that is, being told by MISO that MISO would not pay a positive price for electricity generated by Benton for resale No. 15‐2632 23 by Duke to MISO), MISO would direct Benton not to trans‐ mit electricity to Duke. The result was to curtail Benton’s output and revenues, except insofar as Benton was able to find other buyers for its electricity—an issue not illuminated by the parties’ submissions in this litigation. Duke is arguing that the change in the market caused by wind energy congestion, which in turn caused MISO often to refuse to accept transmission of such energy without being paid to accept it, altered Duke’s obligations under the con‐ tract, which had not contemplated Duke’s having to pay both Benton and MISO for the same electricity—Benton to transmit the electricity at the Point of Metering and MISO to receive it there from Duke. Recall the provision in Duke’s contract with Benton that requires the parties to “reasonably cooperate with each other with respect to the bidding and scheduling with … [MISO] of the Electrical Output to be sold and delivered by [Benton] and accepted and purchased by [Duke].” One possible interpretation of reasonable coop‐ eration is that Duke must buy all the electricity that Benton wants to sell it, but another is that Benton must accept a re‐ duction in the amount of electricity bought from it by Duke in recognition that “reasonable cooperation” requires a com‐ promise in which both parties accept a reduction in compen‐ sation as a result of a development beyond their control— that development in this case being the advent of an unex‐ pected number of new wind energy farms, requiring in turn an alteration in MISO’s purchasing policies. But it’s unlikely that this provision was intended to place limits on the finan‐ cial obligations of the parties to each other in the bidding process—the clause is terribly fuzzy and the liquidated damages clauses deal adequately with the problem. 24 No. 15‐2632 Yet some years ago, Wisconsin Electric Power Co. v. Union Pacific R.R., 557 F.3d 504 (7th Cir. 2009), noted that “the doc‐ trine of impossibility in the common law of contracts excuses performance when it would be unreasonably costly (and sometimes downright impossible) for a party to carry out its contractual obligations. If the doctrine is successfully in‐ voked, the contract is rescinded without liability. The stand‐ ard explanation for the doctrine is that nonperformance is not a breach if it is caused by a circumstance ‘the non‐ occurrence of which was a “basic assumption on which the contract was made.”’” Id. at 505 , quoting Restatement (Second) of Contracts, introductory note to ch. 11, preceding § 261 (1981), quoting UCC § 2–615. Conceivably, to require Duke to pay a positive price to Benton for wind‐powered energy and receive a negative price for the same energy from MISO (that, or pay liquidated damages), resulting in Duke’s ob‐ taining zero or negative revenue, could be regarded as “un‐ reasonably costly” to Duke, requiring a modification of its contract with Benton. But neither in the district court nor in this court has Duke argued impossibility. It did plead as an affirmative defense a provision in the contract which states that enforcement is to be limited by general principles of eq‐ uity, including “concepts of … reasonableness.” But it can’t be that the mere fact that additional wind farms were built in Indiana after the contract was signed made enforcement of the contract, and in particular invocation of the liquidat‐ ed‐damages provision by Benton, unreasonable. So Duke violated the Purchase Power Agreement, and therefore I agree with the majority that the judgment of the district court regarding that branch of the case must be re‐ versed and the case remanded for a determination of the amount of liquidated damages to which Benton is entitled. No. 15‐2632 25 The second agreement between Duke and Benton is the Joint Energy Sharing and Operating Agreement. This agreement requires Duke to buy part of the additional out‐ put of the Benton wind farm resulting from its increasing its production capacity by 30 megawatts, and (in this respect much like the Purchase Power Agreement) denies Duke “the right to curtail or reduce [Benton’s] Total Facility Output,” defined as “the total electrical energy produced by [Benton] … as measured at the Delivery Point,” which is another name for the Point of Metering. Duke violated the contract by using MISO’s competitive bidding process to curtail Ben‐ ton’s production whenever market prices are negative. Since Duke can curtail Benton’s output only as “expressly provid‐ ed” in the Purchase Power Agreement, and the only express provision for reducing output requires Duke to pay liqui‐ dated damages, Duke’s curtailment of Benton’s output without paying liquidated damages is a breach of the second contract between the parties as well as of the first. This is clear enough to require reversal of the district court’s rejection of Benton’s argument that Duke breached the second contract. The majority opinion treats Duke’s breach of the second contract as a duplicate source of the same damages as required by the breach of the first contract. But the second contract determines how much of the ex‐ panded output of the Benton wind farm Duke is required to pay for, and if it fails to pay, how much in damages it will owe Benton. Regarding the second point, the issue of damages, the second contract (the Joint Energy Sharing and Operating Agreement) provides that “each party’s liability hereunder shall be limited to direct actual damages only,” and “neither 26 No. 15‐2632 party shall be liable for … lost profits or other business inter‐ ruption damages.” There is no definition of “direct actual damages,” and the meaning of the term has not been briefed on appeal. Duke may owe Benton less in “direct actual dam‐ ages” for its failure to buy any of the expanded output of the Benton wind farm than it would owe were there a liquidat‐ ed‐damages clause in the second contract. Benton’s losses from not operating its wind farm seem most like “lost profits or other business‐interruption damages.” The amount of damages to which Benton is entitled by Duke’s breach of the second agreement therefore remains to be determined on remand. A further complication is that although there’s no liquidated‐damages clause in the second contract, the con‐ tract refers to the Purchase Power Agreement throughout in such a way as to indicate that the parties may have expected the liquidated‐damages clause to apply, for otherwise, given that direct actual damages are likely to be zero or close to zero, the purpose of the Joint Energy Sharing and Operating Agreement would be defeated. Benton wanted to secure a steady income stream before it began constructing the new turbines, just as it had wanted before constructing the plant in the first place. It had the incentive under both contracts to have fallback protection in the form of a liquidated‐damages clause. I trust that on remand the district judge will be conscious of the “long tradition in contract law of reading contracts sensibly,” not as “parlor games but [as] the means of getting the world’s work done.” Beanstalk Group, Inc. v. AM General Corp., 283 F.3d 856, 860 (7th Cir. 2002), quoting Rhode Island Charities Trust v. Engelhard Corp., 267 F.3d 3, 7 (1st Cir. 2001).