Before JONES, SMITH, and HAYNES, Circuit Judges. JERRY E. SMITH, Circuit Judge: Plaintiffs, Devin Barrios—an employee of Centaur, L.L.C. (“Centaur”)—was injured while offloading a generator from a crew boat to a barge. The crew boat was owned and operated by River Ventures, L.L.C. (“River Ventures”); the barge was leased by Centaur. Barrios sued River Ventures and Centaur for vessel negligence under general maritime law and the Jones Act. River Ventures crossclaimed against Centaur for contractual indemnity. The district court granted summary judgment to Centaur, and River Ventures appeals. We reverse and remand. I. Before Barrios’s accident, non-party United Bulk Terminals Davant, LLC (“UBT”), executed a Master Service Contract (the “MSC”) with Centaur, a small marine construction company. The MSC added Centaur to UBT’s approved vendor list for work at its dock facility adjoining the Mississippi River (the “Davant Facility”). The MSC contained two provisions relevant to this appeal. The first imposed on Centaur an obligation to indemnify UBT and its contractors: CONTRACTOR SHALL RELEASE, DEFEND, INDEMNIFY AND HOLD UBT GROUP (DEFINED AS UBT AND UBT’S OTHER CONTRACTORS AND SUBCONTRACTORS OF ANY TIER . . . ) HARMLESS FROM AND AGAINST ANY AND ALL CLAIMS . . . BROUGHT BY ANY PERSON, PARTY OR ENTITY IN RESPECT OF PERSONAL OR BODILY INJURY TO, SICKNESS, DISEASE OR DEATH OF ANY MEMBER OF CONTRACTOR GROUP (DEFINED AS CONTRACTOR GROUP . . . REGARDLESS OF CAUSE OR FAULT, AND EVEN IF CAUSED IN WHOLE OR IN PART BY THE SOLE, JOINT OR CONCURRENT NEGLIGENCE OR FAULT OF ANY MEMBERS OF THE UBT GROUP OR THE UNSEAWORTHINESS OF ANY VESSELS OWNED, OPERATED OR OTHERWISE UNDER THE CONTROL OF ANY MEMBER OF UBT GROUP. The second required Centaur to obtain insurance covering those same parties: Prior to Contractor commencing Work hereunder for UBT, Contractor shall, but only to the extent of the liabilities assumed by Contractor in this Agreement, obtain from each of its insurers a waiver of subrogation in favor of each of the “UBT Group” . . . and, with the exception of Workers’ Compensation Coverage . . . and the Hull Insurance . . . name each of the UBT Group as additional insured to each insurance policy . . . , but only to the extent of the liabilities assumed by Contractor in this Agreement. . . . Contractor shall ensure that any endorsement naming the UBT Group as additional insureds shall not exclude from coverage the sole negligence of the insureds. Contractor shall be responsible for payment of all deductibles, premiums, retentions and payment for all expenditures incurred under any sue and labor provision. The MSC governed future project-specific work orders between the parties. Centaur and UBT executed several work orders for projects at the Davant Facility. One—for which Centaur submitted a proposal in October 2015—required installation of a concrete containment rail at one of the facility’s docks. The dock was principally used to load and offload ships carrying “dry bulk materials,” including coal and petroleum coke. The containment rail was necessary to prevent those materials from spilling both onto the dock and into the river. Centaur’s proposal indicated that, at an increased cost, both a barge and a tug boat would be required to complete the project. UBT accepted the proposal and issued a purchase order in November 2015. That purchase order and the MSC, in tandem, formed the contract at issue (the “Dock Contract”). To perform the work, Centaur chartered barge DB-582, which was equipped with a crane. Because DB-582 was a “dumb” barge that couldn’t self- navigate, it was moved up and down the river using a tugboat and winch. The Centaur crew used the barge to perform some construction work on the dock, including “drilling holes, cutting rebar, and pouring forms.” It also used the barge to store items, pack and unpack tools, hold safety meetings, take breaks, and eat lunch. Because the dock was most easily accessed by boat, UBT contracted with River Ventures for a crewed vessel—the M/V TROOPER—to transport Centaur’s employees from the parking area to their worksite. Centaur also used the crew boat to ferry tools and equipment in addition to its employees. On the day of the incident, Barrios and other Centaur employees were transporting a portable generator on the crew boat. While attempting to offload the generator, the M/V TROOPER began to separate from DB-582. That movement caused Barrios to fall into the river, where the generator hit him in the head, severely injuring him. Barrios sued River Ventures and Centaur, alleging, inter alia, vessel negligence under general maritime law and the Jones Act.[1] River Ventures— averring that it was a third-party beneficiary of the Dock Contract—cross- claimed against Centaur for contractual indemnity and additional assured status under its insurance policies. Centaur moved for summary judgment on River Ventures’s crossclaim, averring that the Dock Contract was nonmaritime and that its indemnity provision was therefore void under Louisiana law. To determine whether the contract was maritime, the court considered whether “(1) the work Centaur was performing for UBT involve[d] maritime commerce, (2) it involved work from a vessel, and (3) the contract provided or the parties expected that a vessel would play a substantial role in completing the contract.” Applying that test, the court held that the Dock Contract was a “land- based construction contract” governed by Louisiana law. It granted summary judgment because the Louisiana Construction Anti-Indemnity Statute (“LCAIS”) “applie[d] to prohibit the indemnity and insurance provisions.” River Ventures filed a notice of interlocutory appeal challenging the summary judgment, averring that this court had jurisdiction under 28 U.S.C. § 1292(a)(3) because its claims against Centaur arose in an “admiralty case” and determined the “rights and liabilities” between the parties. Centaur moved to dismiss that appeal for lack of jurisdiction, maintaining that the appeal “should not go forward until a Final Judgment is entered by the District Court.” A panel of this court determined that Centaur’s motion should be carried with the case. While the interlocutory appeal was pending, Barrios’s underlying tort claims proceeded to a bench trial. The court ruled for Barrios, holding that River Ventures was liable and that Centaur wasn’t liable because Barrios wasn’t a Jones Act seaman. The court then entered final judgment. River Ventures appealed, reasserting its intent to seek review of the summary judgment. It also filed a notice of appeal of the bench-trial findings, but it voluntarily dismissed that appeal after settling with Barrios. River Ven- tures’s crossclaim against Centaur is the only claim remaining on appeal. II. “[W]e have a constitutional obligation to satisfy ourselves that subject matter jurisdiction is proper before we engage the merits of an appeal.” Ziegler v. Champion Mortg. Co., 913 F.2d 228, 229 (5th Cir. 1990). Therefore, we first consider Centaur’s motion to dismiss the appeal. We need not decide, however, whether we have jurisdiction under § 1292(a)(3). That is because after final judgment was entered, River Ventures filed a renewed notice of appeal related to its indemnity and insurance claims. Because we have jurisdiction over River Ventures’s appeal under 28 U.S.C. § 1291, Centaur’s motion to dismiss for lack of jurisdiction is denied as moot. III. The indemnity dispute presents issues with which this court is familiar. It boils down to what law governs. If federal maritime law controls, then the Dock Contract’s indemnity provision is enforceable. See Hoda v. Rowan Cos., Inc., 419 F.3d 379, 380 (5th Cir. 2005). If Louisiana law applies, then the LCAIS voids the indemnity provision as against public policy. See LA. STAT. Ann. § 9:2780.1. So the question is whether the Dock Contract is maritime. But before we can resolve that, we must identify the proper test for making that determination, a task that has vexed this court for decades. A. From 1990 to 2018, we applied the six-factor test announced in Davis & Sons, Inc. v. Gulf Oil Corp., 919 F.2d 313, 316 (5th Cir. 1990), to determine whether a contract was maritime: 1) what does the specific work order in effect at the time of injury provide? 2) what work did the crew assigned under the work order actually do? 3) was the crew assigned to work aboard a vessel in navigable waters; 4) to what extent did the work being done relate to the mission of that vessel? 5) what was the principal work of the injured worker? and 6) what work was the injured worker actually doing at the time of injury? Though Davis & Sons was intended to provide clear criteria for courts to apply, the test proved unwieldy in practice, with “final result[s] [often] turn[ing] on a minute parsing of the facts.” Hoda, 419 F.3d at 380. Fourteen years after Davis & Sons, the Supreme Court erected a guide- post in Norfolk Southern Railway Co. v. Kirby, 543 U.S. 14 (2004). The Court considered whether a money-damages claim arising from a train derailment fell within its admiralty jurisdiction. The cargo destroyed in the derailment was completing the second, land-based leg of its journey from Australia to Alabama. The first leg had transported the cargo by boat from Australia to Georgia. The two legs of the trip had separate but co-extensive bills of lading. To determine whether the bills of lading were maritime, the Court noted that it could not merely “look to whether a ship or other vessel was involved in the dispute” or “to the place of the contract’s formation or performance.” Id. at 23-24. Geography couldn’t be controlling because “the shore [was] now an artificial place to draw a line.” Id. at 25. Instead, “the answer depends upon the nature and character of the contract, and the true criterion is whether it has reference to maritime service or maritime transactions.” Id. at 24 (cleaned up).[2] That approach vindicated the fundamental interest undergirding maritime jurisdiction: “the protection of maritime commerce.” Id. at 25. Applying those principles, the Court held that the bills of lading were maritime “because their primary objective [was] to accomplish the transportation of goods by sea from Australia to the eastern coast of the United States.” Id. at 24 (emphasis added). “[S]o long as a bill of lading requires substantial carriage of goods by sea, its purpose is to effectuate maritime commerce—and thus it is a maritime contract.” Id. at 27 (emphasis added). That some of the performance was land-based did “not alter the essentially maritime nature of the contracts.” Id. at 24. In In re Larry Doiron, Inc., 879 F.3d 568 (5th Cir.) (en banc), cert. denied, 138 S. Ct. 2033 (2018), we sought to remedy the infirmities of Davis & Sons and harmonize our law with Kirby. We considered whether a “work order . . . to perform ‘flow-back’ services on a gas well in navigable waters” was a maritime contract. Id. at 570. The work didn’t require vessels, and neither party expected to use them. A crane barge was engaged only after the workers “determined that some heavy equipment was needed to complete the job.” Id. A worker was injured when he was struck by the crane. In a unanimous opinion, we adopted a two-question test—centering the inquiry “on the contract and the expectations of the parties,” id. at 576—to determine whether a contract was maritime: First, is the contract one to provide services to facilitate the drilling or production of oil and gas on navigable waters? . . . Second, if the answer to the above question is “yes,” does the contract provide or do the parties expect that a vessel will play a substantial role in the completion of the contract? Id. That standard jettisoned the irrelevant prongs of Davis & Sons and made clear that “contract rather than tort principles” control when determining whether a contract is maritime.[3] Applying the new test, we held that the contract was nonmaritime because “[t]he use of the vessel to lift the equipment was an insubstantial part of the job and not work the parties expected to be performed.” Id. at 577. The crew had involved a vessel only after it had “encountered an unexpected problem that required a vessel and a crane to lift equipment needed to resolve [it].” Id. Even though the vessel’s involvement was important, it wasn’t substantial because its use didn’t comport with the parties’ expectations. Since Doiron, we’ve had only one occasion to apply its standard: Crescent Energy Services, L.L.C. v. Carrizo Oil & Gas, Inc., 896 F.3d 350 (5th Cir.), cert. denied, 139 S. Ct. 642 (2018). There, the contract involved the plugging and abandonment of three offshore oil wells on small fixed platforms. Id. at 352. About half the job involved “wireline work.”[4] To complete the task, Crescent charted three vessels: a crane barge, a tug boat, and a cargo barge. A Crescent employee’s leg was severely injured when a piece of pipe struck him while he was sitting on one of the fixed platforms. Crescent’s insurers, attempting to limit Doiron‘s reach, made two contentions. First, the insurers posited—relying primarily on circuit caselaw stating that work performed on fixed offshore platforms is nonmaritime—that Doiron‘s first prong wasn’t satisfied because “the plugging and abandoning work did not occur on ‘navigable waters.’” Id. at 356. Second, and relatedly, they averred “that Doiron must be read in conjunction with other law,” specifically precedents classifying activities as either maritime or not. Id. at 357. We rejected both theories, affirming that, for analyzing whether a contract was maritime, “Doiron now control[led] that endeavor.” Id. at 358. Because the wells at issue “were located within the territorial inland waters of Louisiana and . . . the vessels involved . . . were able to navigate to them,” the contract “was to facilitate the drilling or production of oil and gas on navigable waters.” Id. at 357. And because the “contract anticipated the constant and substantial use of multiple vessels,” it was maritime. Id. at 361. Outside of the instant case, only one district court that we know of has applied Doiron to a non-oil-and-gas contract. In Lightering LLC v. Teichman Group, LLC, 328 F. Supp. 3d 625, 627-29 (S.D. Tex. 2018), the court considered whether a contract for wharfage, storage, and other dockside services was maritime. In determining how to apply Doiron outside the oil-and-gas sector, the court first observed that “Kirby state[d] that for a contract to be maritime, its principal objective must be maritime commerce.” Id. at 636. Doiron, the court inferred, “applie[d] Kirby to interpret a ‘principal objective.’” Id. Based on that, the court determined that Doiron s first factor—i.e., “is the activity at issue oil and gas?”—was a substitute for Kirbys broader question whether a contract involved maritime commerce and work from a vessel. Id. “Under Doiron and Kirby, determining whether a contract is maritime requires three steps: (1) [T]he activity must be maritime commerce; (2) the activity must involve work from a vessel; and (3) the contract must provide or the parties must expect that a vessel will play a substantial role in completing the contract.” Id. at 637. Though the analysis is seemingly clear-cut, applying that test—and especially the first prong—was far from straightforward. As the court recognized, the caselaw doesn’t clearly define the boundaries of “maritime commerce.” Id. As a result, “most courts resort to a case-by-case approach, relying heavily on precedent.” Id. Utilizing that precedent-focused method, the court determined that the contract’s wharfage and dockside services were subsumed within a wide range of land-based activities that facilitate maritime commerce but that aren’t, themselves, maritime commerce. See id. at 637-38. As a result, the contract was “[nonmaritime] in nature and character.”[5] B. The parties agree that Kirby, Doiron, and their progeny govern, but they read those authorities differently.[6] Their primary disagreement centers on how to apply Doirons first prong outside the oil-and-gas context. River Ventures avers that “this Court should apply a test that the contract be performed or facilitate operations on navigable waters and that the contract provide or the parties expect that a vessel will play a substantial role in the completion of the contract.”[7] Centaur counters that Lightering sets forth the proper test. The district court accepted Centaur’s position and applied Lightering. River Ventures has the better of the argument: Doiron should apply essentially as written. For non-oil-and-gas contracts, Doiron would ask whether (1) the contract is “one to provide services to facilitate [activity] on navigable waters,” and (2) if so, whether “the contract provide[s] or . . . the parties expect that a vessel will play a substantial role in the completion of the contract.” Doiron, 879 F.3d at 576. That test is preferable for two reasons: (1) Doiron was meant to streamline the inquiry regarding whether a contract is maritime; and (2) Doiron‘s rule, even applied to non-oil-and-gas contracts, is consistent with Kirby. In Doiron, the en banc court clarified that its test was intended to simplify the is-this-contract-maritime inquiry, not complicate it. To do that, we abrogated a significant portion of Davis & Sons‘s six-factor standard. Chief among those factors that Doiron jettisoned was the second, which required courts “to parse the precise facts related to the services performed under the contract and determine whether those services were inherently maritime.” Id. at 573. That was true even for mixed-services contracts where none of the services were inherently maritime. Id. That inquiry, Doiron held, was irrelevant to whether a contract was maritime because it didn’t focus on whether the contract required “substantial work to be performed from a vessel.” Id. at 573, 576-77. To the extent that the Davis & Sons factors remained relevant, they were so only as they helped to explain the “scope of the contract” or “the extent to which the parties expect[ed] vessels to be involved in the work.” Id. at 577. Doirons method, in contrast to Davis & Sons, ensures that courts aren’t determining whether some “service work has a more or less salty flavor than other service work when neither type is inherently salty.” Id. Centaur’s position would turn Doiron on its head and effectively return courts to Davis & Sons‘s precedent-laden trudge. Lightering recognized as much.[8] But Doiron and Crescent made clear, and for good reason, that we should be out of that business. “[R]egardless of what other Fifth Circuit caselaw there may be, nothing in such caselaw detracts from the clarity of our 2018 en banc decision in Doiron.’”[9] Centaur contends that applying Doiron outside the oil-and-gas context would run afoul of Kirby s command that the “principal objective” of the contract must be maritime commerce. That is so because Doiron established its two-part test only after it “emphasiz[ed] the importance of first determining whether the activity is ‘commercial maritime activity.’”[10] Doirons first prong, Centaur posits, merely provides “a short-cut for deciding whether a contract’s principal objective is maritime commerce, but only for oil and gas contracts.” “Outside the oil and gas context, the test first requires the court to ask whether the activity involves maritime commerce and work from a vessel.” Lightering, 328 F. Supp. 3d at 637 (quotation marks omitted). But Centaur’s position, though somewhat supported by language in the caselaw, doesn’t adequately grapple with Kirby, Doiron, or Crescent. In Doiron, 879 F.3d at 576, we found “strong support in Kirby” for our two-prong test. Several passages provide those buoys. First, Kirby states that “[t]o ascertain whether a contract is a maritime one, . . . the answer depends upon the nature and character of the contract, and the true criterion is whether it has reference to maritime service or maritime transactions.” Kirby, 543 U.S. at 23-24 (cleaned up). Next, Kirby instructs that “[m]aritime commerce has evolved . . . and is often inseparable from some land-based obligations.” Id. at 25. And finally, Kirby declares that “[c]onceptually, so long as a bill of lading requires substantial carriage of goods by sea, its purpose is to effectuate maritime commerce—and thus it is a maritime contract.”[11] Those statements are entirely consistent with Doiron‘s standard as applied to any mixed-services contract. Doiron‘s first prong—though requiring some nexus to the traditional maritime predicate of activity on navigable waters[12]—doesn’t exclude non-sea-based obligations. And Doirons second prong clarifies that cursory or unexpected vessel involvement, even if important, isn’t enough; the involvement must be substantial. In that sense, both prongs of Doiron stand in for Kirby s requirement that the “principal objective” of the contract be maritime commerce.[13] In short, Doiron‘s two-part test applies as written to all mixed-services contracts. To be maritime, a contract (1) must be for services to facilitate activity on navigable waters and (2) must provide, or the parties must expect, that a vessel will play a substantial role in the completion of the contract. IV. Having fashioned the appropriate test—and because “the interpretation of a maritime contract is a question of law”—we now apply it. See Int’l Marine, L.L.C. v. Integrity Fisheries Inc., 860 F.3d 754, 759 (5th Cir. 2017). 1. The Dock Contract easily satisfies Doiron s first prong. It called for Centaur to install a concrete containment rail at one of the Davant Facility’s docks. That dock extended into the Mississippi River, a waterway on which both DB-582 and the M/V TROOPER were navigated. The dock was used to load and offload ships carrying dry bulk materials. And the containment rail was meant to prevent those materials—principally coal and petroleum coke—from spilling onto the dock or into the river, which would result in adverse effects to both commerce and the environment. Collectively, those facts establish that the Dock Contract required services to be performed to facilitate the loading, offloading, and transportation of coal and petroleum coke via vessels on navigable waters. That some services were also performed on the dock, which was affixed to the land, isn’t dispositive.[14] 2. When considering whether there was substantial involvement of a vessel, “[w]e must remember that the contracting parties’ expectations are central.” Crescent, 896 F.3d at 359. “When work is performed in part on a vessel and in part on a platform or on land, we should consider not only time spent on the vessel but also the relative importance and value of the vessel-based work to completing the contract.” Doiron, 879 F.3d at 576 n.47. Doiron suggests that a rule of thumb similar to the thirty-percent guideline in Jones Act cases might be useful. Id.[15] But that “would not include transportation to and from the job site.”[16] Even significant vessel involvement isn’t enough if that involvement was unexpected. Crescent, 896 F.3d at 359-60. Based on that standard, Doiron s second prong is likewise satisfied. The Dock Contract makes clear that the parties expected DB-582 to play a significant role in the completion of the work. Centaur’s project proposal indicated that the “[p]rice is significantly higher due to having [a] crane barge on site to mix the concrete and pour it for the concrete containment rail.” It also stated that a “[t]ug boat will . . . need to be present to shift the barge as needed.” Far from being “an insubstantial part of the job and not work the parties expected to be performed,” Doiron, 879 F.3d at 577, the proposal shows that the parties expected the barge to play a critically important role. Moreover, Taylor Roy, Centaur’s lead project manager, admitted that “at the end of the day, Centaur could not have done the job properly without [the] crane barge.” That differs materially from Doiron, where the vessel was used only after the “crew encountered an unexpected problem.” Id. Instead, like the situation in Crescent, the parties here recognized that DB-582 provided a necessary work platform, an essential storage space for equipment and tools, and a flexible area for other endeavors related to the construction work. See Crescent, 896 F.3d at 361. The district court’s findings of fact show that the parties’ expectations about the use of the barge were borne out. Just as the proposal indicated, Centaur’s crew used DB-582 to perform construction work for the containment rail, including “drilling holes, cutting rebar, and pouring forms.” The crew also used the barge for several activities related to the construction, including storing and packing tools, holding safety meetings, taking breaks, and eating lunch. That Centaur’s workers may have worked on the dock a majority of the time doesn’t alter that conclusion.[17] **** In sum, the district court misapplied Doiron and erroneously concluded that the Dock Contract was nonmaritime. Because federal maritime law applies, the LCAIS does not. The summary judgment is REVERSED and REMANDED. We place no limitation on the matters that the district court may consider, as appropriate and in its discretion, on remand.
