dissenting:
The majority’s analysis expansively applies precedent that had already expanded the controlling statute to its limits. The result is an outcome that is not only entirely divorced from the plain language of the statute, but that also lacks any connection to the realities of maritime industry and creates needless uncertainty for employers with operations anywhere in the vicinity of a waterway.
The following facts provide the context for understanding the majority’s errors: Within the scope of his employment with NODSI, Zepeda never loaded or unloaded a vessel, never assisted anyone else who was loading or unloading a vessel, never so much as witnessed a vessel being loaded or unloaded; he never set foot on a ship, dock, pier, or wharf; he never worked at a job site that was used to load or unload vessels; and the Chef Yard, Zepeda’s primary workplace while employed by NOD-SI and the only work site at issue, is not connected to or even surrounded by areas used to load or unload vessels, and, critically, is not used to repair or maintain equipment that is used to load or unload ships. Despite these facts, the majority still concludes that NODSI can be held liable under a statute that should apply to him only if Zepeda played an “integral or essential part of loading or unloading a vessel,” Chesapeake & Ohio Ry. Co. v. Schwalb, 493 U.S. 40, 45, 110 S.Ct. 381, 107 L.Ed.2d 278 (1989), and was injured at a maritime situs, 33 U.S.C. § 903(a). Because this conclusion is contrary to law and common sense, I respectfully dissent.
I. Situs
“For a claimant to be eligible for benefits under the LHWCA (1) his injury must occur on a maritime situs, and (2) his status must be that of a maritime employee.” Coastal Prod. Servs. Inc. v. Hudson, 555 F.3d 426, 431 (5th Cir.2009). Because the Zepeda’s injury clearly did not occur “upon the navigable waters of the United States” or on an “adjoining pier, wharf, dry dock, terminal, building way, [or] marine railway,” he is eligible for benefits only if the Chef Yard constitutes an “other adjoining area customarily used by an employer in loading, unloading, repairing, dismantling, or building a vessel.” 33 U.S.C. § 903(a). The respondents concede the Chef Yard has no connection to repairing, dismantling, or building vessels, so the decisive question is whether the Chef Yard is an “adjoining area” that is “customarily used ... in loading [or] unloading.” Even under the framework of our already-expansive precedents, the answer to this question is, quite clearly, no.
Our cases break this “other adjoining area” question into two inquiries: (1) whether the situs in question is sufficiently close to the waterfront (geographic nexus) (2) and whether it has a functional nexus to maritime activity. See Hudson, 555 F.3d at 432. The ALJ defined the relevant “area” as the Chef Yard itself. NOD-SI concedes that, under our caselaw, the Chef Yard satisfies the geographic nexus prong. The relevant inquiry is therefore whether the work done at the Chef Yard, storing and repairing cargo containers, is customarily part of the loading or unloading process.
*411“Loading and unloading” is the work customarily done by longshoremen or stevedores. Precedent requires us to construe the “loading and unloading” broadly to include not only the physical movement of cargo on and off ships but also those additional related functions that contribute to the overall loading or unloading process by making the physical loading and unloading possible. Even construing the loading process exceptionally broadly, however, it still does not encompass container repair, a function that, both customarily and on these specific facts, is operationally separate and distinct. “Loading and unloading” cannot reasonably be interpreted to encompass every step in the overall supply chain surrounding marine shipping. Those conducting loading operations are not customarily involved with the prior step of containering or packaging the ocean-going cargo that they load. In most instances, including this one, marine cargo is packaged well before it is delivered to be loaded. From the point of view of stevedores, then, the container (whatever may be inside it) is simply part of the cargo: it is the thing to be loaded. Moreover, container repair is an additional step removed from the loading process because it happens either before such containers are filled with cargo or after they are empty. It is undisputed that the containers Zepeda repaired arrived at and departed from the Chef Yard completely empty. His work was therefore twice removed from the ship-loading process.
Because there are no facts indicating that the container repair work done at the Chef Yard is part of any loading or unloading process, the yard lacks a functional nexus to maritime activity. To return to the language of the controlling statute, the Chef Yard is emphatically not “customarily used ... in loading [or] unloading.”
Our cases are not to the contrary. In Texports Stevedore Co. v. Winchester, on which the respondents heavily rely, the situs in question was a “gear room” used by a stevedore company to store, repair, and maintain “the equipment used by stevedores to perform the loading operation.” 632 F.2d 504, 506 & n. 1 (5th Cir.1980) (en banc). In upholding the ALJ’s situs determination, the en banc court credited the ALJ’s finding that the functional nexus requirement was satisfied because “the gear room operations were part of the ongoing overall loading process.” Id. at 515. The gear room was therefore critically different from the Chef Yard.
Unlike the Chef Yard, the gear room in Winchester was used by a stevedore company as an indispensable part of its loading operations, and was therefore part of the same “overall loading process.” Operating a gear room or similar facility is essential to the work of stevedores because they must continually store, maintain, and repair their own equipment in order to provide loading and unloading services.1 The Chef Yard was not integrated into any loading process because the work done there — container repair — is not part of the work that must be done by those who are charged with loading vessels.
For the same reason, the language from Hudson on which the majority relies is inapplicable. In Hudson, we stated that “if a particular area is associated with items used as part of the loading process, the area need not itself be directly involved in loading or unloading a vessel or physically connected to the point of loading *412or unloading.” Hudson, 555 F.3d at 434 (emphasis added). As broad as this language is, it still does not apply to the Chef Yard because shipping containers are not “items used as part of the loading process.” . Marine containers are not used to load; they are loaded. This distinction will seem insignificant only if one fails to appreciate the division of labor customarily observed in. marine shipping. From the point of view of those actually involved in the industry, the difference between “items used as part of the loading process” and “items'that are loaded” is both clear and highly important — for stevedores, it is the difference between “my tools” and “my customer’s property.”2
The container repair work performed by Zepeda and others at the Chef Yard is not part of a loading or unloading process. Of course, one could artificially define the loading or unloading process as broadly as suits his purposes. But the statute, quite understandably, ties maritime functions to custom, and the work performed at the Chef Yard is not integral to the customarily-defined process of loading or unloading ships — it falls entirely outside the parameters of what a longshoring company must do or provide in order to load or unload a ship successfully. Accordingly, the Chef Yard is not a maritime situs under the LHWCA.
That the Chef Yard was not part of any “overall loading process” is highlighted by the fact that NODSI’s only customer was not a stevedore company but Evergreen, a shipping company that conducts no loading or unloading operations. This means that whoever was hired to load Evergreen’s containers onto its ships had no connection to or responsibility for repairing containers. Of course, a single employer could conduct both shipping and loading operations, but the separate actors involved illustrate what should have been clear anyway: container repair is functionally separate from the loading process. The ALJ therefore erred in concluding that the functional nexus requirement was satisfied because Evergreen’s containers were stored and repaired at the Chef Yard.
The majority relies on the testimony of a claims adjustor “acknowledging the presence of marine facilities in the area surrounding the Chef yard.” Majority Op. at 407. But this is misleading. The record shows that the Chef Yard was surrounded primarily by non-maritime businesses including a car wash, automotive shops, a warehouse, and manufacturing plants. Even more problematic is that the “presence of marine facilities” would do nothing to demonstrate that the Chef Yard, or anything near it, was' an area customarily used in loading, which is what the evidence must show in order to support the award of benefits. Implicit in the majority’s reasoning is a decision to define the relevant “area” as something larger than merely the Chef Yard, but this means that the adjacent buildings form part of the same “area” for purposes of the situs determination. The consequence of this analysis is that the neighboring car wash would also constitute a maritime situs. The ALJ’s situs determination was not supported by substantial evidence.3
*413The majority cites the principle that the LHWCA should be “liberally construed,” but fails to appreciate that our construction of the statute must still “conform[] with its pin-pose.” Winchester, 632 F.2d at 515 (quoting Voris v. Eikel, 346 U.S. 328, 333, 74 S.Ct. 88, 98 L.Ed. 5 (1953)). One purpose of the 1972 amendments to the LHWCA emphasized in our past decisions was to “reduc[e] the number of employees walking in and out of coverage.” Id. at 511.
Prior to 1972, the [LHWCA] applied only to injuries occurring on navigable waters. Longshoremen loading or unloading a ship were covered on the ship and the gangplank but not shoreward, even though they were performing the same functions whether on or off the ship. Congress acted to obviate this anomaly: § 903(a) extended coverage to the area adjacent to the ship that is normally used for loading and unloading, but restricted the covered activity within that area to maritime employment.
Schwalb, 493 U.S. at 46, 110 S.Ct. 381. But unlike the gear man in Winchester and the janitors and machinist in Schwalb, neither Zepeda nor his coworkers at the Chef Yard performed any portion of their work on the docks or aboard ships, and treating the Chef Yard as a maritime situs is therefore unrelated to “reducing the number of employees walking in and out of coverage.” The majority has not pointed to another purpose of the act that its exceptionally expansive interpretation of the situs requirement promotes, leaving the ever-growing reach of the LHWCA unchecked by any discernible limiting principle.
II. Status
For similar reasons, Zepeda is not a maritime employee under the LHWCA.4 “[A]side from the [occupations specified in the LHWCA], land-based activity ... will be deemed maritime only if it is an integral or essential part of loading or unloading a vessel.” Id. at 45, 110 S.Ct. 381.5 As explained, the container repair work performed by Zepeda for NODSI was not integral to shiploading.
Container repair cannot possibly be “integral” to loading because the two processes happen separately and neither is dependent on the other. Customarily, container repair and ship loading are not conducted, organized, coordinated, overseen, or commissioned by the same person or entity. See Winchester, 632 F.2d at 511 (“The inquiry is whether the claimant is engaged in or spends some of his time in activities traditionally performed by longshoremen.”). Unlike the equipment repair and maintenance work the Supreme Court considered maritime in Schwalb, Zepeda’s work was never necessary to keep any *414loading process going. Schwalb, 493 U.S. at 48, 110 S.Ct. 381 (“The determinative consideration is that the ship loading process could not continue unless the retarder that [respondent] worked on was operating properly. It is notable that the loading actually was stopped while [respondent] made the repairs----” (emphasis added)).
The respondents rely on our decision in Hullinghorst Indus., Inc. v. Carroll, 650 F.2d 750 (5th Cir.1981), in which we upheld a BRB determination that Carroll, a carpenter injured while building scaffolding beneath a pier, was a maritime employee for purposes of the LHWCA. Carroll was an employee of a scaffolding contractor hired by the owner of the pier, whose business included the loading and unloading of ships. The loading company planned to use the scaffolding built by Carroll to “repair ... a turntable (affixed to the pier) used by [its] longshoremen in the loading and unloading of ships,” but Carroll did not repair the turntable itself or otherwise participate in the physical loading of cargo. Id. at 756. Although Carroll had no direct involvement in loading or unloading, the critical fact was that his scaffolding work “directly furthered ... the loading and unloading of ships” by enabling the owner of the pier to perform its core longshoring operations. Id.
Zepeda, on the contrary, was not similarly a part of loading or unloading operations because his work was not something that a loading company would have to arrange or commission in order to load ships. The relevant question is whether the work at issue is within the purview of what a loading company would have to accomplish in order to successfully carry out its ship-loading work. Because a loading company must provide for (whether with its own employees or independent contractors)6 the repair of its piers and loading equipment, but is not responsible for packaging its customers’ cargo, Carroll is readily distinguishable. Zepeda was not “an integral part” of shiploading or any other maritime project and therefore is not a maritime employee within the meaning of the LHWCA.
III. Conclusion
The majority’s reasoning sweeps so broadly that it threatens to swallow every employer with even a tangential relation to the maritime industry. If a worker whose sole responsibility is to repair containers is covered by the LHWCA, why not a factory worker who manufactures those same containers? And if container manufacturing has a sufficient connection to loading, so also will the manufacture of countless other products that have some potential use in maritime activities. For example, a machinist working on an assembly line making cranes, some of which will eventually be used in shiploading by the manufacturer’s stevedore customers, has just as close a connection to loading and unloading operations as did Zepeda during his employment with NODSI, assuming only that the crane factory is located in an industrial district in the general vicinity of a port or harbor. But it does not end there. On the majority’s reasoning, it is not at all clear why a cobbler’s apprentice who repairs shoes for stevedores would not be covered by the LHWCA. The repaired shoes are, unlike repaired containers, “items used as part of the loading process,” and the cobbler’s shop is therefore a *415better candidate for maritime situs than is the Chef Yard.
Many non-maritime employers, including NODSI, prefer to locate their businesses in industrial districts close to waterways for reasons that have little to do with their proximity to the water. For example, property is relatively affordable and industrial uses are often permitted in such locations. But such businesses will now be pushed to move their operations inland because the only remaining way to ensure that they are not unintentionally subjecting themselves to LHWCA liability is to move far enough away from the waterfront to defeat the “geographic nexus” requirement. Unfortunately, even this strategy is fraught with uncertainty because our previous cases have implicitly eliminated any meaningful, independent proximity-based requirement from the statute’s situs provisions. See Winchester, 632 F.2d at 520 (Tjoflat, J., dissenting) (“To approach the situs question as one determined solely by function is effectively to read the situs requirement out of the [LHWCA] .... ”). Just how far inland does one need to go?
Employers should be able to ascertain their potential liability under the LHWCA, and the majority’s opinion will make that more difficult than ever. In this vein, it is important to note that NODSI consulted two separate insurance brokers, who both advised that the company did not need LHWCA insurance. Industry expectations cannot be controlling of our decisions, but we have previously stated that “Congress intended that liability should be imposed only where the employer had real or constructive notice of the likelihood of coverage.” Carroll, 650 F.2d at 757.
I respectfully dissent.
. A stevedore company, of course, could hire a subcontractor to operate a gear room, but whether conducted "in-house” or by third-parties, such operations are the responsibility of stevedores. A marine shipper does not repair the equipment stevedores use to load the shipper's cargo.
. The majority's analysis essentially equates an area "associated with items used as part of the loading process” with an area “associated with items associated with the loading process,” which greatly extends the reach of this already expansive language.
. The majority, understandably, relies heavily on the deferential standard of review, but its approach exemplifies a troubling pattern that has developed in our case law. Rather than analyzing whether there is substantial evidence that the Chef Yard is “used in loading or unloading,” the majority merely concludes *413that the ALJ's situs determination is only a minor expansion of what has previously been accepted as a maritime situs, and therefore approves the expansion. The problem with this approach is that, as the majority essentially concedes, it results in an ever-shifting baseline for evaluating the reasonableness of the BRB's decisions. So long as each case is only incrementally more expansive than the last, it is bound to be upheld. Accordingly, the LHWCA is gradually swallowing every employer in the vicinity of a port.
. Although the “situs” and "status” inquiries are distinct, the functional nexus component of the situs test will naturally converge with the status test where a claimant’s injury is sustained during the ordinary course of his employment at his regular workplace.
. The majority repeats that Zepeda was, during some of his employment by NODSI, a member of a longshoremen union, but this is of no import because this court has already held, en banc, that "[u]nion jurisdiction and the employer’s whim are not definitive on whether activities are 'maritime employment.’ ” Winchester, 632 F.2d at 511.
. Carroll made it clear that it does not matter whether the claimant's employer also has employees directly involved in loading and unloading operations, but that decision did nothing to change the requirement that the claimant must perform work that is "an integral or essential part of loading or unloading a vessel.''