Arbitration Starts Before the Dispute: A Guide for In-House Counsel

When a potential sale or joint venture (“JV”) opportunity comes in, the focus is usually on the substantive terms of the deal. Performance deadlines and profit/cost allocations predominate. However, when the dispute lands on your desk, the arbitration clause buried in the contract suddenly becomes crucial. Whether that clause was carefully negotiated or copied from a template years ago will shape the next two to five years of your professional life—dictating which institution administers the case, where hearings take place, how much the process will cost, and whether the resulting award can actually be enforced.

This primer is designed for in-house counsel who need to get arbitration clauses right at the drafting stage and navigate the strategic crosscurrents—seat selection, enforcement, cost management, confidentiality, and emerging trends—that run through every international arbitration.

Choosing arbitration over litigation is not a contractual formality—it is a strategic decision with legal, financial, and operational consequences. At its best, arbitration can be faster and less expensive than litigation, allows the parties to appoint arbitrators with genuine subject-matter expertise, keeps sensitive business information out of the public record, and—critically—produces awards enforceable in over 170 countries under the New York Convention. In complex cross-border transactions such as mergers and acquisitions (“M&A”) deals and engineering, procurement, and construction (“EPC”) contracts, arbitration can be seen as the obvious choice, as it offers confidentiality, specialized decision-makers, and a degree of predictability that national courts in unfamiliar jurisdictions may not.

But the benefits of arbitration are not guaranteed. None of these advantages materialize automatically—they depend on how the clause is drafted, how the process is managed, and who is making the decisions along the way.

That is where in-house counsel comes in. Your role spans the full arbitration lifecycle—from drafting the dispute-resolution clause, to selecting the tribunal and managing costs, to driving settlement strategy and enforcing the award. Even during initial contract negotiations, when bargaining power may be limited, in-house counsel should lock down the critical variables: seat, language, arbitral institution, and governing law.

The arbitration clause is the jurisdictional foundation of the entire dispute-resolution mechanism. A defective clause does not just create procedural headaches—it can render the mechanism unworkable altogether. The scale of the problem is striking: institutional leaders have estimated that roughly 30 percent of cases involve a threshold fight over arbitrability traceable to poor drafting, and that as many as 70 to 80 percent of arbitration clauses are “pathological” in some respect. The reason is predictable. Because dispute-resolution clauses are typically the last provision negotiated, they become "midnight clauses" drafted in haste or "champagne clauses" finalized amid closing celebrations—and the defects that result can haunt the parties for years.

A "pathological clause" is one that is defective, unclear, or contradictory enough to frustrate the arbitral process. The most common defects include: hybrid or conflicting clauses that leave it unclear whether arbitration is mandatory, inviting preliminary objections and judicial intervention; failure to specify the seat (or confusing “venue” with “seat”), which creates jurisdictional uncertainty because the seat determines the procedural law and supervisory court; references to non-existent or incorrectly named institutions, which generate threshold disputes before the merits are ever reached; bare clauses that state disputes will be arbitrated but omit essential details like seat, language, or appointment mechanism; internally contradictory clauses naming two seats or two institutions; and hybrid institutional arrangements—such as one institution administering under another’s rules—which produce procedural complications and challenges to awards.

Courts generally try to uphold arbitration clauses where the parties clearly intended to arbitrate, but this is not guaranteed—the Swiss Supreme Court, for example, rejected a clause calling for American Arbitration Association (“AAA”) arbitration "or to any other US court" as failing to evince that intention. Relying on judicial rescue is poor risk management. You have control over whether arbitration succeeds or fails at the drafting stage.

The single best safeguard against a pathological clause is to start with a model clause from a reputable arbitral institution. The International Chamber of Commerce (“ICC”), AAA/International Chamber for Dispute Resolution, Singapore International Arbitration Centre, Hong Kong International Arbitration Centre, and Lusaka International Arbitration Centre all publish tested model language designed to be enforceable worldwide. Always confirm you are using the current version—institutions update their model clauses when rules change.

Beyond adopting a model clause, keep these principles in mind:

• Do not treat the clause as boilerplate. Dispute-resolution provisions deserve the same attention as substantive deal terms—draft them deliberately.
• State an unequivocal intent to arbitrate. The clause should make clear that all disputes will be submitted to final and binding arbitration—ambiguity here is what produces hybrid-clause pathologies.
• Lock down the essential elements. At a minimum, specify the seat, the institution and its rules, the number of arbitrators, the language, and the governing law.
• Do not mix institutions. Combining one institution’s administration with another’s rules invites procedural challenges and is a leading cause of pathological clauses.
• Review your standard clauses periodically. Laws and institutional rules evolve; a clause that was enforceable five years ago may not be today.
• Get the clause right before the dispute. Trying to form an arbitration agreement after a dispute has arisen is "fraught and subject to new and special considerations of party advantage.”
• Consider a multi-tiered mechanism. A mediation or dispute-board tier before arbitration can resolve matters early and at lower cost; the ICC publishes standard tiered-clause wording for this purpose.
• Tailor the model clause to your deal. No model clause fits every situation. Adapt it when parties are in different jurisdictions, non-monetary remedies may be needed, or the counterparty is a state or state-owned enterprise.
• Confirm arbitrability and choose your format. Verify that the likely disputes are arbitrable under the applicable law and weigh the trade-off between institutional arbitration (structured rules, administrative support, but less flexibility) and ad hoc arbitration (more party control, but more procedural burden).

The seat is the "legal place" of an international arbitration—the jurisdiction in which the arbitration is deemed to take place—and is usually fixed by the parties in the agreement. Its selection is one of the most important choices parties make and carries significant legal and practical consequences. The seat determines the lex arbitri (the procedural law), which fills gaps on procedural issues the parties have not addressed, and it subjects the arbitration to any mandatory, non-waivable requirements of that jurisdiction. Critically, choosing the seat designates the courts with a "curial role"—only the courts of the seat have jurisdiction to annul or set aside an award—so the choice also determines the grounds on which an award can be challenged.

The seat need not be the jurisdiction whose law governs the merits; a contract governed by New York law may be seated in France, because the choice-of-law clause and the seat are distinct. Because local courts can affect proceedings at every stage—assisting in commencing arbitration, granting provisional relief, aiding evidence-gathering, and hearing set-aside applications—parties should thoughtfully select the seat and understand its implications at the drafting stage. Practical factors include convenience and access, the pool of experienced arbitrators, and the language and legal culture of the seat. Given rising geopolitical risk, parties are increasingly favoring seats perceived as politically neutral with strong, independent, arbitration-friendly judiciaries.

The law governing the arbitration agreement itself is a distinct and often overlooked question. An arbitration agreement may be governed by the law of the seat—for example, English law where the arbitration is seated in England and Wales—absent an express choice. Commentary proposes that an arbitrator should apply the law of the seat where a particular issue falls within the scope of the lex arbitri, and a different law where it does not. This is a live issue in enforcement, because the substantive validity of the arbitration agreement can determine how a pathological clause is interpreted under the New York Convention.

The 1958 New York Convention is the backbone of international commercial arbitration, adopted by more than 170 countries, giving a winning party a streamlined path to convert a private award into an enforceable court judgment in virtually any major trading nation, with a strong presumption favoring enforcement. Under Article III, each contracting state must recognize arbitral awards as binding and enforce them under its rules of procedure. The Convention covers both "foreign" awards (issued in a different country from where enforcement is sought) and "non-domestic" awards (rendered in the enforcing country but treated as foreign due to international elements).

Counsel should be aware of two common reservations: the reciprocity reservation, which limits benefits to awards made in another signatory state, and the commercial reservation, which restricts the Convention to relationships deemed "commercial" under national law. Enforcement can be refused only on the limited grounds in Article V, of which the "public policy" defense under Article V(2)(b) has generated the most discussion, though U.S. courts have interpreted it narrowly consistent with a pro-enforcement policy. Because a losing party may hold assets in multiple countries—each with its own procedural requirements—delay in selecting an enforcement strategy is a real risk. Note also that non-monetary awards, while generally enforceable under the New York Convention (unlike under Article 54 of the International Centre for Settlement of Investment Disputes Convention), can raise practical enforcement issues that have received limited judicial consideration.

Persistent concerns about cost, delay, and "over-lawyering" continue to challenge arbitration's reputation as an efficient mechanism. Recurring drivers of inefficiency include "due process paranoia" (arbitrators permitting overly extensive procedures to avoid challenges), excessive document production influenced by common-law discovery practices, lengthy submissions and hearings, and guerrilla tactics. In-house counsel are advised to pursue procedural efficiency—single-arbitrator or expedited proceedings can help in appropriate cases—but cautiously, balancing speed against evidentiary depth so that efficiency measures fit the circumstances of each case. The ICC's guide for in-house counsel provides a practical toolkit and a checklist of procedural decisions to be made at each phase to conduct arbitration in a time- and cost-effective manner. Agreeing key procedural aspects such as place and language in the clause itself avoids delays from having to resolve them after a dispute arises.

Confidentiality is one of arbitration's central attractions—proceedings are private and not open to public scrutiny—and the ICC offers a standard clause providing that no award or procedural order shall be published where parties want added privacy. The 2025 Queen Mary University of London International Arbitration Survey identifies efficiency, confidentiality, and the use of artificial intelligence (“AI”) as major current trends and offers insights on preferred seats and rules. Beyond these, the landscape is being shaped by legislative reform (such as the English Arbitration Act 2025), geopolitical instability and sanctions, the future of investor-state dispute settlement, and new guidance on the use of AI in arbitration.