Updated posts 2026
The Multilateral Agreement on Investment: The Treaty That Never Was
For three years in the late 1990s, the world’s wealthiest countries tried to write a single rulebook for international investment. They failed — and the way they failed changed how trade and investment negotiations have been conducted ever since.
The Multilateral Agreement on Investment, negotiated at the Organisation for Economic Co-operation and Development between 1995 and 1998, was abandoned without a single signature. It has no legal force and never has had. But the arguments it provoked are still being had, and the system it was meant to replace is still in place.
Why Anyone Wanted It
By the mid-1990s, cross-border investment was governed by an unwieldy patchwork. Thousands of bilateral investment treaties (BITs) — individually negotiated, inconsistent in their terms, varying in what protections they offered and to whom — had accumulated country by country over several decades. A company investing in three markets might operate under three different sets of rules, with three different dispute mechanisms.
The OECD’s proposition was straightforward: replace the patchwork with one agreement. Its members would commit to common standards on how foreign investors were treated, and the resulting clarity would encourage capital to move more freely.
The draft rested on principles familiar from trade law. National treatment obliged governments to treat foreign investors no less favourably than domestic ones. Most-favoured-nation treatment meant any advantage granted to one country’s investors extended to all signatories. Restrictions on performance requirements would have barred governments from conditioning investment on local hiring, local sourcing or technology transfer. And investor-state dispute settlement would have let companies bring claims against governments directly, before international arbitration panels, rather than through national courts.
What Went Wrong
The negotiations were conducted among officials, with little public attention, until a draft text leaked in early 1997 and was posted online.
What followed is now studied as a case in its own right. Campaign groups across OECD countries — environmental organisations, trade unions, development NGOs, cultural lobbies — read the draft and organised against it using email lists and early websites, at a speed the negotiating process had not anticipated. For many participants it was the first transnational advocacy campaign conducted primarily over the internet.
The objections fell into several categories, and they were not all raised by the same people for the same reasons.
Regulatory sovereignty was the central one. Critics argued that ISDS would let foreign companies challenge environmental, health and labour regulations as expropriation of their investments, and that the mere threat of claims would discourage governments from legislating. Supporters replied that the provisions protected against genuine mistreatment and that well-drafted regulation had nothing to fear.
Cultural policy mattered particularly in France and Canada, both of which subsidise and protect domestic film, music and publishing. Neither was willing to accept an agreement that might treat those measures as discrimination against foreign investors.
Development concerns ran through the whole exercise. The MAI was being written by rich countries, among themselves, with the expectation that developing countries would accede later to rules they had no hand in drafting.
Labour and environmental standards were, critics said, addressed weakly or not at all, while investor protections were detailed and enforceable.
The decisive moment came in France. Prime Minister Lionel Jospin commissioned a review from Catherine Lalumière, a former government minister and MEP, whose report in September 1998 concluded that the agreement as drafted was incompatible with French and European interests. France withdrew from the negotiations in October. Without it, the talks had no path forward, and they were formally abandoned before the end of the year.
What Happened Instead
Nothing replaced the bilateral patchwork. There are now well over two thousand investment treaties in force worldwide, still individually negotiated, still inconsistent — the precise situation the MAI was meant to resolve.
The investor-state mechanism that drew the loudest objections in 1998 did not disappear either. It exists in most of those treaties and has been used extensively since, including in cases that bore out what critics had warned about: tobacco companies challenging plain-packaging laws, energy companies claiming against nuclear phase-outs and fossil fuel restrictions, mining firms disputing refused permits.
Those cases have pushed reform back onto the agenda, though along different lines. The European Union has proposed replacing ad hoc arbitration panels with a permanent investment court, with tenured judges and an appeals mechanism, and has written that model into agreements including CETA. The UN Commission on International Trade Law has run a multi-year working group on ISDS reform. The European Court of Justice ruled in 2018 that investment arbitration between EU member states was incompatible with EU law, and intra-EU treaties have been unwound since. Several countries have withdrawn from the Energy Charter Treaty entirely.
The Lasting Lesson
The MAI’s more durable legacy is procedural. Trade and investment negotiations after 1998 were conducted with an awareness that drafts leak, that civil society organisations can mobilise quickly across borders, and that an agreement which cannot survive public scrutiny probably cannot survive at all. The negotiations over the Transatlantic Trade and Investment Partnership fifteen years later ran into recognisably similar opposition, over recognisably similar provisions, and met a similar end.
Whether the MAI was a sensible framework defeated by misunderstanding or an overreach correctly stopped remains contested, and the answer tends to track what a person thinks about investor-state arbitration generally. What is not contested is that it failed, that nothing took its place, and that the questions it raised about the balance between investor protection and the right to regulate are no closer to settled now than they were in 1998.
Relevant article to read: Investment FintechZoom: Transforming the Landscape of Financial Services
Meta description:
The Multilateral Agreement on Investment collapsed in 1998 without a signature. What it proposed, why it failed, and what governs investment instead.
