Investor-State Arbitration: What Governments and Investors Get Wrong

Investor-state arbitration is different from commercial arbitration in ways that matter.

Most of the mistakes made in it come from treating it as if it were not.

The differences begin with the treaty. In commercial arbitration, the parties chose each other, negotiated their contract, and agreed to arbitrate. In investor-state arbitration, the state made a standing offer to arbitrate in a treaty — often years or decades before the dispute arose — and the investor accepted that offer by commencing proceedings. The relationship between the parties was never consensual in the ordinary sense. That shapes everything.

It shapes the applicable law. Treaty standards — fair and equitable treatment, full protection and security, indirect expropriation, the umbrella clause — are not contract terms. They are public international law obligations, interpreted by reference to the treaty's object and purpose, the negotiating history, and a body of arbitral jurisprudence that is extensive, inconsistent, and not binding on any tribunal.

It shapes the arbitrator pool. Investor-state arbitration draws from a smaller, more specialized group of arbitrators than commercial arbitration. The same names appear repeatedly. Their published awards are the primary source of guidance on what the treaty standards mean. Knowing how a particular arbitrator has interpreted fair and equitable treatment — not in the abstract, but in cases that resemble yours — is not optional due diligence. It is the due diligence.

It shapes the political dimension. Governments are not ordinary parties. They have legislatures, elections, and constituencies. A measure that looks like an expropriation from the investor's perspective may look like legitimate regulation from the state's. The tribunal will have to decide which characterization is correct. That decision will be made by arbitrators who are also human beings, operating in a world where the legitimacy of investor-state arbitration is itself contested.

What do investors get wrong?

The most common mistake is overestimating the strength of the treaty claim and underestimating the state's regulatory defense. Investors who have suffered a genuine loss — a cancelled concession, a retroactive tax, a regulatory reversal — often arrive at arbitration with a strong sense of grievance and a weaker case than they believe. The treaty standards are not a guarantee against adverse government action. They are a guarantee against a specific category of it. The line between the two is where most cases are actually decided.

The second mistake is damages optimism. ICSID and UNCITRAL tribunals have become more rigorous about damages methodology over the past decade. DCF models that would have been accepted without serious scrutiny fifteen years ago are now challenged hard. The assumptions buried in the valuation — the discount rate, the terminal value, the counterfactual revenue projections — will be tested. Investors who have not stress-tested their own damages model before the hearing often discover its vulnerabilities at the worst possible moment.

What do governments get wrong?

The most common mistake is underestimating the case until it is too late to settle it cheaply. States often treat an investor's notice of arbitration as a political problem rather than a legal one. By the time the legal team is properly resourced and the strategy is coherent, the procedural calendar has already been set and the early settlement window has closed.

The second mistake is assuming that sovereign immunity from enforcement provides meaningful protection. It does not, in most cases. An ICSID award is enforceable in the domestic courts of all member states. A New York Convention award is enforceable in over 170 countries. A state that loses and refuses to pay will find its commercial assets — bank accounts, aircraft, receivables — subject to attachment in jurisdictions it cannot control.

Both sides share one mistake.

They wait too long to get an independent view of the case.

In investor-state arbitration, the stakes are typically large, the proceedings are long, and the costs are substantial. The decision to commence, to defend, to settle, or to walk away is made under conditions of significant uncertainty — about the applicable standard, about the tribunal's likely reasoning, about the damages range, about the enforcement landscape.

An independent assessment does not eliminate that uncertainty.

It prices it.