Losing Control: Capital Controls and Trade Agreements

Kevin P. Gallagher

Do nations have the policy space to deploy capital controls to prevent and mitigate financial crises? In a new report for the Intergovernmental Group of Twenty-Four on International Monetary Affairs and Development (G-24) I examine the extent to which measures to mitigate the global financial crisis and prevent future crises are permissible under a variety of bi-lateral, regional, and multi-lateral trade and investment agreements. I find that the United States trade and investment agreements, and to a lesser extent the WTO, leave little room to maneuver when it comes to capital controls. This is the case despite the increasing economic evidence showing that capital controls can be useful in preventing or mitigating financial crises.

More specifically, I demonstrate that:

• Capital account liberalization is not associated with economic growth in developing countries;
• Capital controls can help developing countries maintain financial stability.  Indeed, those nations that deployed capital controls in the run-up to the global financial crisis were less hard hit during the crisis.
• The WTO allows for the use of capital controls for those nations that have not made market access commitments in cross-border trade in financial services or foreign investment in financial services.
• WTO members that have made commitments in financial services are not permitted to use capital controls, however untested safeguard mechanisms may apply to prevent and mitigate crises.
• US agreements do not permit restrictions on capital movements of any kind and have no apparent exceptions to this rule.  A handful of recent US agreements have a grace period however, which delays action on capital controls for a certain period of time.
• US agreements stand in stark contrast with the agreements of other capital exporters such as EU nations, Canada, Japan, and even China.  These nations either fully permit the use of capital controls or at minimum have safeguard mechanisms for crises.

These findings present real policy problems for at least three reasons:

• Trade and investment agreements hinder the ability of nations to deploy capital controls at a time when capital controls have been re-legitimized.
• The effectiveness of capital controls will be diluted given that fact that some agreements allow for controls and others do not—implicitly causing discrimination in the use of controls.
• Overlapping regimes that are inconsistent in regard to capital controls present a jurisdictional problem when controls are used.

To some extent the problems identified in this report can be alleviated by:

• Removing short-term debt obligations and portfolio investments from the list of investments covered in treaties. This has been raised as a possibility by parties ranging from the IMF to civil society.
• Creating ‘controlled entry’ Annexes in BITs and FTAs analogous to the Canada-Chile, Canada-Colombia, Chinese, and EU agreements with those nations. Controlled entry grants a nation the full ability to use capital controls on capital account transactions as a nations sees fit.
• Designing a balance-of-payments exception that covers both inflows and outflows such as the provisions found in the Japan-South Korea BIT.
• Resorting to a State-to-State dispute resolution process for claims related to financial crises, analogous to the WTO and the other chapters in most FTAs.

Such measures should be taken under consideration as nations consider new commitments under the GATS under the WTO and consider new FTAs and BITs, especially with the US.

Download the full report

Download the accompanying policy brief

Visit GDAE’s web page on capital controls, trade treaties, and development