The COVID-19 pandemic is the largest public health disaster the world has experienced in a century. Its economic fallout will exceed that of the Global Financial Crisis of 2008. The fallout has exposed weaknesses in national economies and international supply chains.

As a result, we may well witness a retreat from globalisation. Since the most globalised sector in most national economies is finance, any substantial retreat from financial globalisation may well involve controls on capital movements. Capital controls have an interesting history. Accepted orthodoxy in the 1950s and 1960s, they became deeply unpopular by the 1990s, and remain controversial despite their increasing acceptance in mainstream economics.

Capital controls have costs. They restrict access to foreign capital for investment, increase real interest rates and require considerable administration. However, the experience of capital controls in emerging markets—such as Chile and Malaysia—suggests they are a useful policy response for addressing volatile short-term capital flows, particularly during periods of financial crisis.

Capital controls have been used most extensively as a long-term policy measure by China. China’s approach toward capital controls has undergone several distinct phases.

In Phase 1 (1979-1993), capital controls were adopted to implement an ‘easy in and difficult out’ regime to avoid a balance of payments crisis. While restrictions on foreign direct investment (FDI) were gradually relaxed, controls on non-FDI inflows were strictly enforced and some transactions entirely prohibited.

In Phase 2 (1994-2000), China underwent major market-oriented exchange reforms. This included abolishing its foreign exchange retention system and allowing domestic enterprises to buy and sell foreign exchange through designated foreign exchange banks. China also dramatically restructured its banking system and strengthened controls after the 1997 Asian Financial Crisis.

In Phase 3 (2001-2012), China’s entry into the World Trade Organization caused its trade surplus to surge. Speculative capital inflows greatly increased, putting upward pressure on the renminbi (RMB). Accordingly, China’s primary focus was controlling capital inflows. The policy was characterised as ‘difficult in and easy out’. As overseas investment by Chinese enterprises also became less restricted, the monitoring of capital inflows was simultaneously strengthened.

Since President Xi Jinping took office, China has established a more balanced approach to its capital control regime. In October 2019, China introduced several measures to loosen regulatory controls over foreign exchange income payments in respect of both current and capital account cross-border transactions. Portfolio flows that had been restricted and subject to limits on amounts were encouraged, although the State Administration of Foreign Exchange (SAFE) retains its power to impose temporary restrictions on repatriation when required.

China still has a long way to go before it fully liberalizes capital flows and makes the RMB fully convertible, meaning a multitude of capital controls will remain into the foreseeable future.

The use of capital controls raises questions about state compliance with international commitments. While economists have long debated the effectiveness of capital controls, their legality has been virtually ignored in the literature.

In a recent paper we address this gap in the literature.

The absence of a single international law regime governing capital controls means controls are regulated by an intersecting web of monetary, trade and investment law. Accordingly, the legality of a specific control measure depends on the language in a country’s international treaties and in its commitments to international organisations.

While the IMF rules and GATS can be viewed as floors which provide some policy space for members to adopt restrictions on cross-border capital flows, over 3,000 bilateral investment treaties (BITs) and free trade agreements (FTAs) regulate with typically stronger treaty language and higher-level commitments.

For instance, China has entered many FTAs since its accession to the WTO. However, due to different treaty language and commitments made under different FTAs, some Chinese capital control measures may be consistent with some agreements and be inconsistent with commitments under others.

To illustrate, one Chinese control measure prohibits transferring money overseas for the purpose of purchasing life insurance and investment-type insurance. While China made no GATS commitments for these sectors, it did make market access and national treatment commitments in FTAs with both Singapore and Korea on the cross-border supply of  insurance and insurance-related services including: (a) life, health and pension/annuities insurance; and (b) non-life insurance. Obviously, both life insurance and investment-type insurance (pension/annuities insurance) are included in China’s FTA commitments. Thus, China’s prohibition on residents purchasing life and investment-type insurance from non-residents appears inconsistent with its commitments to allow the cross-border supply of such services.

Comparatively, although China did not initially include insurance and insurance-related services in its financial services commitments in the ASEAN-China FTA in 2007, this recently changed under the 2nd package of Chinese commitments. Therefore, the control measure on purchasing foreign insurance can no longer be maintained.

Another issue is the potential for regime conflict in the multilateral governance of capital flows, with the scenario of the IMF recommending the imposition of restrictions on the capital account notwithstanding such measures being inconsistent with a member’s specific commitments under the GATS and/or FTAs and BITs. The uncertainty in the interpretation of various exception clauses in many trade and investment agreements heightens this tension.

In conclusion, it is difficult to definitively determine whether specific controls are consistent with the international legal framework given the patchwork of varying agreements with differing treaty language and levels of commitments. Governments must understand the potential risks of locking in commitments which prevent prudent fiscal and monetary measures being implemented (particularly in times of financial crisis) and negotiate treaties with appropriate language, exceptions and safeguards to ensure capital controls are consistent with those obligations. In the meantime, countries wishing to use some form of capital control to retreat from the globalised financial system need to review the language of their existing treaty and other commitments very carefully.

Bryan Mercurio is the Simon F.S. Li Professor of Law at the Chinese University of Hong Kong - Faculty of Law.

Ross P Buckley is Scientia Professor, and the KPMG Law – King & Wood Mallesons Professor of Disruptive Innovation and Law at UNSW Sydney.

Erin Jiangyuan Fu is Associate Professor at Huazhong University of Science and Technology - School of Law.