IMF: Trade obligations may work against financial stability goals (December 2012)

Rethinking Bretton Woods | Thu, Dec 27, 2012

By Aldo Caliari

An article comments on the IMF's institutional view on management of capital flows.

According to a recent paper by the International Monetary Fund, the obligations of countries under trade and investment agreements may reduce the policy space for countries to implement capital controls necessary to preserve financial and macroeconomic stability. “[M]ost of the current bilateral and regional agreements addressing capital flow liberalization do not take into account macroeconomic and financial stability,” the IMF study finds.[1]

The assertion, coming from an institution that in the late 1990s came close to embedding capital account liberalization as one of its mandates, is an important sign of the state of mainstream knowledge about this aspect of the deregulatory paradigm prevailing in the decades before the crisis. It is bound to lend substantial new impetus to civil society claims that treaties on financial services trade negotiated under that paradigm are in need of deep review, and to its demands that in the meantime such obligations should be suspended to avoid further harm.[2]

More importantly, the paper espousing these findings is part of an IMF’s “institutional view” issued on December 2012,[3] (hereinafter “institutional view”) which means that they have now Board approval. The process leading to the institutional view took almost three years. In 2010, a staff paper reflected positively on the experience of countries that had implemented capital controls to face sudden surges of capital and influence the composition of their external liabilities.[4] The paper ventured the view that capital controls should be part of the policy toolkit to achieve financial stability, even if with many qualifications as to the tools that should be tried before and even concurrently.

Several papers followed – in the process the IMF adopted the language of “capital flow management measures” to refer to capital controls - but the contours of that initial staff contribution can still be recognized in what is now the institutional view.

Some of the limitations that were criticized of that paper can certainly still be criticized in the institutional view, namely, the reluctance to admit a wider role for capital controls. The measures are, in the IMF’s study, embraced only after a number of qualifications and tests that are inconsistent with the weakness of the evidence in favor of capital account liberalization,[5] when not plainly impracticable.[6]

The 2010 staff paper’s expressed concerns about the multilateral repercussions of countries implementing capital controls are, also, quite present in the institutional view. The latter advocates cross-border coordination among recipient countries and between source and recipient countries to “mitigate undesirable spillover effects.”[7] The potential for multilateral spillovers and instability resulting from members’ unilateral and uncoordinated action play a big part in the IMF’s argument why the institutional view should play an important role in global management of capital flows to promote macroeconomic and financial system stability. In the IMF’s words “This dialogue could eventually contribute to reducing the potential volatility and distortions that could result from the current complex patchwork of bilateral, regional and multilateral agreements.”[8]

Granted, nobody would want to trust the fate of global financial stability to the unwieldy juxtaposition of such a plethora of bilateral, regional and multilateral agreements. But this does not mean the objective of stability is better assured by any framework that is global and “consistent.”

While the threat of instability generated by macroeconomic spillovers is quite real, it is equally true that the size of the threat posed by different countries, and the size of the impact countries stand to suffer relative to their economies, can vary considerably.

Then the question becomes: will a “consistent” approach to capital flows management adopted globally restrain the countries that stand to cause the most damage with their unilateral action? Will it protect those that stand to suffer greatest damage? It is hard to imagine how or why it would.

Another precedent to the IMF’s institutional view regarding the clash between obligations in trade agreements and capital controls, with special reference to the GATS agreement on financial services, is in a reference note issued in 2011. That note said: “While capital controls may be considered a legitimate part of the toolkit to manage capital inflows in certain circumstances, they may, in some cases, be inconsistent with GATS obligations.”[9]

The institutional view has a wider coverage, referring in a more general way to the legal obligations on capital flows that member countries have assumed under international agreements:

“As the nature and scope of the Fund‘s proposed institutional view may differ from those of other agreements, there may be circumstances where differences arise. For example, there are likely to be cases in which other (particularly bilateral and regional) agreements establish liberalization obligations that are broader and more accelerated than recommended under the integrated approach, or where obligations to avoid [Capital Flow Management Measures or CFMs] are unqualified in a manner that is not compatible with the policy space for both inflow and outflow CFMs that is recommended under the proposed institutional view. Similarly, the proposed view recognizes that there are circumstances in which residency-based CFMs, although generally much less preferred than non-residency based measures, could nonetheless be maintained, but such maintenance could be at odds with the national treatment provisions under many international agreements.”[10]

The Fund is careful, however, to recognize that the institutional view does not alter members’ rights and obligations under such other international agreements. For example, the Fund says that “even where the proposed Fund institutional view recognizes the use of inflow or outflow CFMs as an appropriate policy response, these measures could still violate a member‘s obligations under other international agreements if those agreements do not have temporary safeguard provisions compatible with the Fund‘s approach.”[11]

While this, alone, should be a cause for alarm about the impacts of trade and investment agreements on financial stability, so should the IMF’s stated ambition that its institutional view play a “vital role” in promoting a more consistent approach to the design of policy space for CFMs under bilateral and regional agreements.[12]

The IMF proposes that “members drafting such agreements in the future . . . could take into account this view in designing the circumstances under which both inflows and outflows CFMs may be imposed within the scope of their agreements.”[13]

There is no contradiction between recognizing the beneficial step forward taken by the IMF and admitting that it is far from a U-turn. Being the result of a consensus among all its members, it will probably need to continue to be that far from a U-turn in the foreseeable future.

Given the substantive limitations that, as noted above, remain in the IMF’s approach to capital controls, using the institutional view would prove an unduly restrictive drafting guide for determining when capital controls are allowed. This effect could be seriously compounded should the institutional view requirements become matters subject to verification by a dispute settlement or, worse, an investment tribunal. Fears of extended and costly litigation or damaging trade consequences would become the main concern for officials in charge of finance, rather than the measures’ pros and cons for financial and macroeconomic stability.

Moreover, there is a risk that the IMF’s role as an arbiter on the validity of certain capital measures could become entrenched – for instance, giving the IMF’s opinion a binding role, such as is now granted to the IMF’s judgment on certain aspects invoked as a justification for Balance of Payment restrictions in GATT-WTO.

Ironically, that would end up giving the IMF more sway on members’ capital accounts than conceivable in the wildest dreams of those supporting the reforms to the IMF Articles of Agreement rejected in the late 1990s – and now widely seen as a bad idea.

After all, the proposed reforms of the 1990s would have been only enforceable through the IMF’s lending and surveillance mechanisms, not (additionally) the sort of sanctions trade or investment panels are able to apply today.

[1] IMF 2012. The Liberalization and Management of Capital Flows: An Institutional View. November 14.

[2] See IWGTF Steering Committee 2009. Submission to Commission of Experts of the President of the UN General Assembly on the Reforms of the International Monetary and Financial System, available at; IWGTF Steering Committee 2012. Statement in the occasion of UNCTAD Trade and Development Board 59th Session, September, available at

[3] IMF 2012.

[4] Ostry, Jonathan et al 2010. Capital Inflows: The Role of Controls. February 19. IMF Staff Position Note.

[5] Gallagher, Kevin 2012. The IMF’s New View on Financial Globalization: A Critical Assessment, available at

[6] Frenkel, Roberto 2012. Managing Capital Inflows in Emerging Markets. G-24 Policy Brief No. 75.

[7] IMF 2012.

[8] IMF 2012.

[9] For further assessment of this note see Caliari, Aldo 2011. IMF issues reference note on trade in financial services, at

[10] IMF 2012.

[11] Ib.

[12] Ib.

[13] Ib.