Does the MetLife Decision Change the Way U.S. Regulators Should be Thinking About Systemic Risk?

by Cristie Ford

A bit of comparative analysis, like this, might help us think about this question.

Across the EU, the US and (at least until the impact of Brexit is felt) the UK, we see varying degrees of agreement about how to address systemic risk. Generally, what agreement there is comes down to developing new macro-prudential tools to prevent the build-up of systemic risk and its contagion. Where the regimes inevitably differ is in how these tools should be implemented. In particular, how should systemic riskiness be assessed? How should a whole new level of systemic risk regulatory authority be integrated with existing regulatory structures? To what degree can pre-existing institutions be expected to wield these new tools and assume these new responsibilities? Understanding systemic risk regulation requires looking at multiple, distinct national and regional regimes in which significant new regulatory powers have effectively been bolted onto existing regulatory architecture.

History and local experience play a role in each place too. For example, the notion of too-big-to-fail financial institutions heavily shaped the regulatory response in the United States toward thinking about institutions as the objects that should be regulated for systemic risk. This approach has come under pressure recently, after MetLife Inc. successfully argued that its designation as a systemically important financial institution, or SIFI, was invalid. The investment fund industry, as well, is lobbying hard these days to demonstrate that investment funds should not be objects of regulation; rather, the products they sell should be. This view is in keeping with a particular understanding that systemic risk resides in networks, markets, products and practices, rather than in institutions per se.

The need to describe systemic risk and to distinguish it from the normal kinds of risks that operate within capital markets is crucial in Canada because of constitutional questions related to our federal division of powers. In late 2011, the Supreme Court of Canada held that the Canadian provinces maintained their historical jurisdiction over the “day to day operations of the securities markets,” while the federal government had jurisdiction over “systemic risk” and “data collection”. Thus, the question of when a risk in the capital markets achieves systemic significance has not just financial consequences, but constitutional implications as well.

I was retained by the Department of Finance, Government of Canada to produce the report above (which is in English with a French language summary at the front) in response to a constitutional challenge by the province of Quebec to proposed federal legislation. The challenged legislation would put in place a systemic risk regulatory regime with oversight of Canadian capital markets, including those in Quebec. The report succinctly compares, with tables and diagrams, the systemic risk regulatory regimes that have been put in place since 2008 in the EU, the UK, the US, and Canada. It also compares existing Canadian provincial securities regulatory regimes with the proposed federal regime, and with international standards and agreements established by the G20, the Financial Stability Board and others.

For non-Canadians, apart from the report’s summaries and diagrams, the most interesting point may be that the proposed Canadian regime is structured to designate and regulate only benchmarks and products as systemically important, and practices as systemically risky. Institutions, no matter how large, and financial market infrastructure, no matter how essential, are not subject to the proposed systemic risk regime (except through the products in which they trade or the practices in which they engage). The Canadian approach is meant to cover the same ground as the institution-based approach in the US and elsewhere, but is more in line with the growing recognition that systemic risk operates primarily at the level of markets and products, not institutions. It also avoids the need to designate institutions as SIFIs, or to designate investment funds (as opposed to the products they sell) as systemically important. Given the challenges associated with an institution-focused determination of systemic risk, it is an approach that merits some further consideration.