“We have listened to the wisdom of an old Russian maxim, doveryai, no proveryai—trust, but verify.” President Ronald Reagan at the signing of the INF Treaty, December 8, 1987.
In July 2010, central bank governors and supervisors from the 28 jurisdictions that make up the Basel Committee membership were hammering out the agreement on new capital and liquidity requirements now known as Basel III. There was a large sticking point. Some members were standing firm on their desire to have higher capital requirements. Others felt that this would make credit more expensive and less plentiful.
Had agreement not been reached, those insisting on more capital might have said: “Go ahead, be permissive. But if you let your banks operate with low levels of capital, we’ll restrict our banks from doing business with them.” Fortunately, it didn’t come to that. Today, “global systemically important banks” (G-SIBs) are required to hold capital equal to at least 10 percent of risk-weighted assets (RWA), with Total Loss Absorbing Capacity (TLAC) of 16 to 20 percent of RWA.
While we are clearly on the side of those who would like higher standards (see here), the point is that the national authorities were able to reach an agreement at all. Like other international agreements on issues ranging from arms control to the environment, this required a level of mutual trust, backed by some ability to monitor (verify) performance. This is a good thing, because without a common prudential standard, the global financial system would balkanize and fragment, diminishing both global growth prospects and opportunities for risk-sharing among residents of different countries.
Unfortunately, the danger of financial fragmentation persists. As argued in a recent paper one us wrote with Paul Tucker, common standards for which international compliance cannot be verified will be insufficient over time to retain the benefits of global finance. Doing so also requires international cooperation and coordination beyond what policymakers currently envision in overseeing the system’s safety and soundness. This is a daunting task.
To understand this conclusion, we need to step back and explain the nature of the challenge.
Modern banks and financial markets are large in size, broad in scope, and global in reach. The 30 largest banks in the world currently have combined assets in excess of $50 trillion, with half of them operating in at least 40 countries. The Chicago-based CME Group, the largest trading platform in the world today, clears about three billion trades annually with a notional value of $1,000 trillion (that’s one quadrillion dollars) in a combination of cash, futures and options in interest rates, equity indexes, foreign exchange, commodities, weather, and real estate. The CME lists products in 18 currencies and has 72 clearing members from all over the world.
The resulting cross-border linkages mean that stresses in one country’s financial system can and do quickly spread to others. Nearly catastrophic examples are easy to find. The most recent is the exposure of European banks to the U.S. dollar assets, especially securities backed by subprime mortgages, that triggered the global crisis of 2007-2009. BIS economists estimate that by 2007 continental European banks had taken dollar positions in excess of US$ 1 trillion that relied on short-term funding. When interbank lending dried up, the survival of these institutions (which lacked direct access to Federal Reserve loans) was at stake. To prevent a complete global meltdown, the Federal Reserve lent upwards of half a trillion dollars in the form of dollar liquidity swaps to foreign central banks, who then lent the funds on to their commercial banks.
Now, it is typical to think of financial stability—nationally and globally—as a public good (like price stability and national defense) that is available to all and which no one can deplete. This public goods aspect of financial stability is an externality that justifies intervention by national governments, but implies limited need for cross-border coordination (other than burden sharing).
But following Tucker, we find it useful to think about financial stability as a problem of the commons. That is, financial stability is a depletable resource that is available to all. From this perspective, the externality is different from that of a standard public good. The classic examples are grazing on public lands or fishing in the open seas. In these cases, individuals have the incentive to do things that degrade the resource for all. Just as an individual farmer has the motivation to over graze his or her cows, diminishing the shared pasture, an intermediary in the financial system has an incentive to behave in ways that deplete its resilience and put others at risk.
In such a global financial system, neither financial resilience nor instability are circumscribed by geographic borders. This truth is already well understood at the national level: the stability of the Ohio and Pennsylvania financial systems is not thought of as distinct from the stability of those in California and Oregon. But, where financial institutions and markets operate globally, national borders do not shield a local financial system from global developments. Indeed, in the European Union, the principle of the single market means that a bank with authorization to operate in one country may also provide services in any of the other 27 member states.
Just as a financial firm inside a country can create instability locally, a large institution (or a host of smaller ones) can create instability globally. This can happen in a variety of ways. Cross-border lending can weaken the balance sheets of borrowers in other countries. A financial institution operating in a large number of countries can get into trouble, threatening counterparties wherever it is present. Or, a bank could have exposures and business lines that are similar to banks in other countries, so that when it hits difficulty, customers and counterparties worry about the entire class of like institutions. In all of these cases, real or imagined problems can spread rapidly without concern for political boundaries.
This ability of one country’s institutions to act in ways that weaken the global financial stability commons means that, in a world with large cross-border firms and global markets, authorities in one country must care deeply about the quality of regulation and supervision elsewhere.
But the assessment of regulatory quality does not stop at the evaluation of the standards and written rules. It is also about their enforcement. Each country’s financial stability authority faces the possibility that parts of the financial system in both their own jurisdiction and in others will find ways around a regime of common standards. If so, the resilience of both the domestic and global system will suffer.
Moreover, cross-border counterparty relationships make it impossible for supervisors in a given jurisdiction to assess the stability and resilience of their banks, or their banking system, without knowing what is going on elsewhere. And, given the nature of modern supervision, where authorities with access to confidential firm-level information watch institutions at close range, it is impossible to see how this can be done effectively without extensive information sharing beyond borders. This constitutes joint surveillance and cooperation at a level that is not currently envisioned.
The need for policy coordination becomes clearer when considering the dynamic responses of regulators to changing levels of systemic risk. When conditions change, maintaining financial stability may require making changes to capital requirements, adjusting risk weights for exposures to particular sectors, altering stress tests, and the like. These are part of the developing “macroprudential” policy toolkit in many countries. Yet, adjustments in one jurisdiction will inevitably have implications for stability and circumstances elsewhere.
Consider, for example the simple case in which authorities in one jurisdiction become concerned about a property price boom. They might choose to raise the regulatory capital risk weight assigned to loans collateralized by real estate. To ensure global systemic stability, it would seem natural for authorities elsewhere to follow suit, raising the risk weight on real estate-related loans made into the initial jurisdiction. If so, then the policy requires more than simple cooperation and information sharing; it requires coordination.
If we are to retain most of the benefits of a global, integrated financial system we see two long-run scenarios. The first is to dramatically expand the international arrangements for preserving financial stability. This would mean that supervisors will need to share confidential information, and policymakers will need to coordinate their actions.
Alternatively, we can envision a system in which most national authorities implement easily verifiable standards that lead to substantially greater resilience than those currently in effect. That is, these jurisdictions implement a system based on regulation by function, where all banking activities (regardless of whether they are being conducted by banks or nonbanks) face much higher capital and liquidity requirements, where virtually all derivatives are centrally cleared, and where automatic mechanisms compel disclosure of innovations that otherwise would allow institutions and investors to evade the regulatory perimeter. The point is that the standards must be both sufficiently high and easily verifiable at low cost to limit any incentive for regulators to conceal the fragility of their home-country institutions. At the same time, intermediaries in these “high-standards jurisdictions” would face restrictions on transactions with firms in “regulatory havens.” The long-run balance between those jurisdictions that are included and those that are not will depend on the relative costs and benefits of international integration.
But, if neither of these comes to pass—say, due to the cost of verification—then the distrust that was evident when Basel III was initially constructed could easily become the norm. Open hostilities among regulators would then lead to a balkanized system where institutions and markets operate primarily inside national boundaries and at least some, if not most, of the benefits of cross-border financial activity would be lost.
Were it not for technological advances that continue to lower the costs of cross-border services, the challenges associated with policy coordination and verification would make the prospects for enhancing global finance rather dim. Even retaining the status quo would be difficult, with large cross-border providers facing heightened costs from the recent wave of local (uncoordinated and potentially inconsistent) national regulatory changes. Perhaps the best we can hope for from regulators is that those who prefer a high-standards regime coordinate their efforts with regard to those aspects of global finance—such as the cross-border payments system—that are most important for sustaining economic growth.
(Note: This post is based extensively on Cecchetti and Tucker, "Is there macroprudential policy without international cooperation?" CEPR Discussion Paper 11042, January 2016. We thank Paul Tucker for his comments and dicussion, but the conclusions here are our own.)