Daniel K Tarullo: Shared responsibility for the regulation of international banks

Speech by Mr Daniel K Tarullo, Member of the Board of Governors of the Federal Reserve System, at the18th Annual International Banking Conference "The Future of Large and Internationally Active Banks", sponsored by the Federal Reserve Bank of Chicago and the World Bank, Chicago, Illinois, 5 November 2015.

The views expressed in this speech are those of the speaker and not the view of the BIS.

Central bank speech  | 
09 November 2015
PDF version
 |  10 pages

The regulation of financial institutions is necessarily a dynamic exercise. Growth or innovations in banking may create new risks that prompt regulatory change. The new requirements, in turn, incentivize or disincentivize certain actions by financial institutions, including shifts in activity that may start anew the process of regulatory response. The regulation of international banking reflects this general pattern, but because internationally active banks can quickly transmit financial problems across national boundaries, it also features the question of who should be doing the regulating in a dynamic financial environment.

Following the financial crisis, during which some internationally active banks posed special problems for both home and host countries, this issue has commanded attention reminiscent of the aftermaths of the Herstatt failure of the 1970s and the Bank of Commerce and Credit International (BCCI) failure of the early 1990s. Unlike those earlier instances, though, this renewed prominence of the "who does the regulating" question has accompanied a major reconsideration of what regulation is appropriate. Today I would like to address both the "who" and the "what" issues in the regulation of international banking. My theme is hardly an original one - namely, that in the absence of either a global regulator or completely insular national banking systems, we must continue to work toward a system of shared responsibilities to assure both home and host regulators that internationally active banks are subject to adequate oversight and controls.

I will begin by reviewing briefly the benefits and risks associated with international banking and then identifying the different models for allocating responsibility for the oversight of international banks. After noting the shortcomings of the system for regulating internationally active banks that prevailed before the financial crisis and developments in the intervening years, I will turn to a consideration of the challenges that remain, with a few suggestions on how we might make more progress.

Benefits and risks of international banking

It is worth at least briefly reminding ourselves of some of the more salient advantages and risks associated with cross-border banking as they help inform development of regulatory options.1

Among the potential advantages are facilitating productive capital flows, diversifying risks associated with growth in host countries, diversifying the earnings and thus the stability of the global bank, offering counter-cyclical lending through support from the parent when host country economic conditions constrain domestic banking operations, enhancing efficiency in financial intermediation in host countries, providing specialized financial services,2and providing price or product competition for banking services in host countries. The magnitude of these benefits obviously differs from country to country, depending on a variety of factors. Some of these benefits can be greater if foreign banks have the freedom to deploy capital and liquidity to whatever markets offer the most attractive opportunities, whereas others are achievable simply through the bank's expertise, existing business relationships, and range of services.

On the other side of the ledger are the risks associated with foreign banking that are distinct from risks associated with banking more generally. These risks tend to be related to the parent bank's capacity to support the larger organization. Reversals in the home market or other significant foreign operations may reduce the ability of the parent to support its foreign affiliates with needed capital and liquidity. Even if the foreign affiliate is not itself under great stress, the weakened overall condition of the parent because of problems in other parts of the world may prompt it to retrench - often rather abruptly - by reducing activity in foreign markets in which it is active. This response may be driven either by bank management itself or at the instance of home country officials who want the bank to continue to lend in its home market. Especially in countries where foreign banks account for a significant part of financial intermediation and where the underlying problems are not idiosyncratic to a specific bank, the result may be a significant diminution in intermediation beyond what would have taken place because of macroeconomic developments.

Even more serious is the risk that the foreign bank will fail, and that the home country will lack the resources or the will to ensure either that it is recapitalized and continues to function or that it fails in an orderly fashion. If the foreign operation has been thinly capitalized and is lacking in liquidity, host country officials may face an unpleasant choice between supporting a foreign bank (including operations for which they have not had consolidated supervisory responsibility) or allowing it to fail in a disorderly fashion, with potentially serious knock-on effects in the host country's financial system.

Thus, what might have been economic advantages for host countries from foreign banks in reasonably good times can turn into substantial disadvantages in periods of idiosyncratic or generalized stress. As with the benefits of foreign banking, the risks vary considerably among host countries. Obviously, countries without well-developed domestic banking systems will both benefit more and be at greater risk. Yet even the most sophisticated domestic financial systems can be affected significantly. For example, the risks can be exacerbated by funding patterns and currency mismatches, as happened in the United States during the financial crisis. Foreign banks that had been using their U.S. branches to raise dollars in short-term markets for lending around the world were suddenly left without access to this funding and, as a result, made substantial and - relative to their assets - disproportionate use of the Federal Reserve's discount window.

Approaches to the regulation of internationally active banks

As I mentioned at the outset, international banking raises the question of who should do the regulating, as well as the question of what regulation is appropriate. The two questions are related, of course. As I will explain shortly, the nature of regulations in part depends upon the perspective and aims of the regulator. There are essentially four models, each of which has benefits and shortcomings.

First, the home jurisdiction can have dominant or exclusive regulatory responsibility for all of its banks' global operations through application of consolidated regulation and supervision. Second, host jurisdictions can have dominant regulatory responsibility for all foreign banking operations within their borders. This approach requires foreign banks to charter locally and to meet the same regulatory and supervisory standards applicable to domestic banks. Third, there can be shared authority between home and host jurisdictions, whereby host countries do some regulating and supervising of foreign banks within their borders but do not require all foreign banking activities to be locally chartered and subject to regulation identical to that of home banks. Finally, there could be one global regulator to oversee all the operations of internationally active banks around the world.

There is an almost unlimited number of variations on the shared approach and, in fact, one or another variants on that approach have been adopted by most jurisdictions during the modern banking era (though there have been instances of countries severely limiting or prohibiting foreign banking altogether). Before turning to a discussion of how the shared approach has evolved and may be further modified, I think it useful to identify both the appeal and the problems with the other three, conceptually purer, approaches.

Both the attraction and limitations of the host country model are fairly apparent. On the one hand, the host country is most likely to be attentive to the risks posed to its financial system by foreign banks. More generally, the host jurisdiction is at least presumptively best positioned to craft a regulatory and supervisory framework to protect its financial system from the particular risks engendered by economic and financial conditions. Having all foreign banking operations meet local capital and other standards helps achieve that end. The risks of abrupt shifts of capital and liquidity out of the country can be minimized, and depositors can be better protected.

On the other hand, a fully local regulatory system would make the costs of entry very high. For example, if no foreign branches were allowed, or were required to operate as if they were separately chartered and capitalized, the commitment of resources needed to enter a foreign market would be considerably higher than those typically associated with opening a branch. In addition, even complete local subsidiarization might not protect a foreign banking operation from suffering some contagion if its parent is under stress. Thus, the quality of home country regulation may have some bearing even under the host country model.

Not surprisingly, the home country model presents essentially the obverse set of advantages and limitations. Having a consolidated set of capital requirements and a single supervisor allows for the quickest deployment of capital and liquidity where it is most in demand, or most needed to relieve stress, and minimizes compliance costs. However, as has often been pointed out, the home country regulator will be most responsive to the impact of both regulation and distress of its banks on its own market. In its regulatory and resolution activities, it is likely to undervalue the potential risks and costs for host countries. In periods of stress, the home country regulator, accountable primarily to home country legislators or government officials, may concentrate on stabilizing its own financial markets and be more inclined to allow, or even demand, a sharp reduction in activity abroad. The result would, at a minimum, be an abrupt decrease in intermediation at particularly sensitive times. At worst, foreign operations could default on obligations and exacerbate financial stress.3

At first glance, it might seem that the home country and global regulatory approaches would yield similar substantive results, since in each case, there would be consolidated regulation and supervision. A global regulator, however, would at least in theory take the interests of all jurisdictions into account in regulating, supervising, and resolving a global bank. Of course, how to balance those interests - particularly in the face of unanticipated circumstances - would be a difficult, and almost invariably political, judgment. This reality raises the thorny issue of the accountability of a global regulator.

The political factor is one of many reasons why jurisdictions are likely to remain unwilling to cede much authority to global, as opposed to international, financial institutions. Indeed, quite apart from political considerations, there may be good reasons not to do so. For one, a single global regulator of internationally active banks would presumably be something close to a regulatory monopolist, whose policies and practices could be inappropriately uniform across quite different national markets and slow to adapt to changing conditions. Also, as with dominant or exclusive reliance on home country consolidated supervision, it seems unlikely that a global regulator - no matter how well-staffed - would be fully informed on the varieties of financial risks posed to regulated institutions across national markets.

A limited exception to the general disinclination to cede financial sovereignty, as in various other areas, lies within the European Union or, more precisely, the euro zone. With the creation of a Single Supervisory Mechanism (SSM) in the European Central Bank and a freestanding Single Resolution Mechanism (SRM), there have been important transfers of authority, though national regulators continue to play a supporting role. Interesting and important as this regional initiative is, however, the unique European arrangement of shared sovereignty makes it less a model for the world as a whole than an extension of the single currency project, responding to some of the difficulties encountered during the financial crisis.

The shared model and lessons of the crisis

The shortcomings of each conceptually "pure" model explains why some version of a shared home/host model has prevailed over time. Given the range of variations in this model, however, it is useful to bear in mind the relative advantages and disadvantages of the cleaner models in choosing the elements of a specific shared approach. In considering recent developments, as well as what remains to be done, it is also useful to begin by recalling the situation that prevailed at the onset of the financial crisis.

In its early years, the Basel Committee on Banking Supervision's work focused on elaborating the responsibilities of home and host regulators of internationally active banks. The principle of consolidated supervision was developed in the early 1980s, and reinforced following the failure of the BCCI in the early 1990s, in an effort to ensure that some regulatory authority had an overview of a global bank's consolidated assets and liabilities. At the same time, though, the Basel Committee set out expectations for host country prudential oversight of foreign banks that would be similar to that for domestic banks.4

The financial crisis painfully demonstrated the inadequacy of both home and host country regulation. Home country regulators of some large, internationally active banks clearly did not appreciate the risks those firms were assuming overseas. Host country regulators, including those in the United States, had not exercised prudential oversight of some foreign bank activities and had not sufficiently appreciated the risks associated with the funding models and other activities of some foreign banks that were subject to consolidated prudential regulation. And there were indeed instances of international bank failures in which the home country authorities seemed to focus on domestic interests to the possible detriment of the interests of host countries.5

Of course, regulatory failures were far more pervasive than inattention to the specifically cross-border activities of banks. The substantive rules governing capital and other requirements for all banks were woefully inadequate, although the fact that most very large banks around the world have significant cross-border operations exacerbated the shortcomings. While banks were growing in size, integrating traditional lending and capital markets in ever more complicated ways and relying increasingly on vulnerable short-term wholesale funding models, many regulators around the world were at best failing to keep up with these changes. At worst, they removed older prudential limitations without substituting new measures designed to address the new realities of banking. The Basel Committee spent most of the decade before the crisis dominantly focused on the Basel II framework, which was intended to reduce somewhat effective regulatory capital levels for large banks in return for their transition to an internal-models-based approach to capital requirements. This was a choice made by national regulators, led by those in the United States, and not a byproduct of the structure of the Basel Committee itself.

Adding the lessons of the 2007-09 financial crisis to those of earlier episodes of financial stress, I think we can infer some guidelines on host and home responsibilities to help shape expectations for practice. For host countries, the overarching guideline is that each jurisdiction should take responsibility for protecting the financial stability of its own markets as its contribution to achieving global financial stability. The extent of this responsibility obviously increases with the size and significance of the jurisdiction's financial markets. Thus the United States and the United Kingdom - which currently have the greatest concentrations of capital markets activities - have a particular obligation to oversee the local activities of both domestic and foreign banks that could pose particular risks to financial stability and are likely to be especially difficult to observe for a home country supervisor less familiar with those markets.

A corollary of this general guideline is that the scope of host country regulation might sensibly vary with the size and systemic importance of foreign banks. This notion is consistent with the principle embodied in the Dodd-Frank Wall Street Reform and Consumer Protection Act that prudential regulation should be progressively more stringent as banks pose greater risks to financial stability. As I will discuss in a moment, this principle also lies behind some of the post-crisis frameworks developed by the Basel Committee and the Financial Stability Board (FSB).

For home countries, the relatively longstanding principle that regulators should exercise effective consolidated supervision remains critical, though I would emphasize that the regulation imposed by home jurisdictions is of equal - if not greater - importance. It is important to emphasize that this obligation is not a substitute for host country regulation and supervision of foreign banking organizations. The home jurisdiction regulatory structure must ensure that the banks are fundamentally safe and sound, and that the parent will generally be able to support its operations around the world. Here, of course, a major shortcoming of the pre-crisis regime becomes apparent, in that capital and other regulatory requirements for internationally active banks were simply not strong enough.

Consolidated supervision must contain the risks to the financial system created by banking activity that is not fully captured by regulations. It must also ensure that banks do not hide problems by shifting assets or liabilities around their global operations and, more generally, that the banks are fundamentally safe and sound so as to forestall possible contagion risk to foreign operations.

Within these admittedly broad guidelines, there is obviously room for host countries to balance the benefits and risks presented by foreign banking organizations in a number of different ways. Their choices will be affected by policy preferences, the characteristics of their domestic financial systems, and the relative importance of foreign banks in those systems. The host country choices will also inevitably be affected by how home countries are carrying out their regulatory and supervisory roles. This consideration includes, of course, not just regulations that are nominally applicable, but the manner in which those regulations are enforced.

Branching presents a particularly instructive example of the tension between assuring financial stability and permitting foreign banking operations that may carry economic benefits. Because branches are not separately chartered and capitalized, a bank can relatively easily enter a foreign market by opening a branch, through which it can make loans - often initially to companies from its home country - using funds from the home bank. Particularly where (as in the United States) foreign branches are forbidden by local law from accepting retail deposits, it might seem that there are minimal risks to the host country if the parent bank (and thus the local branch) fails. However, U.S. experience with foreign branches in the decade prior to the crisis shows the very real risks that can arise when a branch is used to raise funds in the host country (in the United States, in dollars) through short-term wholesale borrowing, and then directs those funds out of the host country for loans or asset purchases by other parts of the bank. As noted earlier, when short-term funding dried up, many foreign branches were left seriously short of liquidity and had to turn to the discount window.

A shared feature of the U.S. and European Union regulatory systems for foreign banks is that branching is permitted without requiring separate capitalization. Many other jurisdictions have similar policies. Thus opportunities for foreign bank entry and market access are provided. In the United States, larger branches do have to meet some liquidity requirements, though they are less restrictive than the standards applicable to domestic banks and intermediate holding companies. This requirement is an important example of a prudential measure that balances financial stability and the benefits of international banking. The degree to which we or, I presume, other jurisdictions will remain comfortable with this balance will depend on two factors. The first factor is the degree to which local branches are used by foreign banks as significant sources of unstable funding or for other risky purposes. While this has been a significant issue in the United States, it is less clear that other jurisdictions face similar risks. The second is the confidence host jurisdiction regulators have that the parent banks are subject to effective regulatory and supervisory oversight.

Developments since the crisis

The profound shift in political and policy environments as a result of the financial crisis has led to considerable strengthening of minimum international standards for internationally active banks, at both national and international levels. Basel III enhanced the quantity and quality of capital requirements and introduced, for the first time, quantitative liquidity standards. Following completion of Basel III, the Basel Committee developed a structure of slightly misleadingly named capital "surcharges," which requires global systemically important banks (GSIBs) to maintain higher capital levels.6

The Federal Reserve supported all these measures but was a particularly strong advocate of the capital surcharges, which established the new principle that some international prudential standards should be progressively more stringent as the systemic importance of a bank increases. In meeting its responsibility to promote domestic financial stability, the Federal Reserve last year followed the lead set by the European Union some years previously and adopted a regulation requiring subsidiaries of GSIBs engaged in traditional banking as well as those engaged in capital markets activities be covered by local capital requirements consistent with Basel III. But neither in the United States nor the European Union do the GSIB capital surcharges imposed at the consolidated level apply to foreign banking operations in their jurisdictions.7 So even if the global bank has local capital requirements for most or all of its foreign operations, the parent still has some flexibility as to where the additional capital buffer can be maintained. More generally, our requirements for other prudential regulations applicable to foreign banking organizations (FBOs) are calibrated to the relative importance of the FBOs in the U.S. financial system. Thus the structure of surcharges also help to create a good mechanism for balancing host country interests in assuring financial stability and in realizing the benefits that can come from global banking.

Even with good standards, regulators in host jurisdictions will want assurance that these standards are being rigorously implemented and enforced. The relative opaqueness of bank balance sheets makes capital, liquidity, and other common banking regulations difficult to monitor effectively. This argues for complementing fairly complex regulation that seeks to track the often-complex activities of large banks with simpler regulations, such as the leverage ratio and a standardized risk-weighted capital floor. But it also argues for existing international fora such as the Basel Committee and the FSB to provide effective monitoring mechanisms. Even with higher standards in place, supervisors in home and host jurisdictions will still face challenges in assessing cross-jurisdiction vulnerabilities. More regular sharing of information and assessments among home and key host jurisdictions both formally and informally should be high on our shared agenda.

At present, both those groups have useful processes for overseeing the implementation of agreed upon international standards. But they tend to be a bit formalistic, concentrating on comparing the language of domestic implementation to that of the international standards, rather than examining whether domestic practice in fact ensures substantive compliance or gaining a shared understanding of the unique risks in each market.

It would, I believe, be counterproductive to establish in either the Basel Committee or the FSB the kind of adversarial dispute settlement process associated, for example, with the World Trade Organization. It is in the interest of all members of those bodies to cooperate in the shared task of overseeing internationally active banks. Thus, the better approach to compliance would be one that simultaneously provides regulators with a way to work with one another and to gain deeper insight into how their counterparts in other jurisdictions are applying prudential standards.

For example, there has been considerable documentation of the interjurisdictional divergence in risk weights for similar exposures under the internal models-based capital approaches of Basel II.While the Basel Committee has been working on this issue,9 I suspect that one of the most effective ways of promoting broadly comparable risk weighting would be to have technically competent supervisory staff from other jurisdictions participate with home regulators in the actual bank model validations, oversight, and related supervisory functions. Similarly, as stress testing becomes a more important global financial stability tool, it will be useful to have staff experienced in stress testing at home participating in the stress testing exercises of other jurisdictions. These kinds of interactions, along with the more traditional device of supervisory colleges, can help foster confidence among host jurisdictions in both the regulatory and supervisory activities of home country authorities.

Another useful practice for furthering mutual confidence would be a program for regular contact among the very top officials of key regulators. The original Basel Committee brought together these officials for what were usually relatively informal meetings. As one of the early participants in those meetings once told me, the relationships he built with his counterparts through these regular contacts served everyone well when issues concerning international banks arose. But with the concentration of the Basel Committee on sometimes highly technical standards, participation has generally drifted down to the senior staff level. The FSB was created in part to compensate for this change in the Basel Committee. And the FSB usually does garner higher levels of participation. However, other features of the FSB - such as including market regulators and finance ministries in order to provide a broader range of views on financial stability issues - mean that the FSB cannot serve the original Basel Committee purpose either. Moreover, even when the right member agencies are represented, the actual individuals participating may not be the most senior officials in the supervisory function of those agencies. Finally, the near doubling in size of both the Basel Committee and the FSB, while again critical for ensuring a representative group to consider financial stability issues, further complicates the matter.

Thus, while a regular, high-level interaction among all key regulators would be optimal, for the foreseeable future, we will probably have to live with something less than optimal. Ad hoc meetings around the fringes of various Basel convocations and bilateral interactions may have to suffice. In this regard, I note the importance of the creation of the SSM within the European Central Bank as the supervisor for all larger banks in the euro zone. The Federal Reserve has already established instructive and productive relationships with the experienced and committed group of supervisors that have been brought in to lead the SSM.

The limits of shared responsibility

It is important to recognize, though, that even with the best of intentions and actions in home country regulatory and supervisory regimes, there will be limits to how much responsibility can appropriately be shared for international banking activities. These limits are most apparent in the context of the possible insolvency of a major foreign banking organization.10 The work of the FSB in promoting effective resolution regimes around the world and in seeking an international framework for building the total loss absorbing capacity (TLAC) of GSIBs are very good examples of cooperative efforts that promote the aim of ensuring that even the largest banks can fail without either causing financial disorder or requiring injection of public capital. My expectation is that the FSB's framework for TLAC will incorporate the principle of an extra buffer of loss absorbing capacity at the consolidated level beyond what may be required in the aggregate at local levels.11 But the margin may be a little thinner here, precisely because of the circumstances in which the loss absorption capacity may be needed.

With respect to going concern prudential requirements such as capital levels, host countries have a continuing opportunity to observe how home country officials are regulating and supervising their banks. Particularly if effective monitoring mechanisms are developed, host countries may become comfortable with limited oversight of some or all domestic operations of foreign banks. If they see the rigor of home consolidated oversight waning, they will have a chance to intensify their own supervision. But with the prospect of a failed bank, there will be no time for such adjustments or, as a practical matter, the capacity to impose new requirements may become limited by the time the prospect of failure is looming. The imposition of requirements in the midst of a crisis would in any event likely exacerbate stress. Even with the best of intentions, today's home country regulators cannot effectively bind their successors' response to the insolvency of one of their globally important banks when political and economic pressures are likely to be high.

The gone-concern loss absorbency requirement for FBO intermediate holding companies proposed by the Federal Reserve Board on October 30 should enhance the prospects for an orderly firmwide global resolution of an FBO by its home country resolution authority through increasing confidence that the U.S. operations of the FBO will obtain their appropriate share of the loss absorbency capacity of the consolidated foreign bank. Past experience suggests that host supervisors are most likely to ringfence assets when there is doubt that the local customers and counterparties of foreign banks will be adequately taken into account. Yet if, for any reason, the home jurisdiction resolution is unsuccessful, the internal long-term debt will be available to U.S. authorities for orderly resolution and recapitalization of the intermediate holding companies.

We have calibrated our proposed internal TLAC requirements slightly below our proposed external TLAC requirements for U.S. GSIBs. This slightly lower calibration for internal TLAC recalls the difference between local going-concern capital requirements and the GSIB surcharge, but the gap is somewhat smaller, reflecting the concerns I mentioned a moment ago. The proposal thus balances support for the preferred resolution strategy of the home resolution authority of the foreign GSIB with assurance of U.S. financial stability if that strategy cannot be executed successfully.12


My view of shared responsibility for overseeing international banks emphasizes the importance of financial stability even as it allows for benefits specific to international banking. For the reasons I have explained, in the end host countries need to make the judgments on the tradeoffs between these goals. But I have also explained how a strong set of international prudential standards and good institutional relationships among regulators could help tilt this balance toward greater flexibility for internationally active banks.

In response to positions akin to what I have presented today, one often hears complaints that the emphasis on financial stability will result in the balkanization of international banking. I would note first that it is not at all clear that developments since the crisis have on net balkanized banking, so much as shifted some international banking assets from the Organisation for Economic Co-operation and Development (OECD) countries whose banks were disproportionately affected by the financial crisis to banks from some emerging market and developing countries.13 This development probably reflects both needed changes in some of the OECD nation banks and a logical reflection of the increasing economic importance of the non-OECD countries.

Second, I wonder how these critics can think that the pre-crisis situation of supposedly consolidated oversight and substantial bank flexibility was a desirable one. At least some of the flexibility enjoyed by banks in shifting capital and liquidity around the globe was deployed in pursuit of unsustainable activity that eventually ran badly aground.

Third, as I suggested earlier, even where concerns about "trapped" capital or liquidity are more sensibly based, reasonable ex ante constraints by host country authorities in pursuit of a sound and stable domestic financial system are likely to be far preferable to ex post constraints - for example, ringfencing - that are imposed when the foreign bank is under the greatest pressure.

1  By "cross-border banking," I mean to refer broadly to activities carried on outside of a bank's home country through subsidiaries and branches. I do not include direct lending or other financial transactions across national borders, whether or not facilitated by an agency in the country of the bank's customer or counterparty. While such activities can raise concerns pertaining to investor protection, the volatility of financial flows, or foreign exchange policies, relevant prudential considerations will generally be limited to the country(ies) from which the bank is initiating the transaction and raising any funds needed to fund it.

For references to many individual studies examining one or more the advantages or risks, see Stijn Claessens and Neeltje van Horen (2014), "The Impact of the Global Financial Crisis on Banking Globalization," IMF Working Papers 14/197 (Washington: International Monetary Fund, October).

2  For an example, see Stijn Claessens, Omar Hassib, and Neeltje van Horen (2015), "The Role of Foreign Banks in Trade," mimeo, (Washington: Board of Governors of the Federal Reserve System and De Nederlandsche Bank, March).

3  Mindful of the considerations lying behind the limitations of both models, Dirk Schoenmaker has offered his theory of the "financial trilemma," which states that a jurisdiction can only have two of the three objectives of a stable financial system, international banking, and national regulatory policies. Professor Schoenmaker introduced his theory at a conference in 2008 and subsequently formalized it in Dirk Schoenmaker (2011), "The Financial Trilemma," Economics Letters, vol. 111 (April), pp. 57-59; and Dirk Schoenmaker (2013), Governance of International Banking: The Financial Trilemma (Oxford: Oxford University Press). See also Richard J. Herring (2007), "Conflicts between Home and Host Country Prudential Supervisors," in Douglas D. Evanoff, George G. Kaufman, and John R. LaBrosse, eds., International Financial Stability: Global Banking and National Regulation, (Hackensack, N.J: World Scientific), pp. 201-20.

4  The current version of this obligation is set forth in Basel Committee on Banking Supervision (2012), "Core Principles for Effective Banking Supervision" (Basel, Switzerland: Bank for International Settlements, September). I have addressed the issue of host state responsibility at somewhat greater length elsewhere. Daniel K. Tarullo (2014), "Regulating Large Foreign Banking Organizations," speech delivered at the Harvard Law School Symposium on Building the Financial System of the Twenty-first Century, Armonk, New York, March 27.

5  For a review of these instances, see Schoenmaker, Governance of International Banking, pp. 72-87.

6  I say "misleadingly" called capital surcharges because that term implies the banks are faced with a "charge" that has to be paid to someone. In fact, of course, the requirement is that the bank retain higher capital buffers in order to increase its resiliency.

See 79 FR 17240 (March 27, 2014); and Tarullo, "Regulating Large Foreign Banking Organizations."

8  See generally Vanessa Le Leslé and Sofiya Avramova (2012), "Revisiting Risk-Weighted Assets," IMF Working Paper 12/90 (Washington: International Monetary Fund, March).

9  See Stefan Ingves (2014), "Finishing the Job: Next Steps for the Basel Committee," keynote address to the Ninth BCBS-FSI High-Level Meeting on "Strengthening financial sector supervision and current regulatory priorities," Cape Town, South Africa, January 30.

10  Indeed, Dirk Schoenmaker developed his notion of the financial trilemma around the conflicts of interest that arise in the context of the insolvency of a global bank.

11  The TLAC proposal of the FSB contemplates that host authorities will set internal TLAC requirements at 75 to 90 percent of applicable external TLAC requirements.

12  Our internal TLAC proposal effects this balance principally by not including parent GSIB surcharges in the calibration for FBO intermediate holding companies but also by lowering the baseline TLAC requirement for U.S. intermediate holding company subsidiaries of single-point-of-entry strategy FBOs from 18 percent of risk-weighted assets to 16 percent of risk-weighted assets.

13  For a discussion of these points, see Claessens and van Horen, "The Impact of the Global Financial Crisis on Banking Globalization."