Speechby Mr Fernando Restoy, Chairman, Financial Stability Institute, Bank for International Settlements, at the IIF Market fragmentation roundtable, Washington DC, United States, 10 April 2019.
Introduction1
One of the key features of global financial market integration is that, wherever they happen to be established, entities are able to offer financial services in other jurisdictions on terms similar to those enjoyed by domestic market participants. The degree of integration could be measured in terms of, for instance, the intensity of cross-border financial flows or the market quota of foreign entities in domestic markets.
Market integration provides a number of social benefits, including broadening the range of financial services and investment opportunities available to consumers and increasing competition in the provision of those services. In addition, integrated financial markets act as private risk-sharing mechanisms that facilitate the smoothing of both economic and financial cycles in domestic economies. Moreover, market integration enables greater risk diversification, thereby contributing to more effective risk management and to financial stability.
Regulation certainly plays a highly relevant role in facilitating market integration. In particular, the hom*ogeneity of financial regulation across jurisdictions and the consistency of the requirements imposed on internationally active entities may provide powerful incentives for cross-border financial activities and operations. By the same token, heterogeneous rules or any type of regulatory discrimination against foreign players in domestic markets tend to inhibit the internationalisation of financial activity.
Regulation: cause or consequence of fragmentation
Yet market integration is by no means the only, let alone the most relevant, policy objective. Indeed, fostering international market integration is not the primary goal of financial regulation. As it is, standard regulatory mandates, such as pursuing financial stability or ensuring consumer protection in a specific jurisdiction, may occasionally conflict with achieving international regulatory consistency.
Insofar as domestic economic or market conditions relevant to the achievement of specific policy goals differ across jurisdictions, regulation may need to be adapted to those conditions, even at the cost of generating regulatory discrepancies across jurisdictions or additional costs for internationally active entities.
In other words, regulatory heterogeneity is at times more the consequence than the cause of the specificities prevailing in different jurisdictions.
As an example, if the failure of a foreign bank's subsidiary generates a systemic impact in the host jurisdiction - without necessarily affecting the viability of the group as a whole - there is a rationale for imposing specific requirements on the local subsidiary, unless the parent company is firmly and credibly committed to supporting its subsidiaries in case of need. This is, of course, the rationale behind different forms of ring-fencing.
Another example is the tailoring of prudential requirements for non-internationally active banks, in application of the principle of proportionality. Depending on the complexity of, or the business model used by, those institutions, some adjustments may be warranted to achieve a proper balance between protecting financial stability and facilitating sufficient competition in domestic markets.
Yet it is clear that there is always a limit to what could be termed an acceptable degree of national regulatory specificity. In the case of ring-fencing, the limit should be set at the point where those domestic requirements started penalising foreign subsidiaries vis-à-vis local players. In the case of proportionality, one of the limits would be to avoid overprotecting smaller institutions from legitimate competitive forces, as this would hinder industry efficiency and the participation of international banks in that market.
The relevant role of international standards
The scope for (warranted) inconsistencies in relation to specific regulatory requirements for internationally active banking groups is certainly limited. These are players that compete globally and should be subject to similar rules. International standards are precisely designed to that very end.
It is often observed that a major source of the regulatory inconsistencies behind market fragmentation is the incomplete implementation of international standards in relevant jurisdictions. That may be the consequence of unjustified delays over the agreed deadlines or the introduction of idiosyncratic adjustments to internationally agreed principles.
Indeed, according to the Basel Committee on Banking Supervision (BCBS (2018)), there is still insufficient progress in the timely adoption of the Basel III framework. In particular, a significant number of jurisdictions have not met the agreed timelines for specific standards. Furthermore, some jurisdictions have reported that their implementation of certain standards has been delayed because of their concerns over the pace of implementation in other jurisdictions. This is specifically the case for the Net Stable Funding Ratio, whose implementation deadline was 1 January 2018, and for which most member jurisdictions have not yet approved final rules.
It could also be argued that there is room for improvement in the way both supervisory colleges and crisis management groups currently function. Improvement in this area would certainly facilitate further consistency of prudential requirements and the establishment of sensible internal total loss-absorbing capacity requirements at the legal entity level within international groups.
What else is needed?
Yet, while a necessary condition, it is not certain that complete timely and consistent implementation of existing international standards would ensure regulatory consistency across jurisdictions. In other words, one should accept, at least hypothetically, that effective harmonisation of relevant regulatory requirements for internationally active entities may require additional policy work at the international level.This is certainly the case for insurance regulation, where the scope of international standards is still quite limited. But it is, to a significant extent, also true in the case of banking regulation, despite the considerable effort made to develop the prudential and resolution frameworks in the context of the post-crisis reforms.
Let's take the example of Basel III. This new framework has done a terrific job of broadening the scope of international standards by establishing common requirements for minimum capital, liquidity coverage and large exposures. But arguably, it still falls short of ensuring complete harmonisation of the relevant prudential regulation for internationally active banks.
While Basel III has secured a common definition of capital and harmonised Pillar 1 capital requirements, the procedures followed to measure regulatory capital still lack hom*ogeneity. Capital is a residual obtained (roughly) by subtracting assets from liabilities, as reflected in financial statements. Thus the insufficient convergence of accounting standards across jurisdictions may imply significant discrepancies in capital measurement.
More importantly, the valuation of assets for prudential purposes does not follow consistent criteria, given insufficient international guidance on the matter. And, as measuring capital is highly sensitive to asset valuations, that inconsistency diminishes the comparability of solvency indicators across jurisdictions.
Specificdiscrepancies affect the measurement of non-performing loans and provisioning practices. For instance, in a number of jurisdictions, supervisors override the accounting code by imposing specific provisioning requirements or introduce a variety of prudential backstops for accounting provisions (Restoy and Zamil (2017)). In addition, criteria for collateral valuations and interest accruals in non-performing loans vary markedly across jurisdictions (Baudino et al (2018)).
In the same vein, while Basel III establishes uniform minimum Pillar 1 capital requirements, these are not the binding constraint for effectively required capital in most jurisdictions. Actual capital requirements are typically established by Pillar 2 capital add-ons derived from Supervisory Review and Evaluation Process analysis and/or supervisory stress tests.
As things stand, there is little international guidance on what criteria supervisors should follow to determine those capital add-ons. In one of our forthcoming publications (Duckwitz et al (2019)), we find that jurisdictions display a number of important differences in determining capital add-ons under Pillar 2. First, there is no consensus on what the Pillar 2 capital add-ons should cover (eg some focus only on risks not covered under Pillar 1, while others also include Pillar 1 risks that may be underestimated; and still others include a systemic risk charge within their Pillar 2 frameworks). Second, there is no uniform approach to how the Pillar 2 add-ons, if imposed, should interact with the new buffer requirements introduced under Basel III. Finally, the approaches followed by supervisors to determine the add-ons also vary, ranging from what we label "guided discretion" methods (eg methodologies that provide some hard-wired parameters around the judgments of supervisors) to other authorities' determining Pillar 2 add-ons through a "holistic assessment" of the institution relying heavily on the informed judgments of supervisory teams. Collectively, these diverse approaches invariably lead to different Pillar 2 capital outcomes across jurisdictions.
As for stress tests, another FSI study (Baudino et al (2018)) shows how authorities in selected jurisdictions design stress tests in different ways across certain key features - covering, among other things, the existence and level of capital thresholds; the number and severity of stress scenarios; the inclusion of feedback effects and balance sheet adjustments; and the restrictions imposed on income components over the stress horizon.
The disparity of the criteria followed to supplement the harmonised Pillar 1 requirements implies that Basel III cannot, by itself, guarantee a complete harmonisation of actual capital obligations across jurisdictions.
In sum
Analysing the relationship between regulation and market fragmentation requires a lot of subtlety. Regulatory heterogeneity is not always the cause - and certainly not the main cause - of fragmentation; and, it is often the consequence of a lack of market integration stemming from other factors.
Nevertheless, regulation could contribute to a higher degree of market integration if existing discrepancies were confined to what is really warranted by the need to accommodate domestic specificities in order to accomplish policy objectives.
Moreover, while international standards play a crucial role in limiting unwarranted fragmentation, there is scope for further strengthening their contribution to market integration by ensuring complete, timely and consistent implementation.
All told, the ambition of ensuring a level playing field for internationally active entities may need to look beyond adequate implementation of existing standards, namely at additional policy work aimed at providing international guidance on high-level issues currently covered by widely disparate approaches.
Such additional policy work will acquire even greater significance against the backdrop of rapid technological developments with the potential to disrupt the financial industry. In particular, new common standards may be required in the forthcoming feature to ensure a coordinated adjustment of the regulatory perimeter to (a) accommodate some of the new providers of financial services and (b) establish consistent rules to deal with financial institutions' increasing reliance on technology.
References
Basel Committee on Banking Supervision (2018): Basel III Monitoring Report, March.
Baudino, P, R Goetschmann, J Henry, K Taniguchi and W Zhu (2018): "Stress-testing banks - a comparative analysis", FSI Insights, no 12, November.
Baudino, P, J Orlandi and R Zamil (2018): "The identification and measurement of non-performing assets: a cross-country comparison", FSI Insights, no 7, April.
Duckwitz, V, S Hohl, K Weissenberg and R Zamil (2019): "Pillar 2, risk-based supervision and proportionality", FSI Insights, forthcoming.
Restoy, F and R Zamil (2017): "Prudential policy considerations under expected loss provisioning: lessons from Asia", FSI Insights, no 5, October.
1 I am grateful to Rodrigo Coelho, Juan Carlos Crisanto and Raihan Zamil for helpful comments and to Christina Paavola for very useful support. The views expressed are my own and do not necessarily reflect those of the BIS.
