International Conference "Central Banks in the 21st Century" hosted by the Banco de Espana in Madrid
COMMENTS ON ”FINANCIAL STABILITY AND GLOBALISATION: GETTING IT RIGHT” BY F. MISHKIN (paper for a Bank of Spain conference, 8-9 June 2006)
On the paper
Professor Frederic Mishkin provides a very insightful and comprehensive paper on potential pitfalls in financial liberalization and globalization, and discusses policy responses to well-manage such a transition – both macro (monetary, fiscal and trade policies) and microeconomic (legal system, corporate governance, information disclosure, financial supervision). On these policies, there is both a balanced account of the literature and some recommendations in the paper. The paper mostly deals with developing countries.
While liberalization and globalization are needed for supporting financial deepening, an almost immediate consequence in several countries was a period of severe financial instability or a full-blown financial crisis. The high cost of these crises makes their prevention and effective resolution a policy priority.
“Anatomy” of crises since early 1980s
There is striking amount similarity in the past crises, even though details of the crises may differ. Most of the financial crises experienced since the early 1980s followed a fairly standard pattern of deregulation, lending boom, asset price booms, business cycle and asset price shocks, and large-scale bank failures, and – eventually – a financial crisis . Crisis was usually triggered by an external shock (e.g. unforeseen recession or ex-change rate collapse). Crisis resolution was often similar as well. Insolvent banks were either temporarily nationalized or forcibly merged, after separating “bad loans” in a state agency. In some cases, a blanket government guarantee was also given that all financial institutions would meet their obligations.
Examples of similar crises include the crises in Latin America (Argentina, Mexico in early 1980s), the US Savings and loans institutions’ crisis (early 1980s), and the three Scandinavian banking crises (Norway, Finland, Sweden in early 1990s). The crises in merging market economies in late 1990s (Korea, Philippines etc.) also reflected the same pattern, by and large.
The Finnish crisis fits well with the major source of instability highlighted in the paper: the danger of accumulating indebtedness in foreign currencies and the resulting currency mismatches in the closed sectors of the economy. Once the Finnish currency depreciated heavily after its flotation, many firms could no longer repay their debt and banks’ ac-crued vast amount of credit losses.
Four viewpoints for discussion
The points below are suggested as complements to the paper – not necessarily as points of disagreement (even though these elements are missing or hidden in the paper):
Financial globalization raises difficult issues related to supervision and crisis management of cross-border and cross-sector financial institutions
Focus also on the quality of supervision by developing risk-based approaches (not only on the institutional framework);
Avoid rigid policy solutions (one-size-fits-all) for countries with different circumstances;
The next crisis could result from different risks and require different preventive or resolution measures. We should be prepared to prevent and handle future crises not the past ones.
1. Financial globalization raises difficult issues related to supervision and crisis management of cross-border and cross-sector financial institutions
The paper argues, rightly, that opening up the domestic markets to foreign financial institutions may promote higher-quality domestic regulation and supervision by, inter alia, putting pressure on the domestic government to make reforms that will make the financial system work more efficiently. However, as is well-known, big market shares of foreign-owned banks in many small countries create potential problems for both prudential financial supervision and the maintenance of financial stability in those countries.
One of the key issues in the international regulatory debate is how to divide the tasks, powers, and responsibility between different countries in supervising cross-border financial institutions and dealing with these institutions in financial distress. In the EU, the di-vision of tasks, powers, and responsibilities is currently based on home-country control. The home country is responsible for supervising financial institutions that are headquartered in that country and its foreign branches. Foreign subsidiaries, on the other hand, are supervised by the host country supervisor. The home country supervisor, in addition, is responsible for the consolidated supervision of the financial group.
Today, financial institutions cross-border and cross-sector activities are increasing at fast speed. Currently, for example, there are at least over 40 banks and banking groups that are active in more than three EU countries. In addition, financial institutions increasingly work along global business lines and less along national borders. For financial groups which are expanding cross-border a part of synergies is derived from centralizing functions along business lines, not necessarily along country borders. Concerning the legal form of banks´ foreign operations, the relative importance of banks´ foreign branches (as opposed to that of foreign subsidiaries) is likely to increase substantially as banks start to take advantage of the European Company Statute and operate as a single entity. As illustrated by these changes, the current and future European banking market is very different from that in which role of banks´ cross-border and cross-sector operations and banks´ foreign branches was negligible.
The problem of the current set-up is the gap, on the other hand, between the legal powers and mandates and, on the other hand, the abilities and responsibilities of the home and host supervisors. This gap becomes a problem once there is a financial institution that is systemically relevant in a host country. The problems are starkest in the event of a crisis in the banking group, as all national authorities have a clear mandate to protect only their depositors and systems. In particular, if the institution´s presence in the host country is a branch, it is the home country supervisor who is responsible for managing a crisis. Although the branch can systemically important in the host country, all its authorities can, to put it bluntly, do is to hope that the home country authorities take the interests of host country into account when dealing with the crisis.
It is clear that because of integration, the EU´s current regulatory set-up will be under increasing pressure in the future. Therefore, it is not a surprise that various alternative models to the current home-host set-up have been presented. The alternatives range from considerably strengthening the role of the home supervisor (lead supervisor model) to creating a European Financial Supervisor with full supervisory powers over branches and subsidiaries of cross-border banks. However, despite the pressures, the EU is likely for the time being to stick to the current European supervisory principles and structures and try to improve the functioning of the current regime by, inter alia, increasing supervisory cooperation and convergence of supervisory methods.
A thorough discussion of the European supervisory arrangements would require a sepa-rate presentation. In this context, it might be sufficient to say that the financial globalization raises very challenging issues related to the supervision of cross-border and cross-sector financial institutions. These issues, in addition to those presented in Mishkin´s paper, should be addressed in any regulatory reform agendas.
2. Focus also on the quality of supervision by developing risk-based approaches
An appropriate institutional framework (deposit insurance, disclosure regime, property rights, independence and accountability of supervisors etc.) is clearly a prerequisite for well-managing liberalization and preventing crises. However, one of the biggest lessons from the crises (also from the Finnish one) is that banks’ risk management and supervisory practices were not yet developed enough to address the risks in the new environment. This resulted in uncontrolled risk taking by banks as such risk taking was made possible by the liberalization. Mishkin discusses as a central issue the need to put in place proper institutional setting (foremost supervision) before liberalization, but does not spend space on the issue of the quality of supervision.
Before the crises, supervision was (and can still sometimes be) mostly legalistic checking of compliance with current regulations, rather than focused on risks. But risk-based supervision is at the core of having preventive (i.e. forward-looking) supervision rather than responding to problems when it is already too late. This is an issue of both supervisory culture and resources (a large enough number of qualified risk analysts working as supervisors).
What is needed is a systematic plan to develop risk-based supervision, which – one should acknowledge – takes many years to institute in a supervisory authority. Even developed countries have further efforts to make in this area.
A key element of such a plan is to put emphasis on institutions’ own risk management, as risks are getting so complex and fast moving that they cannot be observed real-time by supervisors, or prevented through simple supervisory limits. Independent supervisory as-sessment is needed to prevent moral hazard and to avoid problem institutions from shifting risk to the government. Basel II reform supports this development by instituting a regular supervisory risk review (which is the second Pillar of the Basel package).
3. Avoid rigid solutions (one-size-fits-all) for all countries
Policy measures need to be adjusted to individual countries’ circumstances (economic structures and development stage, financial system structure, political systems etc). While lessons on policy mistakes from the crises can be (reasonably) clear, this does not mean that a simple recipe can prevent future crises, or represent otherwise suitable policy options in all circumstances.
The paper does not advocate uniform solutions very strongly; in general it is quite balanced. It only has this flavor at instances. Some examples are below. (International institutions (IMF, World Bank) can have a stronger tendency for one-size-fits- all recommen-dations).
Floating exchange rates (recommended with caveats in the paper) can be more stabilizing than fixed and allow using monetary policy to cool down fast credit growth. However, a currency board could be a good option for “importing” monetary policy and bringing down inflation.
Deposit insurance is often recommended (by World Bank, IMF and in the paper) to be avoided, at least by less institutionally developed countries because of moral hazard and increased risk taking by banks. However, deposit insurance can be a commitment device for governments to rescue only depositors and not other creditors of banks, and thus to support market discipline. Without deposit insurance the true policy adopted in a crisis can be a blanket guarantee for all bank creditors. Market discipline is only possible when some creditors have their money at stake (with some positive ex ante probability) in a crisis. Disclosure of information is needed, but it does not generate market discipline unless the previous condition is met. Hence, explicit and limited deposit insurance – explicitly protecting depositors only and leaving out other creditors – could be a step forward especially in a country (like Scandi-navian countries in late 1980s / early 90s) with a history of blanked guarantees and weak market discipline. Explicit deposit insurance also needs to be coupled with ef-fective supervision to cub any moral hazard effects.
Stringent (and simple?) regulatory capital requirements are recommended (in the paper) for banks in less developed countries. However, a large and simple require-ment (such as a leverage ratio) would not support the development of banks’ own risk management – and ability to function in a liberalized environment – as it is not risk-based. On the other hand, advanced aspects of the highly risk-based Basel II may not be feasible due to lack of expertise and data in banks, but nothing prevents from gradually moving towards a more flexible and risk-based approach in less developed countries (where Basel II is not obligatory and its timing is not pre-specified like in Europe). One could implement e.g. the standard Basel II model for sovereign and bank counterparties based on external ratings and thus avoid the present low risk weight on government bonds under Basel I. Advanced methods for calculating capital charges for firms and households could and should be adopted only over time with more experience in banks and the supervisory authority.
Limiting currency mismatches (recommended in the paper) in closed sectors of the economy would only be feasible by limiting foreign currency-denominated borrow-ing. This would be canceling liberalization, actually, so other indirect measures should probably be considered. What those measures could be will depend on the local tax system (possibility to use taxes to curb lending growth), monetary policy etc. This is an actual issue for many New Members States of the EU with high credit growth, often in foreign currencies.
4. The next crisis could result from different risks and require different preventive or resolution measures
The past crises were mainly due to credit risk and were national in scope. While credit risk often still is the most important risk for banks, the structure of the financial system has been changing in a profound way and the next crisis could come from a different source.
First, a rapid increase in banks’ and other financial institutions’ (e.g. insurance companies and pension funds) financial market-related activities has increased their exposure to financial markets (and counterparty risks to other market players) and earnings sensitivity, suggesting that financial instability may increasingly result from market instability. This refers to equity, bond and derivative markets as distinguished from more traditional exposures of banks to real estate price and exchange rate movements.
Second, the past separation between financial institutions and markets has been replaced by an increasing integration of markets and banks, as well as between banks and other financial institutions. Hence, systemic risks could result from non-bank financial institutions (such as hedge funds, private equity funds etc.), at least to the extent that major fi-nancial institutions are exposed to them.
Third, liquidity conditions and contagion risks play an increasingly important role due to the rapid increase in financial markets’ activities and banks’ increased reliance on market sources of funds.
Fourth, internationalization of the financial system has spread rapidly. Capital controls and restrictions on cross-border bank operations have largely been eliminated. National markets can no longer be viewed as isolated entities, but tend to be embedded in a complex system of interlinks. This holds true especially for EU countries due to the Single Market and spreading of cross-border banking, but also globally and for many developing countries as well, because in their financial systems international banking groups can play an important role. In New EU Members States the ownership of foreign banks in the local banking system is 70-100%.
Policies and measures taken to foster financial stability must take into account these de-velopments in case of both developed and developing economies . Failure to upgrade policies to mach financial system developments can be very costly as the previous lessons show.