Financial Risks, Stability, and Globalization


Omotunde Johnson
Published Date:
April 2002
  • ShareShare
Show Summary Details

The Hamada paper is an interesting attempt to discuss the costs of severe financial crises. The paper’s broad scope—embracing payment system problems, electronic networks, currency crises, and general macroeconomic imbalances—causes some problems for the analysis, however. Nevertheless, this interesting paper succeeds in raising the relevant issues and highlighting the difficulty of estimating the costs of financial crises.

I will first comment briefly on the paper and then present a short analysis of Finland’s economic and financial crises in the first half of the 1990s.

Welfare Cost of Systemic Risk

In the first part of his paper, Hamada explores the possibility of estimating the incidence of systemic risk and the effects from the standpoint of the national economy. The analysis and real world examples are restricted to payment systems and electronic networks.

I find the coverage of this first part somewhat cumbersome and narrow, especially since the purpose is to consider systemic risk. The banking sector’s credit risk and solvency problems are not analyzed although they have been at the “center” of most, if not all, recent financial crises. I do not deny the possible inherent systemic risks of payment and settlement systems, but up to now we have not experienced such problems. International payment and settlement systems are integrating rapidly and they are probably an important channel through which contagion effects can spread between banks and over frontiers. But these systems also have many built-in safeguards, such as use of collateral and loss sharing systems. These should explicitly be taken into account when analyzing the incidence and costs of breakups. Moreover, the increasing use of Real Time Gross Settlement (RTGS) systems contributes to reducing risks in payment systems.

In the first part of Hamada’s paper, the TARGET and FEDWIRE large value payment systems are discussed briefly. The large values of transactions are not, in themselves, an indication of the potential losses in breakup situations. TARGET is based on the use of collateral and should be quite safe against counterparty failures. The situation as regards FEDWIRE is different because it is based on risk pricing. It can generally be said that RTGS payments are risk free to the receiver but the central bank has risks if it grants intraday credits to the payer. The central bank can protect itself against this credit risk either through collateral requirements or risk pricing the intraday credit.

The first part of Hamada’s paper also considers ways to reduce the systemic risks of payment system failures. Some comments on the three proposed measures are given below.

In an RTGS environment, proposal (1)—delay the delivery of funds until interbank settlements are completed—always holds true because in RTGS systems the delivery of funds and interbank settlement by definition always take place at the same time. This is the fundamental principle of RTGS. Proposal (3)—upper limit on daily counterparty exposures—is also always satisfied in the case of collateral based RTGS systems because funds must be available before transactions are processed. For private netting systems, proposal (3) holds for systems that fulfill the Lamfalussy criteria (e.g., European systems). The same is also true for proposal (2): non-sound banks are normally closed out immediately when information on the new status is available to system operators.

As regards technical failures, all major systems, including TARGET and FEDWIRE, have backup systems, which should be taken into account in incidence analysis. For instance, at the Bank of Finland we have a hot backup for the entire production environment in another location. In addition, the PC and manual routines can be used in serious crisis situations in order to maintain a slower service covering all major transactions. This substantially reduces the adverse effects of severe technical problems. There are also other parallel/competing systems that can take over the load of failing systems. If the TARGET component in one country is down, agreed contingency procedures can be used to route critical TARGET payments via another country.

I find some basic problems in the calculation of the costs of the FEDWIRE’s one-week breakdown. The calculation somehow assumes that payments are directly related to GDP. This is hardly the case. Most payments are money, forex and securities market—related—that is, settlements for deals in these markets. RTGS payments are mostly pure interbank payments. Customer payments are executed in batch-based and mostly private systems. If these payments are deferred or cancelled, the underlying deal will be postponed or cancelled in a DVP/PVP environment. Hence, it is only the deal or trade that is at stake here, not the value of the property itself. The delaying or cancellation of the payment may, however, give rise to opportunity costs.

Welfare Costs of Financial Instability and Financial Crises

In general, I have some problems with the distinction between systemic risk, financial instability, and financial crises in the analysis. I would put systemic risks and financial crises in the same category and separate them clearly from general financial market instability. The author basically defines asset price fluctuations as financial instability. As far as these rationally reflect changes in the prospects for economic fundamentals, I would not refer to them as instability. They can be seen as a normal consequence of the fact of life that the future is always uncertain. Economic agents need to insure, hedge, and diversify against fluctuations to cope in an uncertain world. Therefore, I do not think the ABC analysis in the second part of the paper is very relevant. Instead, we should try to identify that part of instability (excessive asset price fluctuations and drying up of liquidity and sources of funding) that stems from problems of asymmetric information being particularly pronounced in the market in some periods. These could cause a failure in the normal functioning of markets, if markets are at least periodically unable to provide the usual financing and insurance and might in the worst case require public intervention.

The second part of the paper considers the costs of alternative currency regimes in terms of the loss in real income. The analysis is very demanding and raises the right issues. However, in order to be able to analyze the real opportunity costs of different forex regimes one needs to use a macro/micro model of the economy. The model should also include the financial markets, banking sector, and outside world. Such models do not really exist, and hence a partial analysis is all that we can do.

I think that the Nordic countries’ experience with financial crises is also relevant in this connection (see Figures 7.5-7.9). I will briefly consider the case of Finland and an important lesson that can be drawn from it. The Finnish crisis in the first half of the 1990s is a typical example of a severe financial crisis that quickly evolved into a macroeconomic, public sector, and banking crisis (see Figures 7.10-7.14). It is also very similar to the recent Asian crisis. GDP fell by almost 15 percent in three years (1991-93), unemployment increased from 4 percent to nearly 20 percent, the public deficit increased rapidly, and the losses of the banking sector were enormous. The fixed exchange rate was replaced by a flexible rate system. Public support had to be given to the banking sector.

Figure 7.5.Rates of Growth in Bank Lending in the Nordic Countries


Source: National central banks.

Figure 7.6.Ratio of Loan Losses to Average Total Assets of Nordic Banks

Source: National central banks.

Figure 7.7.Ratio of Operating Profit to Average Total Assets of Nordic Banks

Source: National central banks.

Figure 7.8.Number of Employees per 1,000 Inhabitants in the Nordic Countries

Source: National central banks.

Figure 7.9.Number of Bank Branches per 1,000 Inhabitants in the Nordic Countries

Source: National central banks.

Figure 7.10.Gross Domestic Product

(Volume index 1970=100)

Figure 7.11.Unemployment Rate


Figure 7.12.Central Government Finances 12-Month Moving Totals

Figure 7.13.Key Interest Rates


Figure 7.14.Bank Lending to the Public, 12-Month Change


The estimated net cost of the support given to the banks was about FIM 50 billion, which is 7 percent of GDP. In gross terms, the original support provided amounted to nearly 15 percent of GDP. By applying very tough economic policies aimed at balancing the public sector and granting support to the banking sector, it was possible to stabilize the economy and the banking sector within a couple of years. Unemployment, however, still remains high. Finland entered the EU in 1994 and European Economic and Monetary Union (EMU) in 1999. Finland now belongs to the Eurosystem and hence it no longer has an independent monetary policy.

The public support to banks was given on stringent terms in order to avoid moral hazard problems and compel the banks to restructure. During the regulated period, considerable overcapacity had emerged, which it was necessary to reduce. These rapid and tough policies, which enjoyed broad political support, succeeded better than expected, as can be seen from the figures. The banking sector is now one of the most profitable and efficient in Europe, with return-on-equity (ROE) of 20 percent and a cost-income ratio of 55 percent. The numbers of both personnel and branches have been reduced by about 50 percent (see Figures 7.8-7.9). Absolute operating costs have declined by 40 percent.

So what is the welfare cost of this severe financial and macroeconomic crisis (“systemic crisis”) to the Finnish economy? The crisis had three main effects—namely, lost output in 1991-94 (see Figure 7.10), higher unemployment (see Figure 7.11), and direct cost of public bank support, equivalent to 7 percent of GDP.

But are these the net costs to the economy? Because of the prompt and stringent policies applied, the economy started a process of restructuring—it received a real boost via the crisis. The banking sector is an example of what restructuring can achieve when circumstances make it necessary. The efficiency of the whole corporate sector has improved enormously. The current account surplus is now running at 5 percent of GDP.

The net cost of the crisis to the Finnish society is much less than the “gross” figures indicate. Of course, this does not mean that we should recommend crisis in order to give impetus to restructuring, since balanced and stable development is always better. But we must realize that crisis situations can also create something new and positive, if the reaction to crisis is quick and appropriate. This is an important lesson—for example, for the Asian countries that are still struggling to recover from their crisis (this also applies to Japan).

    Other Resources Citing This Publication