8 Assessing the Ability of a Financial System to Withstand Financial Stress: A Market Perspective
- Omotunde Johnson
- Published Date:
- April 2002
“Assessing the Ability of a Financial System to Withstand Financial Stress” is an interesting and somewhat challenging topic for a rating agency professional such as myself, since this is similar to, but not exactly the same as, what we do at Moody’s. As you know, rating agencies publish opinions of the credit quality of securities and their issuers. My department, Moody’s Financial Institutions and Sovereign Risk Group, rates financial institutions, sovereigns, supranationals, and subsovereigns. We cover 2,000 issuers in 100 countries. The question that we attempt to answer about each of these issuers is, “How likely is a creditor of this obligor to get all of its money back on a timely basis?” In attempting to answer this question for banks and sovereigns, we are necessarily confronted with the need to analyze the soundness of financial systems as a whole. So it is not unreasonable to ask someone from a rating agency to address this topic.
We have been confronted with acute financial system stress in many countries over the past five years. To list the most obvious such countries: Mexico, Japan, Thailand, Indonesia, Republic of Korea, Russia.
All of these countries have faced a crisis of domestic and international confidence in the safety and soundness of their financial systems resulting in acute stress for individual institutions and for regulators.
Factors in Predicting Stress
What I would like to do today is suggest a method for assessing the ability of a financial system to avoid or to withstand stress. I will begin with a taxonomy of what I consider to be the relevant factors and then I will discuss the interrelationships.
Four factors have predictive power in the identification of vulnerability to stress. But these four factors are necessary but insufficient to cause a financial crisis: systemic solvency; accounting transparency; quality of supervision; and funding stability.
If a large proportion of a system’s banks are insolvent, there is the risk that some spark will trigger a depositor or creditor panic. I do not mean insolvency in an accounting sense, since one would be hard pressed to find a single bank in the world that has annual accounts showing it to be insolvent. Instead, I mean that a bank credit analyst would judge the institution to be economically insolvent. Moody’s defines economic insolvency to mean that an institution’s unrealized credit losses exceed its capital, reserves, and realistic near-term earnings prospects. The critical assumption here, the level of unrealized credit losses, is a professional estimate based on whatever information is available.
Moody’s publishes an index of banking system solvency, which is the average bank financial strength rating—what we refer to as the BFSR—for each of the 75 countries in our universe (see Figure 8.1). The BFSR is an opinion of a bank’s stand-alone financial strength without reference to external support mechanisms or sovereign transfer risk.
Figure 8.1.Average Bank Financial Strength Rating
Opaque and/or bogus accounting increases vulnerability to stress (and reduces the ability to withstand stress) because it prevents depositors and creditors from discriminating between the sound and the unsound. In a period of depositor or creditor panic, there can be no differentiation if accounting is opaque and if the accounts of the weakest banks are indistinguishable from those of the strongest.
The catalog of accounting games used by banks (often abetted by their regulators) to masquerade as solvent ones is endless. Among the most common, I would list:
- minimizing reported delinquencies by lending to finance interest payments or by accruing unpaid interest;
- overvaluing collateral;
- establishing and financing unconsolidated subsidiaries to purchase bad assets from troubled borrowers;
- transferring problem loans to subsidiaries in locations beyond the jurisdiction of the bank’s regulator;
- moving bad assets from branch to branch in advance of the bank examiners; and
- misclassification of problem loans to minimize loan-loss reserves.
Given that probably one-quarter of the banks that Moody’s rates are economically insolvent—and that not one of them publishes accounts that show it to be insolvent—it is fair to say that such practices are routine at hundreds of banks around the world, and are indeed normal in the E and E+ rated banking systems.
Bogus accounting is an artifact of regulation, which is next on the list.
Quality of Supervision
Insolvent banks are able to masquerade as solvent and to publish bogus accounts because their supervisors permit it or lack the authority to prevent it. A central element of the supervisors’ job is to examine a bank’s books, to ensure their accuracy, and—should a bank be found to be insolvent—to remove it from the financial landscape. Unfortunately, many regulators instead see their job as helping the banks to paper over their problems in the hopes that everything will somehow work out in the long run.
A fourth vulnerability factor is excessive reliance upon hot, or non-core, funding. Such funding may come from institutional investors, the interbank market, foreign bank and nonbank depositors, or from commercial paper holders. The problem with such funding is that it is confidence sensitive and subject to withdrawal in the event of credit concerns. The presence of such funding may not be readily apparent on a bank’s books, since many banks make no distinction in their reporting between retail and wholesale deposits.
These four factors that I have enumerated are necessary but, in my opinion, insufficient to cause a crisis. Consequently, I now turn to my next topic.
What Actually Causes Financial Crises?
So far, my catalog of vulnerabilities has been quite familiar and conforms to the prevailing orthodoxy: insolvency, poor transparency, lax supervision, illiquidity. My next assertion, however, may strike a less intuitive note. The foregoing elements are the necessary but insufficient precipitants of a financial crisis. They are the kindling, but they need a spark. This is because weak, opaque, illiquid, and poorly supervised financial systems can (and usually do) survive happily for years without a crisis. Insolvent and illiquid banks don’t have to fail and won’t fail unless some external element intervenes to force a failure. Under normal conditions, the system’s guardians can keep an insolvent bank (or an insolvent banking system) alive indefinitely, so long as they engage in accounting forbearance, provide unlimited liquidity to the system, and provide consistent signals that the entire system is safe and that no bank poses credit risk to any depositor. So long as these signals—and the safety net behind them—are credible, there is no reason why any domestic or foreign creditor should be concerned about solvency, transparency, liquidity, or the quality of supervision. Furthermore, there is nothing to prevent the authorities from conducting prudent supervision including the resolution of troubled banks, so long as they are resolved without loss to depositors or creditors (see Figure 8.2). Thus, weak banking systems are not inherently fragile and financial fragility should not be defined as simply a weak banking system.
Figure 8.2.How Banks Fail
Experience suggests that there are only two things that can be defined as sufficient to cause a banking crisis the abrupt introduction of credit risk into a previously riskless system; and/or an external payments crisis. I will discuss these in turn.
Injudicious Introduction of Market Discipline
Market discipline has an important role to play in the policing of a sound and well-run banking system. The market can play a constructive role by rewarding the prudent and successful, and by punishing the imprudent and the unsuccessful. Market pressure can help to remove an injured player from the playing field. In a healthy system, creditors and uninsured depositors should not believe that they are immune from risk when they hold the obligations of a weak and/or mismanaged bank.
However, what is prophylactic for a healthy banking system can prove lethal when introduced into an unhealthy one. The abrupt introduction of credit risk into a weak but previously riskless financial system will almost surely produce a banking crisis. Market discipline should come after a period during which prudential supervision has resolved the system’s solvency and transparency problems—not before.
What makes a system vulnerable to an abrupt introduction of market discipline? Lately, the introduction of market discipline has often tended to coincide with external payments crises, which is my next topic.
External Payments Crises
Given our recent experiences in the emerging markets, it is certainly not counterintuitive to observe that banking crises and external payments crises often have coincided. What is the relationship between these phenomena?
The common thread among these diverse countries is the introduction of financial liberalization within the context of anti-inflation policies centering on fixed or quasi-fixed exchange rates. Liberalization compels banks to enter into competition for market share in a deregulated environment of which they have little, if any, knowledge and without the proper credit assessment skills.
In a monetary regime where domestic interest rates are generally higher than those abroad (in order to defend the exchange rate), banks and nonbanks increasingly fund themselves by borrowing from abroad, and often short-term. Beyond this, foreign portfolio investors are attracted to the higher interest rates, perceiving minimum exchange rate risk.
If the capital inflows are not fully sterilized and if fiscal policy is not tightened sufficiently, this monetary regime generally leads to credit booms, rapid economic expansion, and asset price appreciation. It also leads to the appreciation of the currency in real terms. This, in turn, hinders exports while encouraging imports, which become the vent for excess internal demand.
Thus, the macroeconomic environment results in a growing economy with declining inflation—every policymaker’s objective. It also sets the stage for the eventual related banking and external payments crises.
Over time, the growing external trade deterioration leads both foreign and domestic investors to question the sustainability of the exchange rate regime. Foreign capital inflows begin to dry up and speculation against the currency by both residents and nonresidents may be encouraged.
Interest rates rise even further to defend the exchange rate, causing economic activity to decelerate and asset prices to decline. Domestic borrowers may find it difficult to repay loans, while banks are left with questionable collateral. At some point, a devaluation of the currency may become inevitable, complicating matters further for the banks. Their foreign liabilities grow proportionally with the devaluation—liabilities that are often mismatched both in terms of currency and maturity structure relative to the banks’ assets.
The links between the external payments and the banking system crises may manifest themselves somewhat differently from how I have described, depending on a number of factors, including the macro-economic policy framework (the consistency of fiscal, monetary, and exchange rate policies), the overall external position of a country, the balance sheet structure of banks (i.e., the scale of currency and liquidity mismatch), and, finally, the degree of banks’ exposure to specific economic sectors (e.g., exports, real estate).
I have shown above the close linkages among macroeconomic policy, financial crisis, and external payments crisis, but I do not want to digress too far into the enormous subject of the etiology of external payments crises. My topic today is not the etiology of external payments crises, but rather the etiology of domestic financial crisis.
In an external payments crisis, two elements can combine to produce a banking panic.
- First, in such a crisis, risk perceptions on the part of domestic and foreign creditors are typically heightened to a fever pitch such that every name in the country suddenly looks very risky. The solvency of every bank is questioned.
- Second, there is a temptation on the part of the authorities to use the crisis as the opportunity to resolve the structural weaknesses in the financial system, by such means as suspending the licenses of hopelessly insolvent banks, securities firms, or finance companies. The explosive combination of heightened risk sensitivity and mixed signals about creditor risk can produce a domestic depositor panic, forcing the authorities to choose between bailing out the hopelessly insolvent or allowing the panic to engulf the entire system.
Thus, an external payments crisis can lead to a domestic banking crisis. Consequently, the degree to which a country is susceptible to an external payments crisis may be properly defined as a potential risk factor for a domestic financial crisis.
Moody’s provides a datapoint that is a reasonable proxy for susceptibility to an external payments crisis: the Country Ceiling for Long-Term Bank Deposits (see Figure 8.3), which measures the risk of a sovereign-imposed deposit moratorium. Here, I would point out that just as financial fragility alone is insufficient to cause a domestic financial crisis, neither is it sufficient to cause an external payments crisis. Just as many insolvent banking systems have survived for years without a banking panic, most countries with insolvent systems have not experienced an external payments crisis.
Figure 8.3.Country Ceiling For Long-Term Bank Deposits
The best example of both phenomena is China. Chinese banks are widely seen as very weak. Chinese accounting is opaque and supervision has historically been lax. But China has not experienced a banking panic because the authorities have kept credit risk out of the banking system. And a weak financial system has not produced an external payments crisis because China has prudently maintained a very strong external liquidity position.
Having proposed a methodology for identifying financial systems that are vulnerable to stress, the next step would be to examine our universe of 75 countries for systems (1) that are vulnerable, in the sense of being weak, and (2) where there is either the risk of an external payments crisis or of the abrupt imposition of market discipline.
As we have seen, the identification of weak systems is possible using the average BFSR, which reveals the banking systems in .1 as extremely weak.
There is no counterintuitive news content here, I’m afraid. So, let us next seek to identify those systems potentially subject to an external payments crisis or the abrupt imposition of market discipline. As we have seen, the Moody’s Deposit Ceiling is a reasonable proxy for the risk of an external payments crisis. The names at the bottom of this list are not surprising either: Pakistan, Ukraine, Ecuador, Cuba, Russia, Romania, Moldova, Indonesia, Vietnam, etc. So we seem to have identified as vulnerable a group of countries that any layman could have come up with.
As measured by the unweighted, average BFSR for the rated banks in Moody’s 70-country bank rating universe.
As measured by the unweighted, average BFSR for the rated banks in Moody’s 70-country bank rating universe.
Alternatively, what about countries that might fit the second scenario: a weak banking system and the risk of the abrupt introduction of market discipline? The Republic of Korea and Indonesia were examples of this in 1997. Two countries that might fit this description today are China and Japan. China’s average BFSR is E+; Japan’s is D. In both cases, the banks are weak and the authorities have flirted with market discipline on the outskirts of the banking system: GITIC in China, and Sanyo Securities in Japan. The authorities in both countries are very much aware of the hazards of market discipline, but it is possible that political developments could lead to further experimentation. The possibility is remote, but potentially very dangerous given the weakness of their banking systems.
I have tried here to change the focus of discussion on the issue of a financial system’s ability to withstand financial stress from the orthodox analysis of systemic weaknesses, which are pandemic to developing countries and which are hard to fix, to two factors that experience has shown to be the sufficient causes of a crisis and that can be more easily corrected: namely, the injudicious introduction of market discipline into a weak financial system and the vulnerability to an external payments crisis.
I do not think that one can realistically expect policymakers in the developing world to rapidly address issues of systemic financial weakness. But I do think it is possible for policymakers to avoid the introduction of market discipline at a time when the banking system is under pressure.
Similarly, policymakers can take steps to reduce their economy’s vulnerability to external pressures. In case after case in recent experience, the appropriateness of macroeconomic policy and the adequacy of a country’s liquidity position stand out as critical to the development of an external payments crisis. I am not saying that underlying structural weaknesses and poor macroeconomic fundamentals do not matter in the long run, but rather that they can be substantially insulated against by appropriate palliative measures. Consequently, systemic weaknesses need not result in stress so long as (1) an external payments crisis is avoided; and (2) the safety net for creditors of the banking system is strong, credible, and unquestioned.