Banks are institutions wherein miracles happen regularly. We rarely entrust our money to anyone but ourselves and our banks. Despite a very chequered history of mismanagement, corruption, false promises and representations, delusions and behavioural inconsistency banks still succeed to motivate us to give them our money. Partly it is the feeling that there is safety in numbers. The fashionable term today is "moral hazard". The implicit guarantees of the state and of other financial institutions moves us to take risks which we would, otherwise, have avoided. Partly it is the sophistication of the banks in marketing and promoting themselves and their products. Glossy brochures, professional computer and video presentations and vast, shrine-like, real estate complexes all serve to enhance the image of the banks as the temples of the new religion of money.

But what is behind all this? How can we judge the soundness of our banks? In other words, how can we tell if our money is safely tucked away in a safe haven?

The reflex is to go to the bank's balance sheets. Banks and balance sheets have been both invented in their modern form in the 15th century. A balance sheet, coupled with other financial statements is supposed to provide us with a true and full picture of the health of the bank, its past and its long-term prospects. The surprising thing is that despite common opinion it does. The less surprising element is that it is rather useless unless you know how to read it.

Financial Statements (Income aka Profit and Loss - Statement, Cash Flow Statement and Balance Sheet) come in many forms. Sometimes they conform to Western accounting standards (the Generally Accepted Accounting Principles, GAAP, or the less rigorous and more fuzzily worded International Accounting Standards, IAS). Otherwise, they conform to local accounting standards, which often leave a lot to be desired. Still, you should look for banks, which make their updated financial reports available to you. The best choice would be a bank that is audited by one of the Big Six Western accounting firms and makes its audit reports publicly available. Such audited financial statements should consolidate the financial results of the bank with the financial results of its subsidiaries or associated companies. A lot often hides in those corners of corporate ownership.

Banks are rated by independent agencies. The most famous and most reliable of the lot is Fitch-IBCA. Another one is Thomson BankWatch-BREE. These agencies assign letter and number combinations to the banks, that reflect their stability. Most agencies differentiate the short term from the long term prospects of the banking institution rated. Some of them even study (and rate) issues, such as the legality of the operations of the bank (legal rating). Ostensibly, all a concerned person has to do, therefore, is to step up to the bank manager, muster courage and ask for the bank's rating. Unfortunately, life is more complicated than rating agencies would like us to believe. They base themselves mostly on the financial results of the bank rated, as a reliable gauge of its financial strength or financial profile. Nothing is further from the truth.

Admittedly, the financial results do contain a few important facts. But one has to look beyond the naked figures to get the real often much less encouraging picture.

Consider the thorny issue of exchange rates. Financial statements are calculated (sometimes stated in USD in addition to the local currency) using the exchange rate prevailing on the 31st of December of the fiscal year (to which the statements refer). In a country with a volatile domestic currency this would tend to completely distort the true picture. This is especially true if a big chunk of the activity preceded this arbitrary date. The same applies to financial statements, which were not inflation-adjusted in high inflation countries. The statements will look inflated and even reflect profits where heavy losses were incurred. "Average amounts" accounting (which makes use of average exchange rates throughout the year) is even more misleading. The only way to truly reflect reality is if the bank were to keep two sets of accounts: one in the local currency and one in USD (or in some other currency of reference). Otherwise, fictitious growth in the asset base (due to inflation or currency fluctuations) could result.

Another example: in many countries, changes in regulations can greatly effect the financial statements of a bank. In 1996, in Russia, to take an example, the Bank of Russia changed the algorithm for calculating an important banking ratio (the capital to risk weighted assets ratio). Unless a Russian bank restated its previous financial statements accordingly, a sharp change in profitability appeared from nowhere.

The net assets themselves are always misstated: the figure refers to the situation on 31/12. A 48-hour loan given to a collaborating firm can inflate the asset base on the crucial date. This misrepresentation is only mildly ameliorated by the introduction of an "average assets" calculus. Moreover, some of the assets can be interest earning and performing others, non-performing. The maturity distribution of the assets is also of prime importance. If most of the bank's assets can be withdrawn by its clients on a very short notice (on demand) it can swiftly find itself in trouble with a run on its assets leading to insolvency.

Another oft-used figure is the net income of the bank. It is important to distinguish interest income from non-interest income. In an open, sophisticated credit market, the income from interest differentials should be minimal and reflect the risk plus a reasonable component of income to the bank. But in many countries (Japan, Russia) the government subsidizes banks by lending to them money cheaply (through the Central Bank or through bonds). The banks then proceed to lend the cheap funds at exorbitant rates to their customers, thus reaping enormous interest income. In many countries the income from government securities is tax free, which represents another form of subsidy. A high income from interest is a sign of weakness, not of health, here today, there tomorrow. The preferred indicator should be income from operations (fees, commissions and other charges).

There are a few key ratios to observe. A relevant question is whether the bank is accredited with international banking agencies. The latter issue regulatory capital requirements and other defined ratios. Compliance with these demands is a minimum in the absence of which, the bank should be regarded as positively dangerous.

The return on the bank's equity (ROE) is the net income divided by its average equity. The return on the bank's assets (ROA) is its net income divided by its average assets. The (tier 1 or total) capital divided by the bank's risk weighted assets a measure of the bank's capital adequacy. Most banks follow the provisions of the Basel Accord as set by the Basel Committee of Bank Supervision (also known as the G10). This could be misleading because the Accord is ill equipped to deal with risks associated with emerging markets, where default rates of 33% and more are the norm. Finally, there is the common stock to total assets ratio. But ratios are not cure-alls. Inasmuch as the quantities that comprise them can be toyed with they can be subject to manipulation and distortion. It is true that it is better to have high ratios than low ones. High ratios are indicative of a bank's underlying strength of reserves and provisions and, thereby, of its ability to expand its business. A strong bank can also participate in various programs, offerings and auctions of the Central Bank or of the Ministry of Finance. The more of the bank's earnings are retained in the bank and not distributed as profits to its shareholders the better these ratios and the bank's resilience to credit risks. Still, these ratios should be taken with more than a grain of salt. Not even the bank's profit margin (the ratio of net income to total income) or its asset utilization coefficient (the ratio of income to average assets) should be relied upon. They could be the result of hidden subsidies by the government and management misjudgement or understatement of credit risks.

To elaborate on the last two points: a bank can borrow cheap money from the Central Bank (or pay low interest to its depositors and savers) and invest it in secure government bonds, earning a much higher interest income from the bonds' coupon payments. The end result: a rise in the bank's income and profitability due to a non-productive, non-lasting arbitrage operation. Otherwise, the bank's management can understate the amounts of bad loans carried on the bank's books, thus decreasing the necessary set-asides and increasing profitability. The financial statements of banks largely reflect the management's appraisal of the business. This is a poor guide to go by.

In the main financial results' page of a bank's books, special attention should be paid to provisions for the devaluation of securities and to the unrealized difference in the currency position. This is especially true if the bank is holding a major part of the assets (in the form of financial investments or of loans) and the equity is invested in securities or in foreign exchange denominated instruments. Separately, a bank can be trading for its own position (the Nostro), either as a market maker or as a trader. The profit (or loss) on securities trading has to be discounted because it is conjectural and incidental to the bank's main activities: deposit taking and loan making.

Most banks deposit some of their assets with other banks. This is normally considered to be a way of spreading the risk. But in highly volatile economies with sickly, underdeveloped financial sectors, all the institutions in the sector are likely to move in tandem (a highly correlated market). Cross deposits among banks only serve to increase the risk of the depositing bank (as the recent affair with Toko Bank in Russia and the banking crisis in South Korea have demonstrated).

Further closer to the bottom line are the bank's operating expenses: salaries, depreciation, fixed or capital assets (real estate and equipment) and administrative expenses. The rule of thumb is: the higher these expenses, the worse. The great historian Toynbee once said that great civilizations collapse immediately after they bequeath to us the most impressive buildings. This is doubly true with banks. If you see a bank fervently engaged in the construction of palatial branches stay away from it.

All considered, banks are risk traders. They live off the mismatch between assets and liabilities. To the best of their ability, they try to second guess the markets and reduce such a mismatch by assuming part of the risks and by engaging in proper portfolio management. For this they charge fees and commissions, interest and profits which constitute their sources of income. If any expertise is attributed to the banking system, it is risk management. Banks are supposed to adequately assess, control and minimize credit risks. They are required to implement credit rating mechanisms (credit analysis), efficient and exclusive information-gathering systems, and to put in place the right lending policies and procedures. Just in case they misread the market risks and these turned into credit risks (which happens only too often), banks are supposed to put aside amounts of money which could realistically offset loans gone sour or non-performing in the future. These are the loan loss reserves and provisions. Loans are supposed to be constantly monitored, reclassified and charges must be made against them as applicable. If you see a bank with zero reclassifications, charge off and recoveries either the bank is lying through its teeth, or it is not taking the business of banking too seriously, or its management is no less than divine in its prescience. What is important to look at is the rate of provision for loan losses as a percentage of the loans outstanding. Then it should be compared to the percentage of non-performing loans out of the loans outstanding. If the two figures are out of kilter, either someone is pulling your leg or the management is incompetent or lying to you. The first thing new owners of a bank do is, usually, improve the placed asset quality (a polite way of saying that they get rid of bad, non-performing loans, whether declared as such or not). They do this by classifying the loans. Most central banks in the world have in place regulations for loan classification and if acted upon, these yield rather more reliable results than any management's "appraisal", no matter how well intentioned. In some countries in the world, the Central Bank (or the Supervision of the Banks) forces banks to set aside provisions against loans of the highest risk categories, even if they are performing. This, by far, should be the preferable method.

Of the two sides of the balance sheet, the assets side should earn the most attention. Within it, the interest earning assets deserve the greatest dedication of time. What percentage of the loans is commercial and what percentage given to individuals? How many lenders are there (risk diversification is inversely proportional to exposure to single borrowers)? How many of the transactions are with "related parties"? How much is in local currency and how much in foreign currencies (and in which)? A large exposure to foreign currency lending is not necessarily healthy. A sharp, unexpected devaluation could move a lot of the borrowers into non-performance and default and, thus, adversely affect the quality of the asset base. In which financial vehicles and instruments is the bank invested? How risky are they? And so on.

No less important is the maturity structure of the assets. It is an integral part of the liquidity (risk) management of the bank. The crucial question is: what are the cash flows projected from the maturity dates of the different assets and liabilities and how likely are they to materialize. A rough matching has to exist between the various maturities of the assets and the liabilities. The cash flows generated by the assets of the bank must be used to finance the cash flows resulting from the banks' liabilities. A distinction has to be made between stable and hot funds (the latter in constant pursuit of higher yields). Liquidity indicators and alerts have to be set in place and calculated a few times daily. Gaps (especially in the short term category) between the bank's assets and its liabilities are a very worrisome sign.

But the bank's macroeconomic environment is as important to the determination of its financial health and of its creditworthiness as any ratio or micro-analysis. The state of the financial markets sometimes has a larger bearing on the bank's soundness than other factors. A fine example is the effect that interest rates or a devaluation have on a bank's profitability and capitalization. The implied (not to mention the explicit) support of the authorities, of other banks and of investors (domestic as well as international) sets the psychological background to any future developments. This is only too logical. In an unstable financial environment, knock-on effects are more likely. Banks deposit money with other banks on a security basis. Still, the value of securities and collaterals is as good as their liquidity and as the market itself. The very ability to do business (for instance, in the syndicated loan market) is influenced by the larger picture. Falling equity markets herald trading losses and loss of income from trading operations and so on.

Perhaps the single most important factor is the general level of interest rates in the economy. It determines the present value of foreign exchange and local currency denominated government debt. It influences the balance between realized and unrealized losses on longer-term (commercial or other) paper. One of the most important liquidity generation instruments is the repurchase agreement (repo). Banks sell their portfolios of government debt with an obligation to buy it back at a later date. If interest rates shoot up the losses on these repos can trigger margin calls (demands to immediately pay the losses or else materialize them by buying the securities back). Margin calls are a drain on liquidity. Thus, in an environment of rising interest rates, repos could absorb liquidity from the banks, deflate rather than inflate. The same principle applies to leverage investment vehicles used by the bank to improve the returns of its securities trading operations. High interest rates here can have an even more painful outcome. As liquidity is crunched, the banks are forced to materialize their trading losses. This is bound to put added pressure on the prices of financial assets, trigger more margin calls and squeeze liquidity further. It is a vicious circle of a monstrous momentum once commenced.

But high interest rates, as we mentioned, also strain the asset side of the balance sheet by applying pressure to borrowers. The same goes for a devaluation. Liabilities connected to foreign exchange grow with a devaluation with no (immediate) corresponding increase in local prices to compensate the borrower. Market risk is thus rapidly transformed to credit risk. Borrowers default on their obligations. Loan loss provisions need to be increased, eating into the bank's liquidity (and profitability) even further. Banks are then tempted to play with their reserve coverage levels in order to increase their reported profits and this, in turn, raises a real concern regarding the adequacy of the levels of loan loss reserves. Only an increase in the equity base can then assuage the (justified) fears of the market but such an increase can come only through foreign investment, in most cases. And foreign investment is usually a last resort, pariah, solution (see Southeast Asia and the Czech Republic for fresh examples in an endless supply of them. Japan and China are, probably, next).

In the past, the thinking was that some of the risk could be ameliorated by hedging in forward markets (=by selling it to willing risk buyers). But a hedge is only as good as the counterparty that provides it and in a market besieged by knock-on insolvencies, the comfort is dubious. In most emerging markets, for instance, there are no natural sellers of foreign exchange (companies prefer to hoard the stuff). So forwards are considered to be a variety of gambling with a default in case of substantial losses a very plausible way out.

Banks depend on lending for their survival. The lending base, in turn, depends on the quality of lending opportunities. In high-risk markets, this depends on the possibility of connected lending and on the quality of the collaterals offered by the borrowers. Whether the borrowers have qualitative collaterals to offer is a direct outcome of the liquidity of the market and on how they use the proceeds of the lending. These two elements are intimately linked with the banking system. Hence the penultimate vicious circle: where no functioning and professional banking system exists no good borrowers will emerge.