Banking Soundness and Monetary Policy

10 Problems of Bank Soundness: Mexico’s Recent Experience

Charles Enoch, and J. Green
Published Date:
September 1997
  • ShareShare
Show Summary Details

The factors behind Mexico’s banking crisis should be identified in order to reach a clear diagnosis and, therefore, a clear prescription for handling problems of this sort or, better still, preventing future ones. With the benefit of hindsight, the origins of the crisis can be traced to a combination of factors that, although difficult to disentangle, contributed to the fragility of the banking sector.

Mexico’s successful stabilization and structural reform in the late 1980s and early 1990s, including the deregulation of financial intermediation and privatization of the banking system—encouraged by international organizations and supported by prevailing market opinion—fueled a large expansion of the supply of loanable funds. Unfortunately, this occurred before banks had developed appropriate internal controls and before prudential regulation and supervision were adequate to contain the increased risk of new or expanded commercial banking activities. No one was fully aware of these weaknesses. This paper takes up each of these factors below.

Rapid Expansion of Bank Credit

Bank credit expanded rapidly from 1988 to 1994. First, the consolidation of public finances implied that the government no longer needed to absorb bank resources in order to finance its deficit. This allowed the gradual elimination of reserve requirements and government borrowing from banks and increased the availability of financial resources to the private sector.

Second, the liberalization of the financial and other sectors, as well as the successful macroeconomic stabilization and the privatization of public enterprises, brought about an important process of financial deepening, attracting very large flows of foreign capital to the country. These policies were widely applauded, mostly by participants in the so-called Washington consensus.

The consolidation of public finances shifted banks’ lending behavior away from financing the government and toward riskier lending to the private sector. As banks were reprivatized in 1991 and 1992, their risk evaluation staffs, inherited from the period when the government owned the banks or brought in by new private owners, were not in all cases prepared to operate in this new environment. During the years when bank lending was primarily directed to the government, the risk evaluation function had been overlooked, and organizational and information systems were not adequate to assess credit and market risks. Neither the public servants administering the commercial banks nor the supervisory authorities worried much about loan portfolio risk. In fact, by the end of the 1980s, when financial liberalization accelerated, bank credit to the private sector represented only a small proportion of banks’ total assets.

As of late 1988, Mexico’s financial system underwent deep structural changes. Banks were allowed to tap deposits at market interest rates and to participate fully in the government securities market. Moreover, quantitative credit controls were eliminated. As a result of both macroeconomic stabilization and financial liberalization, financial deepening took place. The year-end ratio of the stock of the broadest monetary aggregate (M4) to GDP increased from 32.3 percent in 1988 to 51.3 percent in 1994. This meant that the availability of loanable funds increased significantly.

In addition to the greater availability of domestic funds for lending, Mexico received important inflows of capital from 1990 to early 1994 in response to the policy reforms and the global trend of capital flows to emerging markets. During this period, improved expectations about the future course of the economy caused inexperienced bank credit officials and bank managers to confuse good and bad credit risks. In such an environment, too many borrowers seemed to have profitable projects and reasonably good prospects for servicing their debts. In turn, potential borrowers who had forgotten how to use credit prudently or who had never had any experience in paying debts, were not very cautious either.

Between 1988 and 1994, the stock of bank credit to the private sector approximately quadrupled as a percentage of GDP, albeit starting from a very low base by international standards. The ratio, measured at year end, increased from 13.4 percent of GDP in 1988 to 50.7 percent in 1994. At the time, the expanded supply of bank credit to the private sector was viewed, both domestically and abroad, as a highly favorable development. At last, the private sector had access to the resources needed to increase production and improve living standards.

Available funds found ready takers. Private sector demand for bank credit was strong for several reasons:

  • (1) Favorable expectations of Mexican economic performance;
  • (2) The need to modernize and expand Mexico’s industrial plant in light of increased competition resulting from a more open economy;
  • (3) The appalling housing shortage;
  • (4) Higher perceived permanent income;
  • (5) Consumers wishing to replace obsolete stocks of durable goods; and
  • (6) The need to finance the acquisition of public enterprises being privatized.

The factors that increased the supply of bank credit, and supported the private sector’s demand for it, were desirable outcomes of Mexico’s stabilization and structural reforms. Yet together these and other factors led to the surge of nonperforming loans.

Several developments accompanied the expansion of bank credit to the private sector. On the margin, the structure of credits to this sector shifted conspicuously toward financing consumer spending (particularly durables) and housing. Also, credits denominated in dollars grew rapidly through the beginning of 1992, when the central bank imposed restrictions on the increase of foreign currency liabilities of commercial banks.

These developments brought increased exposure to risk. Unlike government debt, private debt is not riskless. In addition, consumer and mortgage credit tend to be riskier than loans to firms, as legal proceedings in case of default are particularly cumbersome in the former cases, often rendering collateral on credits of little use. Moreover, dollar-denominated credits were subject to exchange rate risk.

Several prudential measures were adopted during the period. As previously mentioned, the central bank imposed limits on commercial bank liabilities denominated in foreign currency. Commercial banks were required by the Mexican treasury to observe the Basle capital standards. Stricter provisions for loan losses were prescribed by the National Banking Commission and, by late 1993, moral suasion by the central bank prompted a good number of banks to apply more rigorous standards for credit extension. In general, supervision was tightened. However, the supervisory ability of the National Banking Commission had waned during the long years when banks lent little to the private sector.

By 1993, the net indebtedness of the private sector became increasingly burdensome due to a slowdown of the economy, diminishing employment opportunities, and persistently high loan rates from an all too active lending-borrowing process. As a result, many commercial bank loans that, from the start, were of dubious quality began to deteriorate rapidly. To put additional pressure on an already complex situation, interest rates further increased after March 1994 and the peso depreciated considerably as a consequence of interest rate hikes in the United States and unfortunate events in the Mexican political arena.

The Mexican banking system was already fragile before December 1994. A systemwide insufficiency of capital was becoming evident, a phenomenon explained by the relatively high level of past due loans that had not been adequately provisioned. Moreover, some commercial banks were operating with serious problems, which were not readily noticeable from the information disclosed to financial authorities. In some instances, bank administrators disregarded existing regulations and proper banking standards.

Managing the Banking Crisis

Although the macroeconomic repercussions of the December 1994 devaluation were very damaging to commercial banks, the immediate impact was limited. It is fitting to recall that the Bank of Mexico had already imposed a ceiling on the amount of foreign currency-denominated liabilities that banks could assume. This measure, adopted in early 1992, was in part based on the consideration that the central bank cannot, even if needed, indefinitely perform its function as a lender of last resort with respect to banks’ liabilities denominated in foreign currency. Additionally, it was estimated that limiting the banks’ capacity to contract foreign debt would help moderate the expansion of credit and, thereby, domestic demand and the current account deficit. Consequently, the run on banks’ external obligations in early 1995 was somewhat mitigated by the existence of these regulations. The harmful effects of the devaluation were also lessened by another ordinance that imposed a ceiling on banks’ open foreign exchange positions.

Nonetheless, the devaluation prompted other deleterious effects: inflation and interest rates soared, economic activity collapsed, the burden of servicing credits denominated in both foreign and domestic currency increased, and banks’ capital ratios fell. By February 1995, half of Mexico’s banks were not complying with the Basle capital adequacy standards.

In order to deal with these problems, the monetary and banking authorities, through the Bank Fund for the Protection of Savings (FOBAPROA), implemented several programs to help solve difficulties faced by financial intermediaries that could have had systemic repercussions. These actions were guided by the ultimate goal of protecting the real sector of the economy and the immediate objective of maintaining and strengthening the financial system.

The different schemes to support the banking system and indebted bank clients were developed with the following principles in mind:

  • (1) Prevent systemic risk;
  • (2) Protect the legitimate interest of depositors and other bank creditors;
  • (3) Help debtors strained by the macroeconomic crisis;
  • (4) Resist pressures to bail out the stockholders of financial institutions;
  • (5) Avoid expansion of the central bank’s credit;
  • (6) Minimize the fiscal costs of the crisis management and distribute these costs over a number of years;
  • (7) Interfere the least possible with the normal functioning of markets, but take care, in order to minimize moral hazard problems, not to create perverse incentives for debtors and stockholders; and
  • (8) Implement programs that are both simple and transparent so as to foster confidence in the adopted measures.

Programs and Facilities

The package adopted to deal with the banking crisis in Mexico responded to many specific needs. In early 1995, a dollar liquidity facility and a temporary capitalization program were launched. Thereafter, programs to help debtors directly were made available. The objective was to ease the burden of servicing debt and to foster responsible payment practices by debtors. Direct support to debtors (homeowners, agricultural enterprises, and small and medium-sized firms) was provided through the absorption by the government of a portion of the interest rate and capital burdens.

For troubled yet viable banks, a number of programs were introduced to induce current and potential stockholders to inject fresh capital into these banks. To this end, the government purchased two pesos worth of a qualified loan portfolio for every peso of new capital paid in by stockholders. Nonetheless, commercial banks have continued administering the loans and have shared risks with the government.

Furthermore, restrictions regarding participation in the capital of commercial banks have been eased with the aim of attracting new investors, both foreign and domestic. Nonviable banks are intervened, brought back to health and resold.

Dollar Liquidity Facility

The devaluation of the peso caused a loss of confidence in Mexico’s financial system, and commercial banks faced difficulties in refinancing their dollar-denominated certificates of deposit (CDs) placed abroad. The purpose of the dollar liquidity facility was to stop, and eventually reverse, the run on the external liabilities of commercial banks. Relief took the form of dollar loans granted by the Bank of Mexico to commercial banks to enable them to pay all foreign currency liabilities as they came due. The high interest rates charged on these loans encouraged commercial banks to pay them back as quickly as possible and to look for cheaper sources of financing. In order to further encourage the prompt repayment of these loans, it was established that early repayment of even a portion of these loans would reduce the interest charged on some fraction of the outstanding balance.

Seventeen commercial banks benefitted from this program. At its peak in April 1995, the outstanding credit granted through this facility reached US$3.9 billion. By September of that year, all banks had paid off these debts by taking advantage of the renewed access to international financial markets.

Temporary Capitalization Program

The devaluation of the peso increased the domestic currency value of bank loans denominated in foreign currency. Consequently, the capital/asset ratio fell below the stipulated minimum. In addition, the stricter preventive reserve requirements enacted in February 1995 meant that many banks needed additional capital. To remedy this situation, FOBAPROA purchased subordinated debt instruments issued by commercial banks with capital/asset ratios below 8 percent. These debentures were mandatorily convertible to capital after five years. However, anticipated conversion was to take place should a bank’s capital ratio deteriorate beyond certain parameters. The debt instruments were acquired by FOBAPROA with resources obtained from the central bank. However, in order to prevent an unwarranted expansion of net domestic credit of the Bank of Mexico, commercial banks had to deposit in the central bank the resources thus obtained.

Under this program1 banks enjoyed a relatively long time to obtain fresh capital and pay their debts to FOBAPROA, However, banks’ stockholders might have lost their investment, or seen their participation diluted, had FOBAPROA become an owner. In that eventuality, FOBAPROA would have been a temporary shareholder and immediately proceeded to find new suitable owners.

It is worth mentioning that when faced with the mere possibility of having to rely on PROCAPTE’s resources, some banks made an effort to increase their capital immediately. Five commercial banks obtained support through PROCAPTE, three of which have already paid in full their subordinated debt.

Investment Units

High inflation causes a serious problem that is seldom understood, particularly in countries that have not experienced inflation for a long time. This problem, which is not contractually foreseen, consists of the accelerated amortization of credits in real terms.

Perhaps the easiest way to explain this phenomenon would be to imagine what happens, during inflation, to a loan at perpetuity. It is evident that the value of the principal would remain constant in nominal terms but not in real terms. With inflation, the real value of the principal continuously deteriorates. Except when highly negative real interest rates prevail, the real amortization will be paid by debtors through interest payments. In this regard, it is necessary to remember that at times of high inflation, a large portion of interest payments goes to compensate the creditor for the erosion of the real value of the principal lent.

To deal with this serious problem, Mexico introduced the UDI,2 a constant value unit of account that is used to denominate credits. The daily value of the UDI reflects the behavior of the consumer price index, albeit with a short but inevitable lag. Consequently, the value of credits denominated in UDIs remains constant in real terms with regard to both principal and interest. In other words, credits denominated in UDIs are protected against the accelerated amortization caused by inflation.

The authorities had to step in to facilitate the restructuring of a large number of viable loans and encourage their denomination in UDIs, Thus, the loan restructuring mechanism involves the government—through FOBAPROA—and commercial banks. The latter bear the credit risk, while the government bears the interest rate risk stemming from the conversion of credits into UDIs.

The banks’ loan portfolios are transferred to trusts administered by the banks themselves. The trusts restructure the loans and denominate them in UDIs. To this end, the trusts obtain loans from the government, denominated in UDIs as well. In exchange for their loan portfolios, commercial banks acquire bonds issued by the government. Thus, peso-denominated liabilities of commercial banks remain matched by assets in the same currency.

In turn, the government ends up with domestic currency-denominated liabilities (the bonds acquired by banks) and UDI-denominated assets (the loans to the trusts). Consequently, the government bears the risk that the interest rates charged on its UDI loans will be lower than the real interest rates it pays on the aforementioned bonds.

Loans restructured in UDIs have medium- and long-term maturities, so that the potential fiscal cost is spread over a number of years. By the end of 1996, about Mex$167,000 million of the loan portfolio had been restructured in UDIs and approximately 375,000 debtors had benefited from this program.

Permanent Capitalization Measures

To facilitate bank capitalization and attract new domestic and foreign investors, certain legal restrictions concerning participation in the capital of commercial banks have been eased. The participation of Mexican legal entities in banks has been liberalized, and some of the obstacles for foreign banks based in NAFTA countries (those in the North American Free Trade Agreement) to acquire control over Mexican financial institutions have been removed. As of today, three medium-sized Mexican bank are already controlled by prestigious foreign banks. In addition, two large banks and one medium-sized domestic bank have opted for strategic partnerships with prominent foreign banks. Nevertheless, the new law does not permit full foreign ownership or control of any bank accounting for 6 percent or more of the banking system’s capital. At present, this precludes foreign ownership or control of the three largest Mexican banks.

In addition, a scheme has been implemented to assist institutions to remain sound, with adequate provision and capitalization levels. The federal government, through FOBAPROA, provides an incentive to maintain these levels by offering to acquire a fraction of a bank’s loan portfolio in a two-to-one ratio of new capital paid in by stockholders. In exchange for its portfolio, the bank must buy long-term bonds bearing market interest rates. The responsibility of administering the loans remains with the financial institution.

By the end of 1996, 12 banks had resorted to this facility for a total of Mex$119,300 million. This mechanism has induced banks to increase their capital by about 158 percent as compared to the December 1994 level.

Programs in Support of Debtors

The sharp increase in interest rates in 1994 and 1995, and the recession in the latter year, produced a substantial increase in the debt servicing cost for toll road concessionaires and a pronounced decline in highway traffic. To address this problem, the government introduced a program by which the toll road concessionaires may restructure their loans. The government, the banks, and the concessionaires assess the viability of each toll road with the aim of identifying the sustainable debt level. In the end, the government takes over that portion of debt that cannot be serviced out of the net income produced by the roads.

In August 1995, the Mexican government also introduced an unprecedented debt relief program (ADE)3 targeted at consumer, credit card, small business, agricultural, and mortgage loan debtors. This program addressed the difficult situation faced by a large number of debtors affected by high interest rates and the pronounced fall in economic activity and employment. The program sought to encourage loan restructuring and to foster responsible payment practices by providing debtors with interest relief and a legal truce.

By helping debtors remain current, the program aimed at reducing the systemic risk to the banking system stemming from widespread default, while maintaining a case-by-case approach to debt restructuring negotiations. Within this framework, the government shared losses with banks, seeking to minimize fiscal costs and distribute them over time, while preventing further distortions in credit markets. In addition, the program has entailed no monetary expansion, has helped to improve the quality of commercial banks’ loan portfolios, and has induced payment discipline and reductions in required provisions. The program established clear cut-off dates, since it only covers loans outstanding as of August 31, 1995.

The main financial components of this program were a cap on loan rates for a predetermined period (September 1, 1995 to September 1, 1996) and the restructuring of mortgage loans in UDIs to alleviate cash flow problems. Up to 16 percentage points of the difference between the concessional interest rate, which varied depending on the type of loan, and the market reference rate was to be absorbed by the government. If such limit had been surpassed, commercial banks would have borne half of the additional cost.

The ADE program has worked successfully. A large number of the targeted credits has been restructured. Moreover, the cost of the program has been considerably lower than originally estimated, since the difference between the concessional interest rate and the market rate has been narrower than anticipated.

On May 16, 1996, the government launched another program aimed at relieving pressure in the housing market, the mortgage loan support program.4 The main objective of this program is to ease mortgage borrowers’ heavy burden due to high interest rates and reduced purchasing power as a result of the economic crisis. Because of the recession, property values fell, in many cases below the outstanding amount of the respective loans. Under these circumstances, most mortgage debtors had no incentive to remain current in their payments.

In order to address this situation the program offers several inducements to avoid default:

  • (1) A 30 percent discount is available on payments due in 1996, which gradually declines to 5 percent in 2005 and applies only to the first 500,000 UDIs of the total loan.
  • (2) The amount available to restructure mortgages in UDIs has been increased by 43,000 million UDIs, to a total of 100,000 million UDIs (about US$22.3 billion as of mid-January 1997).
  • (3) A scheme of minimum payments equivalent to rent has been designed to help mortgage debtors that, notwithstanding the benefit of the discount, are unable to service their loans. Under this scheme, the debtor surrenders the property as full payment of the loan but retains the right to repurchase it.
  • (4) The federal government pays banks the difference between the original contractual payment and the new discounted payment. The government covers the difference in either cash or five-year credits bearing an interest rate equal to that on 91-day cetes (treasury bills).

The agricultural sector has also been severely affected by the economic crisis, and on July 23, 1996, the Mexican government established a loan restructuring mechanism (FINAPE)5 with discounts of up to 40 percent, depending on the amount of the outstanding loan. The range goes from Mex$500,000 to Mex$4 million.

As with other programs, the cost of the discounts will be shared between the government and banks. However, since the program also seeks to promote flows of new financial resources to the agricultural sector, the cost borne by the government will depend on the amount of new resources banks inject into the sector. The larger the flow of resources, the greater will be the costs absorbed by the government.

On August 16, 1996, the government launched a financial support program for small- and medium-size firms (FOPYME).6 This program was designed to alleviate the heavy financial burden of small, viable firms and to help them finance their current activities. The discounts under the program range from 17 to 30 percent, depending on the amount of the outstanding loan. To be eligible, loans must be under Mex$6 million and have been contracted as of July 31, 1996.

Under this scheme, banks have agreed to promote economic activity by providing up to Mex$ 13,000 million in new financing to creditworthy firms. Banks are committed to maintaining these lines of credit in real terms for at least three years.

In the case of revolving credits, the program contemplates reductions between 5 and 22 percentage points in the contractual interest rate, depending on the amount outstanding. However, interest rates after the discount cannot be lower than 15 percent.

Other Measures to Cope with the Financial Crisis

In March 1996, Mexico announced the creation of Valuación y Venta de Activos (VVA), an agency similar to the U.S. Resolution Trust Corporation. This agency will appraise and sell the assets FOBAPROA acquires from the banks. In the process, VVA will promote the development of markets for banks’ debt instruments and other financial and nonfinancial assets.

The Mexican monetary authorities have also continued to strengthen the regulatory and supervisory framework through significant reforms in the financial sector. First, the National Banking and Securities Commission has improved banking supervision procedures and methodologies. Inspection procedures are being brought up to international standards. On-site, qualitative analysis of banks is being combined with off-site parametric analysis.

Second, stricter accounting standards are being adopted. In July 1996, banks began reporting their statements (on a preliminary and confidential basis) following a methodology similar to the generally accepted accounting principles in the United States. Beginning January 1997, reporting under this standard is mandatory, although some provisions will be introduced gradually.

Third, loan loss provisions have been tightened. Banks are required to provision the larger of 60 percent of their overdue portfolios or 4 percent of their total loan portfolios.

Finally, methodologies for the assessment of loan portfolios are being improved, and stricter limits have been set on lending to related interests. New capitalization rules, following the Basle criteria, impose capital requirements that take both credit and market risks into consideration.

Fiscal Impact

From the onset of the banking crisis, the government has been careful to quantify the cost of supporting the banking system and its debtors. In keeping with the objective of maintaining sound public finances, the government has effected the necessary adjustments to the budget to make room for these expenditures.

At present, the Finance Ministry has estimated that the total fiscal cost of the programs for debtors and banks will amount to Mex$210,300 million in present-value terms, which is equivalent to 8.4 percent of GDP in 1996. This cost will be borne over the next thirty years, the lifespan of the programs. Nevertheless, the exact cost cannot be known with certainty, since it depends on different contingent factors and since some of the costs may be recovered through future increases in asset prices. Table 1 shows the estimated cost of each program.

Table 1.Estimated Costs of Support Programs for Debtors and Banks
ProgramTotal Costs Net Present Value (Millions of pesos)Percentage of 1996 GDP
Capitalization schemes39,0001.6
Toll roads26,1001.0
Mortgage loans additional benefits27,2001.1
Source: Criterios Generales de Politica Ecóndmica para 1997, SHCP (Ministry of Finance).
Source: Criterios Generales de Politica Ecóndmica para 1997, SHCP (Ministry of Finance).

The fiscal surplus generated by the Government of Mexico in 1995 and 1996, which amounts to 1.6 percent of 1996 GDP, has been applied to the support programs. Consequently, the remaining cost of the support programs is estimated to be 6.8 percent of 1996 GDP. This amount appears manageable considering that Mexico’s ratio of net public debt to GDP is relatively low by international standards. At present, this ratio amounts to 33 percent of GDP.

Monetary Impact

Great care has been taken to ensure that the support granted by the Bank of Mexico to the banking system, as lender of last resort, does not produce monetary expansion. Hence, the central bank absorbs liquidity from the banking system so as to offset the resources it provides to FOBAPROA. Since the Bank of Mexico has had a large creditor position vis-à-vis commercial banks, the sterilization operation has been implemented simply by not renewing some of the loans that mature daily. Recently, the central bank has begun taking deposits from commercial banks, selling cetes or conducting repurchase agreement operations in order to absorb liquidity.

Final Comments

In hindsight, the Mexican financial system was overwhelmed by the volume of resources—foreign and domestic—that became available as a result of the structural reforms undertaken in the country. From an operational and managerial standpoint, banks were ill-prepared to handle the extraordinary expansion of credit during the 1988-94 period. Likewise, bank supervision did not keep abreast of developments in the market. Poor loan portfolios were the end result.

To a certain extent, these problems can be traced to the nationalization of banks in 1982. As a result of the nationalization, commercial banks lost many experienced officers who did not approve of the measure and who did not want the government as an employer. Furthermore, bankers who remained at their posts, or who were hired or appointed by public administrators, did not have access to the best training in the credit business, because throughout most of the 1980s loanable funds were almost entirely channeled to the federal government, whose nominal deficit—albeit not always its real deficit—was enormous.

When banks were reprivatized in the early 1990s, the new “teams” did not always include seasoned bankers. In retrospect, it is easy to condemn the fact that banks were allowed to be managed by these new teams. At the time it was not evident, however, that some of these bankers would be reckless, nor were many other candidates in sight.

Nonetheless, the actions undertaken by Mexico to overcome the financial crisis have yielded positive results. As systemic risk in the financial system has been eliminated, through government support and increased capitalization, the confidence of markets and of the public in financial institutions has been maintained. The debt relief programs have alleviated the difficult situation faced by many debtors, thus reducing or eliminating the growth of nonperforming loans. The fiscal cost of all the schemes to support debtors and the banking system has been limited to massageable proportions and will be distributed over many years. Moreover, the important process of opening Mexico’s financial system, which has deepened since the onset of the crisis, will enhance competition in this sector. In turn, this should lower financing costs and provide a boost to investment.

With all the programs now in place, it is our expectation that the problems of the banking system and its debtors will be gradually solved. The steady recovery of economic activity and the reduction in inflation and interest rates—all of which look promising—will be key factors to overcoming the difficulties in the financial system in the near future.


Programs de Capitalización Temporal (PROCAPTE)


Unidad de Inversión (UDI).


Acuerdo de Apoyo Immediato a los Deudores de la Banca (ADE).


Programa de Beneficios Adicionales a los Deudores de Créditos para Vivienda.


Acuerdo para el Financiamiento del Sector Agropecuario y Pesquero (FINAPE).


Acuerdo de Apoyo Financiero y Fomento a la Micro, Pequeña y Mediana Empresa (FOPYME).

    Other Resources Citing This Publication