This report includes five background studies with emphasis on vulnerabilities and growth, the focus of the 2006 Article IV Consultation with Uruguay. Stocks of key financial balance sheet vulnerabilities are also discussed. With the appreciation of the peso since early 2004, the discussion on competitiveness has intensified. To assess competitiveness, the paper looks at balance of payments trends, the ratio of tradable to nontradable prices, cost and profitability measures, and real exchange rates and their alignment with purchasing power parity (PPP).

Abstract

This report includes five background studies with emphasis on vulnerabilities and growth, the focus of the 2006 Article IV Consultation with Uruguay. Stocks of key financial balance sheet vulnerabilities are also discussed. With the appreciation of the peso since early 2004, the discussion on competitiveness has intensified. To assess competitiveness, the paper looks at balance of payments trends, the ratio of tradable to nontradable prices, cost and profitability measures, and real exchange rates and their alignment with purchasing power parity (PPP).

IV. Post-Crisis Credit: Facts, Lessons, and Prospects4

A. Introduction

1. Since its sharp contraction around the 2002 crisis, bank credit to the private sector has remained low, possibly posing a problem for medium-term growth. The fall has been of about 60 percent in real terms. While post-crisis recoveries often happen with virtually no resumption in credit (see Calvo, Izquierdo, and Tanzi, 2006), a lasting lack of financial intermediation could pose an obstacle to medium-term growth. A growing body of evidence suggests that financial intermediation not only correlates with, but also causes growth (see, for example, Levine, 2004).

2. This paper explores the evolution of credit, drawing lessons from other countries to assess likely prospects for economic activity and its financing. Specifically, the paper asks the following questions: How has bank credit, and to the extent measurable, other sources of financing evolved since the crisis? What are the reasons behind this behavior? Has there been a credit crunch? How does Uruguay’s experience compare with that of other countries suffering banking crises? What do these developments imply for investment and output? What policy lessons can we learn from other countries?

3. The low credit flows reflect both supply and demand factors, and the evidence does not suggest the prevalence of credit rationing. On the supply side, banks have been subject to stricter prudential rules. Most foreign-owned institutions have become risk-averse after the crisis. Banks have been reluctant to lend to indebted borrowers and their lending capacity has decreased with the fall in overall deposits and the rise in the share of short-term deposits. On the demand side, some companies may have been postponing investments in the face of large spare capacity. An econometric approach using a disequilibrium framework suggests that since 2003, demand has constrained credit growth, but the results should be interpreted with caution given data and specification difficulties.

4. Peso credit is rising. The dollarization of credit is falling: while new peso credit flows are positive, dollar-denominated credit is declining. Credit dollarization has fallen more strongly in the nontradable sector, reducing vulnerabilities.

5. The present lack of credit should not endanger the continuation of the recovery. International experience shows that after banking and currency crises, companies typically find other sources of finance. Credit then picks up gradually after about three years. In Uruguay, the economic recovery is likely to continue despite the current low lending levels for several reasons. First, alternative sources of credit are helping, at least in part, to finance the recovery. In retrospect, domestic bank credit has not been the only source of finance for Uruguayan economic activity. Moreover, the impact of the drop in bank lending on activity may not be as severe as in other countries, as industries that typically depend heavily on financing do not explain a large fraction of output in Uruguay. Second, in contrast to the post-crisis experience of some other countries, where undercapitalized banks were unable to lend, banks in Uruguay are well capitalized, with the exception of housing bank BHU.

6. Further institutional reforms and the use of trust funds could help in facilitating access to finance. Further strengthening of the financial system would consolidate confidence, yielding an increase in the maturity of deposits. Improving the credit registry, bankruptcy procedures, and corporate transparency could help improve the credit culture and ultimately financial intermediation. More developed short-term money markets could help banks reduce their liquidity holdings and lend. Dedollarization and the development of hedging instruments would reduce currency mismatches, allowing banks to lend with lower credit risk. Trust funds (fideicomisos) could be useful financing instruments while balance sheets are still recovering.

B. Evolution of Credit to the Private Sector

7. After falling for three years, credit stocks have now stabilized. Bank credit to the private sector has dropped by 60 percent in real terms from its peak in 2001, and has only recently stopped falling.5

A04ufig01

Credit to the Resident Private Sector

Citation: IMF Staff Country Reports 2006, 427; 10.5089/9781451839364.002.A004

8. New credit flows are slowly recovering but remain significantly below precrisis levels. They have displayed a slightly positive trend in real terms since 2003, growing about 4 percent a year, but are still only about half their 2001 levels.

9. After increasing sharply during the crisis, nonperforming-loan ratios (NPL) have fallen significantly. The NPL ratio has dropped from 60 percent in 2002 to 18 percent in 2005. Excluding BHU, which no longer conducts financial intermediation, NPLs have fallen to less than 5 percent from a crisis peak of “only” 25 percent (Figure 1). This reflects mainly (i) an improvement of economic conditions and ensuing collection of provisioned loans, but also (ii) the carving out of the bad loan portfolio from the banking system through restructuring of private and public banks and (iii) a government guarantee on assets in BROU’s balance sheet backed with nonperforming loans allowing BROU to classify them as performing.

Figure 1.

Lending to Private Sector Around Selected Banking Crises

Citation: IMF Staff Country Reports 2006, 427; 10.5089/9781451839364.002.A004

Source: IFS and Caprio and Klingebiel (2005). Levels of bank credit are normalized to equal 100 in the pre-crisis year and calculated at constant exchange rates thereafer.
A04ufig02

Total Real Credit to the Private Sector

(In millions of pesos of 1997)

Citation: IMF Staff Country Reports 2006, 427; 10.5089/9781451839364.002.A004

10. Dollarization of credit is falling, reducing vulnerabilities somewhat. Peso-denominated lending increased by about 14 percent in real terms in 2005, helping bring down the degree of dollarization of credit from 83 percent in 2002 to 69 percent in December 2005 (part of this reduction also reflects peso appreciation). The dollarization of new loans has also followed a similar trend. Excluding BHU, whose stock of loans is indexed to wages, the degree of dollarization has reverted to its precrisis average level of 80 percent. Credit dollarization has fallen more strongly in the nontradable sector, reducing currency mismatches between revenues and costs. The fraction of foreign-currency debt in the construction sector, for example, has fallen from 96 percent in 2002 to 81 percent in 2005, and in commerce, it has dropped from 97 percent to 87 percent. This is due to increases in peso credit as well as a fall in dollar loans.

11. Trade and “other” sources of nonbank credit have become somewhat more important as a source of financing for small firms. The share of trade credit in total debt rose from 17 percent to 22 percent for small companies, and the share of “other” credit (which encompasses debt with private individuals, the government, payables due, provisions, unpaid dividends and early receipts) rose from 43 to 57 percent (Table 1). For medium-sized and large companies, the share of trade credit in total debt fell and the share of “other” credit rose only mildly.

A04ufig03

Credit Flows

(In millions of pesos of 1997)

Citation: IMF Staff Country Reports 2006, 427; 10.5089/9781451839364.002.A004

12. Repatriated capital has likely helped in financing the recovery. Deposits that left the financial system during the 2002 crisis have not returned, remaining as deposits in BIS reporting banks. These deposits, which rose from US$3.6bn prior to the crisis to US$4.5 billion in the second quarter of 2002, kept increasing since then; they could have been used to finance the import of intermediate products and other working capital. However, loans from BIS reporting banks have fallen from US$1.9 billion in early 2002 to about US$1.2 billion in late 2005. Nevertheless, balance of payments data, including the sizeable change in errors and omissions, suggest the return of capital that had fled the domestic financial system, and was partly held “under the mattress.” These funds are likely to have been partly used, without being intermediated through the banking system, to finance consumption and investment.

Financial Structure Indicators of the Corporate Sector in Uruguay: 2001, 2003 and 2004 1/

(In percent)

article image
Source: Staff calculations based on Munyo (2005), Gilli (2005), and De Brun and others (2006).

To capture firm size, we sort each years’s sample of firms into thirds based on total sales. Magnitudes denote median values within each category

C. Has There Been a Credit Crunch?

13. Does a low demand for credit or a limited credit supply explain the low bank credit? To explore this issue, we examine whether demand has exceeded supply at prevailing interest rates. First, we briefly discuss institutional and economic factors that may have contributed to a fall in demand and supply. Next, we explore this question more formally.

Institutional and economic factors

14. On the supply side, banks have been reluctant or restricted to lend:

  • Deposits have fallen slightly. Resident deposits have fallen only mildly since the crisis, with the 20 percent drop in its dollar value offset by a similar real depreciation of the peso over the period.

  • The maturities of deposits have shortened. The share of sight in total nonfinancial private sector deposits has increased from about 30 percent in 2001 to about 65 percent in early 2006, partly because depositors distrust time deposits that were reprogrammed during the crisis. Of course, to some extent this is an endogenous phenomenon, since banks could in principle offer higher rates to attract longer-term deposits. In addition, the maturity of time deposits was not high before the crisis.

  • Banks have faced tighter prudential limits to lending. This applied in particular to public banks, which account for 60 percent of deposits. Since the crisis, the supervisor has forbidden BHU from lending until it complies with prudential rules. BROU and NBC have faced additional liquidity requirements. In general, both peso and dollar reserve requirement rates are now larger than in 2001, but peso rates declined since their peak in 2002. Yet, banks have continued to hold liquid assets exceeding reserve and liquidity requirements.

  • Foreign-owned banks have been under tighter risk-management directives after the crisis. Many foreign banks adopted risk management systems and practices of their parent institutions and anecdotal evidence after the crisis suggests that many foreign-owned banks have become more risk averse.

  • Creditworthy borrowers may still be scarce, and improved prudential regulation has raised the costs of lending to highly indebted borrowers. The rules require banks to classify corporations with nonperforming loans to any single bank as risky. After being hit by the crisis, private firms and households remain indebted, and corporations still have in their balance sheets liabilities to the liquidation funds for the failed banks. They also continue to register loans that banks have written off but remain as off-balance-sheet assets. Only small companies have reduced their indebtedness ratios substantially compared with pre-crisis levels (Table 1). While the debt-to-sales ratio of large firms has remained broadly constant, that of medium-sized firms has risen. Balances sheets now appear to be recovering, making firms gradually more attractive to lenders.6

  • The closure of failed banks may have affected credit supply. Customers of closed institutions may have had difficulties getting credit from other banks.

15. The demand for new bank credit may also have been small, as companie’ willingness to incur further debt has remained limited.

  • a. Output has only now returned to 1998 levels, implying that the demand for investment and credit may have been limited. Companies may have decided to exploit spare capacity, postponing investment decisions, as has happened in many post-crises recoveries.

  • b. After the crisis experience, potential borrowers are limiting leverage. Companies and households may not yet be willing to incur the risks of new debt, particularly in foreign currency.

16. Banking spreads have fallen, suggesting that bank competition has not decreased. Banking spreads tend to rise with the level of interest rates, and rates levels have fallen both for dollar and in peso terms. Nevertheless, for U.S. dollars, deposit rates are close to levels seen in the pre-crisis period, while spreads are now much tighter. While this evidence is not enough to rule out credit rationing, it indicates that competitive pressures may have become stronger.

A04ufig04

Peso Lending and Deposit Rates

(In percent)

Citation: IMF Staff Country Reports 2006, 427; 10.5089/9781451839364.002.A004

A04ufig05

U.S. Dollar Lending and Deposit Rates

(In percent)

Citation: IMF Staff Country Reports 2006, 427; 10.5089/9781451839364.002.A004

An empirical disequilibrium framework

17. This section examines more formally, using a disequilibrium model, the question of whether supply or demand for credit has been the binding constraint. In credit rationing models, such as Stiglitz and Weiss (1981), credit demand can exceed supply at prevailing interest rates because of adverse-selection effects. Banks prefer to ration credit rather than to increase interest rates since higher interest rates induce a selection toward riskier projects. One way to assess empirically whether credit rationinig occurs is the method by Kiefer (1980) and Maddala and Nelson (1974).7 The demand for bank credit CtD need not be equal its supply CtS. The realization of credit is given by:

Ct=min(CtS,CtD)(1)

18. The supply of credit is modeled as a function of interest rates, the health of firms’ balance sheets, lending capacity, and nonperforming loans. Following Ghosh and Ghosh (1999) and Barajas and Steiner (2002), the specification is as follows:

Cts=β0s+β1s(rtlendingrtdeposit)+β2slt+β3sNPLt1+β4sPROVt1+εts,(2)

where:

  • rtlending is the lending rate and rtdeposit is the deposit rate. Both dollar and peso spreads enter the supply function.

  • yt is current output (real GDP, seasonally adjusted), to proxy for the health of firm’ balance sheets

  • lt: Commercial bank’ real lending capacity (defined as the minimum of (i) total bank liabilities plus net worth minus required reserves and liquidity requirements minus equity capital, and (ii) the maximum possible loans permitted by the prevailing capital adequacy requirement and existing bank capital) measured at exchange rates with constant purchasing power.

  • NPLt−1 the ratio of nonperforming loans to total loans in bank books

  • PROVt−1 the ratio of provisions to nonperforming loans

  • and credit is measured as gross credit in the books of banks and cooperatives, at exchange rates with constant purchasing power, deflated by the CPI.

19. The demand for credit is assumed to depend on the real interest rate, expected future output, and macroeconomic conditions. Specifically:

Ctd=β0d+β1drtlending+β2dytgap+β4dΠt+β5dUBIt+εtd,(3)

where:

  • rt: the real lending rate (both for peso and dollar credit).

  • ytgap: the output gap, measured as the deviation of current industrial production from its (Hodrick-Prescott filtered) long-run trend, or, alternatively, the CERES leading indicator of economic activity.

  • IIt: expected inflation (as an indicator of the general macroeconomic environment)

  • UBIt: Country risk, as measured by the Uruguay Bond Index from República AFAP.

As argued in Ghosh and Ghosh (1999), the use of a Hodrick-Prescott filter to estimate trend output provides a simple measure of the output gap. The output gap enters as a proxy for the intuition that, in difficult times, firms may seek more credit. To avoid spurious swings in credit supply stemming from the sharp exchange-rate fluctuations in the sample, we convert foreign currency deposits and loans using an exchange rate that keeps purchasing power of a dollar constant within the sample period. Alternatively, we use a 12-month moving average of the nominal exchange rate to smooth the effect of exchange rate volatility. To allow for a differential behavior during the height of the crisis, in some estimation we include a dummy for this period. Estimation in levels is legitimate as long as the determinants of supply and credit demand form a cointegrating vector, which is true in the present case. Following the notation in Ghosh and Ghosh (1999), the probability that an observation is supply-constrained is given by:

θt=Prob(CtD>CtS)=Φ((CtdCts)/(σs2+σd2))

where σs1 and σs2 are the standard errors of the credit supply and demand functions and Φ(•) is the cumulative normal distribution function. This can be used to derive the unconditional density function of Ct. Maximizing the corresponding likelihood function allows for the estimation of both supply and demand.

A04ufig06

Outstanding Loans and Lending capacity

(In millions of 1997 pesos, at exchange rates with constant purchasing power)

Citation: IMF Staff Country Reports 2006, 427; 10.5089/9781451839364.002.A004

20. The evidence does not suggest that the decline in bank lending after the crisis reflects mainly a rationing phenomenon. As could be expected, lending capacity falls strongly during the crisis due to the difficulties faced by banks and the subsequent tightening of liquidity requirements and increase in demand for liquid assets. In 2003, lending capacity begins to recover, showing a steady and moderate increase, staying below pre-crisis levels and above actual observed loan stocks thereafter. The estimations indicate that since 2003, credit supply at prevailing interest rates has exceeded demand (see figure).8 While both demand and supply fall with the crisis, supply recovers more strongly, driven by the improvement in lending capacity. Details of the estimations can be found in the appendix. The results should be interpreted with caution given specification and data problems. Estimates of individual coefficients vary and the broad qualitative conclusion of the absence of credit rationing in the last 2–3 years appears to be robust to different sample sizes, but not to alternative specifications.9 The estimations were also conducted for peso and dollar credit separately, yielding a similar qualitative picture. The qualitative result also holds when the credit series are adjusted for loan writeoffs and the loan portfolio of the banks liquidated during the crisis (currently in the liquidation funds and under recovery by asset management firms).

A04ufig07

Estimated supply and demand for bank loans

(In millions of 1997 pesos, at constant purchasing power exchange rates)

Citation: IMF Staff Country Reports 2006, 427; 10.5089/9781451839364.002.A004

D. Comparison with International Experience

Macroeconomic developments after banking crises and implications for Uruguay

21. The international experience shows that after a banking crisis, output recovers relatively quickly but bank lending takes longer to resume. Although large variation across country experiences exists, often bank lending takes more than three years after the crisis (Figure 1) to recover (see also Demirgüç-Kunt, Detragiache, and Gupta, 2000, and Hoggarth and Reidhill, 2003). GDP growth typically takes off much quicker (Calvo, Izquierdo, and Talvi, 2006), so credit-to-GDP ratios stay low for an extended time.10 The drop in lending is stronger when the banking crisis coincides with a currency crisis (IDB, 2004) and in dollarized economies (see Gulde and others, 2003).

A04ufig08

Real Private Credit Growth after Banking Crises

(in percent)

Citation: IMF Staff Country Reports 2006, 427; 10.5089/9781451839364.002.A004

Source: IDB and staff

22. The stagnation in credit after banking crises is only partly attributable to a drop in deposits. Following banking crises, deposits fall somewhat, but the fall in bank credit to deposits is much larger (Figure 2). With four years after the onset of the crisis without a recovery of bank credit, the experience of Uruguay appears consistent with that of other countries suffering banking crises.

Figure 2.

Evolution of Deposits and Lending Around Selected Banking Crises.

Citation: IMF Staff Country Reports 2006, 427; 10.5089/9781451839364.002.A004

Source: IFS and Caprio and Klingebiel (2005).

23. The use of spare capacity and the postponement of investment decisions typically explains the recovery of output after crises. Calvo, Izquierdo, and Talvi (2006) show that the capital stock remains relatively constant throughout the collapse-recovery phase, while investment collapses together with output, and recovers weakly when output regains pre-crisis levels. Krueger and Tornell (1999) and Tornell and Westermann (2002) also stress that exporters quickly find access to international credit and capital markets, but producers of nontradable goods needs to use other sources of financing.

24. Several factors suggest the recovery could be sustained, at least for some time:

  • a. Fixed investment has shown an upward trend since early 2003. This suggests the delay in investment decisions and the use of spare capacity have not been the main reasons explaining the low credit demand. Alternative sources of credit must be sustaining, at least in part, the recovery.

  • b. In the past, domestic bank credit was not the only source of finance, and the historical correlation between output and bank credit is not high. While no comprehensive data are available, nondeposit-taking institutions have reportedly provided loans to both consumers and companies.

  • c. In Uruguay, the industries that typically depend substantially on financing, such as machinery and chemicals, are small. Dell’Ariccia, Detragiache, and Rajan (2005) document that the damaging effect of crises on output is stronger for industries that depend more on borrowed funds.

  • d. Except for BHU, Uruguayan banks are well capitalized. Failure to recapitalize banks after crises has sometimes contributed to a prolonged credit crunch (see Gulde and others, 2003).

Seas. Adj. Real Gross Investment

Citation: IMF Staff Country Reports 2006, 427; 10.5089/9781451839364.002.A004

E. What Should Be Done?

25. How can the authorities foster sound credit without undermining market discipline and financial stability? A resumption of credit should without increasing or creating new vulnerabilities. Simply relaxing prudential norms would only plant the seeds for new weaknesses (Lindgren and others, 1999). The literature on how to encourage sound credit in post-crisis environments is scarce. Some recommendations may, however, be made:

  • a. Encourage the dedollarization of the economy and the development of hedging markets. Over 80 percent of deposits are in dollar, and with only a small fraction of companies with foreign-currency denominated earnings, the room for banks to lend without incurring credit risks due to corporate sector mismatches is limited.11

  • b. Improve the credit culture. This will require enforcing debt service obligations across sectors of the economy, extending coverage and dissemination of credit registry information, creating incentives to build a good credit history, and executing collateral in severely nonperforming loans.12 The authorities could explore shortening foreclosure processes, accelerating out-of-court procedures for debt collection, and adopting proceedings that minimize judicial intervention.

  • c. Encourage the development of short-term money markets. When money markets are deeper, banks need to hold less precautionary holdings of liquidity, enabling them to lend more. The development of bonds, certificates of deposits, and commercial paper markets could contribute to foster financial intermediation (see Feyzoglu and Gelos, 2000).

  • d. Assist in the development of trust funds and securitization (fideicomisos). Banks can use these financing instruments while companie’ balance sheets are still recovering since they help to isolate the financing of specific projects from the bad balance sheets of borrowers. In addition, companies can use these instruments to obtain direct financing from the market. So far, however, only a few such securitizations have occurred and it would be useful to reduce tax and regulatory burden that impedes their development.

26. The government has been working on several relevant measures. For example, to encourage soundness in dollar credit, it has modified prudential regulations to internalize risks from dollar lending to unhedged borrowers, including through higher capital and provisioning requirements. To improve the credit culture, it is gradually expanding coverage of the credit registry and submitted to congress a bankruptcy law reform allowing to reorganize viable firms and liquidate nonviable ones faster and more efficiently.

References

  • Barajas, Adolfo, and Roberto Steiner, 2002, “Credit Stagnation in Latin America,” IMF Working Paper 02/53 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Calvo, Guillermo A., Alejandro Izquierdo, and Ernesto Talvi, 2006, “Phoenix Miracles in Emerging Markets: Recovering Without Credit From Systemic Financial Crises” (unpublished manuscript; Washington: Inter-American Development Bank).

    • Search Google Scholar
    • Export Citation
  • Catão, Luis, 1997, “Bank Credit in Argentina in the Aftermath of the Mexican Crisis: Supply or Demand Constrained,” IMF Working Paper 97/32 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • De Brun, Julio, Néstor Gandelman, Herman Kamil, and Arturo Porzecanski, 2006, “The Fixed-Income Market in Uruguay,” mimeo IADB.

  • Demirguc-Kunt, Detragiache, and Poonam Gupta, 2000, “Inside the Crisis: An Empirical Analysis of Banking Systems in Distress,” IMF Working Paper 00/156 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Feyzoglu, Tarhan, and Gaston Gelos, 2002, “Why is Private Sector Credit so Low in Bulgaria,” in Bulgaria: Selected Issues and Statistical Appendix (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Ghosh, Swati R., and Atish Ghosh, 1999, “East Asia in the Aftermath: Was There a Credit Crunch?” IMF Working Paper 99/38 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Gilli, B., 2005, Estructura de Capital de las Empresas Uruguayas en el Marco de la Teoría de Jerarquía de Preferencias,” (Valencia: Universidad de Valencia)

    • Search Google Scholar
    • Export Citation
  • Gulde and others, 2003, “Dealing with Banking Crises in Dollarized Economies,” in Managing Financial Crises: Recent Experience and Lessons for Latin America, ed. by Charles Collyns, and G. Russell Kincaid,IMF Occasional Paper No. 217 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Hoggarth, Glenn, Jack Reidhill, and Peter Sinclar, 2004, “On the Resolution of Banking Crises: Theory and Evidence,” Bank of England Working Paper 229.

    • Search Google Scholar
    • Export Citation
  • Inter-American Development Bank, 2004, “Unlocking Credit—The Quest for Deep and Stable Bank Lending.”

  • Kiefer, Nicholas, 1980, “A Note on Regime Classification in Disequilibrium Models,” Review of Economic Studies, Vol. 17, pp. 637 –639.

    • Search Google Scholar
    • Export Citation
  • Krueger, Anne, and Aaron Tornell, “The Role of Bank Restructuring in Recovering from Crises: Mexico 1995–98,” NBER Working Paper 7042.

    • Search Google Scholar
    • Export Citation
  • Laffont, Jean-Jacques, and Rene García, 1977, “Disequilibrium Econometrics for Business Loans,” Econometrica, Vol. 45.

  • Levine, Ross, 2005, “Finance and Growth: Theory and Evidence,” in Handbook of Economic Growth, ed. by Philippe Aghion and Steven Durlauf (The Netherlands: Elsevier Science).

    • Search Google Scholar
    • Export Citation
  • Licandro, Gerardo, 2006, “Los balances y el crédito bancario,” El País – Opinión (Montevideo), May 15.

  • Lindgren and others, 1999, “Financial Sector Crisis and Restructuring,” IMF Occasional Paper No.188 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Maddala, G., and Forrest Nelson, 1974, “Maximum Likelihood Methods for Models of Markets in Disequilibrium,” Econométrica, Vol. 42, No. 6, pp. 1013 –1030.

    • Search Google Scholar
    • Export Citation
  • Munyo, Ignacio, 2005, “The Determinants of Capital Structure: Evidence from an Economy without a Stock Market,” Draft Universidad de la República (September).

    • Search Google Scholar
    • Export Citation
  • Pazarbasioglu, Ceyla, 1996, “A Credit Crunch? A Case Study of Finland in the Aftermath of the Banking Crisis,” IMF Working Paper 96/135 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Sealey, C.W., 1979, “Credit Rationing in the Commercial Loan Market. Estimates of a Structural Model Under Conditions of Disequilibrium,” Journal of Finance, Vol. 34, pp. 689 –702.

    • Search Google Scholar
    • Export Citation
  • Stiglitz, Joseph, and Andrew Weiss, 1981, “Credit Rationing in Markets with Imperfect Information,” American Economic Review, Vol. 71, pp. 393 –410.

    • Search Google Scholar
    • Export Citation
  • Tornell, Aaron, 1999, “Common Fundamentals in the Tequila and Asian Crises,” NBER Working Paper No. 7139.

  • Tornell, Aaron, and Frank Westermann, 2002, “The Credit Channel in Middle Income Countries,” NBER Working Paper No. 9355.

APPENDIX Estimation Results

Table A1:

Uruguay Credit Supply and Demand Estimations

(Maximum Likelihood disequilibrium estimation)

article image

Estimation period 1998:12 - 2005:12. Number of observations: 84. t-statistics shown in parentheses. Foreign currency credit stock and foreign currency amounts in lending capacity definition are converted using an exchange rate that maintains the purchasing power constant throughout the estimation period.

Credit series in that column was adjusted with loan writeoffs and the loan portfolio of the banks liquidated during the crisis.

Dummy variable captures the level effect on credit demand during the outlier explosive period of local currency lending rate: Dec.2001-Mar.2003.

4

Prepared by Jorge Iván Canales-Kriljenko and Gastón Gelos.

5

The graph shows gross credit, without correcting for loan writeoffs, at exchange rates with a constant purchasing value of the dollar. The picture is similar with gross credit at current exchange rates, but with a pronounced spike in 2002 reflecting the sharp peso depreciation during the crisis.

6

See Licandro (2006).

7

For applications of this approach to the analysis of bank credit, see Laffont and García (1977), Sealey (1979), Pazarbasioglu (1997), Catão (1997), Ghosh and Ghosh (1999), and Barajas and Steiner (2002).

8

The goodness-of-fit can be gauged by how well the minimum of estimated demand and supply tracks actual credit.

9

The conclusion does not hold when using current exchange rates, but, as argued earlier, the use of current exchange rate is likely to produce significant measurement errors. When using a 12-month moving average of the exchange rate, credit ends to be demand-constrained until around mid-2004, becoming supply constrained thereafter.

10

See also IDB (2004).

11

New prudential rules now penalize credit to firms that do not generate dollar income through higher up-front provisions.

12

According to a recent World Bank Doing Business Survey, Uruguay scores relatively well on credit information availability. However, increased coverage would help assess credit risk better.

Uruguay: Selected Issues
Author: International Monetary Fund