## _wp1543 - conclusions.

## Source details

**Canonical URL:** [_wp1543 - conclusions.](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2015/_wp1543.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2015/_wp1543.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2015/_wp1543.pdf.json)

---

### A. Argentina’s Banking System
- Structural changes and market composition (1994–2001):
  - Number of institutions decreased from 168 to 83.
  - Foreign-owned institutions increased from 31 banks in November 1994 to 36 banks in November 2001.
  - By late 1998 the Argentine banking industry ranked second (after Singapore, and tied with Hong Kong) in terms of the quality of its regulatory environment (World Bank).
  - Market share of foreign banks: approximately half of the assets of the system towards the last years of the sample.
  - Cooperative banks: 38 in November 1994; 10 at end-1995; 2 in November 2001.
  - Ownership classification: based on nationality of the controlling group (group with 50 percent or more stock share).
- Bank-run episodes (definition: two or more months when more than one-half of banks are experiencing deposit losses):
  - December 1994–April 1995 (Tequila Crisis)
    - Trigger: fears Argentina would exit the currency board after Mexico’s devaluation on December 20, 1994.
    - Resident deposits fell by about 16 percent within four months.
    - Almost 90 percent of the banks lost deposits at the peak.
    - Most foreign branches, especially the largest ones, were able to increase both peso and dollar denominated deposits.
  - October–November 1997 (Asian crisis)
    - Essentially runs on peso denominated deposits without systemic characteristics.
    - Aggregate deposit change: 2 percent increase in 1997.
    - Redistribution: most foreign branches and subsidiaries gained deposits at the expense of domestic institutions.
  - August–September 1998 (Russian crisis)
    - Essentially runs on peso denominated deposits without systemic characteristics.
    - Aggregate deposit change: 1 percent deposit loss during the 1998 bank run.
    - Redistribution: most foreign branches and subsidiaries gained deposits at the expense of domestic institutions.
  - October 2000–November 2001 (Argentina’s own crisis; systemic)
    - Trigger: resignation of the Argentinean vice-president on October 6, 2000; ended with the deposit freeze on December 1, 2001.
    - Resident deposits fell by about 20 percent; more than 80 percent of the banks faced deposit losses.
    - Foreign banks, especially branches, experienced proportionally higher deposit withdrawals than domestic banks, reaching close to 25 percent of their deposits.
    - Foreign regional banks lost more deposits than any other group but had market share less than 1 percent.
- Selected system structure figures (November 1994 and November 2001) — as reported:
  - Number of Banks: Domestic Institutions: 33 to 31; Foreign Institutions: 34 to 31; Public: 4 to 3; Private: 13 to 11; Branches: 31 to 18; Subsidiaries: 87 to 68; Regional 3/: 83 to 83
  - Market share — Share of assets: 40.5, 30.9, 41.4, 20.8, 10.2, 16.5, 6.1, 31.1, 1.8, 0.7, 100.0, 100.0
  - Share of deposits 1/: 31.7, 29.6, 50.2, 19.2, 10.3, 14.6, 6.7, 35.9, 1.2, 0.7, 100.0, 100.0
  - Depositor base — Residents: 97.4, 100.0, 95.6, 97.4, 89.3, 99.0, 97.0, 95.9, 78.8, 99.1, 95.4, 97.9
  - Depositor base — Non-residents: 2.6, 0.0, 4.4, 2.6, 10.7, 1.0, 3.0, 4.1, 21.2, 0.9, 4.6, 2.1
  - Deposit denomination — Local-currency: 50.5, 27.1, 43.1, 28.7, 34.0, 32.1, 37.3, 22.9, 36.0, 33.9, 44.0, 26.7
  - Deposit denomination — Foreign-currency: 49.5, 72.9, 56.9, 71.3, 66.0, 67.9, 62.7, 77.1, 64.0, 66.1, 56.0, 73.3

### B. Uruguay’s Banking System
- Pre-crisis structure and features:
  - Domestic system: 22 banks and 6 cooperatives, with assets about 110 percent of GDP by end-2001.
  - Offshore system: 11 banks, accounting for an additional 8 percent of GDP in assets.
  - Large presence of foreign banks: about 58 percent market share including 10 percent participation of regional banks.
  - Public domestic banks market share: 39 percent.
  - High degree of dollarization: 91 percent share of foreign currency deposits.
- Non-resident deposits and distribution (end-2001):
  - Non-resident deposits: about 40 percent of total deposits.
  - Non-resident deposit allocation: 77 percent in foreign banks, 10 percent in local private banks, 13 percent in local public banks.
  - Resident deposit allocation: 21 percent in foreign banks, 18 percent in local private banks, 57 percent in local public banks.
  - Non-resident deposits accounted for about 60 percent of the depositor base of foreign banks and about 5 percent of the depositor base of domestic banks and cooperatives.
- Crisis dynamics:
  - Starting in 1999 adverse shocks produced a recession; confidence remained stable until 2001.
  - As Argentina’s crisis developed, Uruguay attracted a large volume of deposits from Argentines; in Q1 2002 public confidence began to erode as Argentine depositors withdrew funds.
  - 2002 financial crisis outcome: bankruptcy of many institutions; unprecedented bank run with loss of about 46 percent of total deposits of the system in seven months (December 2001–July 2002).
- Selected system structure figures (Dec-99 and Dec-01) — as reported:
  - Number of Banks: 43 to 38; 65 to 55; 109 to 65; 53 to 32; 83 to 28
  - Market share — Share of assets: 48.7, 39.2, 1.9, 2.3, 12.9, 15.4, 23.1, 27.1, 13.3, 16.0, 100.0, 100.0
  - Share of deposits 3/: 46.5, 39.4, 2.4, 2.3, 14.4, 15.0, 21.3, 26.6, 15.4, 16.8, 100.0, 100.0
  - Depositor base — Residents: 91.8, 86.4, 99.8, 95.5, 47.5, 43.5, 59.3, 44.3, 72.5, 83.7, 68.6, 59.5
  - Depositor base — Non-residents: 8.2, 13.6, 0.2, 4.5, 52.5, 56.9, 40.7, 55.6, 47.4, 30.7, 31.3, 40.5
  - Deposit currency structure — Local-currency: 17.0, 12.5, 33.2, 28.3, 9.2, 7.1, 11.2, 6.6, 4.9, 4.1, 13.1, 9.0
  - Deposit currency structure — Foreign-currency: 83.0, 87.5, 66.8, 71.7, 90.8, 92.9, 88.8, 93.4, 95.1, 95.9, 86.9, 91.0
- Note: non-resident deposit share partly reflects capital flight from Argentina during the 2001 financial crisis; non-resident deposits were 31 percent of the system at end-1999.

### Descriptive evidence on deposit losses (Annex 1’s Figure B)
- Private domestic banks: deposit losses of about 27 percent (foreign-currency) and 22 percent (local-currency).
- Foreign banks: larger average deposit losses of 54 percent (foreign-currency) and 28 percent (local-currency).
- Non-regional foreign affiliates (aggregate):
  - Subsidiaries: foreign-currency deposit losses of 65 percent.
  - Branches: foreign-currency deposit losses of 57 percent.
- Regional foreign banks, public domestic banks and cooperatives were also largely affected but with significantly lower deposit losses than the above figures.

### Effect of non-resident deposits
- Non-resident deposit runs were mainly concentrated in foreign banks and substantially influenced unconditional statistics.
- Excluding non-resident deposits:
  - Foreign branches’ deposit losses aligned with domestic (public and private) banks — around 30 percent.
  - Foreign subsidiaries had deposit losses of 51 percent.
  - Regional institutions averaged 52 percent of deposits lost, possibly reflecting exposure to Argentine risk and idiosyncratic solvency issues.
- Interpretation: depositors may have discriminated among foreign banks according to legal structure and embedded risks.

### Econometric exercise — methodology
- Objective: assess whether deposit flow patterns vis-à-vis foreign banks reflect safe haven perceptions, controlling for:
  - Bank fundamentals: asset quality, liquidity, profitability, capitalization, size, exposure to exchange rate and sovereign risk.
  - Interest rate responses: average implicit interest paid by each bank.
  - Macroeconomic variables: including devaluation expectations and country risk.
  - Ownership dummies: public banks, foreign branches, foreign subsidiaries, foreign regional banks (proxying safe haven perceptions).
- Estimation approach: system linear generalized method of moments (Arellano-Bover/Blundell-Bond GMM).
- Sample periods and bank run definitions:
  - Argentina: December 1994–November 2001.
  - Uruguay: December 2001–July 2002.
  - Bank run episodes defined as periods with two or more months of deposit losses in the banking system, or when more than one-half of the banks were losing deposits.
  - Identified Argentina bank runs: Dec 94–Apr 95, Oct 98–Nov 98, Sep 01–Nov 01; alternative: Oct 97–Sep 97.
  - Identified Uruguay bank run: Dec 01–July 02.
  - Systemic bank runs: drop in deposits of more than one standard deviation during the entire bank run period: Dec 94–Apr 95 (Argentina), Sep 00–Nov 01 (Argentina), Dec 01–July 02 (Uruguay).
- Lags: three-month lags used generally; one-month lag used in some Uruguay regressions due to data limitations.

### Regression results — Argentina
- Safe-haven perceptions:
  - Foreign branches and foreign regional banks were less affected during the Tequila bank run (Dec 94–Apr 95), suggesting safe-haven perceptions toward these foreign institutions in that episode.
  - Safe-haven perceptions were not significant during the two non-systemic bank runs and reversed in the 2001 systemic episode, where foreign branches experienced proportionally larger withdrawals after controlling for fundamentals.
- Foreign subsidiaries:
  - Gains in deposits by foreign subsidiaries and branches during most bank runs can be largely explained by bank characteristics for subsidiaries and in many cases for branches.
- Possible explanations for 2001 reversal (branches lost more deposits):
  - Branches might have facilitated capital flight more easily in a systemic crisis.
  - Depositors may have anticipated triggering of foreign branches’ ring-fencing provisions.
  - Data limitations prevented adjudicating between these explanations.
- Devaluation expectations:
  - Played a key role in depositors’ withdrawal decisions across most bank runs.
- Bank fundamentals and dynamics:
  - Fundamentals had expected signs; most significant in the 2001 systemic bank run.
  - Public banks benefited from significant positive depositor perceptions during the 2001 bank run.
  - Larger banks seemed perceived as too big to fail in the 1995 bank run.
  - Lag coefficient of the dependent variable: negative and significant in the non-systemic 1998 bank run (suggesting overshooting); positive and significant during the systemic 2001 bank run (reflecting longer episode and growing concerns about currency board sustainability).

### Regression results — Uruguay
- Bank fundamentals: most coefficients had expected signs or were statistically insignificant.
- Aggregate exchange rate risk: major common factor driving deposit withdrawals.
- Dynamics: lag coefficient indicates some overshooting in deposit adjustments even after controlling for interest rates.
- Safe-haven perceptions: no evidence of safe-haven perceptions toward foreign banks after controlling for fundamentals.
  - Estimated coefficients for foreign branches and subsidiaries were statistically insignificant.
  - Larger unconditional withdrawals faced by foreign subsidiaries relative to domestic private banks can be explained by weaker fundamentals.
  - Foreign branches showed similar performance to domestic private banks both unconditionally and when controlling for fundamentals.

### Conclusions and policy implication
- Depositors do not consistently “fly to (foreign bank) quality” across domestic crises after controlling for bank fundamentals and interest rate responses.
- Among the five cases studied, evidence of safe-haven perceptions appears in one case (Argentina, 1995) and only toward foreign branches.
- In one episode (Argentina, 2001) foreign branches experienced larger deposit withdrawals.
- Foreign subsidiaries were not perceived differently in any of the cases.
- Policy implication: favoring entry of foreign branches over foreign subsidiaries cannot be justified on safe-haven considerations alone.

### Appendix highlights — Evolution of deposits and interest rates (selected exact figures)
- Dec 1994–Apr 1995 (Tequila Crisis)
  - Deposit dynamics: resident deposits fell by about 16 percent within four months; from February to April 1995, both peso and dollar-denominated deposits decreased sharply.
  - Bank closures/suspensions: 9 banks closed their doors (5 suspended by the Central Bank and 4 absorbed by other banks); they represented less than 1.60 percent of the total assets of the banking sector.
- Oct 1997–Nov 1997 (Asian Crisis)
  - Deposit dynamics: fall in peso-deposits of 5 percent, offset by a 9 percent increase in dollar-denominated deposits; by end-November 1997 dollar deposits represented 58 percent of total deposits.
  - Bank closures/suspensions: Two banks suspended; they represented less than a quarter percentage point of total banking sector assets.
- Aug 1998–Sep 1998 (Russian Default)
  - Deposit dynamics: fall in peso-denominated deposits of 10 percent in two months; decrease in total deposits was only 1 percentage point.
  - Bank closures/suspensions: Central Bank suspended Mayo Coop Bank (~1 percent of assets).
- Sep 2000–Nov 2001 (Systemic bank run; end of currency board)
  - IMF program: agreement on December 19th, 2000; size of the international aid program was around 40 billion in total.
  - Phase-level dynamics (selected figures):
    - Phase I (Sep 2000–Feb 2001): foreign branches experienced higher falls in deposits (3 percent). Three banks closed, two suspended; they represented around 0.3 percent of banking system assets.
    - Phase II (Mar 2001–May 2001): banking sector lost around 5 percent of deposits during March–April.
    - Phase III (Jun 2001–Aug 2001): around 21 and 9 percent drops in peso and dollar deposits respectively for the banking sector; public banks lost around 37 and 15 percent of their peso and dollar deposits respectively.
    - Phase IV (Sep 2001–Oct 2001): banking sector gained deposits 2 percent (driven by dollar deposits +3.6 percent); foreign branches and foreign subsidiaries lost deposits (8 percent and 2 percent respectively).
    - Phase V (Nov 2001): drops in peso and dollar-denominated deposits were around 8 and 5 percent respectively.
- Summary statistics (selected exact figures from Appendix 1 — Table C and D):
  - Tequila Crisis (Dec 94 - Apr 95), Obs = 684
    - Change in Deposits: Mean -0.08; Std. Dev. 0.17; Min -0.75; Max 2.08
    - Liquidity: Mean 0.10; Std. Dev. 0.06; Min 0.00; Max 0.39
    - Interest Rate on Deposits: Mean 79.53; Std. Dev. 3.60; Min 0.81; Max 26.88
    - Exposure to Exchange Rate Risk: Mean 2.65; Std. Dev. 1.78; Min -12.96; Max 8.56
    - Public Banks (mean): 0.18; Std. Dev. 0.39; Min 0.00; Max 1.00
  - Asian Crisis (Oct 97 - Nov 97), Obs = 192
    - Change in Deposits: Mean 0.00; Std. Dev. 0.07; Min -0.23; Max 0.32
    - Interest Rate on Deposits: Mean 5.76; Std. Dev. 2.24; Min 0.65; Max 16.16
    - Exposure to Exchange Rate Risk: Mean 1.61; Std. Dev. 0.78; Min 0.83; Max 2.38
  - Russian Crisis (Aug 98 - Sep 98), Obs = 168
    - Change in Deposits: Mean -0.01; Std. Dev. 0.10; Min -0.46; Max 0.51
    - Interest Rate on Deposits: Mean 5.75; Std. Dev. 1.98; Min 0.44; Max 11.10
    - Exposure to Exchange Rate Risk: Mean 1.82; Std. Dev. 0.75; Min 1.07; Max 2.56
  - 2001 Crisis (Sep 00 - Nov 01), Obs = 967
    - Change in Deposits: Mean -0.02; Std. Dev. 0.12; Min -1.00; Max 1.39
    - Interest Rate on Deposits: Mean 68.36; Std. Dev. 3.48; Min 0.59; Max 28.93
    - Exposure to Exchange Rate Risk: Mean 5.60; Std. Dev. 4.96; Min -82.54; Max 23.95
  - Uruguay (Jan–July 02), Obs = 114
    - Change in Deposits: Mean -0.06; Std. Dev. 0.10; Min -0.75; Max 0.12
    - Change in Deposits (FC): Mean -0.05; Std. Dev. 0.11; Min -0.80; Max 0.17
    - Liquidity: Mean 0.32; Std. Dev. 0.25; Min 0.04; Max 2.23
    - Non-Performing Loans: Mean 0.19; Std. Dev. 0.12; Min 0.05; Max 0.52
    - Foreign Subsidiary (mean): 0.32; Std. Dev. 0.47; Min 0.00; Max 1.00

### Appendix 3 — Uruguay chronology and policy responses (selected exact actions and dates)
- Key pre-crisis shocks: devaluations of Brazil’s real (January 1999) and the Argentina peso (January 2002); foot-and-mouth disease early 2001; recession beginning 1999.
- Q1 2002: Argentine depositors withdrew funds from Uruguay as Argentina tightened deposit access.
- Major policy responses and institutional actions:
  - Creation of the Fund for the Stabilization of the Banking System (FSBS) to provide full backing of dollar sight and saving deposits of suspended institutions and state-owned banks.
  - Extension of maturity of dollar time deposits of public banks (BROU and BHU) to three years.
  - Declaration of a 5-day bank holiday by end-July 2002.
  - IMF and IDB announced upcoming augmentations of financial assistance to Uruguay (about US$ 3 billion over two years).
  - The IMF increased its Uruguay’s SBA by about US$ 1.5 billion.
  - Passage of the “Ley de Estabilidad Fiscal” (Parliament approval on May 29, 2002).
  - Abandonment of the monetary regime of adjustable bands and allowing the exchange rate to float (June 19, 2002).
  - Government actions to recapitalize and provide financial assistance to troubled banks (e.g., Banco Comercial received assistance through deposits from CND and shareholder recapitalization on April 26, 2002).
  - Congress passed the Banking System Stability Law, creating the FSBS and reprogramming time deposits at BROU and BHU for three years (July 30, 2002).
  - Government announced a bill to restructure suspended banks and create a new bank (Nuevo Banco Comercial) out of assets of Banco Montevideo, Caja Obrera and Banco Comercial (November 26, 2002).
- Chronology of key events (selected exact dates 2001–2002):
  - 2001: December 2 — Argentina imposes a deposit freeze (“corralito”); December 12 — IMF suspends loan disbursements to Argentina; December 23 — Argentina announces the default on its sovereign debt.
  - 2002: January 4 — increase of the exchange rate band from 6 to 12 percent, and acceleration of the rate of depreciation from 1.2 to 2.4 percent per month; January 11 — S&P lowers Uruguay’s outlook to negative; February 13 — Central bank of Uruguay suspends operations of Banco Galicia (Uruguay); March 25 — IMF approves a new SBA for SDR 2.0 billion; May 29 — Parliament approves the “Ley de Estabilidad Fiscal”; June 19 — exchange rate allowed to float; July 30 — 5-day bank holiday declared and Banking System Stability Law passed; August 28 — ceiling for savings account withdrawals increased; December 2 — Argentina terminates the deposit freeze.

*Source: _wp1543 - conclusions.*

### conclusions.

### conclusions.

### A. Argentina’s Banking System
- During 1994-2001 the number of institutions decreased from 168 to 83.
- Foreign-owned institutions increased from 31 banks in November 1994 to 36 banks in November 2001.
- By late 1998 the Argentine banking industry ranked second (after Singapore, and tied with Hong Kong) in terms of the quality of its regulatory environment, according to the World Bank.
- Market share of foreign banks: approximately half of the assets of the system towards the last years of the sample.
- Cooperative banks fell from 38 in November 1994 to 2 in November 2001; at end-1995 there were 10 cooperative banks.
- Argentina experienced four distinct bank run episodes:
  - December 1994-April 1995 (associated with the Mexican crisis/“Tequila Crisis”).
  - October-November 1997 (Asian crisis).
  - August-September 1998 (Russian crisis).
  - October 2000-November 2001 (Argentina’s own crisis).
- Definition used for a bank run: two or more months when more than one-half of banks are experiencing deposit losses.
- December 1994-April 1995 episode:
  - Trigger: fears Argentina would exit the currency board after Mexico’s currency devaluation on December 20, 1994.
  - Resident deposits fell by about 16 percent within four months.
  - Almost 90 percent of the banks lost deposits at the peak.
  - Most foreign branches, especially the largest ones, were able to increase both peso and dollar denominated deposits.
- October-November 1997 and August-September 1998 episodes:
  - Essentially runs on peso denominated deposits without systemic characteristics.
  - Aggregate deposit changes: 1 percent deposit loss during the 1998 bank run, and a 2 percent increase in 1997.
  - Redistribution of deposits across banks: most foreign branches and subsidiaries gained deposits at the expense of domestic institutions.
- October 2000-November 2001 episode:
  - Trigger: resignation of the Argentinean vice-president on October 6, 2000; ended with the deposit freeze on December 1, 2001.
  - Resident deposits fell by about 20 percent; more than 80 percent of the banks faced deposit losses.
  - Foreign banks, especially branches, experienced proportionally higher deposit withdrawals than domestic banks, reaching close to 25 percent of their deposits.
  - Foreign regional banks lost more deposits than any other group but had market share less than 1 percent.
- Table 1 (selected system structure figures, November 1994 and November 2001) as reported:
  - Number of Banks: (Nov-94 to Nov-01) Domestic Institutions: 33 to 31; Foreign Institutions: 34 to 31; Public: 4 to 3; Private: 13 to 11; Branches: 31 to 18; Subsidiaries: 87 to 68; Regional 3/: 83 to 83
  - Market share — Share of assets: 40.5, 30.9, 41.4, 20.8, 10.2, 16.5, 6.1, 31.1, 1.8, 0.7, 100.0, 100.0
  - Share of deposits 1/: 31.7, 29.6, 50.2, 19.2, 10.3, 14.6, 6.7, 35.9, 1.2, 0.7, 100.0, 100.0
  - Depositor base 1/ 2/ — Residents: 97.4, 100.0, 95.6, 97.4, 89.3, 99.0, 97.0, 95.9, 78.8, 99.1, 95.4, 97.9
  - Depositor base 1/ 2/ — Non-residents: 2.6, 0.0, 4.4, 2.6, 10.7, 1.0, 3.0, 4.1, 21.2, 0.9, 4.6, 2.1
  - Deposit denomination 1/ 2/ — Local-currency: 50.5, 27.1, 43.1, 28.7, 34.0, 32.1, 37.3, 22.9, 36.0, 33.9, 44.0, 26.7
  - Deposit denomination 1/ 2/ — Foreign-currency: 49.5, 72.9, 56.9, 71.3, 66.0, 67.9, 62.7, 77.1, 64.0, 66.1, 56.0, 73.3
- Note: ownership classification based on nationality of the controlling group (group with 50 percent or more stock share).

### B. Uruguay’s Banking System
- Uruguay’s banking system in the 1990s and up to the 2002 crisis:
  - Perceived as a safe financial hub in Latin America.
  - Composition: 22 banks and 6 cooperatives (domestic system), with assets about 110 percent of GDP by end-2001.
  - Offshore system: 11 banks, accounting for an additional 8 percent of GDP in assets.
  - Large presence of foreign banks: about 58 percent market share including 10 percent participation of regional banks.
  - Public domestic banks market share: 39 percent.
  - High degree of dollarization: 91 percent share of foreign currency deposits.
- Non-resident deposits:
  - Accounted for about 40 percent of total deposits at end-2001.
  - Non-resident deposit allocation: 77 percent in foreign banks, 10 percent in local private banks, 13 percent in local public banks.
  - Resident deposit allocation: 21 percent in foreign banks, 18 percent in local private banks, 57 percent in local public banks.
  - Non-resident deposits accounted for about 60 percent of the depositor base of foreign banks and about 5 percent of the depositor base of domestic banks and cooperatives.
- Starting in 1999 adverse shocks (devaluation of Brazil’s real, recession in Argentina, foot-and-mouth disease) produced a recession; confidence remained stable until 2001.
- As Argentina’s crisis developed, Uruguay attracted a large volume of deposits from Argentines; in Q1 2002 public confidence began to erode as Argentine depositors withdrew funds.
- The 2002 financial crisis:
  - Resulted in bankruptcy of many institutions.
  - Unprecedented bank run: loss of about 46 percent of total deposits of the system in seven months (December 2001-July 2002).
- Table 2 (selected system structure figures, Dec-99 and Dec-01) as reported:
  - Number of Banks: 43 to 38; 65 to 55; 109 to 65; 53 to 32; 83 to 28
  - Market share — Share of assets: 48.7, 39.2, 1.9, 2.3, 12.9, 15.4, 23.1, 27.1, 13.3, 16.0, 100.0, 100.0
  - Share of deposits 3/: 46.5, 39.4, 2.4, 2.3, 14.4, 15.0, 21.3, 26.6, 15.4, 16.8, 100.0, 100.0
  - Depositor base 3/ 4/ — Residents: 91.8, 86.4, 99.8, 95.5, 47.5, 43.5, 59.3, 44.3, 72.5, 83.7, 68.6, 59.5
  - Depositor base 3/ 4/ — Non-residents: 8.2, 13.6, 0.2, 4.5, 52.5, 56.9, 40.7, 55.6, 47.4, 30.7, 31.3, 40.5
  - Deposit currency structure 3/ 4/ — Local-currency: 17.0, 12.5, 33.2, 28.3, 9.2, 7.1, 11.2, 6.6, 4.9, 4.1, 13.1, 9.0
  - Deposit currency structure 3/ 4/ — Foreign-currency: 83.0, 87.5, 66.8, 71.7, 90.8, 92.9, 88.8, 93.4, 95.1, 95.9, 86.9, 91.0
- Note: non-resident deposit share partly reflects capital flight from Argentina during the 2001 financial crisis; non-resident deposits were 31 percent of the system at end-1999.

*Source: _wp1543 - conclusions.*

### Annex 1’s Figure B, respectively). A first glance at the data shows that private domestic

### _wp1543 - Annex 1’s Figure B, respectively). A first glance at the data shows that private domestic

### Descriptive evidence on deposit losses
- Private domestic banks faced deposit losses of about 27 (22) percent of foreign (local) currency deposits.
- Foreign banks experienced larger average deposit losses of 54 (28) percent.
- In the aggregate (including resident and non-residents deposits) non-regional foreign affiliates were most affected:
  - Subsidiaries: foreign-currency deposit losses of 65 percent.
  - Branches: foreign-currency deposit losses of 57 percent.
- Regional foreign banks, public domestic banks and cooperatives were also largely affected, though deposit losses were significantly lower than the above figures.

### Effect of non-resident deposits
- Non-resident deposit runs were mainly concentrated in foreign banks and substantially influenced the unconditional statistics.
- Excluding non-resident deposits changed the picture:
  - Foreign branches’ deposit losses aligned with domestic (public and private) banks—at around 30 percent.
  - Foreign subsidiaries had deposit losses of 51 percent.
  - Regional institutions averaged 52 percent of deposits lost, possibly reflecting exposure to Argentine risk and idiosyncratic solvency issues.
- The evidence suggests depositors may have discriminated among foreign banks according to legal structure and embedded risks.

### Econometric exercise: methodology
- Objective: assess whether deposit flow patterns vis-à-vis foreign banks reflect safe haven perceptions, controlling for:
  - Bank fundamentals (asset quality, liquidity, profitability, capitalization, size, exposure to exchange rate and sovereign risk).
  - Interest rate responses (average implicit interest paid by each bank).
  - Macroeconomic variables (including devaluation expectations and country risk).
  - Bank-ownership dummies for public banks, foreign branches, foreign subsidiaries, and foreign regional banks (proxying safe haven perceptions).
- Estimation approach: system linear generalized method of moments (Arellano-Bover/Blundell-Bond GMM).
- Sample periods and bank run definitions:
  - Argentina: December 1994–November 2001.
  - Uruguay: December 2001–July 2002.
  - Bank run episodes defined as periods with two or more months of deposit losses in the banking system, or when more than one-half of the banks were losing deposits.
  - Identified Argentina bank runs: Dec 94–Apr 95, Oct 98–Nov 98, Sep 01–Nov 01; additional period Oct 97–Sep 97 selected under alternative criterion.
  - Identified Uruguay bank run: Dec 01–July 02.
  - Systemic bank runs: drop in deposits of more than one standard deviation during the entire bank run period:
    - Dec 94–Apr 95 (Argentina), Sep 00–Nov 01 (Argentina), Dec 01–July 02 (Uruguay).
- Lags: three-month lags used to incorporate publication delays; note that a one-month lag was used in some Uruguay regressions due to data limitations.

### Regression results — Argentina
- Evidence of safe-haven perceptions:
  - Foreign branches and foreign regional banks were less affected (relative to other banks) during the Tequila bank run (Dec 94–Apr 95), suggesting safe-haven perceptions toward these foreign institutions in that episode.
  - Safe-haven perceptions were not significant during the two non-systemic bank runs and reversed in the 2001 systemic episode, where foreign branches experienced proportionally larger withdrawals after controlling for fundamentals.
- Foreign subsidiaries:
  - Gains in deposits by foreign subsidiaries and branches during most bank runs can be largely explained by bank characteristics for subsidiaries and in many cases for branches.
- Possible interpretations for 2001 result where branches lost more deposits:
  - Branches might have facilitated capital flight more easily in a systemic crisis.
  - Depositors may have anticipated triggering of foreign branches’ ring-fencing provisions.
  - Data limitations prevented adjudicating between these explanations.
- Devaluation expectations:
  - Played a key role in depositors’ withdrawal decisions across most bank runs.
- Bank fundamentals:
  - Generally have expected signs; most significant in the 2001 systemic bank run.
  - Public banks benefited from significant positive depositor perceptions during the 2001 bank run.
  - Larger banks seemed perceived as too big to fail in the 1995 bank run.
- Dynamics:
  - Lag coefficient of the dependent variable was negative and significant in the non-systemic 1998 bank run (suggesting overshooting in adjustment).
  - Lag coefficient was positive and significant during the systemic 2001 bank run (reflecting longer episode and growing concerns about currency board sustainability).

### Regression results — Uruguay
- Most coefficients on bank fundamentals had expected signs or were statistically insignificant.
- Aggregate exchange rate risk appeared as a major common factor driving deposit withdrawals.
- Lag coefficient of the dependent variable indicates some overshooting in deposit adjustments even after controlling for interest rates.
- No evidence of safe-haven perceptions toward foreign banks after controlling for fundamentals:
  - Estimated coefficients for foreign branches and subsidiaries were statistically insignificant.
  - Larger unconditional withdrawals faced by foreign subsidiaries relative to domestic private banks can be explained by weaker fundamentals.
  - Foreign branches showed similar performance to domestic private banks both unconditionally and when controlling for fundamentals.

### Conclusions and policy implication
- Depositors do not consistently “fly to (foreign bank) quality” across domestic crises after controlling for bank fundamentals and interest rate responses.
- Among the five cases studied, evidence of safe-haven perceptions appears in one case (Argentina, 1995) and only toward foreign branches.
- In one episode (Argentina, 2001) foreign branches experienced larger deposit withdrawals.
- Foreign subsidiaries were not perceived differently in any of the cases.
- Policy implication: favoring entry of foreign branches over foreign subsidiaries cannot be justified on safe-haven considerations alone.

*Source: Excerpt from the supplied IMF Working Paper content.*

### Appendix 1 – Figure A

### Appendix 1 – Figure A

### Evolution of deposits and interest rates across crisis episodes (figures summary)
- Dec 1994–Apr 1995 (Tequila Crisis)
  - Deposit dynamics: depositors mostly ran on peso-denominated deposits during December 1994 and January 1995; dollar-denominated deposits increased but did not fully offset peso outflows. From February to April 1995, both peso and dollar-denominated deposits decreased sharply.
  - Bank closures/suspensions: 9 banks closed their doors (5 suspended by the Central Bank and 4 absorbed by other banks); they represented less than 1.60 percent of the total assets of the banking sector.
- Oct 1997–Nov 1997 (Asian Crisis)
  - Deposit dynamics: fall in peso-deposits of 5 percent, offset by a 9 percent increase in dollar-denominated deposits; by end-November 1997 dollar deposits represented 58 percent of total deposits.
  - Bank closures/suspensions: Two banks were suspended (and then revoked) by the Central Bank; they represented less than a quarter percentage point of total banking sector assets.
- Aug 1998–Sep 1998 (Russian Default)
  - Deposit dynamics: fall in peso-denominated deposits of 10 percent in two months; increase in dollar deposits did not completely offset the peso drop, but the decrease in total deposits was only 1 percentage point.
  - Bank closures/suspensions: Central Bank suspended Mayo Coop Bank, which had about 1 percent of the banking sector’s assets.
- Sep 2000–Nov 2001 (Systemic bank run; end of currency board)
  - IMF program: agreement on December 19th, 2000; the size of the international aid program was around 40 billion in total.
  - Phase-level dynamics and specific figures:
    - Phase I (Sep 2000–Feb 2001): foreign branches experienced higher falls in deposits (3 percent). Three banks closed, two suspended; they represented around 0.3 percent of banking system assets. Deposits recovered in Jan–Feb led by domestic public banks but did not offset foreign bank losses.
    - Phase II (Mar 2001–May 2001): banking sector lost around 5 percent of deposits during March–April. Foreign branches lost peso deposits (8 percent) and dollar deposits (2 percent) in May. One small foreign branch was suspended (about 0.01 percent of banking assets).
    - Phase III (Jun 2001–Aug 2001): Congress modification of the currency board triggered large outflows: around 21 and 9 percent drops in peso and dollar deposits respectively for the banking sector. Public banks lost around 37 and 15 percent of their peso and dollar deposits respectively. Central Bank suspended one coop bank (~0.02 of banking system assets).
    - Phase IV (Sep 2001–Oct 2001): banking sector gained deposits 2 percent (driven by dollar deposits +3.6 percent); peso-denominated deposits varied -3.7 percent. Foreign branches and foreign subsidiaries lost deposits (8 percent and 2 percent respectively). Chase Manhattan closed its doors due to a parent-bank merger.
    - Phase V (Nov 2001): international assistance closed off; drops in peso and dollar-denominated deposits were around 8 and 5 percent respectively. All bank groups suffered falls; domestic private banks proportionally most affected.

### Appendix 1 — Data definitions (Argentina)
- Resident deposits (BCRA): Private non-financial sector deposits by residents (sight, saving and time deposits).
- Liquidity (BCRA): Liquid assets over total assets.
- Interest rate on deposits (BCRA): Weighted average of dollar and pesos implicit interest rate. Implicit interest rates in each currency are calculated as charges for deposits over stock of deposits.
- Return on Assets (BCRA): Net income over total assets.
- Non-Performing Loans (BCRA): Non-performing loans over total loans.
- Capital (BCRA): Equity over total assets.
- Exposure to exchange rate Risk (BCRA): Loans in US dollars over equity [same definition as Levy-Yeyati et al (2007)].
- Exposure to Sovereign Risk (BCRA): Bonds of- and loans to the government, as a share of bank’s total assets.
- Log of Assets (BCRA): Natural logarithm of Banks assets (measure of banks’ size).
- Ownership dummies and definitions:
  - Domestic Public Dummy (Bankscope, Bankers’ Almanac, and BCRA): Domestic state-owned (majority) bank.
  - Domestic Private Dummy: Bank controlled by private domestic shareholders.
  - Foreign Subsidiary dummy (Bankscope and Bankers’ Almanac): Subsidiary controlled by foreign shareholders from outside Latin America.
  - Foreign Branch dummy (Bankscope and Bankers’ Almanac): Bank branch from a foreign bank incorporated outside Latin America.
  - Regional dummy (Bankscope and Bankers’ Almanac): Foreign bank (subsidiary or branch) controlled by shareholders from Latin America.
- Size (BCRA): Log of total bank assets.
- Macro series:
  - Exchange rate risk (BCRA): Difference between average domestic currency and foreign currency deposit rates (for time deposits).
  - Sovereign Spread (JP Morgan): Spreads on Argentine sovereign bonds over comparable US bonds (EMBI+).

### Appendix 1 — Data definitions (Uruguay)
- Dollar deposits (BCU): Private non-financial sector dollar-denominated deposits (sight, saving and time deposits).
- Peso Deposits (BCU): Private non-financial sector peso-denominated deposits (sight, saving and time deposits).
- Liquidity (BCU): Liquid assets plus government bonds over total assets.
- Interest rate on dollar deposits (BCU): Implicit interest rate: Charges on dollar time deposits over average monthly stock of dollar time deposits.
- Interest rate on peso deposits (BCU): Implicit interest rate: Charges on peso deposits over average monthly stock of peso time deposits.
- Non-Performing Loans (BCU): Non-performing loans over total loans.
- Capital (BCU): Equity over total assets.
- Return on Assets (BCU): Net income over total assets.
- Return on Equity (BCU): Net income over equity.
- Non-resident exposure (BCU): Non-residents deposits over total deposits.
- Foreign exchange exposure (BCU): Dollar loans over total assets.
- Sovereign Risk exposure (BCU): Bonds of- and loans to the government, as a share of bank’s total assets.
- Ownership dummies and definitions:
  - Domestic Public Dummy (Bank’s Almanac and BCU): Domestic state-owned (majority) bank.
  - Domestic Private Dummy: Bank or cooperative with private domestic share majority holder.
  - Foreign Subsidiary dummy: Foreign bank subsidiary.
  - Foreign Branch dummy: Foreign bank branch.
  - Regional dummy: Foreign (public or private) bank with majority of regional capital.
- Size (BCU): Log of total bank assets.
- Macro series:
  - Exchange rate risk (BCU): [Difference between average local currency and foreign currency deposit rate (for time deposits with maturity between 1 and 6 months).]
  - Sovereign Spread (República AFAP): Uruguayan Bond Index (UBI): Spread between Uruguayan bond yield and benchmark U.S. bond.

### Summary statistics (selected exact figures from Appendix 1 — Table C and D)
- Tequila Crisis (Dec 94 - Apr 95), Obs = 684 (Change in Deposits)
  - Change in Deposits: Mean -0.08; Std. Dev. 0.17; Min -0.75; Max 2.08
  - Liquidity: Mean 0.10; Std. Dev. 0.06; Min 0.00; Max 0.39
  - Interest Rate on Deposits: Mean 79.53; Std. Dev. 3.60; Min 0.81; Max 26.88
  - Exposure to Exchange Rate Risk: Mean 2.65; Std. Dev. 1.78; Min -12.96; Max 8.56
  - Public Banks (mean): 0.18; Std. Dev. 0.39; Min 0.00; Max 1.00
- Asian Crisis (Oct 97 - Nov 97), Obs = 192
  - Change in Deposits: Mean 0.00; Std. Dev. 0.07; Min -0.23; Max 0.32
  - Interest Rate on Deposits: Mean 5.76; Std. Dev. 2.24; Min 0.65; Max 16.16
  - Exposure to Exchange Rate Risk: Mean 1.61; Std. Dev. 0.78; Min 0.83; Max 2.38
- Russian Crisis (Aug 98 - Sep 98), Obs = 168
  - Change in Deposits: Mean -0.01; Std. Dev. 0.10; Min -0.46; Max 0.51
  - Interest Rate on Deposits: Mean 5.75; Std. Dev. 1.98; Min 0.44; Max 11.10
  - Exposure to Exchange Rate Risk: Mean 1.82; Std. Dev. 0.75; Min 1.07; Max 2.56
- 2001 Crisis (Sep 00 - Nov 01), Obs = 967
  - Change in Deposits: Mean -0.02; Std. Dev. 0.12; Min -1.00; Max 1.39
  - Interest Rate on Deposits: Mean 68.36; Std. Dev. 3.48; Min 0.59; Max 28.93
  - Exposure to Exchange Rate Risk: Mean 5.60; Std. Dev. 4.96; Min -82.54; Max 23.95
- Uruguay (Jan–July 02), Obs = 114
  - Change in Deposits: Mean -0.06; Std. Dev. 0.10; Min -0.75; Max 0.12
  - Change in Deposits (FC): Mean -0.05; Std. Dev. 0.11; Min -0.80; Max 0.17
  - Liquidity: Mean 0.32; Std. Dev. 0.25; Min 0.04; Max 2.23
  - Non-Performing Loans: Mean 0.19; Std. Dev. 0.12; Min 0.05; Max 0.52
  - Foreign Subsidiary (mean): 0.32; Std. Dev. 0.47; Min 0.00; Max 1.00

### Chronology of main events (Argentina; Appendix 2 — selected facts and exact wording)
- Dec 1994 — Apr 1995: Systemic Bank Run (Tequila Crisis)
  - On December 20th, 1994, Mexico devalued its currency. Two phases: (i) depositors mostly ran on peso-denominated deposits during December 1994 and January 1995; (ii) both peso and dollar-denominated deposits decreased sharply during February to April 1995.
- Oct 1997 — Nov 1997: Non-systemic Bank Run (Asian Crises)
  - First observable consequence in Argentina after the First Attack on the Hong Kong dollar on October 23rd. During this bank run, there was a fall in peso-deposit 5 percent, offset by a 9 percent increase in dollar denominated deposits.
- Aug 1998 — Sep 1998: Non-systemic Bank Run (Russian Default)
  - Russia defaulted and devalued on August 17th, 1998. The fall in peso-denominated deposits was important, 10 percent, in only two months. The decrease in total deposits was only 1 percentage point. Central Bank suspended Mayo Coop Bank (~1 percent of assets).
- Sep 2000 — Nov 2001: Systemic Bank Run (End of Currency Board)
  - Key events: vice-president resignation (October 6th, 2000); IMF agreement on December 19th, 2000 (size of international aid program: around 40 billion in total).
  - Five phases with detailed deposit dynamics and bank suspensions/closures as described above (Phase I through Phase V).

*Appendix and figure content as presented in the source document.*

### Appendix 3. Uruguay - Chronology of main events during the 2002 crisis

### Appendix 3. Uruguay - Chronology of main events during the 2002 crisis

### Overview
- Uruguay’s real economy experienced rapid growth between 1990 and 1998, supported by strong fundamentals, investment grade status and a general perception of being a safe economy within the region.
- Starting in 1999, adverse shocks brought the economy to a recession:
  - The crises in Brazil and Argentina—Uruguay’s main trading partners, jointly accounting for about ½ of total exports—damaged the export sector.
  - An outbreak of foot-and-mouth disease in early 2001 inflicted significant damage on the export sector.
  - Devaluations of Brazil’s real (January 1999) and the Argentina peso (January 2002) led to a significant appreciation of Uruguay’s real exchange rate, casting doubts about the sustainability of the crawling peg framework.
- Until 2001 confidence in the Uruguayan banking system remained intact; as the Argentina crisis developed, Uruguayan banks attracted a large volume of deposits from Argentines, reflecting perception of Uruguay as a safe haven, an implicit and unrestricted government guarantee, and a major presence of foreign banking.
- In Q1 2002 public confidence began to erode as Argentine depositors, unable to access accounts after Argentina’s deposit freeze and pesification, withdrew funds from Uruguay (at the time almost half of deposits were held by nonresidents, largely Argentines).
- Initial problems affected Banco de Galicia (local subsidiary) and Banco Comercial (large domestic private bank with substantial exposure to Argentina and weakened by fraudulent activities). As the Argentina crisis worsened and Argentina tightened the deposit freeze (the “corralon”), withdrawals intensified, spreading to resident deposits and other institutions (including public banks).
- Concerns about the crawling peg and the fate of a highly-dollarized banking system were exacerbated by the downgrading of Uruguay’s sovereign debt—previously investment grade—for the first time in many years.
- Large financial support from international financial institutions was not successful in stemming the deposit and currency run.
- By end-July 2002 a 5-day bank holiday was declared; during the holiday further IFI support was negotiated and operations of several banks were suspended. A law created the Fund for the Stabilization of the Banking System (FSBS) to provide full backing of dollar sight and saving deposits of suspended institutions and state-owned banks; maturity of dollar time deposits of public banks (BROU and BHU) was extended to three years. Following these measures the deposit run gradually receded.

### Major policy responses and institutional actions
- Creation of the Fund for the Stabilization of the Banking System (FSBS) to provide full backing of dollar sight and saving deposits of suspended institutions and state-owned banks.
- Extension of the maturity of dollar time deposits of public banks (BROU and BHU) to three years.
- Declaration of a 5-day bank holiday by end-July 2002; suspension of operations of Banco Montevideo, La Caja Obrera, Banco Comercial and Banco de Credito.
- IMF and IDB announced upcoming augmentations of financial assistance to Uruguay (of about US$ 3 billion over two years).
- The IMF increased its Uruguay’s SBA by about US$ 1.5 billion.
- Passage of the “Ley de Estabilidad Fiscal” (Parliament approval on May 29, 2002).
- Abandonment of the monetary regime of adjustable bands and allowing the exchange rate to float (June 19, 2002).
- Government actions to recapitalize and provide financial assistance to troubled banks (e.g., Banco Comercial received assistance through deposits from CND and shareholder recapitalization on April 26, 2002).
- Congress passed the Banking System Stability Law, creating the FSBS and reprogramming time deposits at BROU and BHU for three years (July 30, 2002).
- Government announced a bill to restructure suspended banks and create a new bank (Nuevo Banco Comercial) out of assets of Banco Montevideo, Caja Obrera and Banco Comercial (November 26, 2002).

### Chronology of key events (2001–2002)
- 2001
  - December 2 — Argentina imposes a deposit freeze (“corralito”).
  - December 12 — IMF suspends loan disbursements to Argentina.
  - December 23 — Argentina announces the default on its sovereign debt.
- 2002
  - January 4 — The government announces an increase of the exchange rate band from 6 to 12 percent, and an acceleration of the rate of depreciation from 1.2 to 2.4 percent per month.
  - January 11 — Standard and Poor’s (S&P) lowers Uruguay’s outlook to negative reflecting concerns regarding the continued contraction of the economy and doubts about the ability to reduce the fiscal deficit.
  - January 15 — Uruguay widens its Crawling Exchange Rate Band.
  - End-January — Rumors about problems at Banco Galicia Uruguay and Banco Comercial surface.
  - February 3 — Argentina tightens the deposit freeze (“corralón”).
  - February 13 — Central bank of Uruguay suspends operations of Banco Galicia (Uruguay).
  - February 15 — S&P lowers Uruguay’s sovereign debt rating to BB+ (below investment grade) on concerns over the fiscal deficit, weak growth prospects and fragility of the domestic financial system.
  - Late-February — Domestic bank problems spread to other institutions. Banco Montevideo and Banco La Caja Obrera are placed under intensive supervision, after experiencing significant deposit runs. Banco Commercial is capitalized by the government.
  - March 13 — Fitch lowers Uruguay’s credit rating below investment grade.
  - March 25 — IMF completes the last review of the 2000 SBA and approves a new SBA for SDR 2.0 billion.
  - April 19-29 — A bank holiday is imposed in Argentina, leading to an acceleration of the deposit run by non-residents in Uruguay.
  - April 26 — Banco Comercial received financial assistance from the government (through deposits from CND) and is recapitalized by shareholders. The central Bank reaffirms its commitment to the exchange rate regime.
  - May 5 — Moody’s lowers Uruguay’s credit rating.
  - May 14 — S&P lowers Uruguay’s sovereign debt rating by two notches, to BB-, on account of fiscal problems and concerns over exchange rat and monetary policies.
  - May 28 — The IMF and the IDB announce upcoming augmentations of financial assistance to Uruguay (of about US$ 3 billion over two years). Fitch lowers Uruguay rating to B+.
  - May 29 — Parliament approves the “Ley de Estabilidad Fiscal”.
  - June 19 — The monetary regime of adjustable bands is abandoned, and the exchange rate is allowed to float.
  - June 21 — Central Bank intervenes Banco Montevideo/Banco La Caja Obrera (Uruguay’s third largest bank).
  - June 25 — The IMF increases its Uruguay’s SBA by about US$ 1.5 billion.
  - July 10 — Moody’s lowers Uruguay’s credit rating from Ba2 to B1.
  - July 23 — The Minister of Finance and Members of the Board of the Central Bank resign.
  - July 26 — S&P lowers Uruguay’s credit rating to B, quoting pressures on the financial system and weak fiscal accounts.
  - July 30 — A 5-day bank holiday is declared. Operations of Banco Montevideo, and La Caja Obrera are suspended for 60 days. Banco Comercial and Banco de Credito are suspended for 30 days. Congress passes the Banking System Stability Law, creating the FSBS and reprogramming time deposits at BROU and BHU for three years.
  - August 5 — The bank holiday is lifted.
  - August 8 — A new Letter of Intent is signed with the IMF.
  - August 20 — Suspended banks start reimbursing sight deposits.
  - August 28 — The ceiling for savings account withdrawals are increased.
  - September–December — Operations of Banco Montevideo, Banco La Caja Obrera, Banco Comercial and Banco de Credito are repeatedly suspended.
  - November 21 — S&P lowers Uruguay’s credit rating to B-.
  - November 26 — The government announces a bill to restructure suspended banks, and creating a new bank (Nuevo Banco Comercial) out of the assets of Banco Montevideo, Caja Obrera and Banco Comercial.
  - December 2 — Argentina terminates the deposit freeze.

### Banking sector specifics and regulatory context
- Prior to the 2002 crisis, the banking regulatory framework in Uruguay did not distinguish between resident and non-resident deposits, and some banks (e.g. Banco Galicia Uruguay) relied heavily on non-resident deposits as their deposit base.
- At the onset of the crisis almost half of deposits were held by nonresidents, largely Argentines.
- Troubled institutions referenced explicitly include Banco de Galicia (Uruguay), Banco Comercial, Banco Montevideo, La Caja Obrera, Banco de Credito, BROU and BHU.
- Actions taken included suspension of operations, intensive supervision, government capitalization, and creation of the FSBS to guarantee dollar sight and saving deposits of suspended institutions and state-owned banks.

*Source: Appendix 3. Uruguay - Chronology of main events during the 2002 crisis*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2015/_wp1543.pdf_