## _wp12117 — Foreign Banks and the Vienna Initiative: Turning Sinners into Saints? (IMF Working Paper WP/12/117)

## Source details

**Canonical URL:** [_wp12117 — Foreign Banks and the Vienna Initiative: Turning Sinners into Saints? (IMF Working Paper WP/12/117)](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp12117.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp12117.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp12117.pdf.json)

---

### Overview and context
- Title and authors:
  - Foreign Banks and the Vienna Initiative: Turning Sinners into Saints?
  - Ralph De Haas, Yevgeniya Korniyenko, Elena Loukoianova, and Alexander Pivovarsky
  - IMF Working Paper WP/12/117; Monetary and Capital Markets Department; April 2012
- Scope:
  - Uses data on 1,294 banks in Central and Eastern Europe (ECA region) to analyze how bank ownership and creditor coordination in the form of the Vienna Initiative affected credit growth during the 2008–09 crisis.
- Background timeline and drivers:
  - Large-scale foreign bank entry into the ECA region began with the start of the transition process in 1989.
  - Interbank liquidity sharply reduced in the second half of 2007; Lehman Brothers collapsed in September 2008, accelerating deleveraging.
  - February 2009: EBRD, EIB, and World Bank Group launched a “Joint IFI Action Plan in support of banking systems and lending to the real economy in Central and Eastern Europe” within the Vienna Initiative.
  - March 2012: Vienna 2.0 principles agreed in Brussels to improve coordination during subsequent stresses (Eurozone sovereign debt crisis).

### The Vienna Initiative: purpose, structure, and instruments
- Genesis and participants:
  - Informal discussions initiated by the Austrian government and multinational banks with high exposures to the ECA region toward the end of 2008.
  - VI meetings included main IFIs, Ministries of Finance, central banks and bank regulators from host and home countries, European Union, and European Central Bank.
- Two main parts:
  - European Bank Coordination Initiative (EBCI):
    - Secured private sector involvement for five countries with IMF-EU programs—Bosnia and Herzegovina, Hungary, Latvia, Romania, and Serbia.
    - 17 parent banks pledged via commitment or comfort letters to maintain exposures and recapitalize subsidiaries for the duration of the IMF-EU programs.
    - Country-specific examples:
      - Hungary: maintain at least 95 percent of their September 2008 exposure and ensure “prudent capitalization of their subsidiaries.”
      - Romania: increase minimum capital adequacy ratio from 8 to 10 percent and fully maintain their March 2009 exposure for the time of the IMF program.
      - Bosnia and Herzegovina and Serbia: rollover 100 percent of their exposure (as of December 2008) and recapitalize subsidiaries if and when needed.
    - Commitment letters signed: Romania and Serbia in March 2009; Hungary in May 2009; Bosnia and Herzegovina in June 2009. Latvia: comfort letters in January 2009 and February 2010; no bilateral commitment letter.
  - Joint IFI Action Plan (launched February 27, 2009):
    - Aggregate IFI commitment: EUR 24.5 billion to support large cross-border banks.
    - By end-September 2009 IFIs had provided EUR 16.3 billion; by end-December 2010 EUR 33.2 billion made available in total.
    - IFI support linked to European Bank Coordination Initiative and conditional on IMF-EU programs.
- Institutional safeguards:
  - Full Forum Meetings for policy discussions.
  - European Commission assurances that banks receiving state support would not be forced to downsize presence in the ECA region.
  - Horizontal meetings (September 2009 and March 2010) to mitigate risk of withdrawals from non-commitment countries.

### Data, sample, and variable definitions
- Main dataset:
  - Balance sheet and income statement data for 1,294 banks in the ECA region during 1999–2009 from Bureau van Dijk’s BankScope database; all data denominated in US dollars.
  - Panel is unbalanced; for the crisis year 2009 dataset contains 1,098 banks.
  - Banks with fewer than three consecutive years of data are disregarded.
- Hand-collected variables:
  - Crisis-related government support (capital injections, bank-specific guarantees, asset sales) from European Commission publications, IMF, Reuters, and bank websites.
  - Pre-crisis lending or equity participation by EBRD, World Bank Group, or EIB.
  - Time-varying ownership structure for each bank (1999–2009).
- Main dependent variable:
  - Annual gross nominal credit growth defined as net loans plus loan loss reserves.
  - One percent observations with the highest loan growth trimmed.
  - Results robust when using net loans as dependent variable.
- Regional coverage:
  - Albania, Armenia, Azerbaijan, Belarus, Bosnia and Herzegovina, Bulgaria, Croatia, Czech Republic, Estonia, Georgia, Hungary, Kazakhstan, Kyrgyzstan, Latvia, Lithuania, FYR of Macedonia, Moldova, Mongolia, Montenegro, Poland, Romania, Russia, Serbia, Slovak Republic, Slovenia, Tajikistan, Turkey, Turkmenistan, Ukraine, and Uzbekistan.

### Stylized facts and descriptive statistics (selected)
- Pre-crisis (1999–2007):
  - Abundant annual growth on average 50 percent.
  - Foreign banks grew somewhat faster than domestic banks before the crisis, especially in SEE.
- Crisis dynamics (2008–2009):
  - Lending slowed markedly in 2008 and further in 2009.
  - Both foreign and domestic banks cut credit growth significantly during the crisis.
  - In SEE foreign banks reduced credit growth relatively fast in 2008; in CEB and SEE credit growth of domestic and foreign banks converged in 2008 and 2009.
  - In CIS, Russia, and Turkey foreign banks performed relatively well during the crisis as domestic banks decelerated at a fast pace.
  - Lending by state banks held up quite well in 2009 in the CIS and SEE (examples: VTB grew by 53 percent in 2009; Gazprombank increased credit by 38 percent; Bulgarian Municipal Bank grew by 13 percent; Banka Celje grew by 11 percent in 2009).
- Sample composition and averages:
  - Ownership composition: 29 percent foreign owned, 63 percent domestic private banks, and 8 percent state-owned banks.
  - About 4 percent of all banks received some form of government support during 2008–09 (excluding the VI).
  - Loan-to-deposit ratio: on average 91.4 percent.
  - Loan loss reserves to gross loans: 4.5 percent for foreign banks (average) and 6.8 percent for state banks (average).
- Regional average annual credit growth by ownership (selected entries from Section 3 tables; values preserved exactly):
  - CEB, Foreign: Average annual credit growth 1999-2009 0.39; 2009 0.06; 1999-2007 0.14; 2008 0.06.
  - SEE, Foreign: Average annual credit growth 1999-2009 0.67; 2009 0.05; 1999-2007 0.26; 2008 0.05.
  - CIS, Foreign: Average annual credit growth 1999-2009 0.66; 2009 0.00; 1999-2007 0.43; 2008 0.00.
  - Russia, Foreign: Average annual credit growth 1999-2009 0.53; 2009 0.15; 1999-2007 0.44; 2008 0.15.
  - Turkey, Foreign: Average annual credit growth 1999-2009 0.44; 2009 0.18; 1999-2007 0.05; 2008 0.18.

### Identification strategy and empirical methodology
- Cross-sectional 2009 specification (foreign subsidiaries in five VI countries):
  - ΔLij (percentage gross credit growth of bank i in country j in 2009) regressed on pre-crisis growth ΔLij,2005-07, VIij dummy, Xijt pre-crisis controls, and country fixed effects jη.
  - Country fixed effects used to control for local credit demand and compare within-country differences between subsidiaries whose parents did and did not sign commitment letters.
- Panel specification (1999–2009):
  - ΔLijt (percentage gross credit growth) with bank fixed effects ijη, year fixed effects tμ.
  - Ownership dummies OWNijt (domestic private control, state, foreign).
  - Vienna participation dummies: VIENNA COUNTRYj, VIENNA PARENTij, VIENNA LETTERij, PARENT SIGNED ELSEWHEREij, NON-VIENNA PARENTij.
  - SUPPORTij dummy for parent bank receiving home-country government support.
  - Crisis dummies for 2008 and 2009 and interactions with ownership and Vienna variables.
  - Estimation by OLS with bank fixed effects; robust standard errors; controls for bank-specific and host-country macro variables.

### Empirical results — cross-section and panel (selected findings and coefficients)
- Cross-sectional 2009 (foreign subsidiaries in five VI countries; No. observations: 54):
  - Vienna letter coefficients on credit growth:
    - Column (1): 0.022 (p-value (0.661))
    - Column (2): 0.156* (p-value (0.059))
    - Column (3): 0.166** (p-value (0.050))
  - Vienna letter coefficients on asset growth:
    - Column (4): 0.122*** (p-value (0.006))
    - Column (5): 0.200*** (p-value (0.001))
    - Column (6): 0.237*** (p-value (0.000))
  - No. of observations: 54; R-squared ranges: 0.12 to 0.72 across columns.
  - Interpretation: Signing commitment letters positively associated with subsidiary credit and asset growth in 2009; government support also associated with faster growth in some specifications.
- Panel regressions (1999–2009; No. observations: 4,805; No. banks: 1,294; R-squared: 0.315 in baseline):
  - Selected coefficients (percent; robust p-values in brackets):
    - Foreign bank: 0.201***  (0.000)
    - Crisis 2009: -0.191***  (0.000)
    - Crisis 2008 * Foreign: -0.137***  (0.000)
    - Crisis 2009 * State: 0.125***  (0.005)
    - GDP growth: 1.764***  (0.000)
    - Inflation: -0.752***  (0.000)
    - Loan loss reserves/gross loans (lag): -0.011***  (0.000)
  - Vienna Initiative interactions (selected, Table 4):
    - Vienna country * 2008: -0.147***  (0.000)
    - Vienna country * 2009: -0.012  (0.761)
    - Vienna parent * 2009: 0.091**  (0.037)
    - Vienna letter * 2008: -0.216***  (0.001)
    - Vienna letter * 2009: -0.024  (0.756)
    - Parent signed elsewhere * 2008: -0.101**  (0.039)
    - Parent signed elsewhere * 2009: 0.087*  (0.067)
    - Non-Vienna parent * 2008: -0.114**  (0.019)
  - Interpretation:
    - In 2008 (pre-VI), credit growth dropped more sharply in the five countries that would be supported (about 14 percentage points sharper).
    - By 2009, VI countries’ credit declines normalized to be similar to other countries.
    - Foreign banks that were part of the VI were about 11 percentage points higher in credit growth in 2009 compared to non-VI foreign banks, all else equal.
    - Banks whose parents signed commitment letters experienced sharper contractions in 2008 but were relatively more stable in 2009.

### Robustness checks (summary of Table 5)
- Estimators and alternative samples:
  - Variables denominated in euros for highly euroized countries.
  - Sample restricted to banks with at least seven years of observations.
  - Panel-corrected standard errors (PCSE); Arellano-Bond GMM; Hausman-Taylor IV estimator.
  - Dependent variable alternatives: net loan growth; growth of total assets.
- Selected interaction coefficients (Vienna letter * 2008 and * 2009, parent signed elsewhere, non-Vienna parent, state interactions) preserved numeric values across specifications:
  - Vienna letter * 2008 estimates (examples): (1) -0.250*** (0.000); (3) -0.308*** (0.000); (4) -0.344*** (0.000).
  - Vienna letter * 2009 estimates (examples): (1) -0.021 (0.739); (3) -0.068 (0.344).
  - Parent signed elsewhere * 2009 estimates (examples): (2) 0.073** (0.050); (4) 0.096** (0.033); (5) 0.112** (0.024).
  - State * 2009 estimates (examples): (1) 0.091** (0.048); (5) 0.144** (0.012).
- Robustness highlights:
  - Stabilizing effect of Vienna participation and of state ownership in 2009 persists across alternative denominators, sample restrictions, and estimators.
  - Replacing gross loan growth with net loan growth or with total-asset growth largely leaves main conclusions intact, except that growth in total assets removes the positive effect of a parent bank that signed in another country.

### Selection into the Vienna Initiative
- Probit results (selection correlates):
  - Larger subsidiaries (asset size) and parents with greater regional exposure (number of subsidiaries in the ECA region) more likely to be part of the VI.
  - Parents of subsidiaries with relatively low loan-quality (high loan-loss reserves) more likely to sign commitment letters.
  - Parent banks with lower tier 1 capital ratios more likely to sign commitment letters.
- Implication:
  - VI banks were on average larger, less well capitalized, and carried more non-performing loans—selection that biases against finding a positive VI impact on credit growth and underscores need to control for bank characteristics.

### Policy conclusions and recommendations
- Effectiveness and role of the VI:
  - The Vienna Initiative was a relatively successful example of catalytic funding where public funds by IMF, EU, and development institutions were complemented by a coordinated (non-coercive) bail-in of private-sector lenders.
  - VI helped close external funding gaps at the macroeconomic level and softened deleveraging in the ECA region, reducing the risk of an uncoordinated ‘rush to the exit’.
- Cross-border externalities and supervisory implications:
  - Cross-border negative externalities arise from rapid liquidity movements within bank groups; better coordination is required even with effective supervision.
  - Policy options to improve cross-border handling of banking stress:
    - Better coordination, cooperation, and information exchange between supervisors (Vienna 2.0 focus).
    - Move toward greater reliance on local sources of funding: development of local capital markets to top up domestic deposit funding with local wholesale funding.
    - Structural measures: resolution fund at EU level or network of national resolution funds; creation of a pan-European supervisor for large groups with adequate capital and liquidity regulation and host-country macroprudential supervision to curb externally funded credit booms.
- Caution on ring-fencing and forced subsidiarisation:
  - Ring-fenced subsidiaries increase aggregate capital needs and reduce benefits of international diversification; they impede efficient internal capital markets.
  - Forced subsidiarisation may be a second-best option reflecting insufficient cross-border cooperation and burden sharing.
- Practical recommendation:
  - Many foreign bank subsidiaries in Central and Eastern Europe will need to move toward funding policies that rely more on local and less on parent-bank funding, requiring development of local capital markets.

### Timeline and institutional milestones (selected chronology)
- January 23, 2009: First Vienna Initiative meeting at the Austrian Ministry of Finance.
- February 27, 2009: Joint IFI Action Plan launched offering up to EUR 24.5 billion.
- March–June 2009: Commitment letters signed for Romania, Serbia, Hungary, and Bosnia and Herzegovina; Latvia issued concluding statements/comfort letters in September 2009.
- September 24, 2009: First ‘horizontal’ Full-Forum meeting in Brussels.
- End-2010: Expiry of the Vienna Initiative Joint IFI Action Plan.
- March 2011: Final Report on implementation of the Joint IFI Action Plan.
- March 2012: Vienna 2.0 principles agreed in Brussels (to improve coordination during subsequent stresses).

### Variable definitions and summary statistics (selected, preserving source terminology)
- Credit growth (1999-2009): Annual percentage growth in gross loans (= net loans plus loan loss reserves). Source: BankScope.
- Asset growth (1999-2009): Annual percentage growth in total assets. Source: BankScope.
- Vienna parent (2009): 1 = subsidiary is owned by a parent bank that signed up to the Vienna Initiative. Source: EBRD.
- Vienna letter (2009): 1 = parent bank signed a commitment letter in the subsidiary's host country. Source: EBRD.
- Support (2008-2009): 1 = bank received government support (asset sale, capital injection, and/or guarantee). Source: Websites.
- Summary statistics (Panel A, full sample, Obs = 647 unless stated):
  - Credit growth: Mean 40.10; Std. Dev. 48.8; Min -66; Max 249.
  - Loan/deposit ratio: Mean 91.4; Std. Dev. 50.6; Min 8; Max 399.
  - Loan quality: Mean 5.7; Std. Dev. 5.6; Min 0; Max 48.
  - Private domestic: Mean 0.63; State bank: Mean 0.08; Support mean 0.04.
- Country-level (Panel B, 1999-2009):
  - GDP growth: Mean 4.2; Std. Dev. 6.0; Min -18.0; Max 30.5.
  - Inflation: Mean 10.2; Std. Dev. 10.1; Min -8.5; Max 293.7.

*Source: IMF Working Paper WP/12/117 — “Foreign Banks and the Vienna Initiative: Turning Sinners into Saints?” (content unit: _wp12117).*

### Section 1

### _wp12117 - Section 1

### Overview and context
- Title and authors:
  - Foreign Banks and the Vienna Initiative: Turning Sinners into Saints?
  - Ralph De Haas, Yevgeniya Korniyenko, Elena Loukoianova, and Alexander Pivovarsky
  - IMF Working Paper WP/12/117; Monetary and Capital Markets Department; April 2012
- Scope:
  - Uses data on 1,294 banks in Central and Eastern Europe (ECA region) to analyze how bank ownership and creditor coordination in the form of the Vienna Initiative affected credit growth during the 2008–09 crisis.
- Background timeline and drivers:
  - Large-scale foreign bank entry into the ECA region began with the start of the transition process in 1989.
  - Interbank liquidity sharply reduced in the second half of 2007; Lehman Brothers collapsed in September 2008, accelerating deleveraging.
  - In February 2009, the EBRD, EIB, and World Bank Group launched a ‘Joint IFI Action Plan in support of banking systems and lending to the real economy in Central and Eastern Europe’ within the Vienna Initiative.
  - Vienna 2.0 principles were agreed in Brussels in March 2012 to improve coordination during subsequent stresses (Eurozone sovereign debt crisis).

### Vienna Initiative (VI): structure and objectives
- Genesis and participants:
  - Informal discussions initiated by the Austrian government and multinational banks with high exposures to the ECA region toward the end of 2008.
  - VI meetings included main IFIs, Ministries of Finance, central banks and bank regulators from host and home countries, European Union, and European Central Bank.
- Instruments and commitments:
  - Multinational banks signed country-specific commitment and comfort letters pledging to maintain cross-border exposures and continue providing credit to firms and households; banks committed to keep subsidiaries adequately capitalized and provide sufficient liquidity.
  - The VI combined macro-financial support by the IMF and the EU (‘bail-out’) with IFI funding and coordinated private creditor commitments (a coordinated ‘bail in’).

### Key empirical findings (Section II — Main findings)
- Aggregate lending behavior:
  - Both foreign and private domestic banks sharply curtailed credit growth during the crisis.
  - State-owned domestic banks were relatively stable lenders.
  - Banks that took part in the VI were relatively stable lenders.
- Vienna Initiative participants vs. non-participants:
  - Foreign banks that participated in the VI were relatively stable lenders compared with others.
  - VI banks did not retrench from non-VI countries in order to maintain exposures to countries where they signed commitment letters.
  - Participation in the VI led to positive rather than negative spillover effects to other countries.
- Government support and retrenchment:
  - No evidence that home-country government support made foreign banks retrench more from the ECA region.
- Comparative dynamics by bank type and period:
  - State-owned banks became somewhat more stable credit sources particularly during the second year of the crisis.
  - Much of the reduction in lending took place in 2009 (in contrast to some earlier studies focusing on 2007–08).

### Contributions to the literature and interpretation
- Foreign banks as stabilizers vs. shock transmitters:
  - Confirms mixed roles of multinational banks: can absorb local shocks (stabilizing) and transmit foreign shocks (destabilizing).
  - Novel contribution: first to show how international policy and creditor coordination (the VI) can partially alleviate cross-border transmission.
- State support and state ownership:
  - Provides systematic comparison of foreign, private domestic, and state banks during the crisis.
  - Contrary to Rose and Wieladek (2011), no negative impact of state support on lending by Western European bank subsidiaries in the ECA region is found.
  - Suggests either strong regional commitment by banks (a “second home market”) or positive effect of European Commission statements that banks receiving state support would not be expected to reduce presence in Central and Eastern Europe.
- Catalytic role of IMF and coordinated support:
  - Analyzes impact of combination of IMF funding and active creditor coordination rather than narrow country-level funding gaps.
  - Relates to theoretical arguments on contingent support acting as a seal of approval and to case studies (e.g., Korea 1997, Brazil 1999) where roll-overs were supported by official sector coordination.

### Data and next steps (as described)
- Dataset:
  - Bank-level dataset covering 1,294 banks in the ECA region used to compare VI participants vs. non-participants and to analyze within-bank lending across countries where individual banks did and did not sign commitment letters.
- Paper structure (forward look):
  - Section 2 provides more details about the Vienna Initiative.
  - Section 3 describes the data.

*Source: IMF Working Paper WP/12/117 — “Foreign Banks and the Vienna Initiative: Turning Sinners into Saints?” (Section 1).*

### Section 2

### _wp12117 - Section 2

### The Vienna Initiative: purpose, structure, and timeline
- Goal: safeguard a continued commitment of parent banks to their subsidiaries in the ECA region.
- Inaugurated on January 23rd 2009 in Austria as a coordination platform for multinational banks, their home and host country supervisors, fiscal authorities, the IMF, and development institutions.
- Trigger: November 2008 letters from a number of pan-European banks to the European Commission (copying the EBRD and EIB) calling for a quick and coordinated response to ensure sufficient funding for banks operating in the region.
- Two main parts:
  - European Bank Coordination Initiative (EBCI): secured private sector involvement for five countries with IMF-EU programs—Bosnia and Herzegovina, Hungary, Latvia, Romania, and Serbia—where substantial amounts of foreign currency debt were maturing and external financing gaps were opening up.
    - A total of 17 parent banks pledged, via so-called commitment or comfort letters, to maintain their exposures and to recapitalize subsidiaries for the duration of the IMF-EU programs.
    - Country-specific commitments varied:
      - Hungary: banks promised to ensure a “prudent capitalization of their subsidiaries” and to maintain at least 95 percent of their September 2008 exposure.
      - Romania: banks promised to “increase the minimum capital adequacy ratio for each subsidiary from 8 to 10 percent” and to fully maintain their March 2009 exposure for the time of the IMF program.
      - Bosnia and Herzegovina and Serbia: banks committed to rollover 100 percent of their exposure (as of December 2008) and to recapitalize subsidiaries if and when needed.
    - Commitment letters were signed for Romania and Serbia in March 2009, Hungary in May 2009, and Bosnia and Herzegovina in June 2009. In the case of Latvia, parent banks issued a concluding statement on September 11, 2009 and a press release on September 14, 2009; bilateral commitment letters were therefore never signed but banks signed “comfort letters” in January 2009 and February 2010 committing to comply with all regulatory requirements.
    - Some commitments were reaffirmed later in 2009; as the crisis subsided, roll-over commitments were lowered by early 2010 in some cases.
    - Concern noted: focusing commitments on five core countries could induce withdrawals from countries without exposure commitments (e.g., Poland and the Czech Republic); this risk was mitigated by agreements extending informal commitments to Central and Eastern Europe more broadly and by “horizontal meetings” in September 2009 and March 2010 focusing on lending to the region as a whole.
  - Joint IFI Action Plan (launched February 27, 2009): EBRD, EIB, and the World Bank Group announced coordinated financial support “to support banking sector stability and lending to the real economy in crisis-hit Central and Eastern Europe.”
    - During spring 2009 these institutions met several times with 17 banking groups covering over 60 percent of all banking assets in the region and produced a “joint needs assessment.”
    - Aggregate IFI commitment: EUR 24.5 billion to support large cross-border banks.
    - By end-September 2009, banks had received EUR 16.3 billion of IFI support in the form of senior loans, tier 1 and 2 capital, trade finance, facilities for small business loans, and syndicated loans.
    - By end-December 2010, EUR 33.2 billion had been made available in total.
    - The Joint IFI Action Plan was closely linked to the European Bank Coordination Initiative, as banks’ commitments were conditional on IMF-EU programs being in place.

### Institutional coordination and safeguards
- The VI included “Full Forum Meetings” as a platform for policy discussions across the region and with western counterparts.
- The European Commission ensured that banks benefiting from state support would not be forced to downsize their presence in the ECA region.
  - March 2009 Emergency Summit of EU leaders confirmed that national bank support packages should not lead to restrictions on banks’ eastern European subsidiaries.
- The VI created cross-country and within-bank variation exploited in the empirical analysis:
  - In each of the five VI countries there were two groups of foreign bank subsidiaries: those with parents that were part of the VI in that country and those with parent banks that were not.
  - Parent banks sometimes signed commitment letters for some countries but not others (examples: Erste Bank signed in Hungary but not in Serbia; NLB Bank committed in Bosnia and Herzegovina but not in Serbia).
  - Parent banks varied in whether they received government support (example: Commerzbank received capital support from the German government whereas Deutsche Bank did not).

### Data and descriptive statistics: sample, definitions, and coverage
- Main dataset: balance sheet and income statement data for 1,294 banks in the ECA region during 1999–2009 from Bureau van Dijk’s BankScope database; all data denominated in US dollars.
- Sample restrictions and panel:
  - Banks with fewer than three consecutive years of data are disregarded.
  - The panel is unbalanced; for the crisis year 2009 the dataset contains 1,098 banks.
  - Combined with macroeconomic information from the IMF International Financial Statistics.
- Hand-collected variables:
  - Crisis-related government support in home and host countries (capital injections, bank-specific guarantees, and asset sales to the government) from European Commission publications, the IMF, Reuters, and bank websites.
  - Whether each bank had pre-crisis lending from or equity participation by EBRD, World Bank Group, or EIB.
  - Time-varying ownership structure of each bank: period 1999–2004 ownership from De Nicolò and Loukoianova (2007); later years from bank websites and annual reports.
    - For foreign bank subsidiaries, the year t when they became part of a group is traced; newly established subsidiaries are used from year t onwards; takeovers are used from year t+1 onwards; parent bank mergers during year t are included from t+1 onwards.
- Main dependent variable:
  - Annual gross nominal credit growth defined as net loans plus loan loss reserves (this corrects for changes in (net) loans due to provisioning and write-offs).
  - One percent observations with the highest loan growth are trimmed to exclude M&A-related observations.
  - Results are robust when using net loans as dependent variable (see Table 5 referenced).
- Regional definition of ECA for this study:
  - Albania, Armenia, Azerbaijan, Belarus, Bosnia and Herzegovina, Bulgaria, Croatia, Czech Republic, Estonia, Georgia, Hungary, Kazakhstan, Kyrgyzstan, Latvia, Lithuania, FYR of Macedonia, Moldova, Mongolia, Montenegro, Poland, Romania, Russia, Serbia, Slovak Republic, Slovenia, Tajikistan, Turkey, Turkmenistan, Ukraine, and Uzbekistan.

### Stylized facts and descriptive findings
- Pre-crisis credit growth:
  - Abundant annual growth during 1999–2007 (on average 50 percent).
  - Foreign banks grew somewhat faster than domestic banks before the crisis, especially in south-eastern Europe (SEE).
  - Foreign bank subsidiaries typically had easier access to foreign funding and were less constrained by local funding availability.
- Crisis dynamics (2008–2009):
  - Lending slowed markedly in 2008 and further in 2009 (Table 1 and Figure 1 referenced).
  - Both foreign and domestic banks had to cut credit growth significantly during the crisis.
  - In SEE, foreign banks reduced credit growth relatively fast in 2008 compared to their very high pre-crisis expansion rates; in absolute terms credit growth of domestic and foreign banks in Central and Eastern Europe & the Baltic States (CEB) and SEE converged in 2008 and 2009.
  - In the Commonwealth of Independent States (CIS), Russia, and Turkey foreign banks performed relatively well during the crisis as domestic banks decelerated at a fast pace; Kazakh, Russian, and Ukrainian domestic banks were heavily leveraged with large amounts of foreign debt that became difficult to roll over.
  - Lending by state banks held up quite well in 2009 in the CIS and SEE (examples: VTB grew by 53 percent in 2009; Gazprombank increased credit by 38 percent; Bulgarian Municipal Bank grew by 13 percent; Slovenian Banka Celje grew by 11 percent in 2009).
- Vienna Initiative participant banks:
  - Foreign banks that became part of the VI in 2009 were among the fastest growing banks in the ECA region before the crisis (difference statistically significant at the 1 percent level).
  - During the crisis, both VI-participating and non-participating foreign banks slowed lending; in 2009 credit contracted for VI banks in CEB.
- Sample composition and key averages:
  - Ownership composition: 29 percent foreign owned, 63 percent domestic private banks, and 8 percent state-owned banks.
  - About 4 percent of all banks received some form of government support (excluding the VI) during the crisis years 2008–09.
  - Banks’ loan-to-deposit ratio (indicator of wholesale funding use): on average 91.4 percent.
  - Loan loss reserves to gross loans: 4.5 percent for foreign banks (on average) and 6.8 percent for state banks (on average).

### Sources of identification and variation exploited in analysis
- Cross-bank and within-bank variation arises from:
  - Variation in whether parent banks signed commitment letters in specific countries.
  - Variation in whether parent banks received government support.
  - Variation over time in ownership changes (timing of entry, takeovers, and parent mergers).
- Tables and figures referenced for further detail (Table 1, Tables 6–10, Figure 1) document credit growth patterns, variable definitions, descriptive statistics, correlation matrix, and lists of banks participating in the VI and the specific countries of commitment.

*Source: _wp12117 - Section 2*

### Section 3

### _wp12117 - Section 3

### Data summary: bank counts and average annual credit growth by region and ownership
- CEB (Central Europe and the Baltic States)
  - Domestic state: No. banks 94; Average annual credit growth 1999-2009 0.27; 2009 0.11; 1999-2007 0.09; 2008 0.11
  - Domestic private: No. banks 497; Average annual credit growth 1999-2009 0.32; 2009 0.07; 1999-2007 0.17; 2008 0.07
  - Foreign: No. banks 748; Average annual credit growth 1999-2009 0.39; 2009 0.06; 1999-2007 0.14; 2008 0.06
  - Vienna: No. banks 74; Average annual credit growth 1999-2009 0.45; 2009 -0.05; 1999-2007 0.13; 2008 -0.05
  - Non-Vienna: No. banks 674; Average annual credit growth 1999-2009 0.38; 2009 0.07; 1999-2007 0.15; 2008 0.07
  - Total: No. banks 1,339; Average annual credit growth 1999-2009 0.36; 2009 0.07; 1999-2007 0.16; 2008 0.07
- SEE (south-eastern Europe)
  - Domestic state: No. banks 79; Average annual credit growth 1999-2009 0.36; 2009 0.19; 1999-2007 0.22; 2008 0.19
  - Domestic private: No. banks 340; Average annual credit growth 1999-2009 0.49; 2009 0.09; 1999-2007 0.21; 2008 0.09
  - Foreign: No. banks 527; Average annual credit growth 1999-2009 0.67; 2009 0.05; 1999-2007 0.26; 2008 0.05
  - Vienna: No. banks 165; Average annual credit growth 1999-2009 0.78; 2009 0.06; 1999-2007 0.20; 2008 0.06
  - Non-Vienna: No. banks 362; Average annual credit growth 1999-2009 0.63; 2009 0.05; 1999-2007 0.28; 2008 0.05
  - Total: No. banks 946; Average annual credit growth 1999-2009 0.57; 2009 0.07; 1999-2007 0.24; 2008 0.07
- CIS (Commonwealth of Independent States)
  - Domestic state: No. banks 129; Average annual credit growth 1999-2009 0.47; 2009 0.14; 1999-2007 0.52; 2008 0.14
  - Domestic private: No. banks 675; Average annual credit growth 1999-2009 0.64; 2009 -0.06; 1999-2007 0.28; 2008 -0.06
  - Foreign: No. banks 311; Average annual credit growth 1999-2009 0.66; 2009 0.00; 1999-2007 0.43; 2008 0.00
  - Total: No. banks 1,115; Average annual credit growth 1999-2009 0.62; 2009 -0.01; 1999-2007 0.36; 2008 -0.01
- Russia
  - Domestic state: No. banks 173; Average annual credit growth 1999-2009 0.53; 2009 0.06; 1999-2007 0.50; 2008 0.06
  - Domestic private: No. banks 2,463; Average annual credit growth 1999-2009 0.51; 2009 0.00; 1999-2007 0.50; 2008 0.00
  - Foreign: No. banks 192; Average annual credit growth 1999-2009 0.53; 2009 0.15; 1999-2007 0.44; 2008 0.15
  - Total: No. banks 2,828; Average annual credit growth 1999-2009 0.51; 2009 0.01; 1999-2007 0.50; 2008 0.01
- Turkey
  - Domestic state: No. banks 25; Average annual credit growth 1999-2009 0.19; 2009 0.22; 1999-2007 0.09; 2008 0.22
  - Domestic private: No. banks 137; Average annual credit growth 1999-2009 0.44; 2009 0.13; 1999-2007 -0.05; 2008 0.13
  - Foreign: No. banks 89; Average annual credit growth 1999-2009 0.44; 2009 0.18; 1999-2007 0.05; 2008 0.18
  - Total: No. banks 251; Average annual credit growth 1999-2009 0.41; 2009 0.16; 1999-2007 0.00; 2008 0.16

- Source: Bankscope database.
- Notes: State banks are more than 30 percent owned by the state. Domestic private banks are majority owned by domestic private shareholders. Foreign banks are majority foreign owned. Vienna banks are subsidiaries of foreign banks that were part of the VI.

### Selection into the Vienna Initiative (VI)
- Determinants of VI participation (probit analysis summary):
  - Larger subsidiaries (asset size) and parents with greater regional exposure (number of subsidiaries in the ECA region) were more likely to be part of the VI.
  - Parents of subsidiaries with relatively low loan-quality (high loan-loss reserves) were more likely to sign commitment letters.
  - Parent banks with lower tier 1 capital ratios were also more likely to sign commitment letters.
- Implication:
  - VI banks were on average larger, less well capitalized, and carried more non-performing loans.
  - These selection effects bias against finding a positive impact of the VI on credit growth, highlighting the need to control for bank characteristics to reduce omitted-variable bias.

### Empirical methodology
- Cross-sectional specification for 2009 (foreign bank subsidiaries in five VI countries)
  - Equation (1):
    ij
    jijijijij
    XVILLεηγγγαγΔ⋅+⋅+Δ⋅+⋅+−2007,32072004,11
  - Definitions:
    - ΔLij is the percentage gross credit (or asset) growth of bank i in country j in 2009.
    - α1 is an intercept term and γ are coefficients or coefficient vectors.
    - ΔLij,2005-07 is the average annual percentage gross credit growth of bank i in country j during 2005-07.
    - VIij is a dummy indicating whether the parent bank of subsidiary i signed a commitment letter in country j.
    - Xijt is a matrix of pre-crisis (2007) control variables for the (parent bank of) subsidiary i.
    - jη are country fixed effects; ijε is the idiosyncratic error, ijε ~ IID(0, σ^2_ε).
  - Identification strategy:
    - Use country fixed effects to control for local credit demand and compare within-country differences between subsidiaries whose parents did and did not sign commitment letters.
- Panel-regression specification for 1999–2009
  - Key features:
    - Time-varying ownership dummies OWNijt distinguishing domestic private banks (control), state banks, and foreign banks.
    - Five time-invariant Vienna participation dummies:
      - VIENNA COUNTRYj: whether country was one of the five VI countries.
      - VIENNA PARENTij: parent bank signed one or more VI commitment letters (in country j or elsewhere).
      - VIENNA LETTERij: parent bank signed a VI commitment letter in country j.
      - PARENT SIGNED ELSEWHEREij: parent bank signed a VI commitment letter but not in country j.
      - NON-VIENNA PARENTij: parent bank did not sign any VI commitment letters.
    - SUPPORTij: dummy for parent bank receiving home-country government support.
    - Two crisis dummies CRISISt for 2008 and 2009; interactions with ownership and Vienna variables to capture crisis-period behavior.
  - Panel equation (2):
    ijtijtijttijtijtijtijt
    XCrisisViennaCrisisOwnOwnLεημγγγγαΔ43211⋅+⋅+⋅+⋅+⋅+⋅++=
  - Definitions:
    - ΔLijt is percentage gross credit growth of bank i in country j in year t.
    - OWNijt, VIENNAij, CRISISt, Xijt as described above.
    - ijη bank fixed effects; tμ year fixed effects; ijtε idiosyncratic error, ijtε ~ IID(0, σ^2_ε).
  - Estimation:
    - OLS with bank-specific fixed effects (Hausman tests indicate correlation between bank effects and regressors).
    - Robust estimators to correct for heteroscedasticity.
    - Year fixed effects to control for global trends.
  - Controls:
    - Bank-specific, time-varying financial characteristics, parent-bank characteristics, and host-country macro variables (including inflation to correct for nominal loan growth and PPP considerations).

### Empirical results: cross-sectional regressions for 2009 (Table 2)
- Sample: foreign bank subsidiaries in the five VI countries: Bosnia and Herzegovina, Hungary, Latvia, Romania, Serbia.
- Main findings:
  - Signing commitment letters is positively associated with subsidiary total credit and asset growth in 2009.
  - Subsidiaries of parent banks that received some form of government support also grew faster in 2009 versus non-supported banks.
  - Control variables behaved as expected: larger subsidiaries and subsidiaries with weaker balance sheets (higher loan loss reserves) had slower credit and asset growth; subsidiaries that grew faster pre-crisis exhibited slower growth in 2009 (mean reversion).
- Selected coefficients and statistics from Table 2 (cross-sectional regressions; columns (1)–(3) credit growth; (4)–(6) asset growth)
  - Vienna letter coefficients:
    - Column (1) credit growth: 0.022 (p-value reported below as (0.661))
    - Column (2) credit growth: 0.156* (p-value (0.059))
    - Column (3) credit growth: 0.166** (p-value (0.050))
    - Column (4) asset growth: 0.122*** (p-value (0.006))
    - Column (5) asset growth: 0.200*** (p-value (0.001))
    - Column (6) asset growth: 0.237*** (p-value (0.000))
  - Support (government support) coefficients:
    - Reported as positive and significant in several specifications; example p-values shown (0.009), (0.013), (0.009), (0.112) in table entries.
  - Pre-crisis average annual credit (asset) growth:
    - Negative coefficients: -0.232** (p-value (0.028)), -0.217** (p-value (0.047)), -0.168** (p-value (0.019)), -0.171** (p-value (0.015))
  - Bank size (lag):
    - Negative and significant: -0.119*** (p-value (0.001)), -0.132*** (p-value (0.001)), -0.096*** (p-value (0.002)), -0.124*** (p-value (0.000))
  - Loan loss reserves/gross loans (lag):
    - Coefficients: -0.017 (p-value (0.138)), -0.016 (p-value (0.154)), -0.011** (p-value (0.044)), -0.011* (p-value (0.050))
  - Equity/net loans (lag):
    - Coefficients: 0.001 (p-value (0.319)), 0.000 (p-value (0.481)), -0.001** (p-value (0.025)), -0.002*** (p-value (0.000))
  - Other controls shown in table: profitability, loan/deposit ratio, liquid assets/customer funding, income to assets, size parent bank, loan/deposit ratio parent bank (with corresponding p-values reported).
  - No. of observations: 54 across all columns.
  - R-squared:
    - Column (1): 0.12
    - Column (2): 0.51
    - Column (3): 0.53
    - Column (4): 0.36
    - Column (5): 0.66
    - Column (6): 0.72
  - Fixed effects: Country FE = Yes in relevant specifications; Bank FE = No; Parent FE = No (as reported in table notes).
- Interpretation:
  - The positive and statistically significant coefficients on Vienna letter and on Support in various specifications suggest that parent-bank commitment letters and government support were associated with stronger subsidiary lending and asset growth during 2009.
  - The results are robust to inclusion of parent and subsidiary controls and country fixed effects, but the sample consists only of foreign bank subsidiaries in the five VI countries and contains 54 observations in the cross-section.

*Source: Authors’ estimates (as reported in the provided section).*

### Section 4

### _wp12117 - Section 4

### Panel evidence on bank ownership and credit growth
- The panel regressions explain about 30 percent of the variation in banks’ annual credit growth rates.
- Before the crisis, foreign banks grew significantly faster than domestic banks, "exceeding their annual rate of growth by as much as 20 percentage points."
- Controls show:
  - Large banks, banks with an already high loan-to-deposit ratio, and banks with high loan loss reserves grew slower on average.
  - More solvent, liquid, and profitable banks expanded credit more quickly.
  - Credit growth was positively correlated with the business cycle (proxy for credit demand at the host-country level).
- Timing of deleveraging:
  - In 2008, foreign banks were the first to sharply curb their credit growth and this brought them back in line with the average growth rate of private domestic banks.
  - Domestic bank lending slowed mainly in 2009.
  - State banks also slowed in 2009, but the reversal was less sharp compared to private banks (possible use of state-owned banks to smooth aggregate lending).

Selected coefficient estimates from Table 3 (yearly credit growth, percent; robust p-values in brackets):
- State bank: -0.103  (0.151)
- Foreign bank: 0.201***  (0.000)
- Crisis 2008: -0.013  (0.635)
- Crisis 2009: -0.191***  (0.000)
- Crisis 2008 * Foreign: -0.137***  (0.000)
- Crisis 2009 * Foreign: 0.023  (0.426)
- Crisis 2009 * State: 0.125***  (0.005)
- Government support during crisis: 0.044  (0.276)
- GDP growth: 1.764***  (0.000)
- Inflation: -0.752***  (0.000)
- Profitability (lag): 0.002***  (0.007)
- Bank size (lag): -0.067***  (0.000)
- Net loans/deposit ratio: -0.002***  (0.000)
- Equity/net loans (lag): 0.004***  (0.002)
- Liquidity (lag): 0.003***  (0.000)
- Loan loss reserves/gross loans (lag): -0.011***  (0.000)
- Cost-to-income ratio (lag): 0.002***  (0.000)
- Constant: 1.005***  (0.000)
- No. of observations: 4,805
- R-squared: 0.315
- No. of banks: 1,294
- Sample period: 1999-2009

Notes:
- Column (1b) replicates column (1a) while including year fixed effects.
- The sum of the coefficient for Foreign bank and Crisis 2008*Foreign is just above or below zero (depending on the inclusion of year fixed effects).

### The Vienna Initiative (VI) and credit growth during 2008–09
- Main finding: In 2008 (before the VI was initiated) bank lending dropped significantly more in (future) VI countries compared to non-VI countries—"the adjustment in credit growth was about 14 percentage points sharper in the five countries that would need to be supported by the IMF and EU later on."
- By 2009, VI countries had “normalized” and the credit decline had become more in line with other countries in the region; no significant differences between VI and non-VI countries in terms of average lending contractions in 2009.
- Vienna parent / VI bank behavior:
  - Banks that signed VI commitment letters in 2009 decelerated less compared to domestic banks and non-VI foreign banks.
  - Compared to non-VI foreign banks, credit growth of foreign banks that were part of the VI was about 11 percentage points higher (all else equal).
- More detailed contrasts:
  - In countries where parent banks eventually signed a commitment letter, foreign bank subsidiaries did worse in 2008: the drop in their credit growth was more than 22 percentage points larger compared to all other banks (column 2b) and about 19 percentage points when compared to other foreign banks (column 4b).
  - These sharper drops in 2008 were not present in 2009, consistent with stabilization efforts by the IMF and IFIs in 2009.

Selected interaction coefficients from Table 4 (interaction terms between crisis years and VI variables; robust p-values in brackets):
- Vienna country * 2008: -0.147***  (0.000)  [All banks, column (1)]
- Vienna country * 2009: -0.012  (0.761)  [All banks, column (1)]
- Vienna parent * 2008: -0.055  (0.189)  [All banks, column (1)]
- Vienna parent * 2009: 0.091**  (0.037)  [All banks, column (1)]
- Vienna letter * 2008: -0.216***  (0.001)  [All banks, column (2a)/(2b)]
- Vienna letter * 2009: -0.024  (0.756)  [All banks, column (2a)/(2b)]
- Parent signed elsewhere * 2008: -0.101**  (0.039)
- Parent signed elsewhere * 2009: 0.087*  (0.067)
- Non-Vienna parent * 2008: -0.114**  (0.019)
- State * 2009: 0.088*  (0.053)
- Government support during crisis: -0.054  (0.168)  [All banks, column (1)]
- No. observations (All banks): 4,805; No. banks: 1,294; R-squared: 0.596
- No. observations (Foreign banks sample): 1,287; No. banks: 292; R-squared: 0.595

Interpretation and implications from the section:
- The Vienna Initiative appears to have mitigated further disproportionate credit contraction in the countries and banks most affected in 2008, with credit dynamics in 2009 becoming more similar to other ECA countries.
- Participation by parent banks (Vienna parent, Vienna letter) is associated with differential credit outcomes across 2008 and 2009: sharper contractions in 2008 in VI-related countries/subsidiaries, and relatively better (less negative) performance for VI banks in 2009.
- State-owned banks displayed relative resilience in 2009 compared to private banks, consistent with a smoothing role during the crisis.

*Source: Authors’ estimates (Section 4, _wp12117 - Section 4).*

### Section 5

### _wp12117 - Section 5

### Empirical setup and sample
- Dependent variable: yearly credit growth (percent).
- Sample period: 1999-2009.
- Bank-level sample size: 1,294 banks.
- Observations in robustness table: 4,805 (column (1) and most columns); 2,571 (column (2)).
- Regressions include GDP growth, inflation, and the same bank-specific control variables as in Table 3 as (unreported) time-varying controls.
- Crisis 2008 (2009): year dummy = '1' in 2008 (2009).
- Robust p-values appear in brackets. ***, **, * correspond to the one, five and ten percent level of significance, respectively.

### Main findings on the Vienna Initiative (VI) and bank behavior
- Foreign bank subsidiaries reduced lending somewhat earlier, already in 2008, bringing their pace of lending back in line with domestic peers.
- Subsidiaries of parent banks that signed commitment letters (Vienna participants) were significantly more stable sources of credit in 2009 than subsidiaries of banks that did not sign such letters in the same country.
- No evidence that VI banks reduced lending in non-signing countries to prop up lending in VI countries. Instead, positive spillover effect observed:
  - Credit growth in 2009 of foreign bank subsidiaries whose parent banks signed commitment letters in one or more other countries exceeded that of all other banks by 9 percentage points (columns 2a,b) and that of other foreign banks by about 11 percentage points (columns 4a,b).
- Subsidiaries of parent banks not part of the VI in any country did not do worse in 2009 when compared to domestic banks, consistent with Table 3 results.
- State banks were a relatively stable source of credit during 2009.
- No evidence that home-country government support made foreign banks retrench more from the ECA region.
- By the second year of the crisis, state-owned domestic banks had become a relatively stable credit source.

### Robustness tests (summary of Table 5 specifications and results)
- Column descriptions:
  - (1) Variables denominated in euros instead of USD for highly euroized countries.
  - (2) Only banks with at least seven years of subsequent observations.
  - (3) Panel-corrected standard errors (PCSE) allowing bank-level heteroscedasticity and AR(1).
  - (4) Arellano-Bond (1991) GMM estimator.
  - (5) Hausman and Taylor (1981) instrumental variable estimator.
  - (6) Dependent variable: net loan growth (adjusted for provisioning).
  - (7) Dependent variable: growth of total assets.
- Selected coefficient estimates and p-values (as reported in Table 5):
  - Vienna letter*2008:
    - (1) -0.250*** (0.000)
    - (2) -0.038 (0.509)
    - (3) -0.308*** (0.000)
    - (4) -0.344*** (0.000)
    - (5) -0.291*** (0.000)
    - (6) -0.235*** (0.000)
    - (7) -0.191*** (0.002)
  - Vienna letter*2009:
    - (1) -0.021 (0.739)
    - (2) -0.051 (0.374)
    - (3) -0.068 (0.344)
    - (4) -0.074 (0.285)
    - (5) -0.049 (0.513)
    - (6) -0.033 (0.573)
    - (7) 0.031 (0.596)
  - Parent signed elsewhere*2008:
    - (1) -0.136*** (0.006)
    - (2) 0.050 (0.254)
    - (3) -0.119*** (0.004)
    - (4) -0.115*** (0.009)
    - (5) -0.088* (0.066)
    - (6) -0.121*** (0.002)
    - (7) -0.184*** (0.000)
  - Parent signed elsewhere*2009:
    - (1) 0.078* (0.095)
    - (2) 0.073** (0.050)
    - (3) 0.081* (0.053)
    - (4) 0.096** (0.033)
    - (5) 0.112** (0.024)
    - (6) 0.068* (0.062)
    - (7) -0.001 (0.969)
  - Non-Vienna parent*2008:
    - (1) -0.141*** (0.004)
    - (2) 0.032 (0.572)
    - (3) -0.099** (0.044)
    - (4) -0.118** (0.016)
    - (5) -0.134*** (0.003)
    - (6) -0.137*** (0.003)
    - (7) -0.160*** (0.000)
  - Non-Vienna parent*2009:
    - (1) 0.002 (0.961)
    - (2) 0.015 (0.719)
    - (3) -0.014 (0.770)
    - (4) -0.012 (0.789)
    - (5) 0.005 (0.909)
    - (6) -0.004 (0.922)
    - (7) 0.047 (0.178)
  - State*2008:
    - (1) -0.038 (0.454)
    - (2) 0.080 (0.223)
    - (3) 0.034 (0.509)
    - (4) -0.007 (0.901)
    - (5) 0.038 (0.511)
    - (6) -0.035 (0.504)
    - (7) -0.048 (0.402)
  - State*2009:
    - (1) 0.091** (0.048)
    - (2) 0.125** (0.020)
    - (3) 0.129** (0.018)
    - (4) 0.109** (0.033)
    - (5) 0.144** (0.012)
    - (6) 0.095** (0.048)
    - (7) 0.112** (0.026)
- Robustness highlights:
  - Results on the stabilizing effect of Vienna participation and of state ownership in 2009 persist when denominating variables in euros, limiting the sample to banks with 7+ years of data, and when using PCSE, GMM, and Hausman-Taylor estimators.
  - Replacing gross loan growth with net loan growth (column 6) leaves results intact.
  - Replacing gross loan growth with growth in total assets (column 7) removes the positive effect of a parent bank that signed in another country, suggesting participating banks may have continued lending while shortening balance sheets in other ways.

### Conclusions and policy implications
- The Vienna Initiative (VI), an ad hoc coordination mechanism, was a relatively successful example of catalytic funding where public funds provided by the IMF, EU, and various development institutions were complemented by a coordinated (but non-coercive) bail-in of private-sector lenders.
  - This helped close external funding gaps at the macroeconomic level in affected countries and softened deleveraging in the ECA region, helping to prevent an uncoordinated ‘rush to the exit’.
- Cross-border negative externalities would likely have occurred even with effective supervision because they are related to fast liquidity movements across borders within the same bank group.
- Policy options to improve cross-border handling of banking stress:
  - Better coordination, cooperation, and information exchange between supervisors (Vienna 2.0 focus).
  - Move the cross-border banking model of emerging Europe toward greater reliance on local sources of funding (development of local capital markets to top up domestic deposit funding with local wholesale funding).
  - Possible structural measures:
    - Resolution fund at the EU level or a network of national resolution funds.
    - Creation of a pan-European supervisor for large groups, supplemented by adequate capital and liquidity regulation and host-country macroprudential supervision to curb externally funded credit booms.
- Caution on forced subsidiarisation / ring-fencing:
  - Ring-fenced subsidiaries are costly to bank groups because the sum of ring-fenced pools of capital will be larger than current group capital, reducing benefits of international diversification.
  - Ring-fencing may impede efficient internal capital markets and the ability of multinational banks to allocate funding to the most worthy investment projects.
  - Forced subsidiarisation may be a second-best option reflecting insufficient cross-border cooperation and burden sharing.
- Practical policy recommendation:
  - In Central and Eastern Europe, many foreign bank subsidiaries will need to move toward funding policies that rely more on local and less on parent-bank funding, requiring development of local capital markets.

### Variable definitions (selected, from Table 6)
- Private domestic bank (1999-2009): 1 = bank is privately domestically owned. Source: BankScope, websites.
- State bank (1999-2009): 1 = bank is >30 per cent owned by the state. Source: BankScope, websites.
- Support (2008-2009): 1 = bank received government support (asset sale, capital injection, and/or guarantee). Source: Websites.
- Vienna parent (2009): 1 = subsidiary is owned by a parent bank that signed up to the Vienna Initiative. Source: EBRD.
- Non-Vienna parent (2009): 1 = subsidiary is owned by a parent bank that did not sign up to the Vienna Initiative. Source: EBRD.
- Vienna country (2009): 1 = subsidiary is based in a country part of the Vienna Initiative. Source: EBRD.
- Vienna letter (2009): 1 = parent bank signed a commitment letter in the subsidiary's host country. Source: EBRD.
- Parent signed elsewhere (2009): 1 = parent bank signed a commitment letter in another but not in the subsidiary's country. Source: EBRD.
- Pre-crisis IFI client (2009): 1 = subsidiary was a client of the EBRD, IFC, or EIB before 2008. Source: EBRD.
- Regional exposure (2007): Number of subsidiaries that a foreign bank owns in the ECA region. Source: BankScope.
- Credit growth (1999-2009): Annual percentage growth in gross loans (= net loans plus loan loss reserves). Source: BankScope.
- Asset growth (1999-2009): Annual percentage growth in total assets. Source: BankScope.
- Deposit growth (1999-2009): Annual percentage growth in deposits. Source: BankScope.
- Profitability (1999-2009): Return on average equity (in percent). Source: BankScope.
- Bank size (1999-2009): Log total assets in thousands of US$. Source: BankScope.
- Loan/deposit ratio (1999-2009): Net loans/short term funding (in percent). Source: BankScope.
- Tier 1 (1999-2009): Tier 1 capital ratio. Source: BankScope.
- Solvency (1999-2009): Equity/net loans (in percent). Source: BankScope.
- Liquidity (1999-2009): Liquid Assets / Dep & ST Funding (in percent). Source: BankScope.
- Net interest margin (1999-2009): Net interest income / Earning assets (in percent). Source: BankScope.
- Efficiency (1999-2009): Cost/income ratio (in percent). Source: BankScope.
- Loan quality (1999-2009): Loan loss reserves/gross loans (in percent). Source: BankScope.
- Country-level: Number of countries = 30.
  - GDP per capita (1999-2009): Lagged log GDP per capita, PPP (constant 2005 international US$). Source: IFS.
  - GDP growth (1999-2009): Real GDP growth (in percent). Source: IFS.
  - Inflation (1999-2009): Change in CPI inflation, end of period (in percent). Source: IFS.
  - Exchange rate change (1999-2009): Change in local currency unit/USD period average (in percent). Source: IFS.

*Source: Authors’ estimates (content unit: _wp12117 - Section 5).*

### Section 6

### _wp12117 - Section 6

### Variable definitions and summary statistics (Table 6)
- Panel A. Bank-level variables (1999-2009). Full sample summary statistics (Obs = 647 unless stated):
  - Average no. banks per year: 411
  - Private domestic: Mean 0.63, Std. Dev. 0.5, Min 0, Max 1
  - State bank: 50, Mean 0.08, Std. Dev. 0.3, Min 0, Max 1
  - Support: Obs 647, Mean 0.04, Std. Dev. 0.2, Min 0, Max 1
  - Credit growth: Obs 647, Mean 40.10, Std. Dev. 48.8, Min -66, Max 249
    - Means by bank-ownership: 40.9 (Foreign Vienna), 43.2 (Foreign non-Vienna), 39.6 (Private domestic), 36.6 (State)
  - Profitability: Obs 647, Mean 11.3, Std. Dev. 13.0, Min -72, Max 67
    - Means by bank-ownership: 11.1 (Foreign Vienna), 8.5 (Foreign non-Vienna), 11.9 (Private domestic), 11.2 (State)
  - Bank size: Obs 647, Mean 12.5, Std. Dev. 1.8, Min 9, Max 19
    - Means by bank-ownership: 14.2, 12.8, 12.0, 13.4
  - Loan/deposit ratio: Obs 647, Mean 91.4, Std. Dev. 50.6, Min 8, Max 399
    - Means by bank-ownership: 78.2, 82.6, 96.2, 93.3
  - Solvency: Obs 647, Mean 34.0, Std. Dev. 30.4, Min 0, Max 294
    - Means by bank-ownership: 21.6, 37.8, 35.5, 38.7
  - Liquidity: Obs 647, Mean 45.9, Std. Dev. 27.5, Min 3, Max 247
    - Means by bank-ownership: 35.9, 44.8, 48.0, 50.4
  - Efficiency: Obs 647, Mean 66.9, Std. Dev. 22.2, Min 4, Max 196
    - Means by bank-ownership: 63.6, 70.3, 67.1, 66.2
  - Loan quality: Obs 647, Mean 5.7, Std. Dev. 5.6, Min 0, Max 48
    - Means by bank-ownership: 4.5, 4.4, 6.2, 6.8

- Panel B. Country-level variables (1999-2009):
  - GDP growth: Obs 6479, Mean 4.2, Std. Dev. 6.0, Min -18.0, Max 30.5
  - Inflation: Obs 6479, Mean 10.2, Std. Dev. 10.1, Min -8.5, Max 293.7

### Pairwise correlations (Table 8)
- Panel A. Bank-level correlations (selected pairs):
  - [1] Private domestic with:
    - [2] State: -0.38
    - [3] Credit growth: -0.01
    - [4] Profitability: 0.07
    - [5] Bank size: -0.39
    - [6] Loan/deposit ratio: 0.12
    - [7] Solvency: 0.06
    - [8] Liquidity: 0.10
    - [9] Loan quality: 0.11
    - [10] Efficiency: 0.01
    - [11] Profitability (appears duplicated in table heading): 0.22
  - [2] State with:
    - [3] Credit growth: -0.02
    - [4] Profitability: 0.00
    - [5] Bank size: 0.14
    - [6] Loan/deposit ratio: 0.01
    - [7] Solvency: 0.04
    - [8] Liquidity: 0.05
    - [9] Loan quality: 0.06
    - [10] Efficiency: -0.01
    - [11] Profitability: -0.04
  - [3] Credit growth correlations (selected): with [4] Profitability 0.13, with [7] Solvency -0.11, with [9] Loan quality -0.23, with [10] Efficiency -0.13, with [11] Profitability -0.15
  - [4] Profitability correlations (selected): with [5] Bank size 0.11, with [10] Efficiency -0.47
  - [5] Bank size correlations (selected): with [7] Solvency -0.33, with [8] Liquidity -0.32, with [10] Efficiency -0.22, with [11] Profitability -0.23
  - [7] Solvency with [8] Liquidity: 0.55
  - [8] Liquidity with [9] Loan quality: 0.12
  - [9] Loan quality with [11] Profitability: 0.25
  - [10] Efficiency with [11] Profitability: 0.28

- Panel B. Country-level correlations:
  - [1] GDP per capita with:
    - [2] GDP growth: -0.22
    - [3] Inflation: -0.10
    - [4] Inflation volatility: -0.31
    - [5] Exchange rate change: -0.08
  - [2] GDP growth with:
    - [3] Inflation: -0.02
    - [4] Inflation volatility: 0.12
    - [5] Exchange rate change: -0.42
  - [3] Inflation with:
    - [4] Inflation volatility: 0.02
    - [5] Exchange rate change: 0.46
  - [4] Inflation volatility with [5] Exchange rate change: 0.06

### Overview of government support and Vienna Initiative participation (Table 9)
- Table 9 provides information on the presence of Vienna Initiative (VI) support and/or government support for parent banks of subsidiaries in countries that were part of the Vienna Initiative.
- Coding: "Y" ("N") indicate that the subsidiary's parent bank received (did not receive) government support in 2008-09.
- The table enumerates banks by country (examples shown):
  - BOSNIA-HERZEGOVINA: entries include Raiffeisen Bank (Y), Intesa SanPaolo Bank (Y), Turkish Ziraat Bank Bosnia (N), among others.
  - HUNGARY: entries include UniCredit Bank (Y), Erste Bank Hungary (Y), K&H Bank (Y), OTP Bank (N), among others.
  - LATVIA: entries include Swedbank (N), SEB Bank (N), DnB Nord Bank (Y), among others.
  - ROMANIA: entries include Alpha Bank Romania (Y), Banca Romaneasca (Y), BRD - Groupe Societe Generale (Y), Banca Comerciala Romana (Y), among others.
  - SERBIA: entries include Société Générale Bank Serbia (Y), Piraeus Bank Beograd (Y), Volksbank Serbia (Y), NLB Bank (N), among others.
- The table source: EBRD and banks' websites.

### Banks participating in the Vienna Initiative and horizontal meetings (Table 10)
- Table 10 lists banks that participated in horizontal meetings of the Vienna Initiative and indicates participation in:
  - Commitment letter, Participation September 2009 Brussels, Participation March 2010 Athens.
- Selected entries (country and parent/subsidiary with participation markers "x"):
  - Hungary: Bayerische Landesbank — MKB Bank: x (Brussels) x (Athens)
  - Hungary: Erste Group — Erste Bank Hungary: x x
  - Hungary: UniCredit Group — UniCredit Bank Hungary: x x
  - Romania: Erste Group — Banca Comercială Română: x x
  - Romania: Raiffeisen International — Raiffeisen Bank: x x
  - Romania: National Bank of Greece — Banca Romaneasca: x x
  - Romania: UniCredit Group — UniCredit Tiriac Bank: x x
  - Romania: Société Générale — BRD: x x
  - Bosnia: Raiffeisen International — Raiffeisen Bank: x x
  - Bosnia: Hypo Alpe-Adria — Hypo Alpe-Adria Bank: x x
  - Serbia: Eurobank EFG — Eurobank EFG: x x
  - Serbia: Intesa SanPaolo Bank — Intesa: x x
  - Latvia: Bank DnB NORD AS — DnB NORD Bank: x x
  - Latvia: Nordea Bank — Nordea branch: x x
  - Other banks listed as participating in meetings include ING Bank and OTP Bank (markers shown).
- Table source: commitment letters and concluding statements with the IMF and European Commission.

### Selection into the Vienna Initiative (Table 11)
- Table 11 reports probit regressions estimating likelihood of participation in the Vienna Initiative:
  - Column (1): probability that the parent bank of subsidiary i signed a commitment letter in country j (subsidiary-level).
  - Column (2): probability that the parent bank of subsidiary i signed a commitment letter in at least one VI country (subsidiary-level).
  - Column (3): probability that a parent bank signed a commitment letter in at least one VI country (parent-bank level).
- Notes:
  - All independent variables are defined in Table A1.
  - Robust p-values appear in brackets.
  - Significance notation: ***, **, * correspond to the one, five and ten percent level of significance, respectively.

*Content derived from _wp12117 - Section 6*

### Section 7

### _wp12117 - Section 7

### Regression results (Vienna letter / Vienna parent / Vienna parent)
- Sample columns labeled (1), (2), (3).
- Credit growth 2004-07:
  - (1) -0.022 (0.362)
  - (2) -0.003 (0.907)
  - (3) 0.801 (0.473)
- GDP growth country j:
  - (2) -3.271 (0.264)
  - (3) -8.175** (0.014)
- Inflation country j:
  - (2) -3.668 (0.248)
  - (3) -8.872** (0.019)
- Profitability subsidiary:
  - (1) 0.009 (0.242)
  - (2) 0.012 (0.126)
- Size subsidiary:
  - (1) 0.218** (0.032)
  - (2) 0.446*** (0.000)
- Loan/deposit ratio subsidiary:
  - (1) 0.001 (0.690)
  - (2) 0.002 (0.249)
- Loan quality subsidiary:
  - (1) 0.082** (0.030)
  - (2) 0.088** (0.026)
- Profitability parent bank:
  - (1) -0.061 (0.620)
  - (2) 0.141 (0.322)
  - (3) -0.100 (0.703)
- Size parent bank:
  - (1) 0.051 (0.543)
  - (2) 0.245** (0.024)
  - (3) 0.076 (0.607)
- Loan/deposit ratio parent bank:
  - (1) 0.005 (0.151)
  - (2) 0.008* (0.052)
  - (3) 0.006 (0.357)
- Tier 1 capital ratio parent:
  - (1) -0.000* (0.063)
  - (2) -0.000*** (0.000)
  - (3) -0.000* (0.060)
- Loan quality parent bank:
  - (1) -0.162 (0.185)
  - (2) -0.127 (0.278)
  - (3) -0.471* (0.055)
- Pre-crisis IFI client:
  - (1) 0.462* (0.074)
  - (2) -0.185 (0.583)
  - (3) 0.434 (0.573)
- Regional exposure:
  - (1) 0.066* (0.055)
  - (2) 0.347*** (0.000)
  - (3) 0.672*** (0.001)
- Constant:
  - (1) -6.090*** (0.002)
  - (2) -13.232*** (0.000)
  - (3) -3.814 (0.174)
- Number of observations:
  - (1) 235
  - (2) 235
  - (3) 66

### Timeline of the Vienna Initiative — key events and actions (chronological)
- October 26, 2008 – IMF approves US$16.5 billion Stand-By Agreement for Ukraine.
- November 6, 2008 – IMF approves €12.3 billion Stand-By Arrangement for Hungary.
- November 27, 2008 – Six bank groups (Raiffeisen, Erste, Intesa SP, Société Générale, KBC, Unicredit) write a letter to the EC on financial stability concerns in emerging Europe; EBRD, EIB and IFC start to put together a joint action plan.
- December 17, 2008 – Six multinational bank groups meet in Vienna to discuss next steps; EBRD and EIB invited.
- December 23, 2008 – IMF announces €1.7 Billion Stand-By Agreement for Latvia, coordinated with the European Commission.
- January 16, 2009 – IMF approves €402.5 million Stand-By Arrangement for Serbia.
- January 23, 2009 – First "Vienna Initiative" meeting at the Austrian Ministry of Finance; IMF to develop principles of burden sharing between home and host-country authorities and banks.
- End-January and early February 2009 – Country meetings for coordinated action in Ukraine and Romania with key subsidiaries, IMF and other IFIs.
- February 27, 2009 – Heads of EBRD, EIB, and the World Bank Group launch the Joint IFI Action Plan offering up to €24.5 billion of support to systemic banks in the region and lending to the real economy.
- March 17, 2009 – Second VI meeting; IMF presents burden sharing rules dividing host and home country responsibilities:
  - Host country responsibilities: prudent macroeconomic policies, support of deposit insurance schemes, supply of local currency liquidity irrespective of bank ownership.
  - Parent banks and home country responsibilities: rollover/maintain exposures to the extent possible, recapitalize subsidiaries following stress tests; home country national bank support packages can be used for supporting subsidiaries.
- March 15–June 15, 2009 – Under the Joint IFI Action Plan the EBRD, EIB, and World Bank Group meet jointly with all 17 main multinational bank groups to assess needs.
- March 26–27, 2009 – First set of parent bank commitment letters signed for Romania and Serbia at the Joint Vienna Institute, Vienna.
- April 25, 2009 – Meeting of IFIs and home and host governments during the IMF-World Bank Spring meetings, Washington DC.
- May 4, 2009 – IMF approves €12.9 billion Stand-By Arrangement for Romania as part of a EUR 20 billion package approved by the European Commission.
- May 7, 2009 – EBRD makes investments worth over €400 million to UniCredit subsidiaries across eight countries in the ECA region.
- May 15, 2009 – Joint IFI Action Plan meeting of key parent banks, home and host governments and IFIs during the EBRD Annual Meetings, London.
- May 20, 2009 – Commitment letter signed for Hungary and Romania in Brussels.
- June 22, 2009 – Joint concluding statement signed for Bosnia and Herzegovina in Vienna.
- July 8, 2009 – IMF €1.52 billion Stand-By Agreement for Bosnia and Herzegovina.
- July 10, 2009 – EBRD sets up a €220 million financing facility to the Hungarian subsidiaries of OTP Bank and €100 million to Erste Bank.
- July 29, 2009 – EBRD announces investment of €400 million in the subsidiaries of Société Générale.
- September 11, 2009 – Parent banks issue a concluding statement for Latvia in Stockholm.
- September 23, 2009 – EBRD approves a €150 million financing package to Raiffeisen Bank subsidiaries in Ukraine, Romania, and Russia.
- September 24, 2009 – First ‘horizontal’ Full-Forum meeting of the Vienna Initiative in Brussels; discussion of deleveraging and recapitalization needs; participants included 17 parent bank groups, home and host supervisors and fiscal authorities, IMF, EC, EBRD, EIB, WB, ECB, CEBS.
- October 5, 2009 – Meeting with CEOs of parent bank groups, home and host governments and heads of IFIs at the IMF-World Bank Annual Meetings, Istanbul; Joint Progress Report issued by the Joint IFI Action Plan participants (EBRD, EIB, WB).
- November 18–19, 2009 – Follow-up meetings with parent banks for Romania and Hungary in Brussels; start of discussions addressing vulnerability of foreign exchange (FX) exposures.
- January 18, 2010 – Coordination meeting with IMF, World Bank, EIB, EC in Vienna; shift from crisis management to addressing vulnerabilities (lack of local capital markets, FX exposures) and legacy of the crisis (balance sheet clean-up, distressed asset management).
- February 25, 2010 – EBRD provides a €100 million financing package to Intesa Sanpaolo subsidiaries in Bosnia and Herzegovina, Serbia and Hungary.
- February 26, 2010 – Follow-up meetings on Serbia and Bosnia in Vienna; exposure commitments relaxed for Serbia on the back of good macroeconomic adjustment and recovery.
- March 18–19, 2010 – Second ‘horizontal’ Full-forum VI meeting in Athens; agreement to set up working groups on local currency market development and on absorption of EU funds; participants: 20 bank groups, home and host supervisors and fiscal authorities, IMF, EC, EBRD, EIB, WB, ECB, and CEBS.
- May 4, July 8, and November 10, 2010 – Meetings of the working group on local currency development under the Vienna Initiative at the EBRD in London.
- May 9, 2010 – IMF approves €30 billion Stand-By Arrangement for Greece.
- June 17, 2010 – Information session and meeting on absorption of EU funds in Sofia.
- July 22, 2010 – Follow-up Vienna parent bank meeting on Romania and Hungary.
- July 26–August 4, 2010 – Joint Commission-IMF-WB mission to Romania to discuss measures to accelerate absorption of EU structural funds.
- October 4, 2010 – EBRD extends just under €1 billion of financing to subsidiaries of Piraeus Bank, National Bank of Greece, Alpha Bank, and EFG Eurobank.
- October 6, 2010 – Progress Report on implementation of the Joint IFI Action Plan.
- End-2010 – Expiry of the Vienna Initiative Joint IFI Action Plan.
- March 2011 – Final Report on implementation of the Joint IFI Action Plan.
- March 17–18, 2011 – Third ‘horizontal’ full forum meeting under the VI in Brussels to evaluate and consider recommendations of the two working groups.
- March 17, 2011 – Follow-up meetings in Brussels with parent banks on Romania.

*Source: _wp12117 - Section 7*

### Section 8

### _wp12117 - Section 8

### Citations listed in this section
- Rosengren, (2000), “Implications of the globalization of the banking sector: The Latin American experience,” New England Economic Review, September/October, 45-63.  
- Pistor, K., (2011), “Governing interdependent financial systems: Lessons from the Vienna Initiative,” Columbia Law and Economics Working Paper No. 396, Columbia University School of Law.  
- Popov, A. and G. Udell, (2012), “Cross-border banking, credit access, and the financial crisis,” Journal of International Economics, forthcoming.  
- Progress Report on the Joint IFI Action Plan, (2009), EBRD, EIB, World Bank, October 5th.  
- Rose, A.K. and T. Wieladek, (2011), “Financial protectionism: the first tests,” External MPV Unit Discussion Paper No. 32, Bank of England, London.  
- Roubini, N. and B. Setser, (2004), “Bailouts or Bail-Ins? Responding to Financial Crises in Emerging Economies,” Institute for International Economics, Washington, D.C.  
- Schnabl, P., (2012), “Financial globalization and the transmission of bank liquidity shocks: Evidence from an emerging market,” Journal of Finance, forthcoming.  
- World Bank, (2009), Global Development Finance. Charting a Global Recovery, Washington D.C.  

*Source: _wp12117 - Section 8*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp12117.pdf_