## Appendix A. The Too Big To Fail Premium

## Source details

**Canonical URL:** [Appendix A. The Too Big To Fail Premium](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2011/_111411.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2011/_111411.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2011/_111411.pdf.json)

---

### I. Introduction — key observations
- Consolidated bank claims on emerging and developing economy banks grew by roughly 3 times the rate of real GDP growth of these economies between 1995 and 2008.
- Cross-border AE–EM banking links developed out of regulatory liberalization, increased economic integration, lower information costs, and higher profit opportunities in EMs.
- Crisis vulnerabilities and transmission:
  - EMs exposed to trade and funding shocks from AEs; feedback loop observed in 2008/09.
  - Banks and countries with exposures to Eastern Europe faced considerable market pressure in 2008/09 requiring broad-based policy responses.
- Four “home countries” analyzed: Austria, Belgium, the Netherlands, and Spain.
  - Austria and Spain expanded into regions with strong historical ties but used different business models; Spain’s cumulative exposures around 30 percent of home GDP.
  - Belgium and the Netherlands expanded across a wider range of EM regions with relatively small bilateral exposures to individual countries.
  - Just prior to the crisis, the four economies had exposures to EMs above 20 percent of GDP; average exposure to EMs was one-sixth of total exposures; for Austria EM exceeded that to AEs; Spain had the lowest EM (and overall cross-border) exposure but was regionally concentrated.
- Analysis sources: interviews with senior bankers, regulators, policy makers, market participants, Financial Stability Reports, and literature.

### II. The expansion phase — bank structure, liquidity, and funding
- Expansion waves and drivers:
  - First wave (1980s and early 1990s): representative offices/branches following corporate clients.
  - Second phase (latter 1990s): deregulations/privatizations; acquisition of EM banks by AE investors rose from some $2 billion in the mid-1990s to $17 billion within 5 years before dropping off.
  - Third phase (about 2003 to just before Lehman): ramp-up in bank exposures to EMs; EM Europe banks expanded through higher loan growth funded from abroad.
- Competitive pressures: single banking license (1989) and the euro encouraged growth in EMs.
- Funding models and parent–subsidiary relationships:
  - Eastern Europe: convergence/euro expectations influenced parent funding; growth in foreign currency loans often encouraged parent funding.
  - Spanish banks used locally funded subsidiaries; most other AE banks centralized liquidity and funding in the home country.
  - Spanish stand-alone subsidiaries rationale:
    - Lowers contagion risks between parent and subsidiary.
    - Over time subsidiaries become stronger and more self-sufficient.
    - Easier to manage financial distress and resolution.
  - Alternative explanation: centralization may reflect excess home deposits seeking returns; Spain’s decentralization may reflect large local lending opportunities and current account deficits.
- Reference comparison: New Zealand—about 30 percent of banks’ funding from nonresidents; largest four banks Australian owned and account for 90 percent of banking assets; Australian and New Zealand banking sectors (both with assets around 200 percent of GDP) are smaller than European counterparts.

### Box 1 — Foreign Currency Lending in EM Europe (findings and policy responses)
- Extent and composition:
  - FX loans grew rapidly in Bulgaria, Croatia, Hungary, and Romania, reaching more than 50 percent of total loans by 2007.
  - FX lending insignificant in the Czech Republic.
  - Foreign loans mostly euro-denominated; Swiss franc loans dominated in Hungary; in Ukraine most FX loans were U.S. dollar-denominated.
- Determinants:
  - Interest rate differential, EU membership, and inflation as main factors (Steiner (2011)).
  - Supply-side: margins on domestic vs FX loans and regulatory quality.
  - Demand-side: booming consumption and housing prices.
  - Bank ownership: foreign bank subsidiaries had better access to parent funding abroad.
- Funding gaps and parent lending:
  - By 2007 domestic savings insufficient to finance credit expansion; funding gaps emerged (Croatia, Hungary longer; Bulgaria, Romania, Poland credit stock above deposits around 2007).
  - Parent lending accounted for a significant and increasing share of financing these gaps; proxy measures (Lahnsteiner (2011)) used to estimate parent lending.
- Vulnerabilities and outcomes:
  - Excessive unhedged FX lending increased dependence on foreign funding, deteriorated asset quality, and raised property exposure.
  - NPL ratios more than doubled since 2007 in Bulgaria and Hungary; reached 41 percent of total loans in Ukraine after sharp hryvnia depreciation and housing price collapse.
  - KBC announced stopping FX lending in Hungary in 2009 but later resumed.
- Policy measures:
  - Administrative limits on FX loans (Romania, Croatia); Hungary introduced differentiated loan-to-value ratios by currency in 2010.
  - Croatia’s voluntary preferential exchange-rate installment scheme; Poland’s prudential regulations reducing foreign-currency borrowing capacity.
  - Hungary allowed repayment of FX mortgages at preferential exchange rates, leaving about 25 percent reduction in debt value to be borne by banks.
  - ESRB recommendations: increase borrower awareness, ensure FX loans only to borrowers able to withstand exchange rate shocks, use debt-to-income and loan-to-value ratios, incorporate FX risks in pricing and capital allocation, consider limits on funding/liquidity risks.

### III. Comparative observations and implications
- Business models and crisis resilience:
  - Austrian centralized funding and EM Europe concentration worsened outcomes.
  - Spanish decentralized, locally funded subsidiaries aimed to reduce contagion risks; Spain experienced the largest domestic credit/housing boom among the four AEs.
  - Dutch and Belgian banks were more diversified; EM exposures a smaller share of overall exposures.
- Pre-crisis exposure patterns (end-2007 snapshots):
  - Four economies had exposures to EMs above 20 percent of GDP.
  - Spain: cumulative exposures around 30 percent of home GDP regionally concentrated in Latin America.
  - Austria: EM exposures exceeded AEs and heavily concentrated in EM Europe.
  - Netherlands and Belgium: intermediate exposure size and diversification; Belgium more concentrated on EM Europe.
- Funding reliance and cross-border risk:
  - Parent lending critical to finance rapid credit growth in CEE; reliance increased when domestic deposits insufficient.
  - Centralized liquidity management created potential funding transmission channels parent↔subsidiary.

### Diversification rationale and empirical evidence
- Rationale:
  - Cross-border expansion used to diversify macroeconomic risks and profit sources when cycles imperfectly correlated.
- Empirical findings:
  - Garcia-Herrero and Vasquez (2007): parent banks with larger foreign asset shares attain higher risk-adjusted returns (bank-level data covering 38 international banks).
  - Adjusted correlation analysis indicates diversification scope strongest in Spain, followed by Belgium and the Netherlands, and lowest in Austria.
  - In all cases, gains exceed diversification gains for Australian banks in New Zealand.

### Changes over time and limits to diversification
- Stock price synchronicity:
  - Stock returns between home and host countries became more synchronized in 2003–07 versus 1995–2002, limiting diversification gains.
- Deepening ties between AEs and EMs gradually narrowed room for diversification.

### Crisis experience and policy interventions (selected measures and amounts)
- Home country interventions and schemes:
  - Austria:
    - Unlimited deposit guarantee introduced.
    - Austrian Clearing Bank established to guarantee interbank market; access to CHF liquidity via Swiss REPO market.
    - Total initial rescue package size for banking sector: 100 billion euros.
    - Bank recapitalization fund: 15 billion euros.
    - Nationalization: Hypo Alpe-Adria (Austrian subsidiary) in December 2009; Kommunalkredit also nationalized.
    - Austrian debt guarantee scheme lapsed at end-2010.
  - Belgium:
    - Bank deposits guaranteed up to 100,000 euros.
    - Guarantee scheme for Dexia SA liabilities.
    - Government recapitalized Dexia and KBC; partially nationalized Fortis; later sold a 75 percent stake in Fortis to BNP Paribas in May 2009.
  - Netherlands:
    - Deposit insurance increased up to 100,000 euros.
    - 200 billion euros fund set up to provide liquidity support via guarantees of inter-bank deposits/loans.
    - 20 billion euros fund set up to recapitalize financial institutions with conditions.
    - Guarantee scheme allowed to expire at end-2010.
  - Spain:
    - Deposit insurance increased up to 100,000 euros.
    - Fee-based scheme to guarantee liabilities of Spanish-owned banks or Spanish subsidiaries of foreign banks (continuously extended).
    - Fund for Orderly Bank Restructuring (FROB) set up in June 2009 with 9 billion euros in funding.
- Multilateral and host-country support:
  - EBRD, EIB, and World Bank Group disbursed EUR 33 billion over the next two years for banking group support (Vienna Initiative related).
  - ECB and SNB provided FX repos in Hungary and Poland at crisis onset.
  - Vienna Initiative: written commitments by banks to maintain exposures and recapitalize subsidiaries; host countries extended liquidity and deposit insurance; home countries agreed bailout money could be used without restrictions for foreign subsidiaries in certain cases.
- Specific case: Dexia
  - Belgium agreed to pay 4 billion euros to take over Dexia Bank (Belgium).
  - Belgium, Luxembourg, and France are guaranteeing 90 billion euros in funding of Dexia SA and Dexia Credit Local.
  - For Belgium, the guarantee amounts to 15 percent of GDP.

### Banks’ reactions and heterogeneity
- Spanish banks:
  - Two large international Spanish banks emerged strongest; Latin American subsidiaries profitable and shielded them.
  - Both expanded in Emerging Europe (Santander in Poland; BBVA in Turkey).
- Austrian banks:
  - Larger banks remained profitable; some smaller banks became insolvent.
  - Austrian sovereign and bank spreads climbed the most at the Lehman outbreak.
- Belgian and Dutch banks:
  - Hard hit by crisis; Dexia wound up most EM municipal lending; KBC negotiated sale of Polish subsidiary; ING Group received public capital and sold businesses; Dutch authorities injected capital and took over remainder of ABN AMRO.
- Vienna Initiative / EBCI commitments:
  - Banks committed to maintain exposures and recapitalize subsidiaries where necessary.
  - IFIs provided EUR 33 billion over two years; commitments linked to IMF and EU stabilization programs in five countries (Bosnia and Herzegovina, Hungary, Latvia, Romania, and Serbia).

### Adaptation of bank business strategies (post-crisis)
- Strategic adaptations identified:
  - Diversification toward deposits as external liquidity dried up; loan-to-deposit ratios generally fell.
  - Reassessment of parent-subsidiary relationships amid FX lending clampdowns and borrower credit-relief measures.
  - Refinement of transfer pricing: stricter curbs on liquidity provision to subsidiaries and greater reliance on market-based pricing.
- Knock-on effects:
  - Banks to raise capital adequacy will issue private capital, shrink balance sheets, or seek government/European assistance.
  - Stronger banks likelier to find private capital; weaker banks may divest EM holdings.
  - Potential flows: capital/liquidity transfers from subsidiaries back to parents and higher dividend payouts to parents.
  - AE-owned banks dependent on EM earnings may overcapitalize subsidiaries in host jurisdictions to limit scrutiny on profit repatriations.

### Rethinking regulatory relationships and cross-border resolution
- Pre-crisis shortcomings:
  - Insufficient home-host cooperation; over-reliance on moral suasion in some home jurisdictions.
  - EM regulators sometimes slow to curb FX lending to unhedged retail customers; prudential measures often ineffective.
- Post-crisis improvements:
  - Improved communication between micro and macro supervisors; regulatory overhauls in Austria and Belgium; more frequent communication with host authorities.
  - Multilateral steps: ESRB, colleges of supervisors, cross-border stability groups, crisis management groups; first joint supervisory assessments toward CRD compliance end-2010.
  - Australia–New Zealand: 2006 legislation requires each jurisdiction to consider the financial stability of the other; harmonized capital, liquidity, and provisioning requirements.
- Cross-border crisis management challenges:
  - Absence of explicit burden-sharing led to group splits (Fortis, Dexia).
  - Obstacles: differences in resolution, corporate, and insolvency laws; full assessment on cross-border group resolution not expected until at least 2014.
- IMF proposal for enhanced coordination on resolution (four elements):
  - Amend laws to require national authorities to coordinate resolution efforts.
  - Framework applicable only to countries with core-coordination standards in place.
  - Specify principles to guide burden-sharing among cooperating authorities.
  - Agree coordination procedures to enable resolution actions in a crisis.

### Central issues in AE–EM banking today (guiding questions and findings)
- Three guiding questions: Do EMs benefit from AE banks? Are cross-border/currency funding scale economies realizable? Do AE banks amplify macro-financial feedback loops?

A. Do emerging markets benefit from foreign banks?
- Operational efficiency gains: exported management, technology, risk-management systems; transformation into modern banks attracting deposits and providing credit to households and SMEs.
- Empirical/observed indicators:
  - Host banking systems showed higher risk-adjusted ROA and declines in NPLs after foreign entry.
  - Noninterest expenses as ratio of average assets declined for most EM subsidiaries.
  - Net interest margins:
    - Latin America: margins remained high despite foreign entry.
    - EM Europe: net interest margins compressed in Southeastern Europe and the CIS post foreign entry.
  - Conclusion: benefits exist but substantial welfare gains accrue to banks given persistence of relatively high interest margins.

B. Are scale economies in funding realizable across borders and currencies?
- Reassessment post-crisis: centralized funding can lower costs but can also transmit shocks; decentralized liquidity management appears more stable outside major currency areas.
- Evidence suggesting overestimation of funding scale economies:
  - Subsidiaries’ ROE vs loan-to-deposit (LTD) ratios show little evidence externally-sourced funding improves long-run performance.
  - Risk-adjusted returns and higher LTD ratios/greater reliance on noncustomer deposits associated with worse performance.
- TBTF (too big to fail) premium role:
  - Parents enjoyed funding cost advantages over local funding before the crisis partly due to under-pricing of bank and country risks.
  - Bank and country credit risks rose during the crisis, narrowing funding advantage in Hungary, Poland, and Mexico; since the crisis cost advantage eliminated in Hungary and Poland.
  - In the Czech Republic externally-sourced funding was not advantageous.
  - In Brazil and Turkey, external funding cost advantage persisted over the last two years (from source reference), though eroded more recently in Turkey.

C. Does AE presence amplify macro-financial loops?
- (Text for section C begins in source but is not included in this excerpt; preceding sections document parent funding amplifying credit cycles and feedback loops.)

### Findings on parent funding and credit expansion
- Parent funding fueled rapid credit expansion in EMs and helped develop domestic banking/capital markets, but amplified credit cycles—cheap abundant funding fed booms and required rescues when markets dried up.
- Expansion accelerated after 2003, culminating in 2007 before external funding markets began to dry up.
- Positive and significant correlation between peak-to-trough output declines and rise in cross-border claims on host economies in the pre-crisis period.
- Transfer price methodology (assumptions used in analysis):
  - Home bank acquires 1-year funding from European interbank market at 1-year euribor.
  - Home bank pays bank credit premium equal to the 1-year home bank CDS spread.
  - Home bank charges subsidiary a country transfer premium represented by the 1-year sovereign CDS spread of the host country.
  - TBTF premium included (see Annex I and Box 4 in source for TBTF calculation).
  - Transfer price converted to implied local-currency rate using forward interest parity.
  - 1-year maturity used to compare implied transfer prices with local-currency interbank rate.
- By 2007, consolidated foreign bank claims on many host economies equaled GDP; bilateral consolidated claims of Austrian banks accounted for a large share.
- Cross-country patterns:
  - Poland and the Czech Republic had moderate output and credit growth with lower fiscal and external deficits; loan growth domestically funded.
  - Romania and Slovakia experienced stronger growth with very rapid credit expansion and significant external imbalances.
- Parent funding, credit growth, and NPLs:
  - Clear link between parent-subsidiary funding, degree of credit growth, and subsequent rise in post-crisis NPLs.
  - Emerging Europe received much higher parent funding, experienced stronger credit booms, and subsequent NPL rises compared with Latin America.
- Proxy for parent funding (Lahnsteiner (2011)): locational minus consolidated BIS claims; caveats noted that on balance tend to understate parental funding.

### Feedback loops to home countries and market perceptions
- Strong parent funding implied greater feedback loops to home countries.
- At Lehman outbreak Austrian sovereign and bank spreads climbed the most; Spanish banks’ spreads increased the least among the four sampled home countries.
- As of source date, bank CDS spreads in sample economies have converged to or exceeded average sovereign CDS spreads of the EMs to which they are exposed—reducing attractiveness of parental funding.

### Conclusions and policy recommendations (selected)
- Benefits:
  - AE banks bring experience, technology, risk management, and potentially more efficient capital allocation to EMs.
  - Joint regulatory and supervisory arrangements can yield operational/cost efficiency and better financial intermediation.
  - EM presence can provide AE banks with growth, returns to equity, and diversification.
- Limits and risks:
  - Welfare gains largely accrue to banks via persistent interest margins.
  - Improperly priced parental funding (including sovereign support) can create adverse macro-financial loops.
  - Parents may not internalize macroeconomic risks created by excessive lending growth.
- Recommended directions:
  - Encourage decentralized and diversified liquidity management to lower buildup/transmission of vulnerabilities.
  - For AE-EM pairs in new/frontier markets without credible liquidity backstop and effective cross-border crisis management, encourage subsidiaries to keep arm’s length from parent (taking EU context into account).
  - Transition to more independent subsidiary funding should be managed to avoid abrupt deleveraging cycles.
  - EMs should develop deep and liquid local markets for local funding.
  - AE banks and regulators should evaluate home-host cycle correlations to complement regulatory risk weights.
  - Small AEs with large EM banking presences should internalize greater sovereign risks and consider topping up international regulatory standards.

### Retail-based expansions, FX lending risks, and regulatory priorities
- Retail-based expansions combined with financial innovation warrant special caution due to foreign-currency lending risks.
- AE–EM regulatory and supervisory relationships need strengthening; supervisory colleges including smaller EM stakeholders are a positive development.
- Macro-prudential measures to curb externally funded credit booms must be shaped by EM supervisors in coordination with AE authorities.
- Cross-border resolution and burden sharing remain high priorities; interim approach: consider enhanced coordination framework recommended by Fund staff.

### Appendix A — TBTF premium: findings and estimates (high-level)
- TBTF concept: systemically important banks perceived as likely recipients of government support enjoy a funding advantage (TBTF premium).
- Ratings practice: Moody’s and Fitch publish an all-in rating (includes government support) and a stand-alone rating (excludes government support); rating support equals their difference.
- Historical measures:
  - Leading into the financial crisis the ratings uplift to large banks in advanced economies was around 2 notches.
  - As the crisis unfolded the ratings support increased to around 3 notches by mid-2009 and remained broadly at this level thereafter.
- Impact on funding costs:
  - In years leading into the crisis banks would have paid credit spreads 60 percent higher had they not benefited from ratings support.
  - TBTF premium peaked in early 2009 at the height of the global financial crisis; despite declines since early 2009, the premium remains very high in absolute level and relative to market credit spreads.
- Cross-country variation noted for U.S., Japan, U.K., and rest of Western Europe; countries with significant increases include Austria, Belgium, Germany, and Ireland.
- Estimation approach: TBTF premium computed as difference between credit spreads without support (stand-alone rating) and market credit spreads (all-in rating) using option-adjusted bond spread indices over government yields by credit rating.

### Appendices B–D: Methods and data highlights
- Appendix B: global common cycle estimated via dynamic factor model (DFM) with Kalman filter to fill missing observations; three common factors, two pervasive common shocks, two VAR lags; adjusted correlations calculated by subtracting global common component.
- Appendix C: stock-return synchronicity mapped using Multidimensional Scaling (MDS); distance = 1 − correlation; two-dimensional maps minimize stress.
- Appendix D: Data appendix includes selected macro indicators (Public Debt, Current Account Balance, Fiscal Balance) across years/periods and host-country risk-factor indicators (Real Credit growth, Housing Index growth, Decline in output, Share in total EM exposures). Selected exact numeric entries preserved in source tables (examples: Austria: 14.3 26.6 7.5 5.3 7.5 (2.2) 5.8; Spain: 7.7 18.9 7.5 (4.5) 11.2 4.0 4.1).
- Bank-level summary table (selected fields preserved in source): parent bank assets (e.g., ING: 1,247 USD billions), EBA EM credit risk exposures, number of EM subsidiaries, ROE, regional EM presence, loan-to-deposit, Core Tier 1 and leverage ratios—full table entries are presented in source data appendix.

*Source: Appendix A. The Too Big To Fail Premium (excerpt from IMF staff paper, content unit _111411).*

### Appendix A. The Too Big To Fail Premium ___________________________________________________ 32

### Appendix A. The Too Big To Fail Premium

### I. Introduction — key observations
- Capital flows from advanced economies (AEs) to emerging markets (EMs) strengthened banking sector links through a wave of mergers and acquisitions; BIS data indicate consolidated bank claims on emerging and developing economy banks grew by roughly 3 times the rate of real GDP growth of these economies between 1995 and 2008.
- Cross-border AE–EM banking links developed out of regulatory liberalization, increased economic integration, lower information costs, and higher profit opportunities in EMs.
- The global financial crisis highlighted macro-financial vulnerabilities in AE–EM banking links:
  - EMs are exposed to trade and funding shocks from AEs; there is also potential for shock transmission from EMs back to AEs (feedback loop observed in 2008/09).
  - Banks and countries with exposures to Eastern Europe faced considerable market pressure in 2008/09, requiring broad-based policy responses.
- The paper examines four “home countries”—Austria, Belgium, the Netherlands, and Spain—selected as medium-sized economies with differing policy backgrounds and crisis experiences:
  - Austria and Spain expanded into regions with strong historical ties but used very different business models: Spanish banks used locally funded subsidiaries; Austrian banks used a more centralized funding model. Spain’s cumulative exposures around 30 percent of home GDP (see Figure 2).
  - Belgium and the Netherlands expanded across a wider range of EM regions with relatively small bilateral exposures to individual countries; EM exposures were a small share of overall exposures prior to the crisis.
  - Just prior to the crisis, the four economies had exposures to EMs above 20 percent of GDP, but varied significantly in size and diversification. The average exposure to EMs was one-sixth of the total exposures; for Austria EM exceeded that to AEs; Spain had the lowest EM (and overall cross-border) exposure but was regionally concentrated.
- The analysis draws on interviews with senior bankers, regulators, policy makers, market participants, Financial Stability Reports of the home countries, and related literature.

### II. The expansion phase — bank structure, liquidity, and funding
- Expansion waves and drivers:
  - First wave (1980s and early 1990s): banks established representative offices or branches following corporate clients.
  - Second phase (latter 1990s): large-scale deregulations and privatizations (particularly in Latin America and Eastern Europe) coincided with large expansion; total acquisition of EM banks by AE investors rose from some $2 billion in the mid-1990s to $17 billion within 5 years before dropping off.
  - Third phase (about 2003 to just before Lehman): general ramp-up in bank exposures to EMs; EM Europe banks expanded balance sheets through higher loan growth funded from abroad and participated in privatizations or strategic partnerships.
- Competitive pressures: increasing competition in EU banking markets (single banking license introduced in 1989, euro a decade later) pushed banks to seek growth in EMs.
- Funding models and parent–subsidiary relationships:
  - In Eastern Europe, expectations of convergence and euro adoption influenced parent funding to subsidiaries; growth in foreign currency loans often encouraged or catalyzed parent funding (see Box 1).
  - Spanish banks’ expansion into Latin America followed the “tequila” crisis; foreign currency retail lending was not seen as viable and was discouraged by regulators.
  - Banks generally attempted to export home business lines: Spanish and Austrian banks replicated retail models; Dutch and Belgian banks exported bancassurance; some banks targeted niche sectors.
- Centralization vs decentralization of liquidity and funding:
  - Except for Spanish banks, liquidity and funding were generally centralized in the home country.
  - Spanish banks and regulators maintained that EM subsidiaries were largely stand alone for capital and funding after initial seeding. Stated advantages of stand-alone subsidiaries:
    - Lowers contagion risks to parent when EM-specific shocks occur, and to subsidiary when AE shocks occur.
    - Over time, subsidiaries become stronger and more self-sufficient due to market discipline.
    - Easier to manage financial distress and resolution in an orderly way.
  - Alternative interpretation: centralization by Austrian, Belgian, and Dutch banks may reflect excess home deposits seeking returns; Spain’s decentralized model may have resulted from large local lending opportunities and current account deficits.
  - As a reference, in New Zealand about 30 percent of banks’ funding has come from nonresidents; the largest four banks are Australian owned and account for 90 percent of banking assets, constituting about 15 percent of Australian bank assets; Australian and New Zealand banking sectors (both with assets around 200 percent of GDP) are smaller than their European counterparts.

### Box 1 — Foreign Currency Lending in EM Europe (summary of findings and policy responses)
- Extent and composition:
  - The share of foreign currency (FX) loans grew rapidly in Bulgaria, Croatia, Hungary, and Romania, reaching more than 50 percent of total loans by 2007.
  - FX lending remained insignificant in the Czech Republic.
  - Foreign loans were mostly euro-denominated; Swiss franc loans dominated in Hungary and were present in Poland (and to a lesser extent in Croatia); in Ukraine most FX loans were U.S. dollar-denominated.
- Determinants of FX lending:
  - Steiner (2011) identifies interest rate differential, EU membership, and inflation as main factors explaining rapid FX credit growth in EU accession economies.
  - Supply-side factors: difference between margins on domestic and FX loans and regulatory quality (measured as regulatory restrictions on foreign currency lending).
  - Demand-side factors: booming consumption and housing prices.
  - Bank ownership: foreign bank subsidiaries have better access to parent funding abroad (Bakker and Gulde (2010); Walko (2008); Lahnsteiner (2011)).
- Funding gaps and parent lending:
  - By 2007 domestic savings became insufficient to finance rapid credit expansion, leading to “funding gaps” in many CEE countries. Croatia and Hungary ran gaps for longer periods; in Bulgaria, Romania, and Poland strong credit growth pushed credit stock above deposits around 2007; Czech Republic deposit coverage remained high throughout the decade.
  - Parent lending accounted for a significant and increasing share of financing these funding gaps, particularly during the run-up to and aftermath of the crisis when other funding sources dried up.
  - Funding gaps are measured as the difference between credit to private sector and domestic deposits; this measure can underestimate the gap because parent funding was sometimes provided through deposits rather than loans. A proxy measure defined by Lahnsteiner (2011) is used to estimate parent lending to subsidiaries.
- Vulnerabilities and outcomes:
  - Excessive FX lending to largely unhedged private sector borrowers built significant vulnerabilities: increased dependence on foreign funding, deteriorating asset quality due to rising loan-to-value ratios, and increasing property market exposures.
  - Banks often assumed FX risk was borne by borrowers and did not adequately provision or hedge.
  - Sharp increases in NPL ratios followed credit deceleration after the crisis: NPL ratios have more than doubled since 2007 in Bulgaria and Hungary, and reached 41 percent of total loans in Ukraine following sharp hryvnia depreciation and housing price collapse.
  - In 2009 KBC (a Belgian bank) announced it would stop FX lending in Hungary but later resumed to remain competitive.
- Policy measures and regulatory responses:
  - Domestic regulators introduced measures to discourage FX lending, increasing risk premia for banks:
    - Administrative schemes limiting supply of FX loans implemented in Romania and broader-based limits in Croatia.
    - In 2010 Hungary introduced differentiated loan-to-value ratios depending on loan currency denomination.
    - Croatia implemented a voluntary scheme allowing borrowers to pay installments based on a preferential exchange rate for a few years.
    - Poland adopted prudential regulations effectively lowering amounts customers can borrow in foreign currency relative to zloty.
    - Hungary introduced remedies allowing borrowers to repay FX mortgages at preferential exchange rates, leaving about 25 percent reduction in debt value to be borne by banks.
  - The European Systemic Risk Board (ESRB) recommendations aim to reduce FX lending risks in a harmonized way to limit cross-border contagion and circumvention of national measures; recommendations include:
    - Increasing borrower awareness of FX lending risks.
    - Ensuring FX loans are extended only to borrowers capable of withstanding exchange rate shocks.
    - Use of debt-to-income and loan-to-value ratios.
    - Better incorporation of FX-related risks in banks’ internal pricing and capital allocation.
    - Consideration of limits on funding and liquidity risks if FX lending leads to excessive overall credit growth, with attention to concentration of funding sources and currency mismatches.

### III. Comparative observations and implications
- Business models influenced vulnerabilities and resilience:
  - Austrian banks’ centralized funding model and concentration in EM Europe contributed to challenging outcomes during the crisis.
  - Spanish banks’ decentralized, locally funded subsidiary model aimed to reduce contagion risks and support stand-alone resilience, though Spain experienced the largest domestic credit/housing boom among the four AEs.
  - Dutch and Belgian banks were more diversified across regions and maintained significant trading and investment banking activities; EM exposures were a smaller share of overall exposures for these banks.
- Pre-crisis exposure patterns (as of end-2007 and around 2007–2010 snapshots in figures):
  - Four selected economies had exposures to EMs above 20 percent of GDP prior to the crisis.
  - Spain: cumulative exposures around 30 percent of home GDP and regional concentration in Latin America.
  - Austria: EM exposures exceeded exposures to AEs and were heavily concentrated in EM Europe.
  - Netherlands and Belgium: intermediate cases in exposure size and diversification; Belgium more concentrated on EM Europe.
- Funding reliance and cross-border risk:
  - Parent lending became critical in financing rapid credit growth in CEE countries; reliance on parent funding increased when domestic deposits were insufficient.
  - Centralized liquidity management by some AE banks implied potential funding transmission channels from parents to subsidiaries and vice versa.

*Source: Appendix A. The Too Big To Fail Premium (excerpt).*

### 10.      Banks have looked at cross-border expansion, particularly into emerging markets,

### 10.      Banks have looked at cross-border expansion, particularly into emerging markets,

### Diversification rationale and empirical evidence
- Rationale:
  - Cross-border expansion used to diversify macroeconomic risks and profit sources.
  - If business cycles are imperfectly correlated across countries, assets of cross-border banks are less exposed to country-specific shocks.
  - If home and host country shocks are imperfectly correlated, banks’ risk-adjusted returns are likely to increase.
- Empirical findings:
  - Using bank-level data covering 38 international banks, Garcia-Herrero and Vasquez (2007) find that parent banks with a larger share of assets allocated to foreign subsidiaries are able to attain higher risk-adjusted returns.
  - Macro data over the past two decades indicate diversification gains were attainable, varying across countries.
  - Figure 3 (described) plots weighted average correlations of real GDP growth, nominal M2 growth, and equity price growth between each home country and respective host countries.
  - Simple correlations do not account for co-movement from a common global cycle or correlation among EM countries; adjusted correlation coefficients (accounting for the global cycle and host dependence) were used to assess diversification scope.
  - Results suggest the scope for diversification was strongest in Spain, followed by Belgium and the Netherlands, and lowest in Austria (Table 1).
  - In all cases, the gains exceed diversification gains for Australian banks operating in New Zealand (reference case).

### Changes over time and limits to diversification
- Stock price synchronicity:
  - Stock price return synchronicity maps (Figure 4) compare 1995–2002 (blue points) and 2003–07 (red points).
  - The maps indicate that stock returns between home and host countries became more synchronized in 2003–07, limiting diversification gains from foreign ventures.
- Deepening ties:
  - Deepening economic and financial ties between advanced economies (AEs) and emerging markets (EMs) may have gradually narrowed room for diversification.

### Crisis experience and risks in AE-EM banking relationships
- 2008 crisis highlights:
  - The crisis highlighted risks inherent in AE-EM banking relationships, particularly vis-à-vis Emerging Europe.
  - Several Eastern European countries were cut off from financial markets; banks with exposures in these countries came under considerable stress.
  - Markets focused on Austria where risks from EM banking could affect the sovereign balance sheet, causing sovereign spreads to rise sharply.
  - Spanish banks fared better due to profitable Latin American subsidiaries; Australia/New Zealand banks were supported by strong capitalization, provisioning, and stronger government finances enabling credible guarantees.
- Table 1 (ranking for diversification gains in EM exposures):
  - Rankings (sample period quarterly data 1995 to 2010; computed based on average diversification ranking of real GDP growth, M2 growth, and stock price returns synchronization):
    - 1: Spain — Simple correlation: Spain; Correlation accounting for the global cycle: Spain; Correlation accounting for host dependence: Belgium
    - 2: Netherlands — Simple correlation: Belgium; Correlation accounting for the global cycle: Belgium; Correlation accounting for host dependence: Spain
    - 3: Belgium — Simple correlation: Netherlands; Correlation accounting for the global cycle: Netherlands; Correlation accounting for host dependence: Netherlands
    - 4: Austria — Simple correlation: Austria; Correlation accounting for the global cycle: Austria; Correlation accounting for host dependence: Austria

### Policy interventions during the crisis
- General:
  - Significant policy intervention in both home and host countries was needed to stop the crisis from spreading.
  - All four home country governments in the sample committed large amounts of capital support for their banking systems, increased deposit and bank debt guarantees, and combined intervened to restructure or nationalize several banks (Tables 2–3).
  - The Netherlands and Belgium had higher direct net costs of financial sector support, compared to Spain and Austria as of end 2010 (IMF, 2011).
- Host country measures:
  - Some host countries increased deposit and bank debt guarantees, entered into FX swap arrangements with AE central banks (Brazil, Mexico, Hungary, and Poland), entered into Fund programs or arrangements (Hungary, Mexico, Poland, Romania, and Ukraine), and a few injected capital into banks.
- External and multilateral support:
  - IMF programs and EU balance of payments support helped alleviate severe FX liquidity strains.
  - The Vienna Initiative (European Bank Coordination Initiative / EBCI) sought written commitments and brought together major stakeholders to prevent disorderly pull-out of banks from Emerging Europe.
  - Parent banks provided swaps to subsidiaries; integration of EM European banks within the European banking system helped subsidiaries receive more stable funding from parents than other EM banks.
  - At the onset of the crisis, the ECB and SNB provided FX repos in Hungary and Poland.
  - Without IMF, EU, and other European institutions’ support, EM European currencies may have depreciated much more, exposing AE bank subsidiaries to greater risk.
- Specific policy measures and amounts (Table 3 and related text):
  - Austria:
    - Introduced an unlimited deposit guarantee.
    - Established an Austrian Clearing Bank to guarantee the interbank market; Austrian banks accessed CHF liquidity through the Swiss REPO market.
    - Total initial rescue package size for the banking sector was 100 billion euros.
    - A bank recapitalization fund of 15 billion euros was established.
    - Nationalization of the Austrian subsidiary of Hypo Alpe-Adria in December 2009; Kommunalkredit also nationalized.
    - Austrian debt guarantee scheme lapsed at end-2010.
  - Belgium:
    - Bank deposits guaranteed up to 100,000 euros.
    - Guarantee scheme for Dexia SA's liabilities provided.
    - Government recapitalized Dexia and KBC; partially nationalized Fortis; later sold a 75 percent stake in Fortis to BNP Paribas in May 2009.
  - Netherlands:
    - Deposit insurance increased up to 100,000 euros.
    - 200 billion euros fund set up to provide liquidity support through guarantees of inter-bank deposits/loans.
    - 20 billion euros fund set up to recapitalize financial institutions with attached conditions.
    - Sovereign spreads declined as the guarantee fell into disuse; guarantee scheme allowed to expire at end-2010.
  - Spain:
    - Deposit insurance increased up to 100,000 euros.
    - Fee-based scheme to guarantee liabilities of Spanish-owned banks or Spanish subsidiaries of foreign banks (continuously extended).
    - Fund for Orderly Bank Restructuring (FROB) set up in June 2009 with 9 billion euros in funding.
  - Multilateral funding and initiatives:
    - EBRD, EIB, and the World Bank Group disbursed EUR 33 billion over the next two years for support of various banking groups (Vienna Initiative related).

### Banks’ reactions and heterogeneity across systems
- Differentiation across banks:
  - Spanish banks:
    - Two large international Spanish banks emerged strongest; profitability of Latin American subsidiaries shielded them from home market difficulties.
    - Both banks expanded presence in Emerging Europe: Santander in Poland; BBVA in Turkey.
  - Austrian banks:
    - Larger banks remained profitable; some smaller banks became insolvent.
    - Banks largely maintained exposures as part of the European Bank Coordination Initiative (EBCI), though some needed to deleverage.
    - One bank (Volksbank) planned to sell CESEE subsidiaries ahead of the EBA stress test but failed; Raiffeisen purchased a branch of Greek EFG Eurobank in Poland (subject to conversion to a full subsidiary); Erste Bank increased control over its Romanian subsidiary.
  - Belgian and Dutch banks:
    - Hard hit by the crisis, not exclusively via EM exposures.
    - Dexia wound up most EM municipal lending after Lehman; KBC in talks to sell Polish subsidiary.
    - Dutch authorities injected significant capital; took over remainder of ABN AMRO that Fortis had acquired.
    - ING Group received public capital and sold businesses, including Latin American insurance businesses and ING Direct in the U.S.
- Vienna Initiative / EBCI commitments (Box 3):
  - Banks committed to maintain exposures to CEE countries and recapitalize subsidiaries where necessary.
  - Host countries committed to reasonable macroeconomic policies per IFI agreements, extend liquidity and deposit insurance to subsidiaries of foreign bank groups and not ring-fence assets.
  - Home countries agreed to make bailout money available to banks without restrictions on use for foreign subsidiaries.
  - IFIs provided funding to banking groups (EUR 33 billion disbursed over next two years by EBRD, EIB, and World Bank Group).
  - Commitments by banks were linked to IMF and EU stabilization programs in five countries (Bosnia and Herzegovina, Hungary, Latvia, Romania, and Serbia) where parent banks pledged to largely maintain exposures and recapitalize subsidiaries while programs remained on track.

*Source: IMF chapter/section titled "10.      Banks have looked at cross-border expansion, particularly into emerging markets," from the provided PDF content.*

### 17.      In all cases, however, banks have been forced to adapt business strategies to the

### _111411 - 17.      In all cases, however, banks have been forced to adapt business strategies to the

### Adaptation of bank business strategies (post-crisis)
- Three areas of strategic adaptation identified:
  - Diversification toward deposits as external liquidity dried up; significant deposit competition after the Lehman shock, easing as loan-to-deposit ratios have generally fallen. Panic-induced deposit withdrawal pressures worsened diversification toward deposits.
  - Reassessment of parent-subsidiary relationships as countries clamped down on foreign currency lending and/or implemented credit-relief measures for borrowers.
  - Refinement of transfer pricing strategies: stricter curbs on liquidity provision to subsidiaries and greater reliance on market-based mechanisms to price liquidity.
- Knock-on effects from difficulties in European advanced-economy (AE) banking sectors:
  - To raise capital adequacy, banks will either issue private capital, shrink balance sheets, or seek government (or European) assistance.
  - Relatively stronger banks are likelier to find private capital; weaker banks may divest EM holdings, presenting acquisition opportunities for AE banks or locally owned EM banks.
  - AE-owned banks may be unable or unwilling to lend in EMs to cap group-level balance sheet size.
  - Potentially controversial flows: capital and liquidity transfers from subsidiaries back to parents, or higher dividend payouts to parents.
  - AE-owned banks dependent on EM earnings may leave subsidiaries significantly overcapitalized in each host jurisdiction to limit scrutiny on profit repatriations.

### Rethinking regulatory relationships (home-host cooperation and crisis management)
- Regulators’ assessments of pre-crisis risk build-up highlighted insufficient home-host cooperation.
  - Spanish regulators: earlier crises led to closer cooperation with Latin American counterparts; in early expansion stages they sometimes required group-level provisioning when host standards were lax.
  - Belgian authorities: absence of a quantitative liquidity requirement plus central treasuries treated as profit centers created incentives to "push liquidity out."
  - Dutch authorities: post-crisis experience may favor locally funded subsidiaries, though difficult for banks with a larger corporate mix.
  - Austrian regulator: over-reliance on moral suasion to curb cross-border risks pre-crisis.
- EM Europe regulators: some were slow to curb FX lending to unhedged retail customers; prudential measures in places like Bulgaria, Croatia, Romania, Serbia were often ineffective.
- Post-crisis improvements:
  - Improved communication between micro and macro supervisors; regulatory overhauls in Austria and Belgium; more frequent communication with host authorities.
  - Multilateral steps: establishment of the ESRB, colleges of supervisors, cross-border stability groups, crisis management groups; first joint supervisory assessment of cross-border banks aiming to achieve compliance with the new Capital Requirements Directive (CRD) that came into effect at the end of 2010.
- Australia–New Zealand example:
  - Enhanced coordination, information-sharing, and harmonization discussed in early 2000s; significant 2006 legislation requires each jurisdiction to consider the financial stability of the other.
  - Capital, liquidity, and provisioning requirements well-harmonized; Basel III-related changes to be carried out in mutual consultation.
- Cross-border crisis management, resolution, and burden-sharing remain difficult:
  - Absence of explicit burden-sharing led to splitting groups along national lines (Fortis, Dexia examples).
  - Dexia resolution: Belgium agreed to pay 4 billion euros to take over Dexia Bank (Belgium); Belgium, Luxembourg, and France are guaranteeing 90 billion euros in funding of Dexia SA and Dexia Credit Local; for Belgium, the size of the guarantee amounts to 15 percent of GDP.
  - Obstacles: differences in bank resolution laws, corporate and insolvency laws; full assessment on cross-border banking group resolution not expected until at least 2014.
- IMF proposal for enhanced coordination on resolution issues (four elements):
  - Countries amend laws to require national authorities to coordinate resolution efforts.
  - Framework applicable only to countries with core-coordination standards in place.
  - Specification of principles to guide burden-sharing among cooperating authorities.
  - Agreed coordination procedures to enable resolution actions in a crisis.

### Central issues in AE–EM banking today
- Three guiding questions: benefits to EM banking systems from AE banks; realizable cross-border/currency funding scale economies; potential for negative macro-financial feedback loops.

A. Do emerging markets benefit from foreign banks?
- Operational efficiency gains from AE-owned banks via exported management, technology, and risk-management systems; transformation of EM entities into modern banks that attract deposits and provide credit intermediation and financial products to households and SMEs.
- Literature:
  - Claessens and others (2001): foreign bank entry improves national banking market functioning (80 countries, 1988–1995).
  - Goldberg (2009): financial sector FDI may only induce limited technology transfers and productivity gains; literature unclear whether gains arise from competition or technology transfer.
- Empirical indicators in sample countries:
  - Host banking systems showed higher risk-adjusted return on assets and declines in nonperforming loans after foreign entry (Figure 5 referenced).
  - Noninterest expenses as ratio of average assets declined for most EM subsidiaries in the sample.
  - Net interest margins often remain wide years after acquisition:
    - Latin America: net interest margins remained high despite foreign bank entry.
    - EM Europe: net interest margins compressed in Southeastern Europe and the CIS since foreign bank entry.
  - Conclusion: benefits exist, but persistence of relatively high interest margins implies substantial welfare gains accrue to banks.

B. Are scale economies in funding realizable across borders and currencies?
- Crisis prompted reassessment of centralized cross-border funding models: potential for AE-owned banks to centrally pool/manage funding and pass lower costs to EM subsidiaries, but also become conduits for shocks from AEs to EMs and back to parents.
- The argument for centralized funding weakens if scale economies in funding costs are overestimated or absent, favoring decentralized liquidity management.
- Literature and analyses:
  - McCauley and others (2010): decentralized multinational model appears more stable, especially outside major currency areas; local claims in local currencies were more stable than cross-border claims post-Lehman.
  - Fiechter and others (2011): optimal organizational structure depends on business-line diversity and stages of financial development; branch structures may suit wholesale banking, decentralized subsidiaries may suit global retail banks.
- Empirical prima facie evidence suggesting overestimation of funding scale economies:
  - Simple performance measures: scatter of subsidiaries’ return on equity (ROE) vs loan-to-deposit (LTD) ratios shows little evidence that externally-sourced funding improves long-run performance (Figure 6).
    - Caveat: transfer pricing mechanisms unknown; subsidiaries’ ROEs may not reflect true profitability if profits booked at parent level.
  - Risk-adjusted performance: within-country comparisons of long-term risk-adjusted returns on equity and assets fail to link centralized funding to better performance; higher LTD ratios and greater reliance on noncustomer deposits associated with worse overall performance (Table 4).
- Cost advantages may be linked to implicit sovereign support (TBTF premium):
  - Stylized comparison of domestic interbank lending rates vs transfer-price-implied local currency interest rates accounting for market-based bank/country risk and TBTF premium (Figure 7 referenced).
  - Findings:
    - Parents enjoyed funding cost advantages over local funding in several countries before the crisis, partly due to markets under-pricing bank and country risks.
    - Bank and country credit risks rose during the crisis, narrowing the funding advantage in Hungary, Poland, and Mexico.
    - Since the crisis, when TBTF premia can be better estimated, cost advantage eliminated in Hungary and Poland.
    - In the Czech Republic, externally-sourced funding was clearly not advantageous.
    - In Brazil and Turkey, external funding cost advantage persisted over the last two years, though eroded more recently in Turkey.

C. Does the presence of AE banks amplify macro-financial loops?
- (Text for section C begins but is not included in the supplied excerpt; analysis up to subsection B is provided in the source content.)

*Source: Excerpt from IMF chapter/section provided in the supplied content unit.*

### 31.      Banks’ relatively easy access to home country funding allowed them to significantly

### _111411 - 31.      Banks’ relatively easy access to home country funding allowed them to significantly

### Findings on parent funding and credit expansion
- Banks’ relatively easy access to home country funding allowed them to significantly expand credit in emerging market economies (EMs).
- This expansion was viewed as helping EM countries catch up by developing domestic banking and capital markets and supporting domestic demand growth.
- However, parental funding amplified credit cycles: acting as a funding conduit when liquidity was abundant and cheap, and necessitating rescue measures when financial markets dried up.
- The process accelerated after 2003 and culminated in 2007, after which external funding markets began to dry up.
- There is a positive and significant correlation between peak-to-trough output declines and the rise in cross-border claims on host economies in the pre-crisis period (Figure 8).

### Evidence and statistics
- The transfer price methodology (Figure 7) assumes:
  - home bank acquires 1-year funding from the European interbank market at a cost of 1-year euribor;
  - home bank pays a bank credit premium equal to the 1-year home bank CDS spread;
  - home bank charges its subsidiary a country transfer premium represented by the 1-year sovereign CDS spread of the host country;
  - a TBTF (too big to fail) premium is included (see Annex I and Box 4 in source for TBTF calculation);
  - the transfer price is converted to an implied local-currency rate using forward interest parity;
  - 1-year maturity is used to compare implied transfer prices with local-currency interbank rate.
- By 2007, consolidated foreign bank claims on many host economies equaled GDP.
- Bilateral consolidated claims (ownership basis) of Austrian banks accounted for a large share of these claims (Figure 9).
- Cross-country patterns:
  - Poland and the Czech Republic maintained a moderate pace of output and credit growth and had lower fiscal and external deficits; loan growth in these countries was domestically funded.
  - Romania and Slovakia experienced stronger growth accompanied by very rapid credit expansion and accumulation of significant external imbalances.
- Relationship between parent-subsidiary funding, credit growth, and NPLs:
  - There is a clear link between parent-subsidiary funding, the degree of credit growth, and the subsequent rise in post-crisis NPLs (Figure 11).
  - Emerging Europe generally received much higher parent funding, experienced a stronger credit boom, and a subsequent NPL rise compared with Latin America.
- Proxy for parent funding to host economies (following Lahnsteiner (2011)):
  - Calculated by subtracting consolidated bank claims on host economy banks from the locational claims on BIS reporting banks.
  - Caveats listed in source: (i) difference in number of reporting countries in locational vs consolidated statistics; (ii) foreign currency lending from subsidiaries to other banks cannot be isolated; (iii) locational statistics incorporate cross-border claims on central banks while consolidated do not; (iv) approach will not capture claims resulting from round tripping via other entities in financial centers.
  - These caveats, on balance, tend to understate rather than overstate the extent of parental funding.

### Country and bank-level observations
- Austria:
  - Degree of Austrian EM exposure, regional concentration, and reliance of Austrian subsidiaries on parental funding appear to have been principal factors contributing to investor perceptions of fiscal risks and potential sovereign rescue obligations.
  - Austrian sovereign and bank spreads climbed the most in the sample at the outbreak of the Lehman crisis.
  - Austria’s spreads eventually normalized following extensive external support to EM Europe.
- Spain:
  - Spanish banks’ spreads increased the least among the four home countries considered at the Lehman outbreak (Table 5).
  - Large Spanish banks, which had a locally funded EM expansion strategy, emerged stronger from their EM business through the crisis despite difficulties in home markets.
- Belgium and the Netherlands:
  - Banks that expanded cross-border business most rapidly (Belgian and Dutch banks) were hard-hit; they faced difficulties similar to Austrian banks in EMs and encountered difficulties with AE exposures.
- Mexico:
  - Banks are deposit rich and do not need external parent funding; Spanish banks reportedly keep subsidiaries independent in terms of liquidity management.
- Australia and New Zealand:
  - Although materially different from AE-EM relationships in the paper, both banking systems shared vulnerabilities: reliance on external funding, high loan-to-deposit ratios, and potential downside risks from a housing boom.
  - Features that helped contain funding difficulties:
    - relatively smaller banking systems not exposed to structured products;
    - sovereigns with very strong finances able to credibly guarantee wholesale funding needs;
    - strong capital and provisioning regulation before the crisis and prompt action to strengthen liquidity management since then;
    - strong supervisory and regulatory cooperation and harmonization prior to the crisis.

### Feedback loops to home countries
- Strong parent funding implied greater feedback loops to home countries.
- At Lehman outbreak, Austrian sovereign and bank spreads climbed the most; Spanish banks’ spreads increased the least among the four sample home countries (Table 5).
- As of the date of the source, bank CDS spreads in all sample economies have either converged to, or exceeded, the average sovereign CDS spreads of the EMs to which they are exposed—reducing the attractiveness of parental funding to fuel EM loan growth.

### Conclusions and policy recommendations
- Benefits of AE banks operating in EMs:
  - AE banks bring international experience, technology, risk management, and potentially more efficient capital allocation.
  - Joint regulatory and supervisory arrangements can yield operational and cost efficiency and better financial intermediation in EMs.
  - EM presence can provide foreign banks with increased growth, returns to equity, and diversification.
- Limits and risks:
  - Welfare gains from increased profitability of EM banking sectors accrue largely to banks.
  - If parentally sourced funding is inappropriately priced (e.g., sovereign support preventing proper risk pricing), macro-financial loops can be adverse.
  - Parents may lack incentives to internalize macroeconomic risks caused by excessive lending growth while acting as conduits for shocks via funding links to subsidiaries.
- Recommended directions:
  - Encourage decentralized and diversified liquidity management to lower buildup and transmission of vulnerabilities.
  - For AE-EM pairs in new or frontier markets without a credible liquidity backstop and effective cross-border crisis management, consider encouraging subsidiaries to keep an arm’s length from the parent—recognizing the EU context (pan-EU regulatory and supervisory bodies) may provide stabilizing influences not available elsewhere.
  - Transitioning to more independent subsidiary funding should not be abrupt to avoid disruptive deleveraging cycles.
  - EMs should develop deep and liquid local markets to enable subsidiaries to seek more local funding.
  - AE banks and regulators should pay more attention to the benefits of EM diversification and evaluate correlations between home and host cycles to complement regulatory risk weights.
  - Small AEs with large EM banking presences should internalize greater sovereign risks their banks may pose and may have incentives to "top-up" international regulatory standards, given reputational pressures to act as backstops in crises.

*Source: IMF staff paper excerpt (figures and tables referenced in the source).*

### 40.      Retail-based expansions into EMs that are combined with financial innovation also

### _111411 - 40.      Retail-based expansions into EMs that are combined with financial innovation also

### Retail-based expansions and foreign-currency lending risks
- The introduction and active promotion of foreign currency lending to retail customers in emerging market economies (EMs) “affords a cautionary tale.”
- Lending at household, corporate, and government levels based on underlying currency bets—such as the euro convergence play—has been a feature in Asian and Latin American financial crises, and “may re-emerge in the future.”
- Finding: Retail-based expansions into EMs combined with financial innovation “warrant special caution.”

### AE-EM regulatory and supervisory relationships
- Finding: Advanced economy–emerging market (AE-EM) regulatory and supervisory relationships “need to strengthen.”
- Progress noted: “All authorities in our sample economies (and beyond) have made encouraging progress with respect to joint supervision.”
- Positive development: “The establishment of supervisory colleges including even the smaller EM stakeholders is particularly welcome.”
- Policy recommendation: Macro-prudential measures to curb future recurrences “must be shaped by EM supervisors in close coordination with AE authorities.”
- Complementary measures: “Encouraging more prudent local liquidity management by banks, and other micro and macro-prudential measures to curb externally funded credit booms are likely to be inadequate, unless accompanied by appropriate tightening of macroeconomic policies.”

### Cross-border resolution and burden sharing
- Priority: “Further demonstrable progress on cross-border resolution and burden sharing remains a high priority.”
- Helpful step: “The establishment of cross-border stability groups is particularly helpful in enhancing crisis-preparedness.”
- Core gap: The “(as yet elusive) solution is an international resolution and burden-sharing scheme which allows for smooth controlled failure of a cross-border bank in the event of a crisis.”
- Observation: AE-EM relationships are “perhaps more fragile, and particularly riddled by uncertainties about burden sharing.”
- Interim approach: “Until progress is made here, regulatory approaches will tend toward protecting perceived national interests.”
- Near-term practical solution: “give close consideration [to] the enhanced coordination framework recommended by Fund staff.”

### Appendix A — The Too Big To Fail (TBTF) premium: findings and estimates
- Concept: Banks perceived as systemically important and likely recipients of government support enjoy a funding advantage—the TBTF premium.
- Finding: “We estimate that the TBTF premium increased sharply during the financial crisis, and, although the premium has declined since the heights of the crisis, it remains at high levels and is substantially higher than before the start of the financial crisis.”
- Rating agency practice: Moody’s and Fitch publish an all-in rating (includes government support) and a stand-alone rating (excludes government support). The rating support equals the difference between all-in and stand-alone ratings.
- Historical measures:
  - “Leading in to the financial crisis the ratings uplift to large banks in advanced economies was around 2 notches.”
  - “As the financial crisis unfolded … the ratings support increased to around 3 notches by mid-2009 and has remained broadly at this level since then.”
- Impact on funding costs:
  - In the years leading into the financial crisis, banks “would have had to pay credit spreads that were 60 percent higher had they not benefited from the ratings support.”
  - The TBTF premium “peaked in early 2009 at the height of the global financial crisis.”
  - Despite declines since early 2009, “the TBTF premium remains very high both in absolute level and relative to the actual market credit spreads faced by large banks.”
- Cross-country variation:
  - U.S. banks: “entered the crisis with the lowest levels of support but experienced the largest increase in ratings support … and … their ratings support remains substantial.”
  - Japanese banks: “ratings support … has been high for some time and was little changed during the crisis.”
  - U.K. banks: “ratings support … increased significantly during the crisis and remains high.”
  - Rest of Western Europe: “ratings support … has changed little since mid-2007,” but “the aggregate euro area TBTF premium masks large variations across euro area countries.”
  - Countries noted with significant increases: Austria, Belgium, Germany, and Ireland.
- Estimation approach:
  - TBTF premium is computed as the difference between the credit spread without support (determined by stand-alone rating) and the market credit spread (determined by all-in rating).
  - Method: Estimate long-term wholesale funding credit spreads for a bank at its all-in and stand-alone ratings from indices of option-adjusted bond spreads over government yields by credit rating for U.S. and Western European banks. Assumption: the bank’s credit rating accurately reflects the market view of the bank’s credit risk.

### Appendix B — Estimating the global common cycle (method and data)
- Goal: Estimate the global common cycle for GDP, M2, and equity price using a dynamic factor model (DFM) to account for uneven data across countries.
- Three-step estimation procedure:
  1. Estimate a DFM using the unbalanced dataset of each country’s annual growth rates.
  2. Use the Kalman filter recursion to help predict missing observations to produce a balanced panel.
  3. Re-estimate factors based on the new balanced panel; iterate steps 2 and 3 until convergence.
- Data: Real GDP, nominal M2, and equity price index for each country taken from the Haver database; includes G7 economies.
- Model specification:
  - Factor representation: Xt = ΛFt + et, et ~ N(0, Σ).
  - Factors follow a VAR(p): A(L)Ft = B ut (notation as in source).
  - Assumptions in the model used: three common factors (k), two pervasive common shocks (q), and two lags for the VAR.
- Estimation procedure details:
  - Use principal components on a shorter balanced panel to obtain initial factors and OLS for VAR coefficients.
  - Apply Kalman filter to the entire dataset (including missing observations), re-estimate factors, and fill missing observations using estimated factors via the factor representation.
  - Repeat until no further change to estimated factors.
- Output: A common component estimated from the balanced panel; “adjusted correlations … calculated by subtracting the global common component from each individual country series.”

### Appendix C — Synchronicity mapping using Multidimensional Scaling (MDS)
- Purpose: Use MDS to display structure of distance-like data as a geometrical picture so similar objects (countries) are nearer on a map.
- Distance measure: Defined as one minus the correlation coefficient of countries i and j’s stock price return, over annual data samples 1995–2002 and 2003–2007.
- Procedure:
  - Construct an N by N distance matrix of pairwise dissimilarities.
  - Choose map dimensionality: set to two for ease of visual interpretation.
  - The MDS algorithm finds an arrangement in two-dimensional space that best represents the original distances by minimizing stress (sum of squared differences between original distances and map distances).
  - Goodness of fit is characterized by the stress number or by plotting d_ij against δ_ij (notation from source).
- Note: MDS advantage is limited assumptions on underlying data, yielding robustness; other ordination techniques include principal components analysis and correspondence analysis.

*Source: _111411 - 40.      Retail-based expansions into EMs that are combined with financial innovation also*

### APPENDIX D. DATA APPENDIX

### APPENDIX D. DATA APPENDIX

### Selected Macroeconomic Indicators for Host Countries
- Indicators presented (in percent of GDP, unless otherwise noted): Public Debt, Current Account Balance, Fiscal Balance for years/periods: 2000, 2004, 2007, 2010, and averages for 2002-04, 2005-07, 2008-10.
- Selected country entries (values listed in source order and precision):
  - Austria: 14.3 26.6 7.5 5.3 7.5 (2.2) 5.8
  - Czech Republic: 10.0 16.3 18.1 24.2 (2.0) 19.6 4.3 11.3 10.6 (1.4) 4.0
  - Romania: 1.0 3.2 15.9 15.7 51.7 51.3 13.9 n.a. n.a. n.a. 8.2
  - Hungary: 10.6 18.7 12.5 13.4 20.5 17.1 5.5 7.9 (0.9) (8.4) 7.0
  - Croatia: 5.2 12.7 12.6 13.3 16.1 15.2 6.3 4.1 4.3 (2.5) 7
  - Slovak Republic: 2.8 8.2 10.7 11.2 4.5 30.5 n.a. n.a. 13 0.4 4.8
  - Belgium: 11.9 22.4 9.3 7.2 10.9 (2.6) 4.6
  - Poland: 6.7 17.6 14.7 16.0 6.1 16.3 14.4 (0.6) 24.7 (0.8) 0.3
  - Turkey: 2.7 2.6 10.0 2.5 36.0 34.3 22.6 n.a. n.a. n.a. 8.5
  - Netherlands: 17.5 24.4 15.6 5.1 7.4 0.7 3.3
  - Brazil: 29.0 25.4 29.1 9.4 10.4 24.9 18.1 - - - 1.9
  - China: 3.1 4.9 6.8 8.7 17.6 11.5 20.0 11.4 8.7 11.3 1.9
  - Russia: 3.1 8.1 8.9 8.2 38.4 41.7 13.7 9.8 18.6 (6.2) 8.2
  - India: 5.1 11.3 10.6 8.0 19.6 22.3 16.8 8.3 10.0 5.3 1.5
  - Mexico: 8.1 6.0 5.3 6.9 9.4 19.6 3.8 - - - 6.3
  - Spain: 7.7 18.9 7.5 (4.5) 11.2 4.0 4.1
  - Chile: 14.1 14.7 14.2 13.5 6.4 13.6 3.3 - - - 2.2
  - Argentina: 12.9 5.6 3.8 3.5 (6.7) 32.2 20.3 (6.4) 13.8 1.8 0.2
  - Colombia: 2.5 3.0 3.5 3.3 12.0 20.5 10.2 (2.4) 8.3 6.4 0.2
- Note: Several entries include "n.a." in the source where data are not available.

### Indicators of Host Countries' Relative "Risk Factors"
- Indicators presented: Real Credit (annual growth, in percent), Housing Index (annual growth, in percent), Decline in output (peak to trough), Share in total EM exposures (in percent).
- Aggregate methodology note from source: "For each AM country aggregated numbers are based on top 80 percent exposures to emerging markets. For each column values are calculated as averages weighted by the relative bilateral exposures. Indicators for countries with 5-highest bilateral claims are also presented separately."

### Average Output Growth and Real Domestic Demand (selected excerpts)
- Table headings shown in source: "Average Output Growth (in percent)" and "Real Domestic Demand (annual growth, in percent)" across periods 2002-04, 2005-07, 2008-10 and for years 2002-04, 2005-07, 2008-10 and 2004, 2007, 2010 in parts of the table.
- Selected country excerpts (exact sequences as in source):
  - Australia: 4.5 2.4 -0.2 6.4 3.7 -0.4 23.2 17.4 31.6 -4.4 -8.1 -4.6 4.4 3.6 -2.8
  - Austria: 4.3 5.9 0.0 4.6 7.2 -0.8 39.7 32.1 43.6 -4.5 -5.7 -3.8 -3.7 -1.6 -3.8
  - Czech Republic (first appearance): 22.2 3.3 6.4 0.2 3.5 4.2 -0.4 30.1 29.0 39.6 -5.7 -2.4 -1.4 -5.4 -2.3 -4.4
  - Romania (first appearance): 15.7 6.3 6.1 -0.3 8.1 12.0 -2.3 22.5 19.8 35.2 -5.8 -10.8 -6.7 -2.7 -1.7 -6.2
  - Hungary (partial): 13.4 4.2 2.5 -1.6 .........59.1 66.1 80.4 -7.8 -7.4 -2.1 -7.5 -7.4 -4.0
  - Croatia: 13.3 4.9 4.8 -1.6 6.9 4.9 -2.9 37.9 33.2 40.0 -6.0 -6.7 -5.5 -4.8 -3.0 -3.9
  - Slovak Republic (first appearance): 11.2 4.8 8.6 1.7 2.7 7.5 0.4 41.5 29.6 42.0 -7.2 -7.2 -4.6 -5.2 -2.4 -5.3
  - Belgium: 4.0 6.4 0.7 3.8 6.5 0.7 39.8 35.5 47.1 -5.1 -2.8 -1.9 -4.7 -1.7 -3.5
  - Poland (first appearance): 16.0 3.6 5.5 3.5 2.7 6.0 3.4 45.7 45.0 55.7 -3.1 -2.9 -3.4 -5.7 -3.5 -6.3
  - Turkey (first appearance): 2.5 6.9 6.7 1.6 9.8 7.4 1.6 59.2 39.4 42.2 -2.1 -5.5 -4.8 -9.3 -0.6 -3.6
  - Netherlands: 4.5 6.3 3.5 4.5 7.2 3.7 52.1 45.6 43.5 0.1 0.5 -1.4 -4.1 -1.8 -4.5
  - Brazil (first appearance): 9.4 3.2 4.4 4.0 1.6 5.6 5.6 70.6 65.2 66.1 0.3 1.0 -1.8 -4.2 -3.3 -2.5
  - China (first appearance): 8.7 9.7 12.7 9.7 9.6 11.0 11.5 18.5 19.6 17.7 2.9 9.0 6.9 -2.0 -0.5 -1.6
  - Russia (first appearance): 8.2 6.4 7.7 0.5 7.9 11.3 8.7 22.3 8.5 9.9 8.9 8.8 5.1 2.4 6.8 -2.0
  - India (first appearance): 8.0 6.5 9.6 7.8 7.6 10.2 7.8 84.1 75.8 72.2 1.0 -1.0 -2.7 -8.6 -5.4 -9.1
  - Mexico (first appearance): 6.9 1.8 3.9 0.3 2.4 5.2 -0.1 41.4 37.8 42.7 -1.2 -0.7 -0.9 -1.5 -1.4 -2.6
  - Spain (first appearance): 2.4 4.9 2.5 2.2 6.9 3.1 44.4 38.7 46.2 0.3 1.5 -0.8 -1.5 -0.4 -2.2
  - Chile: 13.5 4.1 4.9 2.4 .........10.7 4.1 8.8 0.1 3.5 0.5 2.1 7.0 -0.1
  - Argentina: 3.5 3.0 3.7 2.5 4.0 4.8 -2.6 125.8 67.7 47.8 5.5 2.7 1.3 -7.6 -1.7 -2.1
  - Colombia: 3.3 3.9 6.1 3.1 4.2 7.7 3.6 42.9 32.7 36.5 -0.9 -2.0 -2.7 -2.1 -0.5 -2.0
- Sources for these tables in the source file: BIS, GFSR, IFS, WEO, and Fund staff calculations.

### Summary Table of Home Country Parent Banks (selected fields)
- Table headings and notes in source:
  - Parent Banks; Assets as of 12/31/2010 (in USD billions); EBA: EM credit risk exposure 1/ (% of home banking system assets); No. of EM Subsidiaries 2/ (% of total); 2010 EM Net Income (2000-2010 risk-adjusted average 7/); Regional EM presence; ROE (Current, 11/4/2011); Price-to-book; Loan-to-deposit; Non-deposit Funding Ratio; Core Tier 1; Leverage Ratio.
  - Notes: 1/ Exposure at default as of December 31, 2010 reported to EBA. 2/ Subsidiaries where the parent has more than 25% ownership. Does not include non-bank financial entities that are not within an EM subsidiary banking group. 3/ EM credit risk exposure for Raffeisen International. 4/ Unicredit Bank Austria, which contains both Austrian and CEE subsidiaries, is a member of the Italian Unicredit Group. 5/ Pre-tax profit. 6/ Net revenues. 7/ Risk adjusted returns = annual average ROE (ROA) / standard deviation of ROE (ROA) between 2000 - 2010.
- Selected bank entries with exact numbers as presented:
  - Austria (parent banks and selected affiliates):
    - Erste Bank: Assets 275; EBA: 183; EM credit risk exposure 81249; CEESEE regional; ROE 5.0 0.50 0 8.8 14.0
    - Raiffeisen: Assets 182; EBA: 125; EM credit risk exposure 13/2646; CEESEE regional; ROE 2.1 0.40 0 8.3 12.5
    - Unicredit Bank Austria 4/: Assets 258; EBA 171; 9935/CEESEE; ROE 1.4 NA 0 0 10.0 10.8
  - Belgium:
    - Dexia: Assets 757; EBA 384; 4186/Turkey; ROE 0.5 0.10 0 10.3 59.1
    - KBC: Assets 429; EBA 223; 013346/CEESEE; ROE 0.6 0.40 0 12.1 16.5
  - Netherlands:
    - ING: Assets 1,247; EBA 345; 63.05/Mixed; ROE 2.0 NA 0 0 9.6 226.6
    - Rabobank: Assets 872; EBA 245; 8Mixed; ROE 9.9 NA 0 0 14.2 15.6
  - Spain:
    - BBVA: Assets 739; EBA 152; 01044Latam; ROE 4.7 0.80 0 9.1 14.7
    - Santander: Assets 1,627; EBA 322; 0962Latam; ROE 6.8 0.71 47519.4 16.0
- Data sources for bank table in the source file: Bankscope, Bloomberg, SNL, financial stability reports, EBA, company annual reports, IMF staff calculations.

### Sources, Notes, and Methodology Highlights
- Sources cited in the appendix: BIS, GFSR, IFS, WEO, Bankscope, Bloomberg, SNL, financial stability reports, EBA, company annual reports, and IMF staff calculations.
- Methodological notes included in source:
  - "1/ Based on top 80 percent exposures to emerging markets in 2007. For each column values are calculated as averages weighted by the relative bilateral exposures for each period."
  - "1/ For each AM country aggregated numbers are based on top 80 percent exposures to emerging markets. For each column values are calculated as averages weighted by the relative bilateral exposures. Indicators for countries with 5-highest bilateral claims are also presented separately."
  - Bank table footnotes 1–7 provided in source (see table for text and exact numbering).

*Source: APPENDIX D. DATA APPENDIX (IMF PDF content provided).*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2011/_111411.pdf_