## _wp1271

## Source details

**Canonical URL:** [_wp1271](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp1271.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp1271.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp1271.pdf.json)

---

### I. Scope, data and framework
- Objective: examine how credit growth and asset quality in 25 emerging markets (EM) relate to domestic and external factors using dynamic panel regressions (1996-2010) and structural panel Vector Auto Regressions (VARs).
- Sample and period: 25 EM during 1996-2010.
- Main variables considered: GDP growth, the exchange rate, terms of trade, capital flows, equity prices, interest rates, credit growth, aggregate nonperforming loans (NPLs) to total loans.
- Methodology:
  - Dynamic panel regressions (OLS, FE, Arellano-Bond difference GMM (AB), System GMM).
  - Structural panel VARs with recursive identification via a Choleski decomposition.
- JEL Classification Numbers: E32, E44, E51, F30, F32, F34, G2
- Keywords: Macro-financial linkages, Emerging markets, Banks, Nonperforming loans, Capital flows, Exchange rate, Terms of trade, Financial stability, Panel regressions, Time series.

### II. Key empirical findings (high level)
- Determinants associated with higher aggregate NPL levels:
  - A slowdown in economic growth.
  - A weaker exchange rate.
  - A deterioration in terms of trade.
  - Rapid credit growth (with subsample variation).
- Composition of capital flows matters: portfolio and bank-related flows—rather than FDI—are associated with stronger links to loan quality deterioration and credit contractions.
- Feedback effects: worsening asset quality and credit contractions feed back to reduce economic growth and tend to depreciate the exchange rate.

### III. Panel regression findings — selected coefficients and interpretations
- Baseline dynamic panel (lagged dependent variable and one lag of explanatory variables). Selected reported coefficients:
  - nplrat_t-1: OLS = 0.520***; FE = 0.384***; AB = 0.299***; System GMM = 0.326***
  - rgdpgr_t-1: OLS = -0.162**; FE = -0.221***; AB = -0.180*; System GMM = -0.219*
  - pbf_t (portfolio and bank flows % GDP): OLS = -0.491***; FE = -0.543***; AB = -0.605***; System GMM = -0.670***
    - Interpretation: "a 1 percentage point increase in this variable is associated with a reduction in the aggregate NPL ratio of approximately 0.5 percentage points."
  - pvtratgr_t-1 (private credit to GDP growth): OLS = 0.027*; FE = 0.02; AB = 0.018; System GMM = 0.025
  - totgr_t (terms-of-trade growth): OLS = -0.080**; FE = -0.055; AB = -0.075**; System GMM = -0.091**
  - ergr_t-1 (nominal exchange rate growth): OLS = -0.047*; FE = -0.055**; AB = -0.104***; System GMM = -0.099***
  - R-squared (OLS): 0.53; R-squared (FE): 0.52
  - Arellano-Bond AR(1) test p-values: AB = 0.05; System = 0.05; AR(2) p-values: AB = 0.99; System = 0.97
- Robustness and subsample findings:
  - Equity prices (reqgr) and lending rates (lr) add limited explanatory power beyond baseline variables.
  - Subsample comparison (1996–2003 vs 2004–2010): parameter estimates are quantitatively similar except that credit growth plays a stronger, more narrowly associated role with asset quality problems in 1996–2003.

### IV. Capital flows and sensitivity of NPLs
- Composition matters:
  - Portfolio and bank-related flows dominated inflows in a number of countries in 2009-11.
- Estimated sensitivity:
  - A 1 percentage point decrease in portfolio and bank liabilities (% GDP) increases the level of aggregate NPLs by 0.5 percentage points.
- Additional panel-flow coefficients (Table 7 examples):
  - ncf_t (net capital flows): OLS = -0.052; FE = -0.088; AB = -0.176**; System GMM = -0.145*
  - pf_t (foreign portfolio inflows excluding bank flows): OLS = -0.299***; FE = -0.385***; AB = -0.465***; System GMM = -0.621***
  - fdi_t: coefficients small and statistically insignificant across estimators

### V. Structural panel VAR results — feedback loops and variance decomposition
- Baseline VAR ordering (recursive identification): nplrat, pvtratgr, pbf, rgdpgr, ergr.
- Impulse-response findings (qualitative):
  - Decreases in rgdpgr, ergr, totgr, or debt-creating capital inflows lead to reduced private credit and deteriorating loan quality.
  - Increases in NPLs or credit contractions slow economic activity.
  - Banking-sector stress tends to lead to nominal exchange rate depreciation.
  - A spike in NPLs tends to reduce flows; a credit shock does not necessarily reduce flows.
- Variance decomposition (forecast horizon 10 years; percent of forecast error variance explained):
  - For nplrat: nplrat = 91.07; pvtratgr = 0.85; pbf = 0.28; rgdpgr = 5.53; ergr = 2.27
  - For pvtratgr: nplrat = 8.22; pvtratgr = 81.19; pbf = 3.03; rgdpgr = 3.60; ergr = 3.96
  - For pbf: nplrat = 12.13; pvtratgr = 0.80; pbf = 83.89; rgdpgr = 1.21; ergr = 1.97
  - For rgdpgr: nplrat = 8.43; pvtratgr = 2.06; pbf = 7.42; rgdpgr = 82.07; ergr = 0.03
  - For ergr: nplrat = 4.55; pvtratgr = 4.02; pbf = 13.56; rgdpgr = 23.27; ergr = 54.60
- Text summary: "shocks in the financial system account for approximately 9 percent of the total forecast error variance of real growth."

### VI. Robustness checks and VAR extensions
- Robustness checks:
  - Inclusion of equity prices and interest rates: limited additional explanatory power.
  - Varying subsamples: main estimates stable except for credit growth in 1996–2003.
- VAR extensions:
  - Replacing pbf with pf, fdi, or ncf: portfolio and bank flows, and foreign portfolio flows, reduce NPLs; foreign direct investment shows no significant link to loan quality.
  - Debt-creating portfolio flow shocks increase private credit; shocks to FDI or net capital flows do not have the same effect.
  - Including totgr disentangles exchange rate versus terms-of-trade shocks: both have similar qualitative effects on credit and NPLs; specification shows a terms-of-trade shock affects the exchange rate but not vice versa.

### VII. Policy-relevant interpretations and applications
- Monitoring composition of capital flows is crucial: reliance on portfolio and bank-related flows raises vulnerability to sudden reversals that can materially worsen loan quality and contract credit.
- Use of sensitivity estimates:
  - The estimated sensitivity (1 percentage point decrease in portfolio and bank liabilities (% GDP) -> 0.5 percentage point increase in aggregate NPLs) can identify EM with heightened exposure to reversals in these flow categories.
- Macro-financial surveillance emphasis:
  - Early warning indicators tied to rapid credit growth, exchange rate pressures, terms-of-trade deterioration, and shifts in capital flow composition.
  - Plan for potential bank recapitalization if asset quality deteriorates sharply following capital flow reversals.
- Applications:
  - Econometric links can be used to develop stress-testing scenarios. A two-stage exercise is described: use panel VAR to generate scenarios with structural shocks to GDP growth, terms of trade and funding costs; then apply an IMF solvency framework to assess impacts on banks’ capital adequacy ratios.

### VIII. Conclusions and directions for future research
- Core conclusions:
  - Slower growth, weaker exchange rates or terms of trade, and rapid credit growth are independently associated with higher NPLs in EM.
  - Structural shocks that raise NPLs or contract credit have economically significant negative effects on economic growth.
  - Capital flow composition matters for credit dynamics and bank asset quality, with portfolio and bank-related flows posing particular risks.
- Suggested extensions:
  - Incorporate other measures of asset quality beyond NPL ratios (e.g., firm-level measures such as the share of corporate debt for which interest expenses exceed EBITDA).
  - Conduct more granular analysis of funding markets: quantify vulnerabilities associated with a country’s bank funding model, examine currency composition of private and public debt, and measure implications of a changing investor base.
  - Explore the feedback channel on investment: for net capital goods importers, currency depreciation (driven by worse asset quality) raises relative price of imported investment goods, potentially constraining investment during deteriorating private sector balance sheets.

*Chapter 1 of the Global Financial Stability Report (September 2011) and chapter 4 of the World Economic Outlook (2011) — working paper chapter text provided.*

### chapter 1 of the Global Financial Stability Report (September 2011) to help evaluate the

### chapter 1 of the Global Financial Stability Report (September 2011)

### I. Scope, data and framework
- Objective: examine how credit growth and asset quality in 25 emerging markets (EM) relate to domestic and external factors using dynamic panel regressions (1996-2010) and structural panel Vector Auto Regressions (VARs).
- Sample and period: 25 EM during 1996-2010.
- Main variables considered: GDP growth, the exchange rate, terms of trade, capital flows, equity prices, interest rates, credit growth, aggregate nonperforming loans (NPLs) to total loans.
- Methodological notes:
  - Dynamic panel regressions to identify drivers of bank asset quality.
  - Structural panel VARs to quantify real-economy responses to credit contractions or deteriorations in bank balance sheets (measured by NPLs).
  - Structural identification via a Choleski decomposition.
- JEL Classification Numbers: E32, E44, E51, F30, F32, F34, G2
- Keywords: Macro-financial linkages, Emerging markets, Banks, Nonperforming loans, Capital flows, Exchange rate, Terms of trade, Financial stability, Panel regressions, Time series.
- Author contact: rdebock@imf.org ; ademyan1@jhu.edu

### II. Key empirical findings from panel regressions
- Determinants associated with higher aggregate NPL levels:
  - A slowdown in economic growth.
  - A weaker exchange rate.
  - A deterioration in terms of trade.
  - Rapid credit growth.
- Composition of explanatory power:
  - The market value of equity and interest rates have limited explanatory power beyond the baseline model variables.
- Subsample stability:
  - Parameter estimates are quantitatively similar across subsamples, with a notable exception:
    - In the earlier sample period (1996-2003), strong credit growth is more narrowly associated with asset quality problems.

### III. Findings on capital flows and their effects
- Composition matters:
  - Portfolio and bank-related flows (rather than FDI) have dominated inflows in a number of countries in 2009-11.
- Estimated sensitivity of NPLs to abrupt reversals:
  - A 1 percentage point decrease in portfolio and bank liabilities (% GDP) increases the level of aggregate NPLs by 0.5 percentage points.
- Economic context:
  - Net capital flows to EM have not been excessively strong by historical standards, but the dominance of portfolio and bank-related flows in some countries creates vulnerability to sudden reversals.
- Model implications:
  - Abrupt reversal of bank-related and foreign portfolio flows is associated with a sharp contraction of credit and a deterioration of loan quality.
  - Predicted outcomes include a significant contraction of bank lending to the private sector and potential bank recapitalization needs.

### IV. Structural panel VAR results and feedback loops
- Presence of significant feedback effects from the financial sector to the real economy.
- Responses to structural shocks:
  - Economic growth falls significantly after structural shocks that drive nonperforming loans higher.
  - Economic growth also falls significantly after structural shocks that generate a contraction in credit.
- Consistency:
  - VAR results are consistent with panel regression evidence: negative structural shocks to economic growth, the exchange rate, or debt-creating capital inflows tend to bring down private credit while loan quality deteriorates.
- Model composition:
  - Baseline VAR includes variables identified as key drivers of asset quality in panel regressions.

### V. Robustness checks and extensions
- Robustness exercises include:
  - Incorporating additional information from financial markets (equity prices, interest rates).
  - Varying the subsample over which models are estimated.
- Results of robustness checks:
  - Equity prices and interest rates add limited explanatory power beyond baseline variables.
  - Estimates remain similar across subsamples except for the differing role of credit growth in the 1996-2003 period.
- Extensions in VAR analysis:
  - Examined the impact on credit and asset quality of shocks to capital flows by type (bank flows, portfolio flows, FDI).
  - Disentangled exchange rate and terms of trade shocks to evaluate their separate effects on credit and NPLs.

### VI. Policy-relevant interpretations
- Monitoring composition of capital flows is crucial: reliance on portfolio and bank-related flows raises vulnerability to sudden reversals that can materially worsen loan quality and contract credit.
- Country-specific vulnerabilities:
  - The estimated sensitivity (1 percentage point decrease in portfolio and bank liabilities (% GDP) -> 0.5 percentage point increase in aggregate NPLs) can be used to identify EM with heightened exposure to reversals in these flow categories.
- Macro-financial surveillance should emphasize:
  - Early warning indicators tied to rapid credit growth, exchange rate pressures, terms of trade deterioration, and shifts in capital flow composition.
  - The potential need for bank recapitalization planning if asset quality deteriorates sharply following capital flow reversals.

### VII. Conclusion and research directions
- Main conclusions:
  - Slower growth, weaker exchange rates or terms of trade, and rapid credit growth are independently associated with higher NPLs in EM.
  - Structural shocks that raise NPLs or contract credit have economically significant negative effects on economic growth.
  - Capital flow composition matters for credit dynamics and bank asset quality, with portfolio and bank-related flows posing particular risks.
- Future work: further disentangling mechanisms across different EM, and refining policy tools to mitigate vulnerabilities arising from capital flow composition and rapid credit expansion.

*Chapter 1 of the Global Financial Stability Report (September 2011), working paper chapter text provided.*

### Chapter 1 of Global Financial Stability Report (2011) and chapter 4 of World Economic Outlook (2011)

### _wp1271 - Chapter 1 of Global Financial Stability Report (2011) and chapter 4 of World Economic Outlook (2011)

### Literature review: channels linking asset quality, credit and external variables
- Pro-cyclical credit markets
  - Financial accelerator channel: asset-price increases raise net worth and borrowing capacity, amplifying booms and potentially leading to lending against inflated collateral.
  - Herd behavior and fading institutional memory can weaken lending standards during booms.
  - Empirical evidence: credit and credit quality tend to be pro-cyclical; increases in NPLs generate a strong, short-lived negative response in economic activity across advanced economies, emerging markets in Central and Eastern Europe, and the Gulf Cooperation Council (GCC).
- Asset quality and external prices
  - Currency weakening makes foreign-currency debt harder to service (especially for non-tradable producers) and can exacerbate banking-system weaknesses, particularly in dollarized or lightly regulated systems.
  - Banks may incompletely hedge exchange-rate risk due to implicit government bailout expectations.
  - Terms-of-trade shocks are emphasized for commodity exporters; empirical unconditional correlation between real exchange rate and terms of trade can be weak.
- Asset quality and capital flows
  - Different gross capital flows have distinct implications for credit and credit quality.
  - Evidence (Tong and Wei (2010)): large pre-crisis exposure to non-FDI capital inflows is associated with more severe credit crunches; pre-crisis exposure to FDI does not worsen credit crunches.
  - Paper aims to assess vulnerabilities that capital flows pose to emerging-market (EM) financial systems, controlling for business cycle phase and FX performance.

### Data: sample, variables, stationarity, stylized facts
- Sample and coverage
  - Annual observations for 25 emerging markets from 1996 to 2010 where available.
  - Countries: Argentina, Brazil, Bulgaria, Chile, China, Colombia, Dominican Republic, Hungary, India, Indonesia, Israel, Korea, Malaysia, Mexico, Peru, Philippines, Poland, Romania, Russia, South Africa, Thailand, Turkey, Ukraine, Uruguay, and Venezuela.
- Key variables and sources (as labeled in the paper)
  - nplrat: Non-performing loans (share total loans) — Bankscope
  - rgdpgr: Growth rate of real GDP — World Economic Outlook
  - pvtratgr: Growth rate in ratio of private credit to GDP — Haver, IFS, World Economic Outlook
  - pbf: Foreign portfolio and bank flows (% GDP) — IFS, World Economic Outlook
  - ncf: Net capital flows (% GDP) — IFS, World Economic Outlook
  - pf: Foreign portfolio flows (% GDP) — IFS, World Economic Outlook
  - fdi: Foreign direct investment (% GDP) — IFS, World Economic Outlook
  - ergr: Growth U.S. Dollars per national currency — World Economic Outlook
  - rergr: Growth in the real exchange rate — Information Notice System (IMF)
  - totgr: Growth in terms of trade (goods) — World Economic Outlook
- Stationarity
  - Panel unit-root tests (Table 2) generally reject non-stationarity for key variables. Selected reported test statistics:
    - nplrat: Im-Pesaran-Shin = -3.32***; Levin-Lin-Chu = -12.38***; Fisher-ADF = 122.17***; Fisher-PP = 169.62***
    - rgdpgr: Im-Pesaran-Shin = -6.79***; Levin-Lin-Chu = -9.64***; Fisher-ADF = 44.42; Fisher-PP = 168.86***
    - pvtratgr: Im-Pesaran-Shin = -4.22***; Levin-Lin-Chu = -9.55***; Fisher-ADF = 44.73; Fisher-PP = 229.97***
    - pbf: Im-Pesaran-Shin = -4.46***; Levin-Lin-Chu = -11.10***; Fisher-ADF = 65.75*; Fisher-PP = 255.17***
    - rrgr: Im-Pesaran-Shin = -6.96***; Levin-Lin-Chu = -12.25***; Fisher-ADF = 103.74***; Fisher-PP = 201.39***
    - totgr: Im-Pesaran-Shin = -10.94***; Levin-Lin-Chu = -14.59***; Fisher-ADF = 56.92; Fisher-PP = 333.20***
  - Note in text: Fisher-ADF test does not reject for rgdpgr, pvtratgr and totgr (potentially due to small panel sizes).
- Stylized facts and correlations (Table 3 highlights)
  - NPLs are countercyclical; portfolio flows, exchange rates (USD per national currency or real effective), and terms of trade are procyclical.
  - Selected correlations and volatilities from Table 3 (standard deviations on diagonal; correlations off-diagonal):
    - nplrat standard deviation: 7.99
    - rgdpgr standard deviation: 4.37; correlation nplrat–rgdpgr = -0.19**
    - pvtratgr standard deviation: 19.08; correlation nplrat–pvtratgr = -0.19**
    - pbf standard deviation: 3.78; correlation nplrat–pbf = -0.35**
    - ergr standard deviation: 14.77; correlation rgdpgr–ergr = 0.50***
    - rergr standard deviation: 10.16; correlation rergr–ergr = 0.56***
    - totgr standard deviation: 8.29; correlation nplrat–totgr = -0.13*
  - Cross-country variation: e.g., Brazil and Turkey show pro-cyclical financial systems; China experienced rapid credit growth post-2008 (policy-induced lending boom of 2009-10).

### Determinants of bank asset quality: panel regressions
- Baseline dynamic panel specification (one lag of explanatory variables and lagged dependent variable):
  - Dependent variable: nplrat_i,t (NPL ratio)
  - Regressors include lagged nplrat, rgdpgr_t-1, pvtratgr_t-1, pbf_t, totgr_t, ergr_t-1 (and others in extensions).
  - Estimation methods: OLS, country fixed effects (FE), Arellano-Bond difference GMM (AB), System GMM.
- Main panel regression results (Table 4) — reported coefficients and significance
  - nplrat_t-1: OLS = 0.520***; FE = 0.384***; AB = 0.299***; System GMM = 0.326*** (t-stats reported in table)
  - rgdpgr_t-1: OLS = -0.162**; FE = -0.221***; AB = -0.180*; System GMM = -0.219* 
  - pbf_t: OLS = -0.491***; FE = -0.543***; AB = -0.605***; System GMM = -0.670*** 
    - Interpretation in text: "a 1 percentage point increase in this variable is associated with a reduction in the aggregate NPL ratio of approximately 0.5 percentage points."
  - pvtratgr_t-1: OLS = 0.027*; FE = 0.02; AB = 0.018; System GMM = 0.025
  - totgr_t: OLS = -0.080**; FE = -0.055; AB = -0.075**; System GMM = -0.091** 
  - ergr_t-1: OLS = -0.047*; FE = -0.055**; AB = -0.104***; System GMM = -0.099*** 
  - R-squared (OLS): 0.53; R-squared (FE): 0.52
  - Arellano-Bond AR(1) test p-values reported as 0.05 (AB) and 0.05 (System); AR(2) p-values 0.99 and 0.97 respectively.
- Interpretation of core findings
  - Portfolio and bank flows (pbf) have a large, highly significant negative association with aggregate NPL ratios.
  - Economic growth is an important determinant: NPL ratios are countercyclical. Elasticity estimate of about 0.2 is comparable to estimates for advanced economies.
  - Terms of trade appreciation and nominal exchange rate appreciation have similar independent effects in reducing NPLs.
  - Credit growth (pvtratgr) coefficient is small and generally insignificant in baseline; subsample analysis shows parameter variation over time.
- Robustness checks and extensions (selected results)
  - Adding real equity returns (reqgr) and lending rates (lr) (Table 5):
    - reqgr coefficients ~ 0.000 to 0.003 (insignificant)
    - lr coefficients small, mixed signs, generally insignificant in full specification
    - Core benchmark coefficient magnitudes and signs remain roughly unchanged.
  - Subsample stability (Table 6): regressions run for 1996–2003 and 2004–2010; coefficients broadly similar except for credit growth which changes sign across subsamples.
  - Composition of capital flows (Table 7):
    - pbf_t: OLS = -0.491***; FE = -0.543***; AB = -0.605***; System GMM = -0.670***
    - ncf_t (net capital flows): OLS = -0.052; FE = -0.088; AB = -0.176**; System GMM = -0.145*
    - pf_t (foreign portfolio inflows excluding bank flows): OLS = -0.299***; FE = -0.385***; AB = -0.465***; System GMM = -0.621***
    - fdi_t: coefficients small and statistically insignificant across estimators

### Modeling feedback loops: structural panel VAR results
- Baseline panel VAR (variables and ordering)
  - Five-variable baseline VAR: nplrat, pvtratgr, pbf, rgdpgr, ergr (in that order).
  - Reduced-form: Z_it = c0 + B1 Z_{i,t-1} + f_i + ε_it
  - Recursive identification (Choleski): financial variables affect the real economy within a year but not vice versa (ordering chosen accordingly); results robust to ordering of macro variables.
- Impulse-response function (IRF) findings (Figure 4; qualitative summary)
  - Decreases in economic growth, nominal exchange rate, terms of trade, or debt-creating capital inflows lead to reduced private credit and deteriorating loan quality.
  - Positive growth shocks and shocks to portfolio and bank flows tend to lead to exchange rate appreciation.
  - Shocks originating in the financial system matter for the real economy:
    - Increases in NPLs or credit contractions slow economic activity.
    - Banking-sector stress leads to nominal exchange rate depreciation.
    - A spike in NPLs tends to reduce flows, whereas a credit shock does not necessarily reduce flows.
- Variance decomposition (Table 8; percent of forecast error variance explained by shocks in column for each row variable)
  - Forecast horizon 10 years (and identical at 20 years as reported):
    - nplrat forecast error variance explained by: nplrat = 91.07; pvtratgr = 0.85; pbf = 0.28; rgdpgr = 5.53; ergr = 2.27
    - pvtratgr: nplrat = 8.22; pvtratgr = 81.19; pbf = 3.03; rgdpgr = 3.60; ergr = 3.96
    - pbf: nplrat = 12.13; pvtratgr = 0.80; pbf = 83.89; rgdpgr = 1.21; ergr = 1.97
    - rgdpgr: nplrat = 8.43; pvtratgr = 2.06; pbf = 7.42; rgdpgr = 82.07; ergr = 0.03
    - ergr: nplrat = 4.55; pvtratgr = 4.02; pbf = 13.56; rgdpgr = 23.27; ergr = 54.60
  - Text note: "shocks in the financial system account for approximately 9 percent of the total forecast error variance of real growth."
- Extensions within VAR framework
  - Replacing pbf with different capital-flow measures:
    - IRFs show that portfolio and bank flows, and foreign portfolio flows, reduce NPLs; foreign direct investment shows no significant link to loan quality.
    - Net capital flows show mixed results but in the VAR foreign portfolio flows and net capital flows can bring down NPLs.
  - Effects on private credit:
    - Innovations/shocks in debt-creating portfolio flows tend to increase private credit.
    - Shocks to foreign direct investment or net capital flows do not have the same effect on private credit.
    - Implication: gross portfolio flows matter more than net flows for financial-stability risk mapping.
  - Disentangling exchange rate vs. terms-of-trade shocks:
    - Re-estimating VAR including totgr (terms-of-trade growth) shows shocks to terms of trade and exchange rate have similar qualitative effects on financial variables.
    - A terms-of-trade shock affects the exchange rate but not vice versa in this specification; result is robust to ordering and consistent with earlier literature (Mendoza (1995), Cashin et al. (2004)).

### Conclusions, policy relevance, and future research directions
- Key conclusions
  - Statistically and economically significant links exist between macroeconomic aggregates and the banking sector in EM: deteriorating growth prospects, depreciating exchange rates, weaker terms of trade, and a fall in debt-creating capital inflows reduce private credit and worsen loan quality.
  - Feedback loops: worsening asset quality and credit contractions slow economic activity and tend to depreciate the exchange rate.
  - Portfolio and bank flows (and foreign portfolio inflows) have particularly large associations with aggregate NPL ratios; FDI shows little effect on NPLs.
- Policy relevance and applications
  - Econometric links can be used to develop stress-testing scenarios. Related models were applied in chapter 1 of the IMF’s Global Financial Stability Report (2011) to assess EM vulnerabilities to aggregate shocks.
  - Two-stage exercise mentioned: use panel VAR to generate scenarios with structural shocks to GDP growth, terms of trade and funding costs; then apply an IMF solvency framework to assess impacts on banks’ capital adequacy ratios.
- Suggested avenues for future research (three extensions highlighted)
  - Incorporate other measures of asset quality beyond NPL ratios (e.g., firm-level measures such as the share of corporate debt for which interest expenses exceed EBITDA).
  - Conduct more granular analysis of funding markets: quantify vulnerabilities associated with a country’s bank funding model, examine currency composition of private and public debt, and measure implications of a changing investor base.
  - Explore the feedback channel on investment: for net capital goods importers, currency depreciation (driven by worse asset quality) raises relative price of imported investment goods, potentially constraining investment during deteriorating private sector balance sheets.

*Source: Chapter 1 of Global Financial Stability Report (2011) and chapter 4 of World Economic Outlook (2011).*

### REFERENCES

### _wp1271 - REFERENCES

### Methodology and Econometric Techniques
- Arellano, M., and S. Bond (1991), “Some tests of specification for panel data: Monte Carlo evidence and an application to employment equations,” Review of Economic Studies, 58, 277-297.
- Arellano, M., and O. Bover (1995), “Another look at the instrumental variable estimation of error-components models,” Journal of Econometrics, 68, 29-51.
- Blundell, R., and S. Bond (1998), “Initial conditions and moment restrictions in dynamic panel data models,” Journal of Econometrics, 87, 115-143.
- Blundell, R.W. and S.R. Bond and F. Windmeijer (2000), “Estimation in Dynamic Panel Data Models: Improving on the Performance of the Standard GMM estimator,” in B. Baltagi (ed), Advances in Econometrics, Volume 15: Non-Stationary Panels, Panel Co-integration, and Dynamic Panels, JAI Elsevier Science.
- Im, K. S., M. H. Pesaran, and Y. Shin (2003), “Testing for unit roots in heterogeneous panels,” Journal of Econometrics, 115: 53-74.
- Levin, A., C.-F. Lin, and C.-S. J. Chu (2002), “Unit root tests in panel data: Asymptotic and finite-sample properties,” Journal of Econometrics, 108, 1-24.

### Credit, Banking, and Financial Fragility
- Bernanke, Ben and Mark Gertler (1989), “Agency costs, net worth, and business fluctuations,” American Economic Review 79, 14-31.
- Berger, A., and G. Udell (2004), “The institutional memory hypothesis and the procyclicality of bank lending behavior,” Journal of Financial Intermediation, 13(4), 458-95.
- Burnside, Craig, Martin Eichenbaum and Sergio Rebelo (2001), “Hedging and financial fragility in fixed exchange rate regimes,” European Economic Review, 45, 1151-93.
- Carlson, Mark, Thomas King, and Kurt Lewis (2009), “Distress in the Financial Sector and Economic Activity,” Finance and Economics Discussion Series, Federal Reserve Board, Washington DC.
- Chen, Nan-Kuang (2001), “Bank Net Worth, Asset Prices and Economic Activity,” Journal of Monetary Economics, 48, 415-436.
- Demirguc-Kunt, Asli and Enrica Detragiache (1998), "The Determinants of Banking Crises in Developing and Developed Countries,” IMF Staff Papers, Vol. 45.
- Gambacorta, Leonardo (2005), “Inside the Bank Lending Channel,” European Economic Review, 49, 1737-1759.
- Hartelius (2010), “Analyzing nonperforming loans in Central and Eastern Europe based on historical experience in emerging markets,” Global Financial Stability Report, Washington, April 2010.
- Jiminez, Gabriel, and Jesus Saurina (2006), “Credit Cycles, Credit Risk and Prudential Regulation,” International Journal of Central Banking, 2, 65-98.
- Kiyotaki, Nobuhiro and John Moore (1997), “Credit cycles,” Journal of Political Economy 105 (2), 211-48.
- Lown, Cara and Donald Morgan (2006), “The Credit Cycle and the Business Cycle: New Findings using the Loan Officer Opinion Survey,” Journal of Money, Credit and Banking, 38(6), 1575-97.
- Marcucci, Juri and Mario Quagliariello (2008), “Is bank portfolio riskiness procyclical? Evidence from Italy using a vector autorgression,” Journal of International Financial Markets, Institutions and Money, 18, 46-63.
- Meeks, Roland (2011), “Do credit market shocks drive output fluctuations? Evidence from corporate spreads and defaults,” mimeo, Bank of England.
- Nkusu, Mwanza (2011), “Nonperforming Loans and Macrofinancial Vulnerabilities in Advanced Economies,” IMF Working Paper, 161, July 2011.
- Espinoza, Raphael, and Ananthakrishnan Prasad (2010), “Nonperforming Loans in the GCC Banking System and their Macroeconomic Effects,” IMF Working Paper, 224.
- Heytens, Paul and Cem Karacadag (2001), “An Attempt to Profile the Finances of China’s Enterprise Sector,” IMF Working Paper, 182.
- Rajan (1994), “Why Bank Credit Policies Fluctuate: A Theory and Some Evidence,” Quarterly Journal of Economics, 109 (2), 399-441.

### Capital Flows, Exchange Rates, and External Shocks
- Borio, Claudio and Piti Disyatat (2011), “Global imbalances and the financial crisis: Link or no link?,” BIS working paper No 346, Basel.
- Borio, Claudio, Craig Furfine, and Philip Lowe (2001). “Procyclicality of financial systems and financial stability, BIS Papers No 1, Basel.
- Broner, Fernando, Tatiana Didier, Aitor Erce, and Sergio Schmukler (2011), “Gross capital flows: Dynamics and crises,” mimeo, World Bank.
- Cardarelli, Roberto, Selim Elekdag and M. Ayhan Kose (2009), “Capital inflows: macroeconomic implications and policy responses,” IMF Working Paper, 40, 2009.
- De Bock, Reinout (2010), “The Composition and Cyclical Behavior of Trade Flows in Emerging Economies”, IMF Working Paper, 46.
- De Bock, Reinout and Alexander Demyanets (2011), “Macro-financial linkages in emerging markets and impact of shocks on bank capital adequacy ratios,” Global Financial Stability Report, International Monetary Fund, Washington DC, September 2011.
- Dornbusch, Rudiger, Ilan Goldfajn, and Rodrigo Valdes (1995), “Currency Crises and Collapses,” Brooking Papers on Economic Activity, June 1995, 219-70.
- Eaton, Jonathan, and Samuel Kortum (2001), “Trade in Capital Goods”, European Economic Review, 45, 1195-235.
- Forbes, Kristin and Francis Warnock (2011), “Capital flows waves; Surges , Stops, Flight and Retrenchment,” mimeo, MIT.
- Helbling, Thomas, Raju Huidrom, Ayhan Kose, and Christopher Otrok (2011), “Do credit shocks matter? A global perspective,” European Economic Review, 55, 340-353.
- Levchenko, A. and P. Mauro (2007), “Do some forms of financial flows help protect against sudden stops?,” World Bank Economic Review, 21, 389-411.
- Magud, Nicolas, Carmen Reinhart and Esteban Vesperoni (2011), “Capital Inflows, Exchange Rate Flexibility, and Credit Booms,” NBER Working Paper, 17670.
- Pirovano, Mara, Jacques Vanneste, and André Van Poeck (2011), “Portfolio and Short-term Capital Inflows to the New and Potential EU Countries: Patterns and Determinants” in W.Meeusen (ed.), The Economic Crisis and European Integration, Edward Elgar, 2011.
- Tong, Hui and Shang-Jin Wei (2010), “The Composition Matters: Capital Inflows and Liquidity Crunch During a Global Economic Crisis,” Review of Financial Studies, 24(6).
- World Economic Outlook (2011), Tensions from the Two-Speed Recovery: Unemployment, Commodities, and Capital Flows, World Economic and Financial Surveys, International Monetary Fund, Washington, April 2011.
- Folkerts-Landau, David (1997), International Capital Markets: Developments, Prospects, and Key Policy Issues, International Monetary Fund, Washington, November 1997.
- Global Financial Stability Report (2011), Grappling with Crisis Legacies, World Economic and Financial Surveys, International Monetary Fund, Washington, September 2011.

### Macro-Financial Linkages, Asset Prices, and Terms of Trade
- Cashin, Paul, Luis Céspedes, and Ratna Sahay (2004), “Commodity currencies and the real exchange rate,” Journal of Development Economics, 75, 239-68.
- Céspedes, Luis, Roberto Chang and Andrés Velasco (2004), "Balance Sheets and Exchange Rate Policy," American Economic Review, vol. 94(4), pages 1183-93, September.
- Fisher, Irving (1933), “The debt-deflation theory of the great depression,” Econometrica, 1, 337-57.
- Kamil, Herman (2012), “How Do Exchange Rate Regimes Affect Firms’Incentives to Hedge Currency Risk? Micro Evidence for Latin America,” mimeo, International Monetary Fund.
- Mendoza, Enrique (1995), “The Terms of trade, the real exchange rate, and economic fluctuations,” International Economic Review, 36 (1), 101-37.
- Mendoza, Enrique (2010), “Sudden stops, financial crises, and leverage,” American Economic Review, 100, 1941-66.
- Tong, Hui and Shang-Jin Wei (2010), “The Composition Matters: Capital Inflows and Liquidity Crunch During a Global Economic Crisis,” Review of Financial Studies, 24(6).

### Sectoral and Regional Analyses
- Hartelius (2010), “Analyzing nonperforming loans in Central and Eastern Europe based on historical experience in emerging markets,” Global Financial Stability Report, Washington, April 2010.
- Espinoza, Raphael, and Ananthakrishnan Prasad (2010), “Nonperforming Loans in the GCC Banking System and their Macroeconomic Effects,” IMF Working Paper, 224.
- Heytens, Paul and Cem Karacadag (2001), “An Attempt to Profile the Finances of China’s Enterprise Sector,” IMF Working Paper, 182.
- Pirovano, Mara, Jacques Vanneste, and André Van Poeck (2011), “Portfolio and Short-term Capital Inflows to the New and Potential EU Countries: Patterns and Determinants” in W.Meeusen (ed.), The Economic Crisis and European Integration, Edward Elgar, 2011.

*Source: _wp1271 - REFERENCES*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp1271.pdf_