## _wp1596

## Source details

**Canonical URL:** [_wp1596](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2015/_wp1596.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2015/_wp1596.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2015/_wp1596.pdf.json)

---

### I. Introduction & Motivation
- Over the last 10 years GDP growth has been disappointingly slow in Pacific island countries (PICs), barely keeping up with population growth.
- PICs considered: Fiji, Kiribati, Marshall Islands, Micronesia, Palau, Papua New Guinea, Samoa, Solomon Islands, Timor-Leste, Tonga, and Vanuatu (definition context).
- Key empirical observation: spreads between lending and deposit rates in the PIC region are among the highest in the world.
- Empirical sample limited to six Pacific island countries (PIC-6): Fiji, Papua New Guinea, Samoa, Solomon Islands, Tonga, and Vanuatu; panel years 2003-13 (net interest income not available for Papua New Guinea, reducing that measure to five countries).

### II. Banking System in the Pacific Islands — Stylized Facts
- Size and intermediation:
  - Total bank assets in PIC-6 vary from 50 percent of GDP in Solomon Islands to over 100 percent in Vanuatu.
  - At end-2013 commercial bank loans: 18 percent of GDP in Solomon Islands compared with 75 percent in Vanuatu.
  - Banks in PICs extend less credit on average relative to their small-state counterparts; outstanding bank deposits and loans indicate significant underbanking; mobile banking expected to help but PICs lag other small states.
- Presence and concentration:
  - Number of banks varies widely: Fiji and Palau each had seven banks (mid-2014); Kiribati and Tuvalu had one bank.
  - Banking sectors in PIC-6 exceed conventional threshold for “highly concentrated markets” (Herfindahl-Hirschman Index).
  - Large foreign banks dominate; ANZ and Westpac present in all PIC-6 and hold large market shares.
- Profitability and business model:
  - Banks in PIC-6 have relatively high profitability: returns on assets and on equity consistently exceed those in Caribbean small islands and compare favorably with regional banks.
  - In 2013, noninterest income of commercial banks in PIC-6 was on average around two-thirds of interest income; 40 percent of noninterest income derived from foreign exchange transaction fees and commissions.
- National provident funds:
  - Provident funds’ total assets average over one-fourth of GDP and over a third of banks’ total assets in PIC-6.
  - Cross-country variation: in Papua New Guinea assets of the National Superannuation Fund equivalent to 13 percent of banking assets; in Fiji the National Provident Fund’s assets almost 70 percent of banking assets.
  - Provident funds tend to hold assets in long-term government debt securities and bank deposits and can affect deposit rates and credit supply.

### III. Methodology
- Dependent variables:
  - Primary: difference between the weighted average lending rate and the weighted average deposit rate.
  - Alternative: ratio of net interest income to total loans (net interest income divided by loans).
- Sample and timing:
  - Panel sample: six Pacific island countries, years 2003-13 (net interest income not available for Papua New Guinea).
  - Explanatory variables largely lagged one year to mitigate endogeneity.
- Explanatory variable categories:
  - Macroeconomic: GDP, GDP per capita, real GDP growth, inflation (GDP measures in constant PPP U.S. dollars).
  - Banking sector: loans-to-GDP, loan loss provisions, nonperforming loans (NPLs), Herfindahl-Hirschman Index (HHI), capital-to-assets ratio, required reserves ratio (RRR).
  - Institutional: rank in contract enforcement and Index of Economic Freedom (correlated with country fixed effects).
- Estimation:
  - Panel fixed-effects regressions for spreads and net-interest-income-to-loans.
  - Additional specification: lending rate regressions including contemporaneous deposit rate.

### IV. Key Findings and Magnitudes
- Macroeconomic:
  - Log of GDP (PPP) and Log of GDP Per Capita (PPP) coefficients negative and statistically significant in preferred regressions — larger and more developed economies tend to have lower spreads.
  - Real GDP growth: not statistically significant.
  - Inflation: positive coefficient; quantitative effect described as small — “inflation would have to decline by one percentage point or more to reduce the spread by 10 basis points.”
- Banking sector:
  - Asset quality: loan loss provisions and NPLs have positive and significant coefficients — weaker credit quality associated with wider spreads.
  - Concentration: HHI positive and significant in most regressions — greater market concentration associated with larger spreads.
    - Illustration (economic interpretation): moving from monopoly to perfect competition implied spreads would fall by 7 percentage points in a baseline regression.
    - Adding banks example from HHI-derived slope: adding a third bank to two reduces spreads by 120 basis points; adding a fourth by a further 60 basis points; a fifth by 35 basis points; further additions have progressively smaller impacts.
  - Loans-to-GDP: not a robust determinant of spreads (significant in only one specification) but more significant in lending-rate regressions.
  - Capital-to-assets ratio and RRR: not statistically significant.
- Lending rate regressions:
  - Results consistent with spreads regressions: Log of GDP, Log of GDP Per Capita, loan loss provisions, and HHI significant.
  - Inflation coefficient larger and more robustly significant than in spread regressions.
  - Contemporaneous deposit rate coefficient statistically significant but considerably smaller than one.
- Institutions:
  - Country fixed-effects intercepts correlated with institutional measures:
    - Stronger creditor protection (contract enforcement) associated with lower spreads.
    - More economic freedom associated with lower spreads.
  - Communal land ownership and frequent government changes cited as constraints via collateral limits and increased uncertainty.

### V. Policy Implications
- Robust determinants to target:
  - Economic size (GDP) — larger economies associated with lower spreads.
  - Inflation — higher inflation associated with higher spreads (small quantitative effect).
  - Bank balance-sheet quality — higher loan loss provisions or NPLs associated with higher spreads.
  - Banking sector concentration — greater competition associated with lower spreads; nonlinear effects yield large marginal gains from initial increases in number/dispersion of banks where few institutions exist.
- Recommended policy actions:
  - Improve macroeconomic stability to reduce uncertainty.
  - Strengthen bank balance sheets via:
    - more aggressive debt collection,
    - writing off nonperforming assets,
    - eliminating connected and directed lending,
    - better regulation and supervision.
  - Increase competition by leveling the playing field and eliminating barriers to entry where they exist (recognizing natural limits in small economies).
  - Improve economic institutions: clear property rights (including land tenure), appropriate collateral regimes, compilation of credit history, strong contract enforcement.
- Implementation caveats:
  - High vulnerability to natural disasters contributes to macroeconomic instability that is difficult to fully address by policy alone.
  - Policy changes are challenging but can yield high benefits: lower lending rates would improve credit access (including for SMEs) and stimulate economic activity and inclusive growth.
- Further research:
  - Impact of national provident funds on banking sector variables, including interest rate spreads, as improved reporting becomes available.
  - Role of development banks and/or state-owned banks and their impact on the cost of financing.

### VI. Summary Statistics (Table II)
- Interest rate spreads: Obs. 77; Mean 8.46; Std. Dev. 2.41; Min. 5.12; Max. 14.92
- Weighted average lending rate: Obs. 77; Mean 10.99; Std. Dev. 2.24; Min. 5.84; Max. 16.96
- Weighted average deposit rate: Obs. 77; Mean 2.53; Std. Dev. 1.89; Min. 0.21; Max. 6.77
- Net interest income to loans: Obs. 60; Mean 8.96; Std. Dev. 3.14; Min. 4.67; Max. 20.02
- GDP (in constant PPP dollars): Obs. 78; Mean 21.32; Std. Dev. 1.04; Min. 20.15; Max. 23.71
- GDP per capita (in constant PPP dollars): Obs. 78; Mean 8.26; Std. Dev. 0.44; Min. 7.38; Max. 9.00
- Real GDP growth: Obs. 78; Mean 3.01; Std. Dev. 3.74; Min. -7.96; Max. 12.85
- Inflation: Obs. 78; Mean 5.46; Std. Dev. 3.67; Min. -0.21; Max. 17.32
- Loans to GDP: Obs. 73; Mean 36.41; Std. Dev. 17.09; Min. 7.72; Max. 74.98
- Loan loss provision: Obs. 72; Mean 4.24; Std. Dev. 2.70; Min. 1.10; Max. 15.13
- NPL: Obs. 66; Mean 6.69; Std. Dev. 4.50; Min. 0.50; Max. 19.50
- Banking concentration (HHI): Obs. 68; Mean 0.34; Std. Dev. 0.05; Min. 0.22; Max. 0.45
- Capital-to-assets ratio: Obs. 72; Mean 13.67; Std. Dev. 3.27; Min. 7.23; Max. 21.26
- Reserve requirement: Obs. 78; Mean 7.16; Std. Dev. 3.04; Min. 3.00; Max. 15.00

### VII. Selected Regression Coefficients (fixed effects, lagged explanatory variables)
- Impact on Spreads (Table IIIA, Dependent: Lending minus Deposit Rates)
  - Log of GDP (PPP): (1) -2.64*** (0.95); (2) -6.40*** (0.93)
  - Log of GDP Per Capita (PPP): (2) -3.15** (1.35)
  - Inflation: (1) 0.07* (0.04); (2) 0.08** (0.04); (3) 0.1** (0.04)
  - Loans to GDP: (1) 0.04* (0.02)
  - Loan Loss Provision: (1) 0.21*** (0.06); (2) 0.20*** (0.06); (3) 0.15** (0.06)
  - Banking Concentration (HHI): (1) 7.01** (3.10); (2) 6.77** (3.16); (3) 5.28* (3.29)
  - Observations (1)-(3): 57; No. of Countries 6; Fixed Effects Within R2 0.33, 0.31, 0.22
  - OLS Adjusted R2 (1)-(3): 0.87, 0.87, 0.85
- Impact on Net Interest Income to Loans (Table IIIA, columns (4)-(6))
  - Log of GDP Per Capita (PPP): (5) -8.99*** (1.34)
  - Loan Loss Provision: (4) 0.26*** (0.06); (5) 0.28*** (0.06)
  - Banking Concentration: (6) 15.50** (5.99)
  - Observations (4)-(6): 45; No. of Countries 5; Fixed Effects Within R2 0.67, 0.66, 0.27
  - OLS Adjusted R2 (4)-(6): 0.90, 0.89, 0.77
- Alternative specification (Table IIIB highlights)
  - Log of GDP (PPP): (2) -4.88*** (0.91)
  - Log of GDP Per Capita (PPP): (2) -6.72*** (1.30)
  - Inflation: (3) 0.09** (0.04)
  - NPL: (4)-(6) 0.22*** (0.04), 0.22*** (0.04), 0.27*** (0.05)
  - Banking Concentration: (1) 6.63* (3.54); (2) 6.37* (3.58)
  - Observations (1)-(3): 55; No. of Countries 6; Fixed Effects Within R2 0.22, 0.20, 0.18
  - OLS Adjusted R2 (1)-(3): 0.85, 0.85, 0.84
- Lending Rate Determinants (Table IV)
  - Deposit Rate: (1) 0.41** (0.16); (2) 0.48*** (0.16); (3) 0.31* (0.16); (4) 0.37** (0.16)
  - Log of GDP (PPP): (1) -3.82*** (0.90); (3) -3.33*** (0.96)
  - Log of GDP Per Capita (PPP): (1) -4.43*** (1.30); (3) -3.72*** (0.7)
  - Inflation: (1) 0.10*** (0.03); (2) 0.11*** (0.04); (3) 0.10*** (0.04); (4) 0.11*** (0.04)
  - Loans to GDP: (1) 0.06*** (0.02); (2) 0.04** (0.02); (3) 0.06** (0.02); (4) 0.04* (0.02)
  - Loan Loss Provision: (1) 0.12** (0.06); (2) 0.12** (0.06)
  - Banking Concentration (HHI): (1) 5.69** (2.76); (2) 5.43* (2.92)
  - Observations (1)-(2): 57; No. of Countries 6; Fixed Effects Within R2 0.61, 0.56
  - OLS Adjusted R2 (1)-(4): 0.87, 0.86, 0.86, 0.84

*Source: IMF Working Paper content (excerpt)._wp1596*

### References .............................................................................................................

### References

### I. Introduction & Motivation
- Over the last 10 years GDP growth has been disappointingly slow in Pacific island countries (PICs), barely keeping up with population growth.
- PICs considered: Fiji, Kiribati, Marshall Islands, Micronesia, Palau, Papua New Guinea, Samoa, Solomon Islands, Timor-Leste, Tonga, and Vanuatu (definition context).
- Key empirical observation: spreads between lending and deposit rates in the PIC region are among the highest in the world.
- Study focus: determinants of interest rate spreads in PICs using a sample limited to the region to ensure lessons apply to PICs; due to data limitations the empirical analysis focuses on six Pacific island countries (PIC-6): Fiji, Papua New Guinea, Samoa, Solomon Islands, Tonga, and Vanuatu.

### II. Banking System in the Pacific Islands — Stylized Facts
- Size and intermediation:
  - Total bank assets in PIC-6 vary from 50 percent of GDP in Solomon Islands to over 100 percent in Vanuatu (Figure 3).
  - At end-2013 commercial bank loans: 18 percent of GDP in Solomon Islands compared with 75 percent in Vanuatu.
  - Banks in PICs extend less credit on average relative to their small-state counterparts.
  - Outstanding bank deposits and loans indicate significant underbanking; mobile banking expected to help but PICs lag other small states.
- Presence and concentration:
  - Number of banks: Fiji and Palau each had seven banks (mid-2014); Kiribati and Tuvalu had one bank.
  - Market concentration measured by Herfindahl-Hirschman Index (HHI) indicates banking sectors in PIC-6 exceed conventional threshold for “highly concentrated markets.”
  - Large foreign banks dominate; ANZ and Westpac present in all PIC-6 and hold large market shares.
- Profitability and business model:
  - Banks in PIC-6 have relatively high profitability: returns on assets and on equity consistently exceed those in Caribbean small islands and compare favorably with regional banks (Figures 8 and 9).
  - In 2013, noninterest income of commercial banks in PIC-6 was on average around two-thirds of interest income; 40 percent of noninterest income derived from foreign exchange transaction fees and commissions.
- National provident funds:
  - National provident funds’ total assets average over one-fourth of GDP and over a third of banks’ total assets in PIC-6.
  - Cross-country variation: in Papua New Guinea assets of the National Superannuation Fund equivalent to 13 percent of banking assets; in Fiji the National Provident Fund’s assets almost 70 percent of banking assets.
  - Provident funds tend to hold assets in long-term government debt securities and bank deposits and can affect deposit rates and credit supply.

### III. Methodology
- Dependent variables:
  - Primary: difference between the weighted average lending rate and the weighted average deposit rate.
  - Alternative: ratio of net interest income to total loans (net interest income divided by loans).
- Sample and timing:
  - Panel sample: six Pacific island countries, years 2003-13 (net interest income not available for Papua New Guinea, reducing that measure to five countries).
  - Explanatory variables largely lagged one year to mitigate endogeneity.
- Categories of explanatory variables:
  - Macroeconomic: GDP, GDP per capita, real GDP growth, inflation (GDP measures in constant PPP U.S. dollars).
  - Banking sector: loans-to-GDP (size), loan loss provisions, nonperforming loans (NPLs), Herfindahl-Hirschman Index (HHI) for concentration, capital-to-assets ratio, required reserves ratio (RRR) (some used in robustness checks).
  - Institutional: rank in contract enforcement (World Bank Doing Business) and Index of Economic Freedom (Heritage Foundation) — not included in fixed-effects regressions due to low time variation; instead correlated with country fixed effects.
- Estimation approach:
  - Panel fixed-effects regressions for spreads and net-interest-income-to-loans.
  - Additional specification: lending rate as dependent variable with contemporaneous deposit rate as explanatory variable.

### IV. Results — Key Findings and Magnitudes
- Macroeconomic variables:
  - GDP level and GDP per capita: coefficients negative and statistically significant in preferred regressions — larger and more developed economies tend to have lower spreads.
  - Real GDP growth: not statistically significant (offsetting channels).
  - Inflation: positive coefficient, statistically significant for interest rate differential regressions; quantitative effect small — “inflation would have to decline by one percentage point or more to reduce the spread by 10 basis points.”
- Banking sector variables:
  - Asset quality: loan loss provisions and NPLs have positive and significant coefficients — weaker credit quality associated with wider spreads.
  - Concentration: HHI has positive and significant coefficient in almost all regressions — greater market concentration associated with larger spreads.
    - Economic interpretation: if a banking system moved from monopoly to perfect competition, baseline regression implies spreads would fall by 7 percentage points (illustrative extreme).
    - Illustration using equal-sized banks and HHI-derived slope: adding a third bank to two reduces spreads by 120 basis points; adding a fourth by a further 60 basis points; a fifth by 35 basis points; further additions have progressively smaller impacts.
  - Loans-to-GDP ratio: not a robust determinant of spreads (significant in only one specification) but shows a higher, more significant coefficient in lending-rate regressions, suggesting loan portfolio expansions may be demand-driven and raise lending rates.
  - Capital-to-assets ratio and required reserves ratio (RRR): not statistically significant in regressions; possible explanations include excess liquidity, low time variation, or offsetting effects.
- Lending rate regressions:
  - Results largely consistent with spreads regressions: total GDP and GDP per capita, loan loss provisions, and HHI significant.
  - Inflation coefficient larger and more robustly significant than in spread regressions (greater passthrough into lending than deposit rates).
  - Contemporaneous deposit rate coefficient: highly significant but considerably smaller than one — deposits important but not sole funding source for lending.
- Institutions:
  - Country-specific fixed-effects intercepts correlated with institutional measures:
    - Positive correlation between intercepts and average rankings in contract enforcement implies stronger creditor protection associated with lower spreads.
    - Negative correlation between intercepts and average Index of Economic Freedom implies more economic freedom associated with lower spreads.
  - Communal land ownership and frequent government changes noted as potential contributors to higher spreads via limits on collateral availability and increased political/economic uncertainty.

### V. Conclusions and Policy Implications
- Robust determinants identified:
  - Economic size (GDP) — larger economies associated with lower spreads.
  - Inflation — higher inflation associated with higher spreads (small quantitative effect).
  - Bank balance-sheet quality — higher loan loss provisions or NPLs associated with higher spreads.
  - Banking sector concentration — greater competition associated with lower spreads; nonlinear effects imply large marginal gains from initial increases in number/dispersion of banks where few institutions exist.
- Policy recommendations to reduce spreads:
  - Improving macroeconomic stability to reduce uncertainty.
  - Strengthening bank balance sheets via:
    - more aggressive debt collection,
    - writing off nonperforming assets,
    - eliminating connected and directed lending,
    - better regulation and supervision.
  - Increasing competition in the banking system by leveling the playing field and eliminating barriers to entry where they exist (noting natural limits to sustainable number of banks in small economies).
  - Improving economic institutions, particularly:
    - ensuring clear property rights (including land tenure),
    - appropriate collateral regimes,
    - compilation of credit history,
    - strong contract enforcement.
- Implementation caveats:
  - PICs’ high vulnerability to natural disasters contributes to macroeconomic instability that is difficult to fully address by policy alone.
  - Policy changes are challenging to implement but potentially yield high benefits: lower lending rates would improve access to credit (including for small and medium-sized enterprises) and stimulate economic activity and inclusive growth.
- Suggested further research:
  - Better understanding of the impact of national provident funds on banking sector variables, including interest rate spreads, as improved reporting and data become available.
  - Investigation of the role played by development banks and/or state-owned banks and their impact on the cost of financing.

*Source: IMF Working Paper content (excerpt)._wp1596 - References*

### References

### _wp1596 - References

### References
- Cabezon, Ezequiel, Leni Hunter, Patrizia Tumbarello, and Yiqun Wu, forthcoming, “Strengthening Macro-Fiscal Resilience to Natural Disasters and Climate Change in the Small States of the Pacific”
- Crowley, Joseph, 2007, “Interest Rate Spreads in English-Speaking African Countries,” IMF Working Paper 07/101
- Davies, Matt, John Vaught, and Ezequiel Cabezon, forthcoming, “Interest Rates, Bank Profitability, and Capital Adequacy in the South Pacific,” International Monetary Fund, manuscript.
- Demirguc-Kunt, Asli, and Harry Huizinga, 1999, “Determinants of Commercial Bank Interest Margins and Profitability: Some International Evidence,” The World Bank Economic Review, Vol. 13, No. 2, pp. 379-408.
- Fairbairn, Te’o I. J. and DeLisle Worrel, 1996, “South Pacific and Caribbean Island Economies: A Comparative Study”, Brisbane, Australia and Suva, Fiji: The Foundation for Development Cooporation and South Pacific Forum Secretariat.
- Greenwood, Jeremy, Juan M. Sanchez, and Cheng Wang, 2013, “Quantifying the Impact of Financial Development on Economic Development”, Review of Economic Dynamics, Elsevier for the Society for Economic Dynamics, Vol. 16, No. 1, pp.194-215.
- Heritage Foundation, 2014, “2014 Index of Economic Freedom”, http://www.heritage.org/index/
- International Monetary Fund, 2013, “Asia and Pacific Small States: Raising Potential Growth and Enhancing Resilience to Shocks”, (Washington: International Monetary Fund). Available via the internet at http://www.imf.org/external/np/pp/eng/2013/022013a.pdf.
- Jayaraman, T.K. and Rajesh Sharma, 2003, “Determinants of Interest Rate Spread in the Pacific Island Countries: Some Evidence from Fiji”, USPEC Working Paper Series, No. 2003/4 (Suva, Fiji: University of the South Pacific).
- Levine, Ross, 1997, “Financial Development and Economic Growth: Views and Agenda”, Journal of Economic Literature, Vol. 35, No. 2, pp. 688-726.
- Moore, Winston, and Roland Craigwell, 2002, “Market Power and Interest Rate Spreads in the Caribbean”, International Review of Applied Economics, Taylor & Francis Journals, Vol. 16, No. 4, pp.391-405.
- Mugizi, Francisco M. P., Ndanshau, Michael O.A., and Jehovaness Aikaeli, “The Determinants of Interest Rate Spreads in Developing Countries: Evidence on Tanzania, 1991-2009” (July 24, 2011). University of Dar es Salaam Department of Economics Working Paper 02/11.
- Randall, Ruby, 1998, “Interest Rate Spreads in the Eastern Caribbean”, IMF Working Paper 98/59, (Washington: International Monetary Fund).
- Rebei, Nooman, 2014; “Determinants of Interest Rate Spreads in Solomon Islands”; IMF Working Paper 14/105 (Washington: International Monetary Fund).
- Sheridan, Niamh, Patrizia Tumbarello, and Yiqun Wu, 2012, “Global and Regional Spillovers to Pacific Island Countries,” IMF Working Paper 12/154.
- Tennant, David and Abiodun Folawewo, 2009, “Macroeconomic and Market Determinants of Interest Rate Spreads in Low and Middle Income Countries”, Applied Financial Economics, Vol. 19, No. 6, pp.489-507.
- U.S. Department of Justice and the Federal Trade Commission, 2010, “Horizontal Merger Guidelines”. Available via the internet at http://www.justice.gov/atr/public/guidelines/hmg-2010.pdf.
- World Bank, 2013, Doing Business 2014: Understanding Regulations for Small and Medium-Size Enterprises, October 29 (Washington: World Bank).
- Yang, Yongzheng, Matt Davies, Shengzu Wang, Jonathan Dunn, and Yiqun Wu, 2011, “Monetary Policy Transmission Mechanisms in Pacific Island Countries,” IMF Working Paper 11/96, April (Washington: International Monetary Fund).

### Data Definitions and Sources (Table I)
- Interest rate spreads: Difference between weighted average rates on loans and weighted average rates on deposits. Data: IMF, International Financial Statistics, country authorities.
- Weighted average lending rate: Average interest rates on loans weighted by the volume of loans extended across different sectors. Data: IMF, International Financial Statistics, country authorities.
- Weighted average deposit rate: Average interest rates on deposits weighted by different types of deposits. Data: IMF, International Financial Statistics, country authorities.
- Net interest income to loans: Net interest income divided by average assets, net of depreciation and provisions. Data: Country authorities.
- GDP (in constant PPP dollars): Sum of gross value added by all resident producers in the economy plus any product taxes and minus any subsidies not included in the value of the products, in constant PPP dollars. Data: World Bank Databank.
- GDP per capita (in constant PPP dollars): Gross domestic product divided by midyear population in constant PPP dollars. Data: World Bank Databank.
- Real GDP growth: Annual change in real GDP. Data: IMF, World Economic Outlook Database.
- Inflation: Percentage change in CPI. Data: IMF, World Economic Outlook Database.
- Loans to GDP: Amount of bank loans as percentage of nominal GDP. Data: Country authorities; IMF, World Economic Database; Financial Access Survey.
- Loan loss provision: Banks’ reserves allocated against loan loss divided by gross loans. Data: Country authorities.
- NPL: Nonperforming loans as a percentage of total loans. Data: Country authorities.
- Banking concentration (HHI): The sum of the squared values of the market shares of all banks within a given country’s financial sector. Market share is calculated using a bank’s share of total loans or deposits. Data: Country authorities.
- Capital-to-assets ratio: Bank capital as a ratio of banks’ total assets. Data: Country authorities.
- Reserve requirement: The minimum percentage of customer deposits that banks are obligated by statute to hold as reserves. Data sources: Countries’ central bank websites; IMF country staff reports; Carmen Reinhart: www.carmenreinhart.com/user_uploads/Reservereq.xls, accessed on October 28, 2014.

### Summary Statistics (Table II)
- Interest rate spreads: Obs. 77; Mean 8.46; Std. Dev. 2.41; Min. 5.12; Max. 14.92
- Weighted average lending rate: Obs. 77; Mean 10.99; Std. Dev. 2.24; Min. 5.84; Max. 16.96
- Weighted average deposit rate: Obs. 77; Mean 2.53; Std. Dev. 1.89; Min. 0.21; Max. 6.77
- Net interest income to loans: Obs. 60; Mean 8.96; Std. Dev. 3.14; Min. 4.67; Max. 20.02
- GDP (in constant PPP dollars): Obs. 78; Mean 21.32; Std. Dev. 1.04; Min. 20.15; Max. 23.71
- GDP per capita (in constant PPP dollars): Obs. 78; Mean 8.26; Std. Dev. 0.44; Min. 7.38; Max. 9.00
- Real GDP growth: Obs. 78; Mean 3.01; Std. Dev. 3.74; Min. -7.96; Max. 12.85
- Inflation: Obs. 78; Mean 5.46; Std. Dev. 3.67; Min. -0.21; Max. 17.32
- Loans to GDP: Obs. 73; Mean 36.41; Std. Dev. 17.09; Min. 7.72; Max. 74.98
- Loan loss provision: Obs. 72; Mean 4.24; Std. Dev. 2.70; Min. 1.10; Max. 15.13
- NPL: Obs. 66; Mean 6.69; Std. Dev. 4.50; Min. 0.50; Max. 19.50
- Banking concentration (HHI): Obs. 68; Mean 0.34; Std. Dev. 0.05; Min. 0.22; Max. 0.45
- Capital-to-assets ratio: Obs. 72; Mean 13.67; Std. Dev. 3.27; Min. 7.23; Max. 21.26
- Reserve requirement: Obs. 78; Mean 7.16; Std. Dev. 3.04; Min. 3.00; Max. 15.00

### Regression Results — Impact on Spreads (Table IIIA)
Note: Regressions estimated using fixed effects model on country-level data across six countries for the 2001-2013 time period. All explanatory variables are one-period lagged. Standard errors in parentheses. Significance levels: * 10 percent; ** 5 percent; *** 1 percent.

Dependent Variable: Difference between Lending and Deposit Rates (columns (1)-(3)) and Net Interest Income to Loans (columns (4)-(6))
- Log of GDP (PPP):
  - (1) Coefficient -2.64*** (0.95)
  - (2) Coefficient -6.40*** (0.93)
- Log of GDP Per Capita (PPP):
  - (2) Coefficient -3.15** (1.35)
  - (5) Coefficient -8.99*** (1.34)
- Real GDP Growth:
  - (1) Coefficient -0.00 (0.04)
  - (2) Coefficient -0.08 (0.06)
- Inflation:
  - (1) Coefficient 0.07* (0.04)
  - (2) Coefficient 0.08** (0.04)
  - (3) Coefficient 0.1** (0.04)
  - (4) Coefficient 0.05 (0.05)
  - (5) Coefficient 0.05 (0.05)
  - (6) Coefficient 0.09 (0.08)
- Loans to GDP:
  - (1) Coefficient 0.04* (0.02)
  - (2) Coefficient 0.03 (0.02)
  - (3) Coefficient -0.01 (0.01)
- Loan Loss Provision:
  - (1) Coefficient 0.21*** (0.06)
  - (2) Coefficient 0.20*** (0.06)
  - (3) Coefficient 0.15** (0.06)
  - (4) Coefficient 0.26*** (0.06)
  - (5) Coefficient 0.28*** (0.06)
  - (6) Coefficient 0.21** (0.10)
- Banking Concentration:
  - (1) Coefficient 7.01** (3.10)
  - (2) Coefficient 6.77** (3.16)
  - (3) Coefficient 5.28* (3.29)
  - (4) Coefficient 6.73 (4.26)
  - (5) Coefficient 8.35* (4.24)
  - (6) Coefficient 15.50** (5.99)
- Observations:
  - (1)-(3) Observations 57; No. of Countries 6; Fixed Effects Within R2 0.33, 0.31, 0.22 respectively.
  - (4)-(6) Observations 45; No. of Countries 5; Fixed Effects Within R2 0.67, 0.66, 0.27 respectively.
- OLS Adjusted R2 reported:
  - Columns (1)-(3): 0.87, 0.87, 0.85
  - Columns (4)-(6): 0.90, 0.89, 0.77

### Regression Results — Impact on Spreads (Table IIIB)
Note: Regressions estimated using fixed effects model on country-level data across six countries for the 2001-13 time period. All explanatory variables are one-period lagged. Standard errors in parentheses. Significance levels: * 10 percent; ** 5 percent; *** 1 percent.

Dependent Variable: Difference between Lending and Deposit Rates (columns (1)-(3)) and Net Interest Income to Loans (columns (4)-(6))
- Log of GDP (PPP):
  - (1) Coefficient -1.74 (1.05)
  - (2) Coefficient -4.88*** (0.91)
- Log of GDP Per Capita (PPP):
  - (1) Coefficient -1.88 (1.46)
  - (2) Coefficient -6.72*** (1.30)
- Real GDP Growth:
  - (1) Coefficient 0.01 (0.04)
  - (2) Coefficient -0.07 (0.05)
- Inflation:
  - (1) Coefficient 0.06 (0.04)
  - (2) Coefficient 0.07* (0.04)
  - (3) Coefficient 0.09** (0.04)
  - (4) Coefficient 0.07 (0.05)
  - (5) Coefficient 0.08 (0.05)
  - (6) Coefficient 0.13** (0.06)
- Loans to GDP:
  - (1) Coefficient 0.03 (0.02)
  - (2) Coefficient 0.02 (0.02)
  - (3) Coefficient -0.00 (0.02)
- NPL:
  - (1) Coefficient 0.06 (0.04)
  - (2) Coefficient 0.07* (0.04)
  - (3) Coefficient 0.07* (0.04)
  - (4) Coefficient 0.22*** (0.04)
  - (5) Coefficient 0.22*** (0.04)
  - (6) Coefficient 0.27*** (0.05)
- Banking Concentration:
  - (1) Coefficient 6.63* (3.54)
  - (2) Coefficient 6.37* (3.58)
  - (3) Coefficient 5.32 (3.57)
  - (4) Coefficient 1.68 (4.22)
  - (5) Coefficient 2.79 (4.25)
  - (6) Coefficient 5.06 (5.48)
- Observations:
  - (1)-(3) Observations 55; No. of Countries 6; Fixed Effects Within R2 0.22, 0.20, 0.18 respectively.
  - (4)-(6) Observations 43; No. of Countries 5; Fixed Effects Within R2 0.72, 0.71, 0.51 respectively.
- OLS Adjusted R2 reported:
  - Columns (1)-(3): 0.85, 0.85, 0.84
  - Columns (4)-(6): 0.91, 0.91, 0.85

### Regression Results — Lending Rate Determinants (Table IV)
Note: Regressions are estimated using fixed effects model on the country-level data across six countries for the 2001-13 time period. All explanatory variables are one-period lagged. Standard errors in parentheses. Significance levels: * 10 percent; ** 5 percent; *** 1 percent.

Dependent Variable: Weighted Average Lending Rate (columns (1)-(4))
- Deposit Rate:
  - (1) Coefficient 0.41** (0.16)
  - (2) Coefficient 0.48*** (0.16)
  - (3) Coefficient 0.31* (0.16)
  - (4) Coefficient 0.37** (0.16)
- Log of GDP (PPP):
  - (1) Coefficient -3.82*** (0.90)
  - (3) Coefficient -3.33*** (0.96)
- Log of GDP Per Capita (PPP):
  - (1) Coefficient -4.43*** (1.30)
  - (3) Coefficient -3.72*** (0.7)
- Inflation:
  - (1) Coefficient 0.10*** (0.03)
  - (2) Coefficient 0.11*** (0.04)
  - (3) Coefficient 0.10*** (0.04)
  - (4) Coefficient 0.11*** (0.04)
- Loans to GDP:
  - (1) Coefficient 0.06*** (0.02)
  - (2) Coefficient 0.04** (0.02)
  - (3) Coefficient 0.06** (0.02)
  - (4) Coefficient 0.04* (0.02)
- Loan Loss Provision:
  - (1) Coefficient 0.12** (0.06)
  - (2) Coefficient 0.12** (0.06)
- NPL:
  - (3) Coefficient 0.05 (0.04)
  - (4) Coefficient 0.06 (0.04)
- Banking Concentration (HHI):
  - (1) Coefficient 5.69** (2.76)
  - (2) Coefficient 5.43* (2.92)
  - (3) Coefficient 4.81 (3.01)
  - (4) Coefficient 4.58 (3.14)
- Observations:
  - Columns (1)-(2) Observations 57; No. of Countries 6; Fixed Effects Within R2 0.61, 0.56 respectively.
  - Columns (3)-(4) Observations 55; No. of Countries 6; Fixed Effects Within R2 0.55, 0.51 respectively.
- OLS Adjusted R2:
  - (1) 0.87; (2) 0.86; (3) 0.86; (4) 0.84

*Source: _wp1596 - References*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2015/_wp1596.pdf_