## 7. Interaction Between Monetary Policy and Lending Standards ## Source details **Canonical URL:** [7. Interaction Between Monetary Policy and Lending Standards](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2016/_wp1604.pdf) ## Other formats - [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2016/_wp1604.pdf.md) - [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2016/_wp1604.pdf.json) --- ### Introduction and objectives - Focus: use the Eurosystem Bank Lending Survey (BLS) to estimate empirically the likely effectiveness of selected macro-prudential policies, their transmission channels, and interactions with monetary policy, with specific focus on the real estate / mortgage market. - Data scope: - BLS quarterly panel between 2003q1 and 2010q4 covering 13 euro area countries: Austria, Belgium, Finland, France, Germany, Greece, Ireland, Italy, Luxembourg, Netherlands, Portugal, Slovenia, Spain. - The BLS sample comprises around 140 banks. - Key empirical questions: - How do policy-driven changes in the cost of bank capital or in bank liquidity positions transmit to credit supply? - Are transmissions to credit supply mainly through non-price factors (LTVs, collateral, maturity) or price factors (margins, fees)? - Can limits on LTVs slow house price appreciation and mortgage loan growth, and how do they interact with monetary policy? ### Data and empirical framework - Main dataset and measures: - Eurosystem Bank Lending Survey (BLS) monitoring corporate lending, loans for house purchases, and consumer lending; reports net changes in lending conditions and contributions of factors (costs of funds and balance sheet constraints; competition; risk perceptions; demand). - Observable BLS measures used: - Six lending-standard components: margins on risky loans, margins on average loans, maturity, loan-to-value ratios, non-interest fees, collateral requirements. - Contributions to lending-standard changes: cost of capital, liquidity position, access to wholesale funding (used to map to macro-prudential instruments). - Sample limitations and identification challenges: - BLS reports perceived drivers but not the exact nature of shocks; possible omitted variable bias, reverse causality, and measurement error. - Mortgage-specific questionnaire lacks the same breakdown of balance-sheet contributions as the enterprise questionnaire; enterprise-questionnaire subcomponents are used as proxies for mortgage loans (measurement error acknowledged). - Structural relationships estimated (schematic): - Equation A: determinants of changes in lending standards (price and non-price components) as functions of bank balance-sheet characteristics, financial-market conditions, expectations, and EONIA, with country fixed effects and time dummies. - Equation B: mortgage credit growth as function of lending standards (price and non-price) and demand factors. - Equation C / reduced forms: combine A and B to link determinants of lending standards to mortgage loan growth and to house price appreciation. ### Methodology - Estimators and specifications: - Dynamic panel GMM (Arellano and Bond, difference GMM with lags t-2 to t-4 as instruments); specification tests reported include AR(1), AR(2), and Sargan. - Instrumental variables (2SLS): instruments are growth rates of claims of US banks and UK banks on banks of each euro area country (from BIS Consolidated Banking Statistics) and their lags up to 6 quarters. - Panel VAR (Love and Zicchino, 2006) with Helmert transformation and system GMM to study dynamics and impulse responses; panel VAR uses 4 lags. - Controls in regressions: - housing-market prospects (BLS), demand for mortgage loans (BLS), household debt/GDP, house price appreciation (when relevant), real GDP growth, yield-curve measure, inflation rate, EONIA, country fixed effects, quarterly dummies. - Endogeneity treatment: - Lending standards and balance-sheet contributions treated as endogenous and instrumented by lags in GMM. - IV strategy exploits cross-border interbank flow shocks from US/UK banks as exogenous to domestic mortgage outcomes conditional on controls. ### Main empirical findings — effectiveness and channels - High-level findings: - Measures that increase the cost of bank capital are effective in slowing mortgage credit growth and house price appreciation. - Changes in LTV also impact credit growth and house price appreciation but their impact tends to be more moderate than capital-cost measures. - Macro-prudential policies affecting the cost of capital are transmitted mainly through price margins, with very little impact on LTV ratios or other non-price mortgage characteristics. - Tightening of LTVs is more effective in slowing credit growth and house price appreciation when monetary policy is “too loose”. - Quantitative magnitudes and illustrative thought experiments: - To reduce credit growth by 10 percentage points (holding other macro variables constant), the estimated required increase in the contribution of the cost of capital to lending standards is a 70 percent contribution. - The authors state this tightening is about the same order of magnitude as the actual average tightening of euro area lending standards due to cost of capital during the year following the start of the global financial crisis (2008Q4 to 2009 Q4). - A similar tightening in the cost of capital would result in a 5 percentage points reduction in house price appreciation (per the authors’ calculation). - Decomposition of transmission: - Shocks to balance-sheet factors (cost of funds, liquidity position, access to market financing) affect multiple lending-standard dimensions, but the largest and most statistically significant transmission is to price margins (margins on average loans and margins on risky loans). - Pass-through from balance-sheet shocks to LTVs and other non-price dimensions is relatively weak in the estimated models. - Lending standards and credit volumes: - Net tightening of margins on average loans and on risky loans is significantly and negatively associated with mortgage credit growth after two to four quarters. - A 100 percent net tightening of margins on risky loans is estimated to reduce credit growth by about 7 percentage points after four quarters. - Tightening of loan-to-value ratios or mortgage fees is also negatively associated with mortgage loan growth after four quarters. - LTV and interaction with monetary policy: - Interaction regressions use a monetary-policy “gap” measure (implied policy rate from a Taylor rule minus EONIA). - Results imply LTV limits are more effective in containing credit growth and house price appreciation when monetary policy is loose; conversely, a loose monetary policy amplifies the effect of relaxed LTVs on credit growth and house prices. - Panel VAR evidence: - Impulse-response analysis indicates: - Monetary policy shocks have delayed but large direct impacts on house price appreciation and mortgage loan growth (effect delayed to the third quarter). - Monetary policy also has indirect effects on mortgage credit and house prices through impacts on macroprudential / non-price lending standards (LTVs and capital-cost measures). - Direct impact of LTV limits on house price appreciation and mortgage loan growth is significant but relatively short-lived. - Direct impact of changes in banks’ capital position is significant, immediate and persistent (first and third quarter). - Higher mortgage loan growth drives house prices higher, but not vice versa (causality direction per VAR responses). ### Robustness, specification tests and limitations - Specification and instrument tests: - Difference-GMM AR(1) and AR(2) tests, Sargan tests, Hansen J tests, and Anderson-Rubin Wald tests are reported across regressions; in many cases tests do not reject instrument validity, though some exceptions are noted (e.g., IV2 estimation for lending standards due to liquidity position). - Measurement and identification caveats: - BLS does not identify whether changes in balance-sheet contributions are due to policy, endogenous assessments, or other shocks — measurement error and endogeneity remain concerns. - Using enterprise-questionnaire subcomponents as proxies for mortgage lending’s balance-sheet contributions introduces measurement error. - Macro-prudential policy actions themselves are endogenous to the cycle (e.g., countercyclical buffers rise during credit booms), complicating causal interpretation. - The sample ends in 2010q4 and thus stops before the euro area systemic phase of the crisis in 2011; findings are based on cross-country heterogeneity 2003q1–2010q4. ### Policy implications and recommendations - Monetary policy and macro-prudential capital instruments: - Policies related to bank capital (capital buffers, sectoral/time-contingent risk weights, dynamic provisioning, restrictions on profit distribution) affect the cost of loans and operate through similar banking channels as monetary policy; they are likely to reinforce each other. - Because capital-related measures transmit mainly via price margins, they can be effective in slowing mortgage credit growth and dampening house price appreciation. - Complementarity of instruments: - Macro-prudential instruments affecting cost of capital or liquidity position could usefully be complemented by instruments targeting non-price mortgage dimensions (e.g., limits on LTVs) because banks’ endogenous response to liability shocks tends to be via margins rather than LTVs. - Limits on LTVs are particularly valuable when monetary policy is too loose. - Implementation and data: - Enhancing direct data on macroprudential policy actions would help identification in future research; until then, BLS-based approaches can be extended to other regions or to broader financial-cycle analysis. *Italic source: IMF Working Paper — section "7. Interaction Between Monetary Policy and Lending Standards" (excerpt).* ### 1. The Eurosystem Bank Lending Survey ______________________________________________ 9 ### 1. The Eurosystem Bank Lending Survey ### Major Sections (as listed) - 1. The Eurosystem Bank Lending Survey ______________________________________________ 9 - 2. Cumulative Net Tightening of LTVs since 2003/1 _________________________________ 10 - 3. Evolution of Price and non-Price Lending Standard, By Country __________________ 11 - 4. Impulse-Responses of the Panel VAR Analysis ____________________________________ 28 ### Tables (as listed) - 1. Pairwise Correlation Among Components of Lending Standards and Credit Demand _____________________________________________________________________________ 12 - 2. Impact of Balance Sheet Contributions of Lending on Mortgage Loan Growth and on House Price Appreciation ________________________________________________________ 18 - 3. Impact of Balance Sheet Factors on Price and non-Price Components of Lending Standards for Mortgage _____________________________________________________________ 20 - 4. Impact of Balance Sheet Factors on Price and non-Price Components of Lending Standards for Mortgage _____________________________________________________________ 21 - 5. Lending Standards and Mortgage Loan Growth __________________________________ 23 - 6. Lending Standards and House Price Appreciation __________________________________ 24 *Source: _wp1604 - 1. The Eurosystem Bank Lending Survey (PDF chapter/section).* ### 7. Interaction Between Monetary Policy and Lending Standards ____________________ 25 ### 7. Interaction Between Monetary Policy and Lending Standards ### Introduction and objectives - Focus: use the Euro-system Bank Lending Survey (BLS) to estimate empirically the likely effectiveness of selected macro-prudential policies, their transmission channels, and interactions with monetary policy, with specific focus on the real estate / mortgage market. - Data scope: BLS quarterly panel between 2003q1 and 2010q4 covering 13 euro area countries (Austria, Belgium, Finland, France, Germany, Greece, Ireland, Italy, Luxembourg, Netherlands, Portugal, Slovenia, Spain); the BLS sample comprises around 140 banks. - Key empirical questions: - How do policy-driven changes in the cost of bank capital or in bank liquidity positions transmit to credit supply? - Are transmissions to credit supply mainly through non-price factors (LTVs, collateral, maturity) or price factors (margins, fees)? - Can limits on LTVs slow house price appreciation and mortgage loan growth, and how do they interact with monetary policy? ### Data and empirical framework - Main dataset: Eurosystem Bank Lending Survey (BLS) monitoring corporate lending, loans for house purchases, and consumer lending; reports net changes in lending conditions and contributions of factors (costs of funds and balance sheet constraints; competition; risk perceptions; demand). - Observable BLS measures used: - Six lending-standard components (margins on risky loans, margins on average loans, maturity, loan-to-value ratios, non-interest fees, collateral requirements). - Contributions to lending-standard changes: cost of capital, liquidity position, access to wholesale funding (used to map to macro-prudential instruments). - Sample limitations and identification challenges: - BLS reports perceived drivers but not the exact nature of shocks; possible omitted variable bias, reverse causality, and measurement error. - Mortgage-specific questionnaire lacks the same breakdown of balance-sheet contributions as the enterprise questionnaire; authors use enterprise-questionnaire subcomponents as proxies for mortgage loans (acknowledged measurement error). - Structural relationships estimated (schematic): - Equation A: determinants of changes in lending standards (price and non-price components) as functions of bank balance-sheet characteristics, financial-market conditions, expectations, and EONIA, with country fixed effects and time dummies. - Equation B: mortgage credit growth as function of lending standards (price and non-price) and demand factors. - Equation C / reduced forms: combine A and B to link determinants of lending standards to mortgage loan growth and to house price appreciation. ### Methodology - Estimators: - Dynamic panel GMM (Arellano and Bond, difference GMM with lags t-2 to t-4 as instruments); tests reported include AR(1), AR(2), and Sargan. - Instrumental variables (2SLS): instruments are growth rates of claims of US banks and UK banks on banks of each euro area country (from BIS Consolidated Banking Statistics) and their lags up to 6 quarters. - Panel VAR (Love and Zicchino, 2006) with Helmert transformation and system GMM to study dynamics and impulse responses. - Controls in regressions: housing-market prospects (BLS), demand for mortgage loans (BLS), household debt/GDP, house price appreciation (when relevant), real GDP growth, yield-curve measure, inflation rate, EONIA, country fixed effects, quarterly dummies. - Endogeneity treatment: - Lending standards and balance-sheet contributions treated as endogenous and instrumented by lags in GMM. - IV strategy exploits cross-border interbank flow shocks from US/UK banks as exogenous to domestic mortgage outcomes conditional on controls. ### Main empirical findings — effectiveness and channels - High-level findings: - Measures that increase the cost of bank capital are effective in slowing mortgage credit growth and house price appreciation. - Changes in LTV also impact credit growth and house price appreciation but their impact tends to be more moderate than capital-cost measures. - Macro-prudential policies affecting the cost of capital are transmitted mainly through price margins, with very little impact on LTV ratios or other non-price mortgage characteristics. - Tightening of LTVs is more effective in slowing credit growth and house price appreciation when monetary policy is “too loose”. - Quantitative magnitudes and illustrative thought experiments: - A thought experiment: to reduce credit growth by 10 percentage points (holding other macro variables constant), the estimated required increase in the contribution of the cost of capital to lending standards is a 70 percent contribution. The authors state this tightening is about the same order of magnitude as the actual average tightening of euro area lending standards due to cost of capital during the year following the start of the global financial crisis (2008Q4 to 2009 Q4). - A similar tightening in the cost of capital would result in a 5 percentage points reduction in house price appreciation (per the authors’ calculation). - Decomposition of transmission: - Shocks to balance-sheet factors (cost of funds, liquidity position, access to market financing) affect multiple lending-standard dimensions, but the largest and most statistically significant transmission is to price margins (margins on average loans and margins on risky loans). - Pass-through from balance-sheet shocks to LTVs and other non-price dimensions is relatively weak in the estimated models. - Lending standards and credit volumes: - Net tightening of margins on average loans and on risky loans is significantly and negatively associated with mortgage credit growth after two to four quarters. - A 100 percent net tightening of margins on risky loans is estimated to reduce credit growth by about 7 percentage points after four quarters (authors’ interpretation of coefficients). - Tightening of loan-to-value ratios or mortgage fees is also negatively associated with mortgage loan growth after four quarters. - LTV and interaction with monetary policy: - Interaction regressions (Table 7) use a monetary-policy “gap” measure (implied policy rate from a Taylor rule minus EONIA). - Results imply LTV limits are more effective in containing credit growth and house price appreciation when monetary policy is loose; conversely, a loose monetary policy amplifies the effect of relaxed LTVs on credit growth and house prices. - Panel VAR evidence: - Impulse-response analysis (panel VAR with 4 lags) indicates: - Monetary policy shocks have delayed but large direct impacts on house price appreciation and mortgage loan growth (effect delayed to the third quarter). - Monetary policy also has indirect effects on mortgage credit and house prices through impacts on macroprudential / non-price lending standards (LTVs and capital-cost measures). - Direct impact of LTV limits on house price appreciation and mortgage loan growth is significant but relatively short-lived. - Direct impact of changes in banks’ capital position is significant, immediate and persistent (first and third quarter). - Higher mortgage loan growth drives house prices higher, but not vice versa (causality direction per VAR responses). ### Robustness, specification tests and limitations - Specification and instrument tests: - Difference-GMM AR(1) and AR(2) tests, Sargan tests, Hansen J tests, and Anderson-Rubin Wald tests are reported across regressions; in many cases tests do not reject instrument validity, though some exceptions are noted (e.g., IV2 estimation for lending standards due to liquidity position). - Measurement and identification caveats (authors’ explicit cautions): - BLS does not identify whether changes in balance-sheet contributions are due to policy, endogenous assessments, or other shocks — measurement error and endogeneity remain concerns. - Using enterprise-questionnaire subcomponents as proxies for mortgage lending’s balance-sheet contributions introduces measurement error. - Macro-prudential policy actions themselves are endogenous to the cycle (e.g., countercyclical buffers rise during credit booms), complicating causal interpretation. - The sample ends in 2010q4 and thus stops before the euro area systemic phase of the crisis in 2011; findings are based on cross-country heterogeneity 2003q1–2010q4. ### Policy implications and recommendations - Monetary policy and macro-prudential capital instruments: - Policies related to bank capital (capital buffers, sectoral/time-contingent risk weights, dynamic provisioning, restrictions on profit distribution) affect the cost of loans and operate through similar banking channels as monetary policy; they are likely to reinforce each other. - Because capital-related measures transmit mainly via price margins, they can be effective in slowing mortgage credit growth and dampening house price appreciation. - Complementarity of instruments: - Macro-prudential instruments affecting cost of capital or liquidity position could usefully be complemented by instruments targeting non-price mortgage dimensions (e.g., limits on LTVs) because banks’ endogenous response to liability shocks tends to be via margins rather than LTVs. - Limits on LTVs are particularly valuable when monetary policy is too loose (they are more effective in such states). - Implementation and data: - Enhancing direct data on macroprudential policy actions would help identification in future research; until then, BLS-based approaches can be extended to other regions or to broader financial-cycle analysis. *Italic source:* IMF Working Paper — section "7. Interaction Between Monetary Policy and Lending Standards" (excerpt). --- _Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2016/_wp1604.pdf_