Banque nationale de Belgique SA · EURONEXT:BNB
Banque nationale de Belgique SA — Technical Presentation (February 2018)
36 slides from Banque nationale de Belgique SA's technical presentation, February 2018
Slide 1: Some borrowers are more equal than others: Bank funding shocks and credit reallocation
Some borrowers are more equal than others: Bank funding shocks and credit reallocation O.De Jonghe H.Dewachter K.Mulier S.Ongena G.Schepens (National Bank of Belgium) (Ghent Uni., NBB) (Uni. of Zurich) (ECB) 5th Research Workshop MPT (Brussels, NBB) February 1-2, 2018 The views expressed in this pre
Slide 2: Funding shock and credit supply: literature
Funding shock and credit supply: literature
Slide 3: Funding shock and credit supply: literature
Funding shock and credit supply: literature I Absorbing the funding shock: domestic credit I Ivashina and Scharfstein (2010), Puri et al. (2011), Iyer et al. (2014)
Slide 4: Funding shock and credit supply: literature
Funding shock and credit supply: literature I Absorbing the funding shock: domestic credit I Ivashina and Scharfstein (2010), Puri et al. (2011), Iyer et al. (2014) I Cross border spillovers of funding shocks I Cetorelli and Goldberg (2011, 2012), Claessens and van Horen (2013), De Haas and van Lely
Slide 5: Funding shock and credit supply: literature
Funding shock and credit supply: literature I Absorbing the funding shock: domestic credit I Ivashina and Scharfstein (2010), Puri et al. (2011), Iyer et al. (2014) I Cross border spillovers of funding shocks I Cetorelli and Goldberg (2011, 2012), Claessens and van Horen (2013), De Haas and van Lely
Slide 6: Motivation: Are all borrowers equal?
Motivation: Are all borrowers equal? (1 + E [RK ]) = p ∗ (1 + RL ) + (1 − p) ∗ γ ∗ (1 + RL ) − c
Slide 7: Motivation: Are all borrowers equal?
Motivation: Are all borrowers equal? (1 + E [RK ]) = p ∗ (1 + RL ) + (1 − p) ∗ γ ∗ (1 + RL ) − c 1. Sector presence? (bank’s share in a sector) I Higher presence → market power → higher interest rates I Rent extraction (Klein (1971), Monti (1972))
Slide 8: Motivation: Are all borrowers equal?
Motivation: Are all borrowers equal? (1 + E [RK ]) = p ∗ (1 + RL ) + (1 − p) ∗ γ ∗ (1 + RL ) − c 1. Sector presence? (bank’s share in a sector) I Higher presence → market power → higher interest rates I Rent extraction (Klein (1971), Monti (1972)) 2. Sector specialization? (sector’s share in a bank)
Slide 9: Motivation: Are all borrowers equal?
Motivation: Are all borrowers equal? (1 + E [RK ]) = p ∗ (1 + RL ) + (1 − p) ∗ γ ∗ (1 + RL ) − c 1. Sector presence? (bank’s share in a sector) I Higher presence → market power → higher interest rates I Rent extraction (Klein (1971), Monti (1972)) 2. Sector specialization? (sector’s share in a bank)
Slide 10: The aggregate shock
The aggregate shock Funding 1000-4 lV 900+ 8005 i alors 600 + T T T T 2006m1 2008m1 2010m1 2012m1
Slide 11: The aggregate shock
The aggregate shock I But, huge cross-sectional variation across banks! I ∆% IBLb (2009:08 - 2008:08): -0.043 (mean), 0.11 (st.dev)
Slide 12: Data
Data I Bank-firm-level credit data: Central Corporate Credit Register I Bank data: Regulatory Bank Balance Sheets and Income Statements I Firm data: Central Balance Sheet Office ⇒ 1 year before and after the Lehman collapse in Belgium
Slide 13: Data
Data I Bank-firm-level credit data: Central Corporate Credit Register I Bank data: Regulatory Bank Balance Sheets and Income Statements I Firm data: Central Balance Sheet Office ⇒ 1 year before and after the Lehman collapse in Belgium CREDITbf = β1 Sector Presencebs ∗ ∆%Fundingb + β2 Sector Speciali
Slide 14: Data
Data I Bank-firm-level credit data: Central Corporate Credit Register I Bank data: Regulatory Bank Balance Sheets and Income Statements I Firm data: Central Balance Sheet Office ⇒ 1 year before and after the Lehman collapse in Belgium CREDITbf = β1 Sector Presencebs ∗ ∆%Fundingb + β2 Sector Speciali
Slide 15: Main findings
Main findings
Slide 16: Main findings
Main findings I Does bank funding affect credit supply? YES! I A 9.2 percent shock (sample average) leads to a 2.3 percentage points decrease in loan growth
Slide 17: Main findings
Main findings I Does bank funding affect credit supply? YES! I A 9.2 percent shock (sample average) leads to a 2.3 percentage points decrease in loan growth I Are all borrowers equal? NO! I Firms can partly offset credit rationing by 1. matching with a bank with larger sector presence (std. ⇑ reduce
Slide 18: Funding shocks and credit supply
Funding shocks and credit supply (1) (2) (3) Increase in Large decrease ∆% Creditbf creditbf in creditbf Panel A ∆% Fundingb 0.259*** 0.267* -0.355*** (0.0806) (0.135) (0.130) Bank controls YES YES YES S Firm FE YES YES YES Observations 47,205 47,205 47,205 R-squared 0.455 0.463 0.481 Panel B ∆% Fun
Slide 19: Funding shocks and credit supply
Funding shocks and credit supply (1) (2) (3) Increase in Large decrease ∆% Creditbf creditbf in creditbf Panel A ∆% Fundingb 0.259*** 0.267* -0.355*** (0.0806) (0.135) (0.130) Bank controls YES YES YES Firm FE YES YES YES Observations 47,205 47,205 47,205 R-squared 0.455 0.463 0.481 Panel B ∆% Fundi
Slide 20: Funding shocks and credit supply
Funding shocks and credit supply (1) (2) (3) Increase in Large decrease ∆% Creditbf creditbf in creditbf Panel A ∆% Fundingb 0.259*** 0.267* -0.355*** (0.0806) (0.135) (0.130) Bank controls YES YES YES Firm FE YES YES YES Observations 47,205 47,205 47,205 R-squared 0.455 0.463 0.481 Panel B ∆% Fundi
Slide 21: Funding shock impact and credit reallocation
Funding shock impact and credit reallocation CREDITbf = β1 Sector Presencebs ∗ ∆%Fundingb + β2 Sector Specializebs ∗ ∆%Fundingb + β3 Sector Presencebs + β4 Sector Specializebs + αf + υb + bf (1) (2) (3) (4) Increase Large decrease ∆% Creditbf ∆% Creditbf in creditbf in creditbf ∆% Fundingb 0.250***
Slide 22: Funding shock impact and credit reallocation
Funding shock impact and credit reallocation CREDITbf = β1 Sector Presencebs ∗ ∆%Fundingb + β2 Sector Specializebs ∗ ∆%Fundingb + β3 Sector Presencebs + β4 Sector Specializebs + αf + υb + bf (1) (2) (3) (4) Increase Large decrease ∆% Creditbf ∆% Creditbf in creditbf in creditbf ∆% Fundingb 0.250***
Slide 23: Further channels and implications
Further channels and implications I What role for firm risk, size and age? Further redistribution? I What about the real impact for firms?
Slide 24: Further channels and implications
Further channels and implications RISK REALLOCATION REAL EFFECTS (1) (2) (3) ∆ % Creditbf ∆% Fixed assetsf ∆% Assetsf ∆% Fundingb Sec presencebs * ∆% Fundingb -0.520*** (0.132) Sec specializationbs * ∆% Fundingb -0.252*** (0.081) Total assetsf * ∆% Fundingb 0.003 (0.021) Agef * ∆% Fundingb 0.001 (0.
Slide 25: Further channels and implications
Further channels and implications RISK REALLOCATION REAL EFFECTS (1) (2) (3) ∆ % Creditbf ∆% Fixed assetsf ∆% Assetsf ∆% Fundingb 0.730** 0.771* (0.311) (0.454) Sec presencebs * ∆% Fundingb -0.520*** -0.517* -0.549 (0.132) (0.264) (0.335) Sec specializationbs * ∆% Fundingb -0.252*** -0.004 0.170 (0.
Slide 26: Robustness
Robustness What about potential bias due to correlation of sector position with alternative explanations? I Bank fixed effects rule out bank specific events (e.g. bank recapilization). I Control for average loan maturity of a bank in a given sector (share ≥ 1y). I Control for geographical specializa
Slide 27: Robustness
Robustness What about potential bias due to correlation of sector position with alternative explanations? I Bank fixed effects rule out bank specific events (e.g. bank recapilization). I Control for average loan maturity of a bank in a given sector (share ≥ 1y). I Control for geographical specializa
Slide 28: Robustness
Robustness What about potential bias due to correlation of sector position with alternative explanations? I Bank fixed effects rule out bank specific events (e.g. bank recapilization). I Control for average loan maturity of a bank in a given sector (share ≥ 1y). I Control for geographical specializa
Slide 29: Robustness
Robustness What about potential bias due to correlation of sector position with alternative explanations? I Bank fixed effects rule out bank specific events (e.g. bank recapilization). I Control for average loan maturity of a bank in a given sector (share ≥ 1y). I Control for geographical specializa
Slide 30: Conclusions
Conclusions I Does bank funding affect credit supply? YES! I Are all borrowers equal? NO!
Slide 31: Conclusions
Conclusions I Does bank funding affect credit supply? YES! I Are all borrowers equal? NO!
Slide 32: Conclusions
Conclusions I Does bank funding affect credit supply? YES! I Are all borrowers equal? NO!
Slide 33: Policy Implications
Policy Implications I Bank market power and credit supply I Higher cost of credit/lower credit volumes m Stability of access to credit in times of crisis I Focus on geographical dimension m Focus on sectoral dimension I Lending concentration and credit supply (Basel Committee, 2006) I Portfolio conc
Slide 34: Thank you for your attention
Thank you for your attention
Slide 35: Rent extraction
Rent extraction Table: Sector presence, sector specialization and rent seeking (1) (2) (3) VARIABLES Debt burdenf Debt burdenf Debt burdenf Sector presencebs 0.0444** 0.0408* 0.0347* (0.0218) (0.0235) (0.0188) Sector specializationbs 0.00447 -0.000661 0.0126 (0.0296) (0.0253) (0.0280) Observations 8
Slide 36: Identifying credit supply
Identifying credit supply Empirical setup . With funding outflow = BankA &B A Creditia;—) BANKA 49% of ta Controls: -deposits + | |-liquidity FIRM 1 -profitability -provisions . With funding outflow = -capital A Credity, BANKB 9% of Ta => We isolate credit supply (from credit demand) by investigatin
Banque Nationale De Belgique — Technical Presentation (February 2018) on 6ix