The Supply Side of Household Finance (original) (raw)

2019, Review of Financial Studies

Abstract

BIS Working Papers are written by members of the Monetary and Economic Department of the Bank for International Settlements, and from time to time by other economists, and are published by the Bank. The papers are on subjects of topical interest and are technical in character. The views expressed in them are those of their authors and not necessarily the views of the BIS. This publication is available on the BIS website (www.bis.org).

Figures (16)

Figure 1. Correlation between the ARM share and alternative measures of the “FRM risk premium” Note: The figure shows the correlation between alternative measures of the FRM risk premium and the ARM share. The blue bars are correlations computed on aggregate data; black bars using data at the bank-client level. The FRM risk premium is given by the difference between the FRM rate and the expected value of the interbank rate. This is calculated under various assumptions about the horizon: a forward-looking horizon of 1 year (F1), the actual value (0), a backward-looking horizon of 1, 2, 3, 4, and 5 years (L1 to L5) and an infinite horizon (00) approximated using the whole sample. The correlation at 0 is the correlation with the current FRM/AMR. Correlations are calculated over the period January 2004 through Dec. 2010.

Figure 1. Correlation between the ARM share and alternative measures of the “FRM risk premium” Note: The figure shows the correlation between alternative measures of the FRM risk premium and the ARM share. The blue bars are correlations computed on aggregate data; black bars using data at the bank-client level. The FRM risk premium is given by the difference between the FRM rate and the expected value of the interbank rate. This is calculated under various assumptions about the horizon: a forward-looking horizon of 1 year (F1), the actual value (0), a backward-looking horizon of 1, 2, 3, 4, and 5 years (L1 to L5) and an infinite horizon (00) approximated using the whole sample. The correlation at 0 is the correlation with the current FRM/AMR. Correlations are calculated over the period January 2004 through Dec. 2010.

Figure 2. Aggregate share of ARM and alternative “Long term financial premium” measure: Note: The red solid line shows the Adjustable Rate Mortgage (ARM) share in Italy (values on the left axis). The blue dashed line is the spread between the FRM and the ARM interest rates (values on the right axis); the dashed green line shows the FRM risk premium computed as the difference between the FRM rate and the one year moving average of the one month interbank rate (a proxy for the expected value of the ARM rate). Data are monthly from January 2004 to December 2010.

Figure 2. Aggregate share of ARM and alternative “Long term financial premium” measure: Note: The red solid line shows the Adjustable Rate Mortgage (ARM) share in Italy (values on the left axis). The blue dashed line is the spread between the FRM and the ARM interest rates (values on the right axis); the dashed green line shows the FRM risk premium computed as the difference between the FRM rate and the one year moving average of the one month interbank rate (a proxy for the expected value of the ARM rate). Data are monthly from January 2004 to December 2010.

Note: The figure shows the cross sectional distribution of the number of quarters of price inaction (spread unchanged between two quarters). In principle the distribution ranges between 0 (no inaction) and 28 (no spread adjustment over the samples). Inaction at bank b is an indicator set to | if the absolute value of the quarterly change of the spread (Cree = rfk™)) is lower that 1/3 of the standard deviation of the change (i.e. the change falls between + Sto <> ) where the standard deviation is specific to each bank.

Note: The figure shows the cross sectional distribution of the number of quarters of price inaction (spread unchanged between two quarters). In principle the distribution ranges between 0 (no inaction) and 28 (no spread adjustment over the samples). Inaction at bank b is an indicator set to | if the absolute value of the quarterly change of the spread (Cree = rfk™)) is lower that 1/3 of the standard deviation of the change (i.e. the change falls between + Sto <> ) where the standard deviation is specific to each bank.

Note: Figure shows the Kaplan and Meier's survival probability of keeping the price invariant, for the baseline measure of price inaction (blue line) and when using the tighter threshold (red line).. The Kaplan— Meier estimator is the nonparametric maximum likelihood estimate of the probability that a bank does not change the FRM/ARM spread in a quarter exceeding t.

Note: Figure shows the Kaplan and Meier's survival probability of keeping the price invariant, for the baseline measure of price inaction (blue line) and when using the tighter threshold (red line).. The Kaplan— Meier estimator is the nonparametric maximum likelihood estimate of the probability that a bank does not change the FRM/ARM spread in a quarter exceeding t.

Note: The figure shows the distribution of the bank fixed effects obtained from the regression in Table 3 column (V). Banks in the bottom decile of the distribution (13 banks, 9.6% of the market) are defined as specialized in ARM mortgages; banks in the top decile of the distribution (13 banks accounting for 14.1% of the market) are defined as specialized in FRM. Figure 7. Pattern of bank specialization in the mortgage market

Note: The figure shows the distribution of the bank fixed effects obtained from the regression in Table 3 column (V). Banks in the bottom decile of the distribution (13 banks, 9.6% of the market) are defined as specialized in ARM mortgages; banks in the top decile of the distribution (13 banks accounting for 14.1% of the market) are defined as specialized in FRM. Figure 7. Pattern of bank specialization in the mortgage market

Notes. (1) Difference between the FRM rate and the ARM rate. (2) Difference between the FRM rate and expectation of the ARM rate. The latter is based on the one year moving average of the one month interbank rate. (3) Average across individuals in the case of joint mortgages. (4) In case of joint mortgage. (6) Deposits over total liabilities. (6) Dummy that takes the value of | if the bank is active in the securitization market in a given quarter. (7) Tier! capital over total assets. (8) Bad loans over total loans. (9) Dummy that takes the value of 1 if the bank takes part to the “Patti Chiari” initiative, whose main objective is to simplify bank-borrower relationship. (10) We control for the distance between the lending bank headquarters and household residence by four dummy variables: DIST1 is equal to | if borrower k has his residence in the same province where bank j has its headquarters; DIST2 is equal to 1 if: a) DIST1=0 and b) firm k is resident in the same region where bank j has its headquarters; DIST3 is equal to | if: a) DIST2=0 and b) borrower k is resident in the same geographical area where bank j has its headquarters; DIST4 is equal to 1 if DIST3=0. (11) Market share of the first 5 banking groups in each province. Not reported Dummy banks, dummy provinces. (12) At the regional level; in thousands euros. Table 1. Descriptive statistics of the main variables used in the estimation

Notes. (1) Difference between the FRM rate and the ARM rate. (2) Difference between the FRM rate and expectation of the ARM rate. The latter is based on the one year moving average of the one month interbank rate. (3) Average across individuals in the case of joint mortgages. (4) In case of joint mortgage. (6) Deposits over total liabilities. (6) Dummy that takes the value of | if the bank is active in the securitization market in a given quarter. (7) Tier! capital over total assets. (8) Bad loans over total loans. (9) Dummy that takes the value of 1 if the bank takes part to the “Patti Chiari” initiative, whose main objective is to simplify bank-borrower relationship. (10) We control for the distance between the lending bank headquarters and household residence by four dummy variables: DIST1 is equal to | if borrower k has his residence in the same province where bank j has its headquarters; DIST2 is equal to 1 if: a) DIST1=0 and b) firm k is resident in the same region where bank j has its headquarters; DIST3 is equal to | if: a) DIST2=0 and b) borrower k is resident in the same geographical area where bank j has its headquarters; DIST4 is equal to 1 if DIST3=0. (11) Market share of the first 5 banking groups in each province. Not reported Dummy banks, dummy provinces. (12) At the regional level; in thousands euros. Table 1. Descriptive statistics of the main variables used in the estimation

Table 2. Do lender characteristics affect mortgage choice? Notes: The table shows the parameter estimates of a linear probability model of mortgage type choice;. The left hand side variable is a dummy =1 if the borrower chooses a FRM, zero otherwise. Robust standard errors clustered at bank level are reported in brackets. *, **, and *** represent significance levels of 10%, 5%, and 1% respectively. Coefficients for dummies and fixed effects are not reported. (1) In columns II-V the LTFP is the difference between the FRM rate and the expected ARM rate based on borrower's actual ARM rate and one year moving average of the one month interbank rate (2) In column 5 the LTFP is the difference between the FRM rate and current the ARM rate. (3) Include: i) GDP per capita at the regional level; ii) a Bersani Law dummy= 1 from the second quarter of 2007 onwards; iii) a dummy if the bank participates in the “Patti Chiari” initiative; iv) dummies to control for the distance between the lending bank headquarters and household residence.

Table 2. Do lender characteristics affect mortgage choice? Notes: The table shows the parameter estimates of a linear probability model of mortgage type choice;. The left hand side variable is a dummy =1 if the borrower chooses a FRM, zero otherwise. Robust standard errors clustered at bank level are reported in brackets. *, **, and *** represent significance levels of 10%, 5%, and 1% respectively. Coefficients for dummies and fixed effects are not reported. (1) In columns II-V the LTFP is the difference between the FRM rate and the expected ARM rate based on borrower's actual ARM rate and one year moving average of the one month interbank rate (2) In column 5 the LTFP is the difference between the FRM rate and current the ARM rate. (3) Include: i) GDP per capita at the regional level; ii) a Bersani Law dummy= 1 from the second quarter of 2007 onwards; iii) a dummy if the bank participates in the “Patti Chiari” initiative; iv) dummies to control for the distance between the lending bank headquarters and household residence.

Note: The table shows the first and second moment of borrowers observable characteristics for three types of banks. a) Banks specialized in ARM; b) non-specialised banks; c) ban specialised in FRM. The three groups have been identified based on the method described in Figure 7. Banks in the first decile of the distribution (13 banks, 9.6% of the market) are defined specialized in ARM mortgages; banks in the last decile of the distribution (13 banks accounting for 14.1% of the market) are defined as specialized in FRM. The others are non-specialized. values of the test that the mean (or the variance) in group (a) is equal to that in group (c) are reported in parenthesis.

Note: The table shows the first and second moment of borrowers observable characteristics for three types of banks. a) Banks specialized in ARM; b) non-specialised banks; c) ban specialised in FRM. The three groups have been identified based on the method described in Figure 7. Banks in the first decile of the distribution (13 banks, 9.6% of the market) are defined specialized in ARM mortgages; banks in the last decile of the distribution (13 banks accounting for 14.1% of the market) are defined as specialized in FRM. The others are non-specialized. values of the test that the mean (or the variance) in group (a) is equal to that in group (c) are reported in parenthesis.

Table 4. Time-varying bank characteristics and mortgage choice Notes: The table shows linear probability estimates of mortgage choice. The left hand side variable is a dummy=1 if a FRM is chosen; zero otherwise. Robust standard errors (clustered at bank level) are reported in brackets. *, **, and *** represent significance levels of 10%, 5%, and 1% respectively. Coefficients for borrowers’ characteristics and fixed effects are not reported. (1) The Long Term Financial Premium (LTFP) is the difference between the FRM rate and the expected ARM rate based on borrowers’ actual ARM rate and one year moving average of the one month interbank rate. (2) Difference between the cost of fixed rate bank bonds and variable rate bonds. (3) Dummy equal to one if the bank is active in the securitization market, 0 elsewhere. (4) Deposits over total liabilities. (5) The bank concentration index is equal to the market share of the first 5 banking groups in each province. (6) Include: i) GDP per capita at the regional level; ii) a Bersani Law dummy= 1 from the second quarter of 2007 onwards; iii) a dummy if the bank participates in the “Patti Chiari” initiative; iv) dummies to control for the distance between the lending bank headquarters and household residence.

Table 4. Time-varying bank characteristics and mortgage choice Notes: The table shows linear probability estimates of mortgage choice. The left hand side variable is a dummy=1 if a FRM is chosen; zero otherwise. Robust standard errors (clustered at bank level) are reported in brackets. *, **, and *** represent significance levels of 10%, 5%, and 1% respectively. Coefficients for borrowers’ characteristics and fixed effects are not reported. (1) The Long Term Financial Premium (LTFP) is the difference between the FRM rate and the expected ARM rate based on borrowers’ actual ARM rate and one year moving average of the one month interbank rate. (2) Difference between the cost of fixed rate bank bonds and variable rate bonds. (3) Dummy equal to one if the bank is active in the securitization market, 0 elsewhere. (4) Deposits over total liabilities. (5) The bank concentration index is equal to the market share of the first 5 banking groups in each province. (6) Include: i) GDP per capita at the regional level; ii) a Bersani Law dummy= 1 from the second quarter of 2007 onwards; iii) a dummy if the bank participates in the “Patti Chiari” initiative; iv) dummies to control for the distance between the lending bank headquarters and household residence.

Table 5. A test for the presence of “dynamic” sorting

Table 5. A test for the presence of “dynamic” sorting

A. Main definition of price inaction (threshold (+ aie

A. Main definition of price inaction (threshold (+ aie

Notes: The table shows linear probability estimates of mortgage choice. The left hand side variable is a dummy=1 if a FRM is chosen; zero otherwise. Robust standard errors (clustered at bank level) are reported in brackets. *, ** and *** represent significance levels of 10%, 5%, and 1% respectively. Coefficients for borrowers’ characteristics and fixed effects are not reported. (1) The Long Term Financial Premium (LTFP) is the difference between the FRM rate and the expected ARM rate based on borrowers’ actual ARM rate and one year moving average of the one month interbank rate. (2) Difference between the cost of fixed rate bank bonds and variable rate bonds. (3) Dummy equal to one if the bank is active in the securitization market, 0 elsewhere. (4) Deposits over total liabilities. (5) Price inaction: in panel A, dummy Dj, =1 in quarters where bank b the change in the FRM/ ARM spread fall in the rage + = where the standard deviation is specific to each bank.; in panel B if change I the spread fall in the rage +@ (6) The bank concentration index is equal to the market share of the first 5 banking groups in each province. (7) Include: i) GDP per capita at the regional level; ii) a Bersani Law dummy= | from the second quarter of 2007 onwards; iii) a dummy if the bank participates in the “Patti Chiari” initiative; iv) dummies to control for the distance between the lending bank headquarters and household residence. B. Tighter definition of price inaction (threshold (+ — ae

Notes: The table shows linear probability estimates of mortgage choice. The left hand side variable is a dummy=1 if a FRM is chosen; zero otherwise. Robust standard errors (clustered at bank level) are reported in brackets. *, ** and *** represent significance levels of 10%, 5%, and 1% respectively. Coefficients for borrowers’ characteristics and fixed effects are not reported. (1) The Long Term Financial Premium (LTFP) is the difference between the FRM rate and the expected ARM rate based on borrowers’ actual ARM rate and one year moving average of the one month interbank rate. (2) Difference between the cost of fixed rate bank bonds and variable rate bonds. (3) Dummy equal to one if the bank is active in the securitization market, 0 elsewhere. (4) Deposits over total liabilities. (5) Price inaction: in panel A, dummy Dj, =1 in quarters where bank b the change in the FRM/ ARM spread fall in the rage + = where the standard deviation is specific to each bank.; in panel B if change I the spread fall in the rage +@ (6) The bank concentration index is equal to the market share of the first 5 banking groups in each province. (7) Include: i) GDP per capita at the regional level; ii) a Bersani Law dummy= | from the second quarter of 2007 onwards; iii) a dummy if the bank participates in the “Patti Chiari” initiative; iv) dummies to control for the distance between the lending bank headquarters and household residence. B. Tighter definition of price inaction (threshold (+ — ae

Notes: The table shows linear probability estimates of mortgage choice. The left hand side variable is a dummy=1 if a FRM is chosen; zero otherwise. Robust standard errors (clustered at bank level) are reported in brackets. *, **, and *** represent significance levels of 10%, 5%, and 1% respectively. Coefficients for borrowers’ characteristics and fixed effects are not reported. (1) The Long Term Financial Premium (LTFP) is the difference between the FRM rate and the expected ARM rate based on borrowers’ actual ARM rate and one year moving average of the one month interbank rate. (2) Difference between the cost of fixed rate bank bonds and variable rate bonds. (3) Dummy equal to one if the bank is active in the securitization market, 0 elsewhere. (4) Deposits over total liabilities. (5) Price inaction: dummy Dj;, =1 in quarters where bank b the change in the FRM/ ARM spread fall in the rage + < where the standard deviation is specific to each bank. (6) The bank concentration index is equal to the market share of the first 5 banking groups in each province. (7) Include: i) GDP per capita at the regional level; ii) a Bersani Law dummy= | from the second quarter of 2007 onwards; iii) a dummy if the bank participates in the “Patti Chiari” initiative; iv) dummies to control for the distance between the lending bank headquarters and household residence.

Notes: The table shows linear probability estimates of mortgage choice. The left hand side variable is a dummy=1 if a FRM is chosen; zero otherwise. Robust standard errors (clustered at bank level) are reported in brackets. *, **, and *** represent significance levels of 10%, 5%, and 1% respectively. Coefficients for borrowers’ characteristics and fixed effects are not reported. (1) The Long Term Financial Premium (LTFP) is the difference between the FRM rate and the expected ARM rate based on borrowers’ actual ARM rate and one year moving average of the one month interbank rate. (2) Difference between the cost of fixed rate bank bonds and variable rate bonds. (3) Dummy equal to one if the bank is active in the securitization market, 0 elsewhere. (4) Deposits over total liabilities. (5) Price inaction: dummy Dj;, =1 in quarters where bank b the change in the FRM/ ARM spread fall in the rage + < where the standard deviation is specific to each bank. (6) The bank concentration index is equal to the market share of the first 5 banking groups in each province. (7) Include: i) GDP per capita at the regional level; ii) a Bersani Law dummy= | from the second quarter of 2007 onwards; iii) a dummy if the bank participates in the “Patti Chiari” initiative; iv) dummies to control for the distance between the lending bank headquarters and household residence.

Loading...

Loading Preview

Sorry, preview is currently unavailable. You can download the paper by clicking the button above.

References (34)

  1. Anagol, Santosh, Shawn Cole and Shayak Sarkar (2012). "Understand- ing the Advice of Commissions-Motivated Agents: Evidence from the Indian Life Insurance Market."Harvard Business School Working Paper, No. 12-055.
  2. Badarinza, Christian, Campbell John Y. and Tarun Ramadorai (2014). "What Calls to ARMs? International Evidence on Interest Rates and the Choice of Adjustable -Rate Mortgages", Harvard University, mimeo.
  3. Battacharya, Utpal, Andreas Hackethal, Simon Kaesler, Benjamin Loos and Ste¤en Meyer (2012). "Is Unbiased Financial Advice to Retail In- vestors Su¢ cient? Answers from a Large Field Study", Review of Fin- ancial Studies, 25 (4): 975-1032.
  4. Berlin, Mitchell and Loretta J. Mester (1999). "Deposits and Relation- ship Lending", Review of Financial Studies, 12(3), 579-607.
  5. Chater, Nick, Ste¤valaen Huck and Roman Inderst (2010). "Consumer Decision-Making in Retail Investment Services: A Behavioral Economics Perspective", Report to the European Commission/SANCO.
  6. Calvet, Laurent E., Campbell John Y. and Paolo Sodini (2009). "Meas- uring the Financial Sophistication of Households,"American Economic Review, 99(2), 393-98.
  7. Campbell, John Y. and Joao F. Cocco (2003). "Household Risk Man- agement and Optimal Mortgage Choice", The Quarterly Journal of Eco- nomics, 118(4), 1449-1494.
  8. Carlin, Bruce I., and Gustavo Manso (2011). "Obfuscation, Learning, and the Evolution of Investor Sophistication", Review of Financial Stud- ies, 24 (3), 754-785.
  9. Crawford, Vincent P. and Joel Sobel (1982). "Strategic Information Transmission"Econometrica, 50(6), 1431-1451.
  10. Foerster, Stephen, Juhani Linnainmaa, Brian T. Meltzer and Alessandro Previtero (2014). "Retail Financial Advice: Does One Size Fit All?", Chicago booth Paper 38.
  11. Fuster, Andreas and James Vickery (2014). "Securitization and the Fixed-Rate Mortgage", Federal Reserve Bank of New York, Sta¤ Re- port n. 594.
  12. Gennaioli, Nicola, Andrei Shleifer and Robert Vishny (2015), "Money Doctors", Journal of Finance, 70 (1), 91-114
  13. Guiso, Luigi and Paolo Sodini (2012). "Household Finance. An Emer- ging Field", in Handbook of the Economics of Finance, edited by Con- standinides, G., M. Harris and R. Stulz, Elsevier Science.
  14. Guiso, Luigi, Luigi Pistaferri and Fabiano Schivardi (2013). "The Firm as a Credit Market", Review of Economic Studies, 80 (1), 211-247.
  15. Gurun, Umit G., Gregor Matvos and Amit Seru (2015), "Advertising Expensive Mortgages", Journal of Finance, forthcoming.
  16. Hackethal, Andreas, Roman Inderst and Ste¤en Meyer (2010). "Trading or Advice", CEPR DP 8091.
  17. Hackethal, Andreas, Michael Haliassos and Tullio Jappelli (2012). "Fin- ancial Advisors: A Case of Babysitters?", Journal of Banking and Fin- ance, 36(2), 509-24.
  18. Inderst, Roman (2010). "Irresponsible Lending with a Better Informed Lender", Economic Journal, 118, 1499-1519.
  19. Inderst, Roman and Marco Ottaviani (2009). "Misselling through Agents", American Economic Review, 99, 883-908.
  20. Inderst, Roman and Marco Ottaviani (2012a). "How (not) to Pay for Advice: A Framework for Consumer Protection", Journal of Financial Economics, 105(2), 393-411.
  21. Inderst, Roman and Marco Ottaviani (2012b). "Competition through Commissions and Kickbacks", American Economic Review,102 (2),780- 809.
  22. Ivashina, Victoria, and David S. Scharfstein. (2010) "Bank Lending Dur- ing the Financial Crisis of 2008", Journal of Financial Economics, 97, 319-38.
  23. Koijen, Ralph S.J., Otto Van Hemert and Stijn Van Nieuwerburgh (2009). "Mortgage Timing", Journal of Financial Economics, 93, 292- 324.
  24. Levy, Avi and Andrea Zaghini (2010). "The Management of Interest Rate Risk during the Crisis: Evidence from Italian Banks", Tema di Discussione n. 753, Banca d'Italia.
  25. Lancaster, Tony (2000), "The Incidental Parameter Problem since 1948", Journal of Econometrics, 95, 391-413.
  26. Liberti, Josè M. and Atif R. Mian (2009)."Estimating the E¤ect of Hier- archies on Information Use", Review of Financial Studies, 22, 4057-4090.
  27. Mullainathan, Sendhil, Markus Nöth, and Antoinette Schoar (2012). "The Market for Financial Advice: An Audit Study", NBER Working Paper, 17929.
  28. Ottaviani, Marco and Francesco Squintani (2006). "Naïve Audience and Communication Bias", International Journal of Game Theory, 12(1), 129-150
  29. Shapira, Zur and Itzhak Venezia (2001). "Patterns of Behavior of Pro- fessionally Managed and Independent Investors", Journal of Banking and Finance, 25, 1573-587.
  30. Woodward, Susan E., and Robert E. Hall (2012). "Diagnosing Consumer Confusion and Sub-Optimal Shopping E¤ort: Theory and Mortgage- Market Evidence", American Economic Review, 102(7), 3249-3276.
  31. Sample period 2004:Q1- 2010:Q4 2004:Q1- 2010:Q4 2004:Q1- 2010:Q4 2004:Q1- 2010:Q4 2004:Q1- 2010:Q4 2004:Q1- 2010:Q4 Notes: The table shows the parameter estimates of a linear probability model of mortgage type choice;. The left hand side variable is a dummy =1 if the borrower chooses a FRM, zero otherwise. Robust standard errors clustered at bank level are reported in brackets. *, **, and *** represent significance levels of 10%, 5%, and 1% respectively. Coefficients for dummies and fixed effects are not reported. (1) In columns II-V the LTFP is the difference between the FRM rate and the expected ARM rate based on borrower's actual ARM rate and one year moving average of the one month interbank rate (2) In column 5 the LTFP is the difference between the FRM rate and current the ARM rate. (3) Include: i) GDP per capita at the regional level;
  32. a Bersani Law dummy= 1 from the second quarter of 2007 onwards;
  33. a dummy if the bank participates in the "Patti Chiari" initiative; iv) dummies to control for the distance between the lending bank headquarters and household residence. 2004:Q1-2010:Q4 2004:Q1-2010:Q4 2004:Q1- 2010:Q4 2004:Q1- 2010:Q4 2004:Q1-2010:Q4 Notes: The table shows linear probability estimates of mortgage choice. The left hand side variable is a dummy=1 if a FRM is chosen; zero otherwise. Robust standard errors (clustered at bank level) are reported in brackets. *, **, and *** represent significance levels of 10%, 5%, and 1% respectively. Coefficients for borrowers' characteristics and fixed effects are not reported. (1) The Long Term Financial Premium (LTFP) is the difference between the FRM rate and the expected ARM rate based on borrowers' actual ARM rate and one year moving average of the one month interbank rate. (2) Difference between the cost of fixed rate bank bonds and variable rate bonds. (3) Dummy equal to one if the bank is active in the securitization market, 0 elsewhere. (4) Deposits over total liabilities. (5) The bank concentration index is equal to the market share of the first 5 banking groups in each (6) Include: i) GDP per capita at the regional level;
  34. a Bersani Law dummy= 1 from the second quarter of 2007 onwards; iii) a dummy if the bank participates in the "Patti Chiari" initiative; iv) dummies to control for the distance between the lending bank headquarters and household residence. Sample period 2004:Q1- 2010:Q4 2004:Q1- 2010:Q4 2004:Q1- 2010:Q4 2004:Q1- 2010:Q4