Are Real Estate Banks More Affected by Real Estate Market Dynamics? Evidence from the Main European

Countries

Lucia Gibilaro, University of Bergamolucia.gibilaro@unibg.it

Gianluca Mattarocci, University of Rome “Tor Vergata” gianluca.mattarocci@uniroma2.it

Vienna – July 3rd-6th, 2013

• Introduction

• Literature review

• Empirical analysis:

• Sample

• Methodology

• Results

• Conclusions

Index

Introduction

The effect on the credit market could be driven by the increase of the collateral value of lending exposure and decrease the bank riskiness if the demand of the real estate financing does not change (Kiyotaki and Moore, 1997) while cause an increase of the probability of default of the bank if debtors increases their exposure due the lower cost of lending and the easier access to financing opportunities (Koetter and Poghosyan, 2010).

Studies on the main market players in the banking sector demonstrate that real estate banks could be more riskier respect to other banks (Blasko and Sinkey, 2006) even if results could change with the proxy used for evaluation the bank risk change (Giannotti, Gibilaro and Mattarocci, 2011) and the criterion for identifying the real estate banks (Eisenbeis et al. 1996).

Introduction

Research questions:

- What is the role of real estate market trends in explaining the higher or lower riskiness and profitability of banks?

- Are real estate banks over-exposed to the real estate market trend?

• Introduction

• Literature review

• Empirical analysis:

• Sample

• Methodology

• Results

• Conclusions

Index

Literature review (1/2)

Empirical evidences demonstrate that the market price of bank shares discount also the risk related to the real estate market trend and the sensitivity to the market could be different on the basis of some bank features (i.a. size) (i.a. Allen et al., 1995) Empirical evidences demonstrate that the real estate price dynamics affects the amount of lending offered by the banks even if the relationship could be more or less significant on the basis of the market analysed and the time horizon considered (i.a. Inoguchi, 2011). The characteristics of the bank could explain an higher or lower sensitivity to the real estate market dynamics and normally the effect is higher the worse are the bank fundamentals (i.a. Peek and Rosengren, 1994)

Literature review (2/2)

Preliminary evidence proposed in literature demonstrates that real estate banks could be riskier respect to other banks (Blasko and Sinkey, 2006) but results are not always confirmed when the time horizon considered and the proxy selected for evaluation the bank risk change (Giannotti, Gibilaro and Mattarocci, 2011) or when we adopt a different criterion for identifying the real estate banks (Eisenbeis et al. 1996). Only few analyses proposed in literature consider the role of the real estate market trend in determining the risk and revenues of the banks (Igan and Pinheiro, 2010) and there are no studies that allow evaluating if a specialized real estate bank is more or less affected by the market trend with respect to all other banks.

• Introduction

• Literature review

• Empirical analysis:

• Sample

• Methodology

• Results

• Conclusions

Index

Empirical analysis: Sample

Database: Bankscope Time horizon: 2007-2011 Frequency of data: yearly

Real Estate Banks 2007 2008 2009 2010 2011

Number 43 161 163 162 166Total assets 3,239,729.8 3,977,323.4 3,238,684.7 3,476,400.4 4,467,402.4

Average Total Assets 75,342.55 24,703.87 19,869.23 21,459.26 27,075.17

Not Real Estate Banks           2007 2008 2009 2010 2011

Number 940 823 822 826 822Total assets 56,404,342 70,861,970 81,591,761 96,917,431 120,437,628

Average Total Assets 59,372.99 85,170.64 98,303.33 116,207.95 144,930.96

Following the approach proposed by Eisenbes and Kwast (1991), we classify in the year t a bank as a real estate bank if the role of real estate lending % Real Estateit is higher than 40%. Summary statistics on the two subsamples (REBs and not REBs) for each year are provided in table 1.

 

Empirical analysis: Methodology (1/2)

Following the approach proposed by Blasko and Sinkey (2006), we compute a measure of the default risk for each bank in the sample using the following formula:

The leverage measure is the ratio between the equity liabilities and total assets at time t for the bank iThe TIER 1 capital requirement at time t for the bank i defined on the basis of the amount and quality of outstanding debt

Return on Assets at time t for the bank iReturn on Equity at time t for the bank iNet Interest Income with the respect to the overall income at time t for the bank iLoan loss provisions with respect to the overall loans at time t for the bank iTotal of past due credits over 90 day with respect to the overall loans at time t for the bank iAmount of derivative exposure with respect to total assets at time t for the bank iDifference between rate-sensitive assets and rate sensitive liabilities with respect to total asset at time t for the bank i

𝑍𝑅𝑖𝑠𝑘𝑖𝑡=𝛼𝑖𝑡+∑𝑘=1

𝑛

𝛽𝑘𝐵𝑎𝑛𝑘𝐹𝑒𝑎𝑡𝑢𝑟𝑒𝑖𝑡𝑘 +∑

𝑙=1

𝑚

𝛽𝑘𝐶𝑜𝑢𝑛𝑡𝑟𝑦 𝐷𝑢𝑚𝑚𝑦 𝑖𝑡𝑙 +𝜀𝑖𝑡

Left hand side variable

Panel regression model

𝑍𝑅𝑖𝑠𝑘𝑡=𝐴𝑣𝑒𝑟𝑎𝑔𝑒𝑅𝑂𝐴𝑡− 3 ,𝑡+𝐶𝐴𝑃 𝑡

𝜎𝑅𝑂𝐴𝑡 −3 , 𝑡

𝑍𝑅𝑖𝑠𝑘𝑖𝑡=𝛼𝑖𝑡+∑𝑘=1

𝑛

𝛽𝑘𝐵𝑎𝑛𝑘𝐹𝑒𝑎𝑡𝑢𝑟𝑒𝑖𝑡𝑘 +∑

𝑙=1

𝑚

𝛽𝑘𝐶𝑜𝑢𝑛𝑡𝑟𝑦 𝐷𝑢𝑚𝑚𝑦 𝑖𝑡𝑙 +𝛾 𝑖𝑡𝐷𝑢𝑚𝑚𝑦 𝑅𝑒𝑎𝑙 𝐸𝑠𝑡𝑎𝑡𝑒𝑖𝑡+𝜀𝑖𝑡

Empirical analysis: Methodology (2/2)

𝑍𝑅𝑖𝑠𝑘𝑖𝑡=𝛼𝑖𝑡+∑𝑘=1

𝑛

𝛽𝑘𝐵𝑎𝑛𝑘𝐹𝑒𝑎𝑡𝑢𝑟𝑒𝑖𝑡𝑘 +∑

𝑙=1

𝑚

𝛽𝑘𝐶𝑜𝑢𝑛𝑡𝑟𝑦 𝐷𝑢𝑚𝑚𝑦 𝑖𝑡𝑙 +𝛾 𝑖𝑡% 𝑅𝑒𝑎𝑙 𝐸𝑠𝑡𝑎𝑡𝑒𝑖𝑡+𝜀𝑖𝑡

𝑍𝑅𝑖𝑠𝑘𝑖𝑡=𝛼𝑖𝑡+∑𝑘=1

𝑛

𝛽𝑘𝐵𝑎𝑛𝑘𝐹𝑒𝑎𝑡𝑢𝑟𝑒𝑖𝑡𝑘 +∑

𝑙=1

𝑚

𝛽𝑘𝐶𝑜𝑢𝑛𝑡𝑟𝑦 𝐷𝑢𝑚𝑚𝑦 𝑖𝑡𝑙 +¿𝛾𝑖𝑡𝐷𝑢𝑚𝑚𝑦 𝑅𝑒𝑎𝑙 𝐸𝑠𝑡𝑎𝑡𝑒𝑖𝑡+𝜏 𝑖𝑡𝑅𝑒𝑎𝑙 𝐸𝑠𝑡𝑎𝑡𝑒𝑀𝑘𝑡𝑡+𝜀𝑖𝑡 ¿

𝑍𝑅𝑖𝑠𝑘𝑖𝑡=𝛼𝑖𝑡+∑𝑘=1

𝑛

𝛽𝑘𝐵𝑎𝑛𝑘𝐹𝑒𝑎𝑡𝑢𝑟𝑒𝑖𝑡𝑘 +∑

𝑙=1

𝑚

𝛽𝑘𝐶𝑜𝑢𝑛𝑡𝑟𝑦 𝐷𝑢𝑚𝑚𝑦 𝑖𝑡𝑙 +¿𝛾𝑖𝑡% 𝑅𝑒𝑎𝑙 𝐸𝑠𝑡𝑎𝑡𝑒𝑖𝑡+𝜏 𝑖𝑡𝑅𝑒𝑎𝑙 𝐸𝑠𝑡𝑎𝑡𝑒𝑀𝑘𝑡 𝑡+𝜀𝑖𝑡¿

(3)

(4)

(5)

(6)

+𝐷𝑢𝑚𝑚𝑦 𝑅𝑒𝑎𝑙𝐸𝑠𝑡𝑎𝑡𝑒𝑖𝑡×𝜕𝑖𝑡𝑅𝑒𝑎𝑙𝐸𝑠𝑡𝑎𝑡𝑒𝑀𝑘𝑡𝑡

𝑍𝑅𝑖𝑠𝑘𝑖𝑡=𝛼𝑖𝑡+∑𝑘=1

𝑛

𝛽𝑘𝐵𝑎𝑛𝑘𝐹𝑒𝑎𝑡𝑢𝑟𝑒𝑖𝑡𝑘 +∑

𝑙=1

𝑚

𝛽𝑘𝐶𝑜𝑢𝑛𝑡𝑟𝑦 𝐷𝑢𝑚𝑚𝑦 𝑖𝑡𝑙 +𝐷𝑢𝑚𝑚𝑦𝑁𝑜𝑡𝑅𝑒𝑎𝑙 𝐸𝑠𝑡𝑎𝑡𝑒𝑖𝑡×𝜕𝑖𝑡𝑅𝑒𝑎𝑙 𝐸𝑠𝑡𝑎𝑡𝑒𝑀𝑘𝑡𝑡+¿¿

+%𝑅𝑒𝑎𝑙𝐸𝑠𝑡𝑎𝑡𝑒𝑖𝑡×𝜕𝑖𝑡𝑅𝑒𝑎𝑙𝐸𝑠𝑡𝑎𝑡𝑒𝑀𝑘𝑡 𝑡

𝑍𝑅𝑖𝑠𝑘𝑖𝑡=𝛼𝑖𝑡+∑𝑘=1

𝑛

𝛽𝑘𝐵𝑎𝑛𝑘𝐹𝑒𝑎𝑡𝑢𝑟𝑒𝑖𝑡𝑘 +∑

𝑙=1

𝑚

𝛽𝑘𝐶𝑜𝑢𝑛𝑡𝑟𝑦 𝐷𝑢𝑚𝑚𝑦 𝑖𝑡𝑙 +% 𝑁𝑜𝑡 𝑅𝑒𝑎𝑙 𝐸𝑠𝑡𝑎𝑡𝑒𝑖𝑡×𝜕𝑖𝑡𝑅𝑒𝑎𝑙 𝐸𝑠𝑡𝑎𝑡𝑒𝑀𝑘𝑡𝑡+¿¿

Empirical analysis: Results (1/3)

REBs Not REBS Kolmogorov Smirnov test

Mean Median Dev.St. Mean Median Dev.St. Value Test15.05 3.52 34.84 6.53 3.10 20.48 0.17 0.000.07 0.07 0.06 0.12 0.10 0.13 0.44 0.000.11 0.10 0.09 0.16 0.14 0.12 0.29 0.000.00 0.00 0.01 0.01 0.01 0.05 0.32 0.000.03 0.03 0.21 0.05 0.05 0.22 0.17 0.000.02 0.01 0.05 0.02 0.03 0.09 0.14 0.000.02 0.01 0.05 0.03 0.02 0.04 0.32 0.000.00 0.00 0.04 0.00 0.00 0.01 0.01 1.000.01 0.00 0.03 0.03 0.00 0.77 0.28 0.00-0.10 -0.06 0.18 -0.02 0.00 1.99 0.49 0.00

Empirical analysis: Results (2/3)

(3) (4) (5) (6)8.21 7.55 7.47 6.73-2.76 -2.68 -2.71 -2.6256.09* 57.18* 50.09 51.382.74 2.64 1.93 1.82

240.57*** 240.84*** 217.18*** 217.47***-18.94* -19.37* -16.97 -17.40-47.01 -49.32 -38.82 -41.45-0.12 -0.11 -0.11 -0.100.00 0.00 0.00 -0.013.27* - 3.26* -- 2.01 - 1.76- - 30.46*** 30.31***

-3.38 -3.40 -3.18 -3.18Country Dummies Yes Yes Yes YesObservations 2982 2982 2982 2982Groups 672 672 672 672R^2 0.10 0.09 0.10 0.10

Empirical analysis: Results (3/3)

(5) (6) (7) (8)7.47 6.73 7.27 6.5-2.71 -2.62 -2.71 -2.6150.09 51.38 52.9 54.691.93 1.82 1.70 1.55

217.18** 217.47** 218.07** 218.29**-16.97 -17.40 -16.82 -17.19*-38.82 -41.45 -39.29 -42.06-0.11 -0.10 -0.10 -0.090.00 -0.01 0.00 0.003.26 - - -- 1.76 - -

30.46** 30.31** - -- - 38.35*- - 28.83**

39.47*28.44**

-3.18 -3.18 -3.15 -3.12Country Dummies Yes Yes Yes YesObservations 2982 2982 2982 2982R^2 0.10 0.10 0.10 0.10

• Introduction

• Literature review

• Empirical analysis:

• Sample

• Methodology

• Results

• Conclusions

Index

• The role of the market trend is not independent with the respect of the specialization of the bank in the real estate sector and, due the higher expertise in the sector, normally Real estate banks are those that are less affected by any positive or negative market dynamics.

• Looking at the literature on the diversification of the lending portfolio of a bank, evidences provided support the hypothesis hypnotized in literature that the reduction of the bank risk is not always related only to the degree of diversification (i.a. Demsetz and Strahan, 1997) and REBs can reduce their risk exposure using their specific knowledge in the real estate lending.

Conclusions

Contact Information

Lucia GibilaroUniversity of Bergamoe-mail: lucia.gibilaro@unibg.it

Gianluca MattarocciUniversity of Rome Tor Vergatae-mail: gianluca.mattarocci@uniroma2.it

Thanks for you attention