Risk and the Organization of Bank Foreign Affiliates Giovanni Dell’Ariccia IMF and CEPR Robert Marquez Arizona State University
Motivation • Recent trend towards financial liberalization and increased cross-border activities for banks • Growing literature on the merits and pitfalls of an extensive foreign bank presence in emerging markets • Yet, little attention has been paid to how that presence is established: – Branches – Subsidiaries – Cross-border flows
Differences across structures • Subsidiaries: – Locally incorporated, stand-alone entities – Separately capitalized – Protected by limited liability at affiliate level • Branches: – No independent legal personality – Capital for regulatory purposes can be supplied at the consolidated level – Liabilities of foreign branches represent real claims on parent bank
Why this matters 1. May affect competitive structure of local banking systems: • • Threatening profits and market share of domestic banks Affecting price and quality of bank services in hostcountry 2. Involves different levels of parent bank responsibility and financial support with implications for: • • • The parent banks, who care about their exposure Local regulators, who care about domestic stability Local depositors, who care about safety of their savings
Our focus: Risk • We argue that risk is an important determinant of banks’ organizational choice • Two sources of risk: – Macroeconomic risk: Economic activity, interest rates, credit risk, etc. – Political risk: Expropriation of foreign banks, infringement of banks’ property rights
Results • Branch structure preferred when: – Political risk is primary source of uncertainty for bank • Bank benefits from keeping its capital at home and avoids being expropriated • Subsidiary structure preferred when: – Economic risk is of primary importance • Fragmented liability structure protects parent bank from losses at affiliate • Extensions to discuss cross-country correlation of risks, pricing of risk by investors, and competitive conditions of banking markets
Related Literature • There is a growing empirical literature studying what factors drive banks abroad and what determines location choices: Claessens et al. (2000), Focarelli and Pozzolo (2001 and 2005), Buch (2000 and 2003), Cerutti et al. (2006) • While theoretical work on this topic is still limited: Kahn and Winton (2004). . . • . . . a few papers have examined the related issue of how to regulate multinational banks: Calzolari and Loranth (2003), Dalen and Olsen (2003), and Harr and Ronde (2005)
Model • Bank has foreign affiliates in countries i = 1, …, n • Revenue for each affiliate: Li. Pi = Li. Riei, where – Li = loan quantity – Pi = realized return on loan – Ri = average interest rate on loan – ei = credit risk • The shock ei is distributed iid across countries, takes a value in [0, 1], and has distribution
Political risk and capital • Political risk: Probability i that host government appropriates foreign bank’s revenue and capital • Capital: Bank has capital E that can be used to satisfy capital requirement – For branches, need – For subsidiaries, need
High political risk • Result: There exists a value of i < 1 such that the bank’s overall profits are higher with a branch structure than with a subsidiary structure • Intuition: When political risk is very high ( i close to 1), parent bank is better off keeping all capital at home and financing branch with foreign deposits
Low political risk • Result: For i = 0, the bank’s profit is higher with a subsidiary structure than with a branch structure • Intuition: When there is no political risk, there is no value to keeping capital at home • However, macroeconomic risk creates an incentive for the bank to operate internationally via a subsidiary network – Limited liability at each affiliate prevents losses from spilling over to parent bank
Comparison of organizational forms • Main result: There is a value that – A subsidiary structure is preferred for – A branch structure is preferred for such
Extension: Correlation of risks • Suppose that macroeconomic risks are correlated across countries – Corresponds to ei being positively correlated • Result: The threshold value is decreasing in the degree of cross-country correlation • Intuition: When economic risk is correlated across countries, the shielding effect from limited liability of subsidiaries is reduced • Hence, the advantage of a branch structure is likely to be preferred
Other extensions • Have so far taken the cost of deposits and other liabilities as exogenous – However, in practice this cost should reflect the different risks and organizational forms • We also look at endogenizing the rate of return on bank assets – Rate of return should be a function of competitiveness of market – Required return to bank should also be risk-adjusted
Conclusion and implications • Branch vs. Subsidiary: Cerutti et al. (2006) find that banks are more likely to set up branches when political risk is relatively high • Correlation of risks: They also find that banks are more likely to use branches within-country, but subsidiaries for cross-border activities
To do • Much to do! • For instance, need to understand how the scale of entry may help drive organizational form • Banks tend to have both branches and subsidiaries – Model may need to be expanded to handle both
Скачать презентацию Risk and the Organization of Bank Foreign Affiliates
47b98337b25ddb445a938483dff11a57.ppt