Banks as Liquidity providers Flashcards
What is the structure of the Diamond-Dybvig Model?
- Three period
- One homogenous good consumed either in first or second period: c_1,c_2
- Individuals → initial endowment of unit 1 in T = 0
- Individuals can either store endowment or invest in project characterized by long-term production technology
How is the production technology defined?
Production technology requires investment in T = 0
• Yields R > 1 in T = 2 for each unit invested
Or
• Returns liquidation value equal to initial investment when production aborted in T = 1
Intuition: LT investments irreversible and liquidating early may result in losses
What are the two types of individuals?
Individuals identical in T = 0 but in T = 1 learn in private whether type 1 or type 2
- Type 1: early consumers and only car about c_1
- Type 2: patient and care about c_2
Intuition: Individuals face private and stochastically independent risk of sudden liquidity needs
What is the intuition behind the First Best assumption?
Individuals share risk of being early consumer optimally
- Exante: individuals smooth consumption across states
- Expost: type 1/2 is better/worse off than under autarky
What happens under optimal risk sharing when it cannot be achieved via financial markets?
- Claims only contingent on publicly available and verifiable information
- But optimal risk sharing require claims contingent on type-membership which = private information
→ Contingent claims allowing for optimal risk sharing are unavailable in financial markets
What happens under optimal risk sharing when it cannot be achieved using self-selection of types?
Financial intermediary offer contract based on self.seletion of types that achieves optimal risk sharing
What is the sequential service constraint?
Depositors willing to withdraw = served in sequential order they appear at bank’s desk
How is it possible that the demand deposit contract achieves optimal risk sharing ?
- Risk t of being of type 1 = stochastically independent
- According to law of large numbers the share of early withdrawers bank faces very close to t
→ bank can stay solvent while engaging in maturity transformation
What happens if the bank invest the rest in illiquid assets ?
It renders bank illiquid = cannot serve more than expected t early without liquidating assets.
→ equilibrium relies entirely on type 2 not withdrawing early
→banks inherently fragile and prone to liquidity risk
What happens when some type 2 lose confidence and withdraw early ?
• Bank pays them c_1* in sequential order (liquidation value smaller than actual payout so bank = insolvent)
→ All individuals run and try to withdraw as early as possible since latecomers get nothing
→ Great economic damage: all investments liquidated = output lower than under autarky
What is the intuition behind bank run equilibrium?
Type 2 face coordination problem:
- Either none withdraws early = efficient equilibrium with optimal risk sharing
- All withdraw early = inefficient equilibrium with bank run
→ Even weak signals can cause loss of confidence and switch from efficient to inefficient equilibrium
→ Bank run may spread easily throughout whole banking system
What is narrow banking ?
Requires bank to hold enough liquidity to withstand bank run
→ Runs never occur
→ Drawback: Benefits from maturity transformation eliminated and optimal risk sharing no longer possible
What are the various techniques to avoid bank runs?
- Suspension of convertibility
- Deposit insurance
- Lending of last resort
What is suspension of convertibility?
Bank pay out at most tc_1* in T = 1
- Only first t depositors can withdraw early
- Bank can surely meet obligations in T = 2
→ removes incentive of type 2 to run
→ drawback: It random, some type 1 may not be able to withdraw
What is deposit insurance?
Guarantees promised return to all who withdraw even if bank fails
• Removes sequential service constraint as payouts independent of depositors’ place in line
→ type 2 no incentive to run
• government natural provider of deposit insurance due to ability to tax