Transcript The Capital Stock and Equilibrium Unemployme - Aix

Deposit Insurance without Commitment:
Wall St. versus Main St. – Discussion
Sujit Kapadia
Bank of England
Workshop on
“Asset Prices, Credit and Macroeconomic Policies”
Marseille, 26 March 2011
These slides and the associated talk represent the views of the speaker and should not be
thought to represent those of the Bank of England or Financial Policy Committee members.
1
High-Level Overview (1)
• Core idea: benefits of ‘bailout’ (deposit insurance) v
costs (adverse redistribution effects)
• A very important question – link to policy on
systemically important financial institutions
• Very well executed paper
2
High-Level Overview (2)
• But how should we interpret the paper in the context of
the policy debate? Small banks or big banks?
• Small (not systemically important) banks; retail depositors
– model setup seems plausible
– but not clear why risk-based levies cannot be imposed ex ante
• Large (systemically important) banks; wholesale
depositors
– may not be possible (or desirable) to commit to bail out ex ante
– but bailout is not necessarily the same as deposit insurance
– and there are important considerations outside the model
3
Deposit Insurance for Small Retail Banks (1)
• Deposit insurance yields insurance gains
• Redistributional costs (poor to rich) vary depending
on how taxes are levied
• Important insights on impact of different tax policies:
– with ex post taxation, government can undo undesirable
redistributional effects (isn’t this the most realistic case?)
– ex ante type independent taxes  adverse redistributional
effects
– ex ante type dependent taxes can lower redistribution costs
4
Deposit Insurance for Small Retail Banks (2)
• But is ex ante commitment to provide deposit
insurance really an issue for small retail banks?
• What about an ex ante deposit insurance fund built
using risk-based levies, plus credible resolution?
–
–
–
–
riskier banks (more likely to draw on DI) pay more
richer households (benefit more from DI) pay more
banks are wound up in an orderly fashion
might this eliminate adverse distributional effects?
• If model is about small banks, would be useful to
consider deposit insurance with commitment
5
Big Banks: Bailout or Deposit Insurance?
• Paper uses Wall Street ‘bailout’ as a motivation
– ex ante commitment to bail out may not be possible /
desirable
– but deposit insurance is not the same as a bailout
– Fed and ECB provided very widespread liquidity support
during the crisis through discount window
• Bailouts linked to solvency risk
– UK ‘bailout’: RBS / LBG (HBOS) recapitalisation
– (Northern Rock is a red herring)
• To explore bailout, would be interesting to see
solvency issues discussed more clearly in the model
6
Wall St. v Main St. (1)
• Model: ‘Bailouts’ yield insurance gains but may have
adverse distributional effects
• But externalities of bank failure are the key reason for
bailout: network effects (liquidity hoarding, default
contagion); asset fire sales; credit crunch effects
– to analyse policy implications on Wall St. bailout, need this
rationale
7
Wall St. v Main St. (2)
• Tax effects are probably second-order for Main St:
what really matters is the GDP cost of crises
– Present value cost of this crisis may be around 90% of GDP
(Haldane, 2010)
• In terms of distributional issues
– what about deposit insurance caps?
– why not just address via tax system? (bank levies and
bonus tax; UK increase in top rate of income tax)
• If about big banks, model could usefully be enriched
to take these considerations into account
8