Aucun titre de diapositive

1
MACROECONOMIC SHOCKS AND BANKING
SUPERVISION Jean-Charles ROCHET (IDEI, Toulouse
University, France) PREPARED FOR THE 2nd
WORKSHOP OF THE LATIN AMERICAN FINANCE NETWORK,
Cartagena, December 2004
2

• 1-INTRODUCTION
• Many Latin American countries (and more generally
  emerging countries) have recently been hurt by
  severe banking and financial crises but they do
  not have a monopoly for such crises
• In the last 25 years, an impressive number of
  countries (developed, developing and emerging)
  have also been involved in such crises
• Among IMF member countries more than 130 out
  of 180
• have experienced crises or serious banking
  problems
• Let us not forget that the cost of Savings and
  Loans debacle in the USA (late 1980s) was
  probably much superior to the cumulated loss of
  all failed US banks during Great Depression (in
  real terms)

3
No significant banking problem/Insufficient inform
ation
Banking crisis
Significant banking problems
4

• PRUDENTIAL INSTITUTIONS HAVE BEEN REFORMED TO
  AVOID OR MITIGATE FURTHER CRISES
• DEPOSIT INSURANCE FUNDS
• USA FDIC (created in 1934) still quite generous
  (100,000) but now funded by risk-sensitive
  premiums
• UK more recent and less generous 75 of the
  first 20,000
• EU minimum harmonization have to be explicit,
  funded by premiums
• Minimum coverage Euros 20,000
• Explicit schemes have been introduced in many
  countries (Garcia 1999)
• It has been argued that they provoke moral
  hazard bankers could be incited to take too much
  risks since they are not disciplined by small
  depositors. Demirguc-Kunt and Detragiache (1997)
  find evidence in this direction.

5

• SOLVENCY REGULATIONS
• Basel accord (1988) minimum capital requirement
  ( 8 of the sum of risk weighted assets) for all
  internationally active banks of G-10 countries.
  Later amended to incorporate interest rate risk
  and market risk.
• USA FDICIA ( 1991) capital requirement plus
  Prompt Corrective Action reform introduced after
  the Saving and Loans crisis supervisors are
  forced to intervene before it is too late (CAMELS
  ratings)
• EU Capital Adequacy Directives similar to the
  Basel accord, which has also been adopted in many
  countries
• OTHER REFORMS
• Creation of integrated supervisory authorities
  (FSA), independent from central banks, in the UK
  and other countries

6

• The Basel Committee on Banking Supervision has
  elaborated a reform of the Basel accord (Basel 2)
• This reform relies on three pillars
• A refined capital requirement with very complex
  weights
• A more pro-active role of banking supervisors
• An increased recourse to market discipline
• The problem is that supervisors have a tendency
  to interfere too much when the banks are well run
  and intervene too less when they have problems

7
In fact, empirical evidence on the resolution of
bank defaults suggests that failing banks are
more often rescued than liquidated. Goodhart
and Schoenmaker (1995)effective methods of
resolving banking problems vary a lot from
country to country but in most cases result in
bail outs Out of a sample of 104 failing banks
they find that 73 resulted in rescue and 31 in
liquidation Santomero and Hoffman (1998) in
the USA, the discount window was often used
(improperly) to rescue banks that subsequently
failed
8

• OBJECTIVES OF THIS PRESENTATION
• Show that deposit insurance and capital
  requirements are not enough to deal with
  macro-shocks
• Study how regulatory/supervisory systems should
  be reformed,
• so as to deal properly not only with individual
  bank failures (micro- prudential regulation)
  but also with systemic crises(
  macro-prudential regulation)
• Try to see whether these conclusions also apply
  to Latin America
• PUNCH LINE
• Central bank independence for monetary policy
  should be extended to lender of last resort
  activities
• Liquidity assistance to the banks in trouble
  should be provided under the strict control of
  independent supervisory authorities

9
STRUCTURE OF THIS PRESENTATION 1
INTRODUCTION 2 A SIMPLE MODEL OF PRUDENTIAL
REGULATION 3 HOW TO DEAL WITH MACRO SHOCKS? 4
IS MARKET DISCIPLINE USEFUL? 5 AN AGENDA FOR
MACRO-PRUDENTIAL REGULATION
10
2- A SIMPLE MODEL OF PRUDENTIAL
REGULATION Adaptation to the banking sector of
Holmström-Tirole (1997, 1998) (model of corporate
financing) 2 dates t 0,1 for the moment a
single bank (micro view point), lends volume L
of loans, financed by deposits D and equity E.
Interest rate normalized to zero. Deposits
insured by Deposit Insurance Fund (DIF) for a
premium P .
Banker gets
E
Returns
Moral Hazard
Lending
Success
t 1
Failure
t 0
D
0
Banker repays depositors
Bank pays P
DIF repays
depositors
11

• Specificities of banks (w.r.t. corporations as in
  Holmström-Tirole)
• A (large) fraction of loans is financed by
  deposits.
• Deposits are insured
• Loans have to be monitored by the banker (Moral
  Hazard)
• good loans high probability of success p
• bad loans probability ( p - ?p)
  private benefit BL.
• Social value of the bank gt NPV loans
• vL additional social value (payment
  activities...).
• (for the moment , could be 0).
• Banking supervision
• contract between banker and Deposit Insurance
  Fund (DIF)
• specifies L and D (and insurance premium P) as
  a function of E

12
Assumptions (A1) (loans have positive social
value only when monitored ) Banks
play a crucial role by monitoring borrowers
notice that (A1) ? (A2)
Banks can obtain large private benefits by
shirking without A2 banks do not need
capital, see below
13

• Optimal contract
• Specifies the volume of loans and the deposit
  insurance premium that maximize the economic
  value added by the bank under 2 constraints
• Bankers monitor the borrowers (MH)
• The D.I.F. breaks its budget
• Proposition 1 Optimal organization of the
  banking sector
• Deposit insurance financed by fair ( risk- based)
  premiums
• Capital ratio
• Capital ratio k gt0 by A2, ? in B (intensity of
  moral hazard)
• and ? in (pR - 1) (quality of assets).

14

• Implementation
• In the absence of macro shocks there are (in the
  model) 2 equivalent solutions
• Public authority regulates banks capital and
  DIF premiums
• Banking supervisors close banks that do not
  comply with capital
• requirements
• Private contracting between competing DIFs and
  the bank
• deposits limited to a maximum of
• insurance premiums set by competition between
  DIFs.
• In practice there may be problems associated with
  private bank insurance, especially during
  systemic crises

15
4- INTRODUCING MACROECONOMIC SHOCKS Crucial
feature of banking sensitivity to macro shocks,
captured by risk of recession (devaluation)
additional liquidity needs at t 1/2
specific to each bank exposure to macro shocks,
observable ex-ante by banking supervisors
(dollarization ratio?) We now adopt banking
system view point distribution F( )
E
t0
t1/2 Shock?
t1
Moral Hazard
Success
refinance
Failure
close
Lending
D
0
0
Additional question which banks are bailed out
(x1) and which banks are closed (x0) in the
event of a shock at date t 1/2 ?
16

• Optimal regulation contract
• For each bank, as a function of its exposure to
  macro shocks, it specifies
• the volume of loans the bank is allowed to
  grant,
• the deposit insurance premium it has to pay and
• whether it is eligible to the LLR or not
• NB Liquidity requirements are suboptimal in this
  model
• We show that
• Banks have to be separated in two categories in
  case of a macro shock,
• those with a low macro exposure are rescued,
• those with a high exposure are closed.

17

• Proposition 2
• Optimal regulation in the presence of
  macroeconomic shocks
• Banks such that (small
  exposure to macro shocks) should not be closed in
  case of macro shock but should be subject to an
  additional capital charge or loan loss provision
  (increasing with macro exposure)
• Banks such that (large
  exposure to macro shocks) should be closed if a
  recession occurs and should be subject to a flat
  capital ratio

18

• Implementation different roles of private
  investors and public regulators
• Private investors only interested by future
  cash flows
• refinance the bank?
  (too many closures)
• Public regulators subject to political pressure
  (efficient lobbying)
• refinance the bank?
  (too few closures).
• Public intervention is needed to avoid too many
  bank closures
• but it leads to forbearance and overinvestment

19
LLR
SBC

• Need for a Lender of Last Resort (LLR)
• But risk of Soft Budget Constraint (SBC)

Possible solutions Market discipline banks
forced to issue subordinated debt held by private
investors, who limit scope for moral hazard
(Section 5) Independent regulatory agency and
lender of last resort (Section 6)
20
5- IS MARKET DISCIPLINE USEFUL? External
monitor
Reduces private benefit of bankers B ? b B -
?
Unit cost ?
Regulation contract plus
remuneration of external monitor in case of
success financial involvement of monitor at
t 1 (subordinated debt)
21
OPTIMAL COMBINATION OF MARKET DISCIPLINE AND
REGULATION

• For each bank, as a function of its exposure to
  macro shocks,
• the regulatory contract specifies
• the minimum capital ratio,
• the deposit insurance premium
• whether it is eligible to the LLR, and
• the volume of sub-debt it has to issue
• the maximum interest rate on sub-debt

Proposition 3 The presence of external monitors
increases the optimal closure threshold (less
closures) But in the absence of commitment power
by the government, the effective closure
threshold remains unchanged Capital
requirements are reduced but the impact on social
surplus is ambiguous.
22

• If the supervisor can commit, the introduction of
  an external monitor
• is not necessary beneficial (depends on cost of
  uninsured debt).
• The consequences of using an external monitor are
• to reduce the need for a LLR,
• to reduce the commitment problem (Soft Budget
  Constraint)
• of the public supervisor is
  unchanged, while

SBC
LLR
23
6- AN AGENDA FOR MACRO PRUDENTIAL REGULATION As
a complement to market discipline I propose to
solve the market failure associated with banks
exposure to macro shocks by reforming prudential
regulation. Two elements are needed for
implementation of the optimal allocation Interv
ention of the central bank as a lender of last
resort for providing liquidity assistance ( in
case of a recession) to the banks characterized
by low macro exposure. Preventing extension of
this liquidity assistance to other banks, with
higher macro exposure. For these banks ex post
continuation value is positive (from a social
point of view) but bailing them out would be
welfare decreasing from an ex-ante perspective.
24

• The optimal contract (characterized in
  Proposition 2) can be implemented by the
  following organization of the regulatory system
• for each commercial bank, supervisory
  authorities evaluate the bank's exposure to
  macroeconomic shocks
• Banks with a small exposure are backed by the
  DIF and receive liquidity assistance by the
  Central Bank in case of a macro shock.
• They face a capital adequacy requirement and a
  deposit insurance premium that increase with
  macro exposure

25

• Banks with a large exposure to macro-shocks
  are not backed by the DIF they are not eligible
  to receive liquidity assistance by the Central
  Bank.
• They face a flat capital requirement and a
  flat deposit insurance premium
• The lender of last resort activities of the
  central bank are made independent from political
  powers the central bank only provides liquidity
  assistance to the banks that are backed by
  supervisory authorities.
• Central bank loans are fully insured by the DIF.

26

• CONCLUSIONS (1)
• The main reason behind the frequency and
  magnitude of recent banking crises is not deposit
  insurance, bad regulation, or incompetence of
  supervisors. It is the commitment problem of
  political authorities, who are likely to exert
  pressure for bailing out insolvent banks and
  delay crisis resolution.
• The remedy to political pressure on banks
  supervisors
• is not to substitute supervision by market
  discipline
• market discipline can only be effective if
  absence of government intervention is
  anticipated.
• Instead the way to restore credibility is to
  ensure
• independence and accountability of bank
  supervisors.

27

• CONCLUSIONS (2)
• The other key reform is to restrict liquidity
  assistance by the central bank to the banks with
  low exposure to macro shocks, that are backed by
  the independent supervisors (alternative cap on
  macro exposure).
• Supervisors should be in charge of selecting
  these banks, who then would face a capital
  requirement and a deposit insurance premium that
  both increase with their macro exposure
• By contrast, banks with a too high macro
  exposure are not backed by the supervisors and
  do not receive liquidity assistance in case of
  macro shocks
• Central bank loans should be insured by the DIF.

28

• APPLICABILITY TO LATIN AMERICAN ECONOMIES?
• Devaluation risk major source of macro shock
  (but not the only one)
• Full dollarization would trivially eliminate this
  risk but would prevent the action of the central
  bank as a LLR
• Liquid asset requirements maybe useful but they
  are costly
• A LLR is needed but neither a purely private
  solution (CCL) or a government controlled
  solution are good
• Like everywhere else, independence and
  accountability of banking authorities (DIF, FSA
  and Central Bank) are fundamental
• Solvency regulations and deposit insurance
  premiums should be modified to account for
  currency risk exposures of banks