Surviving Financial Turbulence: Tales from and for Latin America

1
Surviving Financial Turbulence Tales from and
for Latin America

• Andrew Powell
• Inter American Development Bank
• Pontificia Universidad
• Católica del Perú
• Lima, November 9th 2011

Opinions are strictly those of the author and are
not necessarily those of the IDB or any other
institution.
2
Outline

• A) Latin America has learned important lessons
  from its financial history, and escaped the worst
  of the crisis
• B) The G20 has set an ambitious agenda regarding
  financial sector reform in response to the crisis
• C) Regulatory innovations in the North
  (Dodd-Frank (US), ICB (UK) Euroland overtaken
  by events).
• D) Given capital inflows and strong growth in
  credit, LAC needs a strong financial sector!
• E) What is LAC doing, what needs to be done?
• F) A Proposal Lima I

3
Lessons from LAC Financial History
4
On LAC s (Recent) Financial History

• The debt crisis of the 1980s US interest
  rates, commodity prices, sovereign debt, banking
  crises.
• Tequila in the 1990s liquidity, financial
  crises
• The 1997 Asian crisis 1998 Russian default
  contagion!
• The Argentine crisis of 2002 fixed money with
  not so flexible prices/wages, sovereign debt
  banks and innovative resolution asymmetric
  pessification, aggressive default, deep
  recession but then growth...

5
Financial vs. Other Crises

• Financial crises are crises of stocks, balance of
  payments crises are related to flows
• Rheinhart and Rogoff (This time is different
  eight centuries of financial folly), document
  that financial crises are deeper and more
  persistent
• Financial crises resolution frequently requires
  significant use of public money and may provoke
  redistributions of wealth

6
LAC Has Strengthened Its Defenses

• Reserves
• Financial Systems
• Regulation and Supervision
• International Insurance

7
Reserves Pre-Crisis and End of 2010
Reserves as of GDP, end of period
8
Bank (Regulatory) Capital
Average requirements (exigencias) are around 11,
the average regulatory ratio (integracion) is
around 16. Source IMF, GFSR.
9
Regulation and Supervision

• LAC regulators have become more professional
• Supervision has improved
• Monitoring has been useful
• e.g. IMF/WB FSAP program

10
International Insurance
B) Countries with access to ILOR in Lehman with
same credit ratings as A
A) LAC Countries with no access to ILOR during
Lehman
C) LAC Countries with access to ILOLR during
Lehman
Source Bloomberg and Staff estimations
11
The Ambitious G20Reform Agenda and Regulatory
Innovations from the North
12
Elements of the G20 Agenda

• 1. Bank Capital
• 2. Bank Liquidity
• 3. Complexity and cross border issues
• Other Issues
• Systemic institutions intensive supervision,
  additional capital, liquidity, and other
  prudential requirements
• Derivatives standardized OTC contracts should be
  traded on exchanges or electronic trading
  platforms, where appropriate, and cleared through
  central counterparties
• Rating Agencies, oversight of the ratings
  business
• Compensation packages
• A single set of global accounting standards
• Countermeasures against tax havens
• Other LA issues I will not talk about today
  (dollarization, lending to public sector, related
  lending)

Basel N (N1,2,3)
13
Dodd-Frank (US)

• Includes a serious attempt to limit the
  Executives powers to bail-out financial
  institutions
• A failing institution to be liquidated or
  resolved by title 2 resolution
• Under Title 2, the State has the power to
  create a good and bad bank and then sell the good
  bank (to some extent similar to some LAC bank
  resolution rules)
• But rules on market concentration may prevent
  selling the good bank to another large existing
  entity.
• The idea is that limiting bail out powers ex ante
  will create discipline
• Significant uncertainty as to how would be
  applied in practice, especially in a potentially
  systemic crisis
• See a simulation of a major US bank failure under
  Dodd-Frank at a recent Economist conference in
  NY. www.economist.com/blogs/freeexchange/2011/11/b
  anking-regulation

14
Independent Commission on Banking, ICB (Vickers
Report, UK)

• Separation of Investment and Retail Banking with
  Chinese walls
• Prohibited Services in Retail include
• any service which would result in a trading book
  asset
• any service which would result in a requirement
  to hold regulatory capital against market risk
• any service which results in an exposure to a
  non-ring-fenced bank or a non-bank financial
  organization, except those associated with the
  provision of payments services where the
  regulator has deemed this appropriate
• the purchase or origination of derivatives or
  other contracts
• which would result in a requirement to hold
  regulatory capital against counterparty credit
  risk and services relating to secondary markets
  activity including the purchase of loans or
  securities.
• Authors note A ring-fence of this kind would
  also have the benefit that ring-fenced banks
  would be more straightforward than some existing
  banking structures and thus easier to manage,
  monitor and regulate.
• See Chapter 3, Independent Commission on Banking
  (2011) Final Report Recommendations, London
  ICB.
• Commission also recommends UK following
  Switzerlands lead and introduce the use of Cocos

15
The Latin American ContextNow is the time to Act
16
Latin America Capital Inflows
2008 was an interruption of the current episode
of GDP
Inflows
Net Flows
Outlfows
For LAC 7 Economies
17
The Changing Composition of Inflows
Now More Capital Inflows are Portfolio and Bank
in Other
In 2006, FDI was gt2/3, now lt1/3
18
Strong Growth in Domestic Credit
Source National Sources
19
Strong Appreciation Pressures
Mexico the only depreciator in this sample
Significant depreciations in the crisis, save
Uruguay and Peru
Large appreciations post crisis
Source Staff on National Sources
20
Current Accounts Now Mostly in Negative Territory
Source WEO, IMF
21
Fiscal Balances have Deteriorated
Fiscal Balance of GDP
Source WEO, IMF
22
Structural Balances Deteriorated
23
Ongoing analysis of Emerging Economy Capital
Inflow Episodes

• Of the 91 Capital Inflow Episodes
• 29 Ended in a Banking Crisis
• 53 Ended in a Recession
• 59 Ended in a Banking Crisis or a Recession

24
    Banking Crisis Banking Crisis Banking Crisis Banking Crisis Banking Crisis   Recession Recession Recession Recession Recession
    Def 1   Def 2   Def 3   Def 1   Def 2   Def 3
Inflow Characteristics                        
                         
Total Inflows                        
                         
Portfolio Inflows                        
                         
Debt Securities                        
                         
Banking Flows                        
                         
Banking Characteristics                        
                         
Directed Credit                        
                         
Banking Supervision                        
                         
Increase in Credit                        
                         
                         
Inflows and their Impacts                        
                         
Outflows/GDP                        
                         
Real Appreciation                        
                         
Change in Reserves                        
Key Red implies greater risk of a crisis higher
is variable
E.g. Higher portfolio inflows imply greater
risks, as does faster increase in credit and
appreciation
25
On LAC Financial ReformCapital, Liquidity,
Complexity, Discipline
26
1. Bank Capital

• Level and Quality
• Cyclical Behaviour
• Basel II Approaches
• Provisions

27
Bank Capital where was the crisis?
Regulatory Capital to Assets at Risk. Source IMF
GFSR
28
Quality Not QuantityCapital Increases by Type
2000-2008

• Source Archyra et al (2009) for a sample of
  major banks in each region

29
Basel II vs. Basel IIISchematic Representation
of Requirements
National and Individual Cyclical Buffer (0-2.5)
ea.
Capital Conservation Buffer (0-2.5)
Minimum Total Requirement 8
Tier 2
Additional Tier 1
6
Tier 1
4.5
Core Tier 1
30
Quality of Capital in Brazil and Mexico
50 of these Financial Systems operate with Tier
1 Buffer of about 5 (i.e. 13.5 of assets at
risk 6 2.5 5)
31
What Should LAC do?

• LAC should publish core tier 1 and tier 1 ratios
• And adopt Basel III perhaps with stricter rules
  (as LAC has adopted Basel I and II with stricter
  ratios)
• Analytical work required to consider what those
  stricter ratios should be

32
On Anti-Cyclical Capital Rules

• Basel IIIs cyclical buffers force banks to hold
  more capital in the good times, with an
  expectation it will be used in the bad
• But LAC banks already hold significant capital
  cushions, perhaps to avoid hitting the
  requirement levels.
• Anti-cyclical rules may have less effect than may
  be imagined, depending on why we think banks hold
  these buffers, as banks may reduce the optimal
  buffers held
• Aliaga-Diaz, Olivero and Powell (2011) show this
  formally in a SDGE model calibrated to LAC
  countries.
• Bottom Line If Basel III rules are applied on
  total capital, they will have little effect.
  Assuming they are applied to Tier 1 Capital (as
  they should), and banks are expected to increase
  tier 1 capital by 2.5 in the best simulations,
  we find consumption volatility may be reduced by
  4-5.

33
What Should LAC do?

• Adopt anti-cyclical capital rules
• These should be specified on core tier 1
• Given the amplitude of cycles in LA, these rules
  should be more aggressive than Basel III
• Work needs to be done to calibrate the rules
  appropriately

34
On the Basel II Approaches

• LAC countries have taken different views
• IDB Project on Basel II Implementation in the
  Region
• Brazil Simplified Standardized and IRB for
  qualifying banks (no Credit Ratings)
• Chile Standardized Approach (i.e. Using
  External Ratings)
• Colombia Initially IRB, now more SA
• Uruguay (and others) Softly-Softly, gradually
  adopting measures consistent with Basel II
• Are the Basel II approaches a good fit for
  Emerging Economies?

Basel II Pillar 1 Approaches Simplified
Standardized Approach (SSA) Basel
1 Standardized Approach (SA) Use of Credit
Ratings and a table to map to requirements Interna
l Rating Based Approach (IRB) - Banks rating
methodologies and estimated PDs and a standard
curve
35
Sailing through the Sea of Standards

• Powell (2002, 2004) and Majnoni and Powell
  (2005)1 made the following arguments
• A lack of rated claims reduces the value of
  Standardized Approach
• Arguably, the Internal Rating Based Approach
• gives too much autonomy to banks
• requires very significant supervisory resources
  beyond many countries
• is too much of a black box and the calibration is
  suspect for LAC (correlation assumptions and
  99.9 tolerance)
• At the same time, many countries have centralized
  rating systems to determine provisions
• Majnoni and Powells proposal
• A Centralized Rating Based (CRB) approach would
  have banks rate claims according to a standard
  scale, this has a cost as banks must use a common
  scale but a huge benefit in terms of ease of
  supervision
• Should be calibrated appropriately

1 Majnoni, G. and A. Powell (2005). Reforming
Bank Capital Requirements Economia, Spring 2005.
See also Powell (2002) and Powell (2004).
36
On Provisions

• Many countries in Latin America maintain
  provisions on the basis of losses incurred
• Basel III now calls explicitly for forward
  looking (expected loss) provisions and
  negotiations are underway with the accountancy
  standard setters.
• Some countries have made progress on
  Anti-Cyclical provisions (e.g. Bolivia,
  Colombia, Peru, Uruguay)

37
Galindo and Rojas-Suarez propose an index of
theQuality of the Provisioning Regulation

• Micro-prudential questions
• Does it reflect expected loss (and if so over
  which categories of loans)?
• Does it specify provisions over other assets or
  only loans (investments, contingent claims, other
  assets)?
• Macro-prudential questions
• Does it have an anti-cyclical component?
• Are provisions counted as capital?

38
Survey Results
Source Galindo and Rojas-Suarez (IDB,
forthcoming)
39
2. Liquidity

• Heralded as part of Basel IIIs macro-prudential
  tools.
• Suggested characteristics of liquidity rules to
  be macro-prudential
• Should be held in assets that remain liquid even
  if there is macro (systemic) stress.
• Should be held and used for systemic purposes
• This indicates a system of requirements in a
  central bank or centralized custodian with tough
  rules on their composition
• Basel III liquidity rules do not appear to
  satisfy these characteristics

40
Liquidity vs. CapitalAlternative Views

• 1) Two Risks Solvency Risk vs. Liquidity Risk
• 2) Size Matters capital requirements limit size,
  liquidity rules are a restriction on asset
  composition
• 3) Capital (the difference between of assets and
  liabilities) may be volatile and may disappear,
  if liquidity is cash in the central bank, its
  actually more real
• Appropriate liquidity policy depends on what you
  think is the problem and the role of liquidity

41
And what does/should LA do?

• Several countries have liquidity requirements
  (remunerated, so less of a tax) and banks must
  satisfy in the central bank and in some more
  sophisticated cases may satisfy in other
  centralized custodians
• Flexible tools can be adjusted over the cycle,
  remuneration rates can penalize anti-social
  behaviours such as lending in dollars or riskier
  (more flighty) types of liabilities
• Work needed to determine the appropriate level
  (Charlie Calomiris advocates 20) for developed
  countries

42
3. Complexity Cross Border

• Basel N (N1,23), Pillar II, still contains the
  myth that the role of a Supervisor is to
  supervise a Bank.
• But Lehmann Bros. had 2985 legal entities
  operating in 50 countries with many supervisors.
• Pillar II focuses on a lead Supervisor
  supervising a consolidated entity. Majnoni and
  Powell (2005, 2007) argued that Consolidated
  Supervision may be necessary but it is NOT
  sufficient.
• Pillar 2 should state explicitly what the role of
  the different supervisors should be when there
  are many significant entities with different
  regulators

43
Complexity Cross Border

• If a Subsidiary (or a Branch) is significant for
  the Host that entity should be jointly supervised
  and the Host should be the primary agent to
  Monitor and Supervise.
• The Host should have access to any information
  from the Home (consolidated or lead) supervisor
  that it needs to be able to execute its duties.
• When times get tough, voluntary regulatory
  cooperation will break down, Colleges are not
  enough.
• The Board of Subsidiaries in host countries must
  act only in the interests of the subsidiary
• G20/BCBS needs to work more here as international
  law is inconsistent (single vs. multiple
  entity resolution) and the legal position
  regarding international banks responsibilities
  to their local depositors remains murky.

44
What Should LAC do?

• LAC is host to many legal entities in complex
  corporate structures where its unclear how a
  Subsidiary fits into a structure, action is
  required
• LAC should make very explicit governance rules
  for subsidiaries and board members fiduciary
  duties
• Host supervisors should know how any actions by
  the home supervisor (such as Living Wills,
  Chinese Walls or other measure may affect
  subsidiaries in their country)
• More analytical work required.

45
South-South Complexity

• LAC needs to pay attention to increased
  South-South cross border banking, in South
  America, Central America and the Caribbean.
• LAC Supervisors may wish to agree rules on cross
  border supervision.
• An interesting recent example of the dangers come
  from the Caribbean The CLICO Group (CL
  Financial) in Trinidad and Tobago

46
Complex structure across many countries
Source CCMF Report on CL Financial
47
Cross Border Issues Abound in the Caribbean
Financial System (Commercial Banks Only)
Haiti, Guyana and Suriname in the Periphery
TT in the Center
By ownership by market
Network visualization by Pajek software for a
short description see Batagelj V., Mrvar A.
Pajek - Analysis and Visualization of Large
Networks. in Jünger, M., Mutzel, P., (Eds.) Graph
Drawing Software. Springer, Berlin 2003. p.
77-103
48
Complexity InstrumentsStandardization vs.
Innovation

• The recent crisis has amply demonstrated that
  much financial innovation is motivated by
  regulatory (and tax) arbitrage
• LA countries perhaps given the French Law
  tradition and limited liquidity have favored
  Standardization
• The world may now focus more on Standardization
  and less on innovation banks more akin to
  utility companies
• Standardization also eases Supervision and the
  possibility to collect meaningful data -
  information.
• Indeed non-standard instruments, if permitted,
  should attract higher capital and/or liquidity
  requirements.

49
4. Market Supervisory Discipline

• Market discipline and supervisory discipline
  intimately related, the crisis was a failure of
  both.
• Financial innovation allowed practices that were
  not detected by regulators, heightened the
  information asymmetries Powell, Miller and
  Maier 2011.
• Perhaps market has some information not available
  to supervisors
• Market and supervisory discipline are complements
  (re Argentinas BASIC system in the 1990s
  Information, Auditing, Credit Rating, Bonds,
  Supervision).
• An automatic market trigger may also limit
  forbearance
• Cocos provide additional capital and a market
  trigger Switzerland has two triggers, preemptive
  and crowd-in level.

50
A Proposal Lima 1
51
A Proposal Lima 1

• Given the unprecedented crisis in the North,
  Basel N (N1,2,3) has been driven by the North,
  arguably the sequence is moving further away from
  a good fit for LAC
• There is a danger that a standard is being lost
  across LAC
• More importantly LAC has a lot to offer
• It has learned from its own history
• It has implemented many macro prudential
  measures
• It is blessed with good data (and economists)
• At a meeting of ASBA held in Lima, I proposed a
  LAC Accord - Lima 1
• This is meant to be fully consistent with Basel N
  which are minimum standards.

52
Selected Proposed elements of Lima 1

• A LAC-calibrated basic requirement, likely to be
  gt8, and with appropriate core tier 1/tier 1
  requirements likely to be higher than Basel III.
• A Centralized Rating Based approach consistent
  with IRB procedures but calibrated to LAC
  (subject to the Basel minimum) and facilitating
  supervision.
• A LAC standard on forward-looking, anti-cyclical
  provisions
• Anti-cyclical capital rules, more aggressive than
  Basel III
• Standardization of contracts or possibly higher
  capital (and/or liquidity) requirements on
  non-standard products
• Possible use of Cocos in LAC
• Systemic remunerated liquidity requirements,
  lower remuneration rates for foreign currency and
  short term non-core liabilities
• Clear statements on what LAC supervisors expect
  from home supervisors to supervise the
  subsidiaries and branches of foreign banks and
  what LAC authorities expect from the Board of LAC
  banks that happen to be part of complex groups
• These elements additional or compatible with
  Basel N