Reflections on Financial Sector Restructuring Andrew Sheng - - PowerPoint PPT Presentation

1

World Bank/IMF/SEACEN Regional Seminar Comparative Experiences in Confronting Banking Sector Problems in the Asia/Pacific Region

2 – 3 December 2003 Kuala Lumpur, Malaysia

Reflections on Financial Sector Restructuring

Andrew Sheng Chairman Securities and Futures Commission, Hong Kong

Paper co-authored with Ms GL Tan – all views and errors are personal to authors

2

Historical Overview of Financial Crises

• 1980s and 1990s were decades of financial crises

– 1980s revealed external debt crises of Latin America and first signs of endemic NPLs in LDC banks – 1990s saw Mexican crisis, followed by Asian crisis, and then problems in Argentina – Asian crisis was not public debt crisis, but private debt crisis, hence consensus on key role of corporate governance – Asian crisis saw rapid contagion – local pain can cause global pain, hence strengthening of architecture

• Since then, considerable cumulative experience in crisis

management

– International financial architecture/surveillance strengthened at both domestic and IFI level, e.g. FSF and FSAPs – World Bank/IMF/BIS/OECD etc have greater technical experience and working together better – Greater awareness of importance of global standards, transparency and accountability at all levels

3

Restructuring, Recovery and Fatigue

• Y2K and 9/11 2001 revealed operational risks from globally

linked markets – Markets are networks – Systemic failures are actually network failures

• Asian Restructuring since 1997 has been painful

– Costs up to 50% of GDP – Basically, private sector NPLs were carved out to AMCs and funded by public debt – Major emphasis on corporate governance reforms

• Generally Asia is in better shape after post-crisis reforms

and current recovery is gaining strength

• But why are NPLs still high despite growth?
• Are the problems more fundamental and structural?

4

Banking Crises and Resolution

• Crisis is an event, the culmination of many

factors interacting together

• Bank restructuring and resolution is a process
• Process involves four major steps:

– Diagnosis – Damage Control – Loss Allocation – Rebuilding Profitability and Getting the Incentives Right

• How have we performed since Asian Crisis?

5

What is a Financial Crisis?

• The eruption of an event (e.g. failure of a

corporation or financial institution) that triggers a systemic distress, panic or wealth loss that spreads within domestic markets or abroad

• Crises are caused by internal frailties or

weaknesses that allow systemic breakdown as a result of internal or external shocks

• Because financial system is a network, need to

distinguish between liquidity crisis [flow] from solvency crisis [stock]

• Banking system need lender of last resort,

precisely because Central Bank can step in to prevent a liquidity crisis from triggering a solvency crisis.

6

Stocks and Flows of National Economy

• Financial system is “blood circulation system” of national
• economy. Financial network links the following four other

sectors together:

– Corporate sector – Household sector – Public sector – External Sector

• Each has its balance sheet and flow statements
• Financial sector fulfils four basic functions:

– Resource allocation – Price discovery – Risk Management – Corporate Governance

• Crisis occurs when weaknesses in real and financial sectors

are exposed by event/shock (globalization creates network shock)

7

Vicious Cycle of Financial Distress

Asian crisis: private debt mismatches Latin American crisis: excessive public debt and inflation

Enterprises experience foreign exchange and loan losses Devaluation

• ccurs

The fiscal impact creates a need for bank recapitalization Commercial banks are decapitalized Central bank refinancing leads to money creation Inflation rises higher than the world rate

8

Ultimately, Financial Crisis Ends Up as Quasi-fiscal Deficit

• Banks have implicit or explicit deposit insurance, i.e. moral

hazard risks

• Depositors or foreign creditors cannot absorb losses

without huge political implications

• Both banks or borrowers can be “Too Large to Fail” –

government is concerned that failure can have systemic problems

• Hence, banks or borrowers transfer their stock or flow

losses to the Government via Government guarantees, Asset Management Companies (AMC), or bail-outs

• Relationship is known as Troika model
• Borrower insolvency passed on to become bank insolvency

(NPLs) – if LOLR also perceived as insolvent (i.e. no dollars to settle local convertible currency), then capital flight

9

The Troika Model

1

BUDGET BANKS CORPORATES

Tax and borrow from banks Tax derived from corporations Recapitalize through AMCs Banks exposed to corporate debt deposits Trade credit Enterprises

10

Much Written on Bank Restructuring

World Bank:

• Financial Crisis and Financial Restructuring

– 18 papers presently posted on website

http://www1.worldbank.org/finance/html/fp-financial_crises_and_financ.html

• World Bank Finance Research: Bankruptcy and Resolution of Financial

Distress – 3 papers currently posted

http://econ.worldbank.org/programs/finance/topic/bankruptcy/

IMF:

• Lessons from Systemic Bank Restructuring

Claudia Dziobek and Ceyla Pazarbasioglu

http://econ.worldbank.org/programs/finance/topic/bankruptcy/

• FSAPs useful source on financial health

http://www.imf.org/external/np/fsap/fsap.asp

• David S. Hoelscher and Marc Quintyn, Managing Systemic Banking Crises,

IMF Occasional Paper (forthcoming)

• McKinsey (2003): Banking in Asia: Acquiring a Profit Mindset (2nd Edition)

11

Crises have both Macro and Micro Origins

• Poor macro policies, e.g. fiscal deficits, inflation, balance of

payments deficits

• Weak institutional structures

– Lack of deep debt and capital markets – Lack of credit culture – Outdated laws, weak judicial systems – Poor corporate governance

• Lax enforcement, poor risk management, connected or directed

lending, weak loan recovery, deposit guarantees, and distorted tax policies all show up in weak balance sheets

• Weak players, low standards of performance and efficiency,

insufficient oversight and unclear rules of game, make financial network vulnerable to shocks

12

Lessons of the 1980s’ Banking Crises

• Financial stability rests on maintaining stable currency
• Banks fail because of losses in real sector, compounded by poor

risk management and fraud

• Liberalisation overlooks wealth effects of relative price changes;

losses worsened by inadequate supervision

• Bank losses often become quasi-fiscal deficits
• Failure recognition is important as banking crisis is a solvency

problem, not a liquidity issue

• Stopping the flow of future losses is critical
• Method of loss allocation determines success of restructuring
• Success depends on sufficient real sector resources to absorb

losses, financial sector reforms, and budget’s ability to tax “winners” and wind down “losers” with monetary stability

• Good policies, reliable management and strong institutional

framework needed to rebuild safe and profitable banking systems

• Time and timing are of the essence

Source: Sheng (ed) Bank Restructuring: Lessons from the 1980s, The World Bank, 1996

13

Asian Crisis vs 1980s lessons

• Stable currency vital – volatility concern in 1990s
• Real sector issues – crony capitalism blamed for poor corporate

governance

• Hasty liberalisation – balance sheet effects of interest rate and

exchange rate volatility overlooked

• Bank losses often became quasi-fiscal deficits – cost of resolution

up to 50% of GDP in crisis economies

• Solvency problem, not a liquidity issue – yes
• Stopping future losses critical – did not ensure bank standstill in

time

• Loss allocation – basically governments underwrote loss through

debt swap or AMC

• Reform success – by and large crisis over in three years
• Rebuilding safe and profitable banking systems – profitability back,

but so are bad habits

• Time and timing are of the essence – cannot be complacent

14

How Good is Diagnosis?

• Institutional strengthening

– Coordinated surveillance by IFIs, FSF, FSAP – International standards set by BIS, IOSCO, IAIS, IMF – Strengthened supervisory framework and cooperation at national and international level

• Adoption and implementation of international standards by local

networks would strengthen overall network, eg IAS, OECD corporate governance codes, insolvency laws

• Generally getting better at diagnosis, but informational

difficulties remain:

– Asset valuation is difficult as there are no market prices for loans – Collateral valuation also serious problem in estimating provision needs – Inadequate provisioning due to tax changes that need reform – Banks generally reluctant to reveal extent of losses, unless forced by crisis or incentive to shift loss to AMC

• NPL estimates by market vary and at least double official data

15

What is NPL?

• NPL is loss due to bad risk management of credit to customers

– Weak credit assessment – Poor collateral, either bad assets, fraud, or inability to recover assets due to poor insolvency laws – Loan forbearance, including provision of further loans hoping that borrower will recover

• NPL is revelation of economic losses, after application of IAS

and stringent loan provisioning according to Basle standards

• In most cases, NPLs reflect losses that have already incurred

[stock losses]. Forbearance means unwilling to face revelation.

• Damage Control means that banks or supervisors must have

systems in place to minimize further losses and exposures to problem borrowers

• Consequently, NPLs are actually revelation of inefficiencies or

losses of bank creditors, when they get marked to market using international standards. The quality of corporate governance determines the level of inefficiencies and hence NPLs.

• Alternatively, NPLs arise because of unclear property rights, so

that borrowers pass losses to banks and then to the State

16

NPLs in Asia Still High

As at the end of 2001* $ billion % of total net loans China 345 25 Japan 346 9

• S. Korea

8 4 Indonesia 4 12 Malaysia 14 11 Philippines 6 17 Thailand 9 11 India 14 11 Taiwan 32

7

Source: McKinsey & Company’s Asia Financial Institutions and Corporate Finance &Strategy practices (2003), Banking in Asia Acquiring a Profit Mindset (2nd Edition)

17

NPLs Likely to Increase

• In 2002, Ernst and Young [NPL Loan Report: Asia 2002] estimates

Asia’s NPLs at $2 trillion, up 33% and double official estimates – Japan accounts for $1.2 trillion or 60% – China accounts for $480 million or 25%

• As a percentage of GDP, the estimated NPLs range from about

12% to 45% of GDP

18

Damage Control

• Because NPLs derive from real sector exposures, they

can be minimized in two ways:- – (a) Restrict further bank lending (quick fix) – (b) Work on changes in corporate governance and debt recovery (long haul solution)

• You must stop both the stock and flow losses
• Faster IFI response with better recognition that no

“one size fits all” solution can apply

• Selective bank closures or foreign entry helps to keep

domestic banks on their toes

• The fact that NPLs have been reduced in some

economies does not mean that they will not revive if structural issues (e.g. poor borrower solvency, connected lending, weak corporate governance, policy distortions, weak institutions, political instability) are not resolved.

19

Stemming New NPLs

• In China, McKinsey studies suggest that despite strong growth, 5-

8% of loans turn sour within first year (pp 15).

• Unless credit culture improves, NPLs will rise, particularly with

focus on collateral rather than cash flow

– 80% of bank retail accounts in Japan lose money (McKinsey, pp 48)

• Incentives must be right, e.g. separate social lending from

commercial lending so that management is accountable

• Market discipline must be strengthened –

– Weak companies can be taken over, i.e. markets must be

• contestible. High PE ratios discourage takeovers.

– Mispricing of risk: spreads are too small relative to risk. McKinsey suggests bank spreads in Japan must double from 125 basis points to 250 basis points (pp 48) – Make profits in line with International norm for ROA of 1.2%. McKinsey studies: 0.4% in China, 0.8% in S. Korea, overall weighted average ROA of 0.9% for Indonesia, Thailand, Malaysia and Philippines (pp 61, 106, 189) – Greater transparency and disclosure: less than 15% of top companies in Asia have credit ratings (McKinsey, pp 17)

20

Loss Allocation

• Political economy of loss allocation:

– Who should bear loss – shareholders, borrowers, banks, employees, depositors, external creditors, government? Ultimately, state bore most of losses [e.g. in one Indonesian bank, state took out US$19.8 bn out of $25 bn assets, before it was sold to foreign joint venture]

• Liquidate or restructure banks:

– market solution or government intervention? Mixed approaches: closures, mergers, nationalisation, foreign investment and government bailouts

• Key concern is fiscal sustainability:

– unresolved NPLs, as quasi-fiscal deficits increases contingent liability of budget and therefore become potential future tax burden

21

Why is NPL Resolution Slow?

• Rule of thumb:

– Outstanding (stock) NPLs of 5% is acceptable; 10% considered a problem; 15% and above, a disaster – New (flow) NPLs of around 2% p.a. normal

• Low interest rates make banks reluctant to dispose collateral, and

also help sustain non-viable firms, delaying corporate restructuring or exit. E.g. McKinsey estimates that 20% of Japanese corporations have not repaid principal in over a decade and loans are rolled over (pp 48)

• No incentive for banks to restructure once loans provisioned
• Organisational culture that focuses on blame rather than

corrective action results in compromised decisions that do not deal decisively with problems in order to evade responsibility.

• Strong resistance from vested interests
• Hope to grow out of problem. McKinsey estimates China needs to

grow 20% p.a. for the rest of decade to bring NPL down to 5% of

• utstanding loans (pp 110)

22

Other Views on Asian NPL Resolution

• S & P [China Banking Outlook 2003-2004]

– Many banks in Asia still have weak assets, with poor earnings and capital base still weaker than pre-crisis levels. – China’s “Big Four” unlikely to reduce NPL ratio of 26.1% to 15% by 2005. To reduce NPL to 5% likely to take 10-20 years and an estimated $500 billion or 40% of 2002 GDP.

• FSSA on Japan based on end-March 2002 data show that

capital of 7 city banks would suffer big loss under market stress:

– 44% loss on shareholder equity if equity prices fall by 20% – 33% loss for 100 basis points increase in yields – 94% loss for 3% credit loss on loan book

And average Tier-I ratio of major Japanese banks as at March 2003

would fall to 2% from 5% if deferred tax assets are deducted from capital http://www.imf.org/external/pubs/cat/longres.cfm?sk=16865.0

23

Fiscal Sustainability

• inflation, growth and budget
• Fiscal sustainability of debt, and hence success of bank

restructuring, depends on rate of inflation, growth and level

• f budget deficit.
• The change in government debt to GNP ratio (d):

Change in d = (primary deficit/GNP) – (seigniorage/GNP) + d (real interest rate – growth rate) [Fisher and Easterly]

• Low inflation, high growth and low fiscal deficits or small

surplus economies will see d decline over time. E.g. Malaysia, Spain and Chile

• Large primary deficits and excessive real interest rates

result in unsustainably huge debt ratios, leading to

• hyperinflation. E.g. Argentina and Yugoslavia
• Change in d also depends on whether NPLs are checked

and bank losses stemmed

24

Fiscal Sustainability

• NPLs, external debt amid low interest rates

DSGAsia Estimates, 10 November 2003

25

Rebuilding Profitability and Getting

Incentives Right

• Retaining old bank management in place, and simply carving out

NPLs without changing incentives could lead to new NPLs. McKinsey study suggests that Asian focus has been on consolidation rather than profit.

• Success of restructuring depends on implementation of the right

incentives for self discipline and market discipline to work based

• n risk return considerations, supported by regulatory discipline,

i.e. effective enforcement that promotes observance of rules and regulations.

• So far, performance of AMCs mixed. Historically 1 of 3

restructuring agencies successful and 2 of 4 rapid asset disposition agencies successful [Daniela Klingebiel, World Bank]

• Progress on enterprise or corporate restructuring has been
• uneven. Risk that bad borrowing habits are creeping back in.

26

NPLs Reflect Asian Imbalanced Growth Strategy

• Asia’s Growth Model relied on bank-dominated financial

system to finance its competitive export sector and growth. But NPLs mostly in weaker protected domestic industries and services sectors

– This led to distortions in resource allocation and prices, weak risk management and poor corporate governance

• NPLs reflect these distortions and inefficiencies
• To strengthen its global competitiveness

– There is a need for more complete markets that allow for better risk management, and systems for better protection of property rights of savers – Asia needs to correct its structural imbalance

• Balanced growth strategy means that we must treat global

resource allocation as a national risk diversification policy

• See Sheng (2003) Stanford paper :“The Future of Capital Markets in

East Asia: Implications for China’s Equity Markets”

27

Incomplete Financial Development Increases Risk in the Economy

• Asia’s capital markets not fully developed due to inadequate

protection of property rights, institutions, and laws, i.e. incomplete markets.

• Banks dominate as they are able to bridge informational

asymmetries through bank-firm relationships.

• Result is high leverage, and studies show that banks tend to

assume higher risk compared to capital markets where risk-taking is disciplined by the required rate of return. (Allen and Gale, 1994)*

• Studies show that corporations exhibit riskier financing patterns

where property rights are weaker. (Rajan and Zingales, 1999)*

*Source: Stijn Claessens, Simeon Djankov, Tatiana Nenova (2000), Corporate Risk Around the World, World Bank Policy Research Working Paper No. 2271, January

28

Markets, Networks and Property Rights

• The economy is a network of markets – banks, capital markets,

retirement funds, real estate, consumers and investors, producers, government services, etc – that interact with one another to generate output, employment and growth

• Market is a network across which property rights are defined,

delineated, exchanged, protected and enforced by market participants (inc. SROs and regulators)

• Network stability depends on integrity [solvency] of participants,

strict network standards, clear rules of game, and network

• versight [supervision/regulation] to ensure network operates

efficiently and remains stable

• As domestic networks converge with global networks, differences

in standards, conduct, and weaknesses in interconnectivity, interoperability creates shocks in system that can be inter-active

• In other words, market stability requires network stability, and

depends on property rights integrity!

29

Market/Network Architecture and Property Rights

• Foundation for well-functioning markets and networks is therefore a

property rights infrastructure (PRI), that comprises standards, codes, rules, laws, and the institutional framework, such as judiciary, regulators, service providers/intermediaries that define, delineate, transfer and protect property rights.

• Efficiency of financial markets depend on the quality of 6 “I’s” and 1

“E”: – Accurate, timely and accessible information (auditing, accounting, disclosure standards) – Properly aligned incentives – Educated investor – Efficient and competent intermediaries (lawyers, auditors, accountants, analysts, brokers, financial experts etc) – Strong issuer – Efficient and robust trading infrastructure – Strong prudential framework with enforcement

30

NPLs Reflect Weak PRI

• Why would borrowers want to repay debts if:

– Property rights are unclear? – If businesses can pass losses to government? – Courts do not enforce collateral? – Bankers can be bribed? – Banks would not pursue them?

• Why would banks want to recover loans or recognise bad loans if:

– If bank staff are underpaid, under-trained and cannot enforce loan recovery? – Borrowers have strong political influence? – Loan classification rules are unclear, and auditing and accounting standards are not stringent? – Supervision and enforcement are weak? – The state will bail them out?

• Hence, NPLs are a reflection of weaknesses in the PRI

– Bank restructuring process assumes that well functioning PRI exists to minimize NPLs after carve out. – If PRI is still dysfunctional, NPLs will still flow.

31

Efficient Market Economy Means Building a Robust Property Rights Infrastructure (PRI)

• 1. Delineation of property rights

– Property rights need to be clearly defined and legally protected

• 2. Enforcement of property rights

– Property rights need to be protected and enforced efficiently, fairly, and predictably through independent judiciary and regulatory systems

• 3. Culture in respecting law, property rights and contracts

– Credit culture means respect of law, property rights and contracts, and they are rewarded for doing so

• 4. Reform of market institutions that protect rights

– The systems of accounting, regulation, judiciary and property registration (e.g. land and equity registration systems) need to brought up to international standards in order for markets to perform efficiently, fairly and transparently

32

Architecture of PRI

• Delineation of property rights: Information, Standards, Codes,

Rules, Laws, Contracts, lodged in custody: by owner, agent/trustee, bank, registry

• Exchange of property rights: DvP – holder must notify broker,

registrars [e.g. land registry, share registry], custodians, banks, lawyers etc

– Trading [e.g. stock exchange or OTC] – Clearing, settlement and payment infrastructure – clearing house and payment system

• Protection of Property rights:

– Intermediaries have to be regulated for discipline against misconduct – Clear Rules of Game – norms, standards, codes, regulations, law, transparent to all participants – Enforcement Infrastructure – enforcement costs cannot exceed benefits to market – Independent and transparent judiciary to adjudicate property disputes

33

Service Sector is mainly PRI

• Delineation of property rights: Accountants, Lawyers, Media,

Analysts, Rating Agencies all help interpret, confirm or assure information on property rights,

• Exchange of property rights: Stock and derivative exchanges,

OTC markets or even physical markets require specialist intermediaries, eg brokers, banks, retirement funds, registrars, custodians, to facilitate smooth exchange of rights, with lower risks and hopefully, lower transaction costs

• Protection of Property rights:

– Self-Regulated Organizations + NGOs all engage in protection

• f their members’ property rights

– Regulators, Judiciary and Government agencies’ duty is to enforce and protect property rights by – setting clear rules of the game, enforcing regulatory discipline and minimizing regulatory costs

• In global competitive game, most robust network with lowest

transaction cost and protected property rights wins!

34

Information Component of PRI

• Markets need accurate, timely and readily accessible

information to make good decisions and assess risks

• Quality of information depends on:
• Accounting standards (eg IAS)
• Disclosure and transparency rules
• Quality of audit and corporate governance
• Good disclosure enhances accountability and checks

and balances against abuses-

• Misconduct or incompetence of corporate managers
• Corruption or incompetence of public officials
• Corruption or misconduct of bank staff

35

Restructuring Financial Systems = Rebuilding Networks

• Economies are a web of interlocking networks, the

efficient functioning of which can be affected by the weakest part of the network

• Policy objective is to strengthen all networks so that a

malfunctioning in one sector will not cause system wide failure, i.e. gridlock or a national blackout. The backup systems should kick in to contain and minimise damage.

• In financial markets, this entails:

– Recapitalising banks – Enhancing standards of efficiency, transparency, measurement to make the networks of markets more robust – Ensuring that institutions and systems are in place to ensure that networks operate by the rules of the game

36

Process to Manage Reform

• Asia has implemented many reforms, but outcomes may

not always be on target.

• Reform fatigue could have set in resulting in no follow

through to ensure successful implementation.

• Law of unintended consequences may frustrate reform

efforts and generate reform resistance.

• Reform is a process, but we need a process to manage

the reform process so that it stays the course and departures from path can be put back on track through set procedures.

• Reform needs ownership of the need for change.
• Process to manage change is an important area that

deserves more attention by policy makers.