II. FAILURE OF REGULATION AND SUPERVISION 1
II. Failure of Regulation and Supervision Recurrence and severity of financial crises The 2007-09 financial crisis has a number of lessons, many common to previous episodes No doubt there were market/private sector failures; discussed in previous section But regulation is supposed to correct those failures also failed miserably!!! Monetary Policy also an important culprit 2
3
A Narrative of the crisis (cont.) 4
5
Immediate Causes of the Crisis Crisis was caused by an excessive expansion of credit and a price bubble in the housing sector The expansion of credit/leverage happened through the growth of secured lending and structured products (not in M2!!!) Credit growth fuelled the housing price bubble, leading to weak lending standards, and a large misallocation of resources Jorge Roldós Causes of Crises 6
Causes of Crises Jorge Roldós 7
Jorge Roldós Causes of Crises 8
250 1000 900 200 800 700 Index or Interest Rate 150 100 Building Costs Home Prices Population 600 500 400 300 Population in Millions 50 200 Interest Rates 100 0 0 1880 1900 1920 1940 1960 1980 2000 2020 Year Source: R. Shiller Jorge Roldós 9
Deeper Fundamental Causes Why did this huge misallocation of resources happen? Confluence of a number of forces/factors We organize the many causes into three categories: 1. Monetary Policy Mistakes 2. Market/private sector failures 3. Regulatory Failures Jorge Roldós Causes of Crises 10
1. Monetary Policy Mistakes The crisis started after a period of low macroeconomic volatility and very low interest rates (dubbed the Great Moderation ) Fed easing from 2001 until 2004 led to very low short term interest rates (next chart) Low risk-premia (for both market and credit risk) the so-called connondrum produced very low longer term interest rates Prompted a search for yield by investors worldwide Jorge Roldós Causes of Crises 11
Fed s Easing Beyond Taylor Rule Source: J. Taylor, 2007 Jorge Roldós 12
1. Monetary Policy Mistakes Difference between actual policy rates and optimal ones is even larger if one adjusts for cyclically-low risk-premia Market risk Credit risk Emerging market central banks (especially Asia) reinforced this by intervening in FX markets (to resist appreciation) and accumulating large amount of int l reserves Global imbalances also driven by U.S. financial system intermediating needs of over-stimulated US consumers and under-stimulated surplus countries consumers Jorge Roldós Causes of Crises 13
1. Monetary Policy Mistakes Monetary policy models ignored asset prices [more in L-7] Difficulties diagnosing bubble [L-3], led to policy of cleaning-after-the-fact Reinforced by Greenspan put (an implicit floor for asset prices), an invitation not to store enough liquidity for bad times (moral hazard) Jorge Roldós 14
2. Market/Private Sector Failures We can classify the main market/private sector failures into three main classes: a. Incentive problems (a result of imperfect information/contracts) b. Incomplete markets (say, for liquidity in some states of the world, shorting housing) c. Herding, coordination problems, and externalities Perfect financial markets would correct some of these failures; regulation should take care of others Jorge Roldós Causes of Crises 15
2a. Incentive problems (1) A number of incentive problems were created by the shift in the financial system to an originateto-distribute (O2D) model Securitization segments the credit process, with the originator of the loans selling them to an arranger, who repackages them into securities assessed by rating agencies Distance from investor to debtor, and lack of transparency in some of the stages, weakens ability to evaluate/monitor creditworthiness Jorge Roldós Causes of Crises 16
2a. Incentive problems (2) Rating agencies are paid for by issuer: incentive to do ratings shopping Weaker lending standards also related to use of credit scoring models (reliance only on a few indicators, no soft/character information) Agency problems (shareholders versus managers/traders/bankers) were aggravated by compensation systems: one-sided bonuses linked to short-term returns Jorge Roldós Causes of Crises 17
2a. Incentive problems (3) Product complexity and the associated lack of transparency impeded market and regulatory discipline on the financial system Excessive leverage magnified the above incentive problems, with FIs engineering and undertaking a large amount of tail risk The fact that they kept a large share of that risk directly or indirectly on their balance sheets is a sign that they expected a (too-bigto-fail, TBTF) bailout (moral hazard) Jorge Roldós 18
2a. Incentive problems (4) Fannie Mae and Freddie Mac were privatelyowned companies with a public-policy objective When asked to increase their share of lowincome house-finance, bought subprime-abs Some estimate that, adding the FHA, gov t mandated loans amount to almost 2/3 of all junk mortgages Some pension funds and foreign banks based their investment decisions purely on ratings, with out further due diligence Jorge Roldós Causes of Crises 19
2.b Incomplete Markets When short-term creditors do not roll-over funds ( run on the FI), or require higher margins, FIs have to raise funds when markets are closed to them In other words, markets for liquidity insurance in bad times are non-existent Fire-sale externalities can be damaging to FI balance sheets Counterparty or network externalities Few opportunities to short housing market means prices have tendency to rise excessively (bubbles? [L-3]) Limits to arbitrage could lead to persistent mispricing of assets (bubbles?) Jorge Roldós Causes of Crises 20
Bottom line: Financial institutions and markets are supposed to correct/control information problems Since they are likely to be unable to correct all incentive problems and externalities, there is a role for regulation But regulation may also fail and may even aggravate some of the incentive problems it is meant to solve Jorge Roldós Causes of Crises 21
3. Regulatory Failure Regulations failed for a number of reasons: 1. Lack of resources to understand LCFI strategies and incentive problems, to extend regulation to all relevant parties, and to enforce the existing ones 2. Regulatory capture by the industry or politicians 3. Regulatory mistakes or gaps Examples from the 2007-09 crisis: Jorge Roldós Causes of Crises 22
3.a Lack of regulatory resources The regulatory perimeter (L-8) in the U.S. was restricted, in part because of lack of resources, in part because of underestimation of linkages Some states did not license mortgage brokers, much less monitor their behavior Finance companies largely unregulated The complexity of products and business strategies could not be understood by a few regulators checking the balance sheets of LCFIs Explicit recognition of head of the OCC Jorge Roldós 23
3.a Lack of regulatory resources 1992 legislation allowed Fannie&Freddie to hold less capital than other FIs Also determined that their new regulator, an office within HUD with little financial sector experience, was subject to congressional appropriation Jorge Roldós Causes of Crises 25
3.b Regulatory capture LCFIs lobbied U.S. authorities to repeal Glass- Steagall Act, allowing commercial banks into investment banking activities Investment banks lobbied the SEC to have a voluntary net-capital rule, using own risk models; allowed for unrestricted leverage Fannie&Freddie mandate to fund low-income mortgages expanded during Clinton ( Affordable Housing ) and Bush ( Ownership Society ) administrations Jorge Roldós Causes of Crises 26
3.c Regulatory gaps/mistakes Basel rules allowed for regulatory arbitrage between the banking and trading books, using guarantees or reducing RWA (L-2) A number of explicit and implicit guarantees were underpriced Deposit insurance Too-big-to-fail Lack of an orderly resolution mechanism for LCFI and other non-bank FIs Jorge Roldós Causes of Crises 27
3. c Regulatory gaps/mistakes Lack of regulation on liquidity mismatches (L-2/3) OTC derivatives markets were allowed to grow without transparency or central clearing Regulation focused on individual institutions rather than on systemic risk contribution [L-6] Regulators did not internalize that risk-taking is endogenous Jorge Roldós Causes of Crises 28
Deeper Fault Lines (R. Rajan) Political-economy arguments: facilitate credit to buy houses when politicians cannot solve problem of stagnant middle-class Intersection of politics and finance: although some housing subsidies are OK, when you give a wall of money to financiers Global imbalances are a result of overconsumption/financial deepening in US, opposite in surplus countries Jorge Roldós Causes of Crises 29
III. SYSTEMIC RISK REQUIRES NEW INSTRUMENTS AND INSTITUTIONS Jorge Roldós 30
III. Systemic Risk Requires New Instruments and Institutions Improvement in Supervision and Regulation (Sup&Reg) framework requires a definition of macro-prudential policies and systemic risk More fundamental diagnosis of Sup&Reg failures suggest new, improved instruments Central issue: regulation has to focus on system (macro-pru), not just institutions alone (micro-pru) Sup&Reg framework also requires new institutional arrangements 31
A. Definition and Objectives The objective of macro-prudential policy is to promote financial stability by limiting systemic risk Financial stability means a stable provision of financial intermediation services, i.e. a stable supply and cost of credit, insurance, others Realistic: cannot expect to eliminate credit cycles and/or target asset prices
Systemic Risk Systemic risk can be defined as the risk of disruption to financial services that is caused by an impairment of all or parts of the financial system and has the potential to have serious negative consequences for the real economy (FSB, IMF, BIS, 2009) The sources of systemic risk are basically the ones that we identified as causes of the current crisis in the previous section (also BoE, 2009)
B. Macro-Prudential Instruments Micro-Prudential regulation based on deposit insurance (DI) and capital regulation [L-2] Goal: internalize losses on bank assets to minimize moral hazard and protect DI fund Key element: prompt-corrective-action (PCA) to restore capital adequacy ratio (CAR) Next we provide two examples of why this framework proved inadequate in last crisis 34
An example from Morris and Shin (2008) Bank 2 in the next Figure is a very safe bank by Basel I-II standards: 1. Assets are reverse repos, collateralized 2. Liabilities (repos) are matched with assets Bank 2 could increase its leverage beyond 30X However, Bank s 2 assets are systemically important, because they mirror Bank 1 liabilities
An example from Morris and Shin (2008)
An example from Morris and Shin (2008) If Bank 2 does not rollover reverse repo with Bank 1 (a prudent policy for B2), and B1 assets are illiquid, then B2 s prudent decision is like a run on B1 and causes a fire-sale of assets Bank 2 assets are safe (requiring low CAR under Basel for the individual bank) but are systemically important: they should NOT have a low risk weight
An example from Morris and Shin (2008) This example shows the importance of interconnectedness of the different Financial Intermediaries (FI), not just banks, in the system One tool to identify this interconnectedness is the network approach (to be discussed in L-4 and W-1 of the course) Another approach is through the measurement of Co-VaR, co-movement of bank portfolio returns whenever another bank is under stress (a tail risk event, L-5 and W-2) 38
An example from Hanson, Kashyap, and Stein (2011) Financial intermediaries (FI) amplify or magnify business cycles In the boom phase, asset values increase and so does equity (by a multiple), inducing procyclical behavior (as in the FA + Firesales) Pro-cyclicality means borrowing more to take advantage of the higher value of assets (or quality of loans): the endogenous response of risk-taking is important
increase in value of securities increase in equity Final balance sheet assets equity debt assets equity debt assets equity debt initial balance sheet After q shock new purchase of securities new borrowing
An example from Hanson, Kashyap, and Stein (2011) Hanson, Kashyap, and Stein (HKS, 2011) argue that the problem is not just with capital adequacy ratios (CAR) at the individual level per se Equity CAR = = Assets The problem with the micro-prudential approach is with prompt corrective action (PCA): regulator does not care whether bank adjusts via the numerator (raise equity) or the denominator (lower assets) E A
An example from Hanson, Kashyap, and Stein (2011) If bank adjusts by shrinking assets, causes firesale and/or reduction in lending to real economy Macro-prudential sup® objective then is to minimize the social costs of balance-sheet shrinkage ( fire sales and credit crunches ) This alternative theory of regulation does not rely on the existence of deposit insurance and suggests it has to be applied to more than just deposit-taking institutions
A Set of Macro-prudential Tools In line with their proposal to focus the regulatory system in restricting socially excessive balance-sheet shrinkage, HKS(2011) discuss six sets of tools that can be useful: 1. Time-varying capital requirements (in L-3) Higher (lower) in good (bad) times Exceed the market-imposed standard in bad times 2. Higher-quality capital (in L-2) Common equity better than preferred stock
A Set of Macro-prudential Tools 3. PCA targeted at dollars of capital (not CAR) Relative to lagged assets As in the U.S. Supervisory Capital Assessment Program 4. Contingent Capital (in L-2) Contingent Convertibles Capital Insurance Economize on equity, requires regulatory blessing
A Set of Macro-prudential Tools 5. Regulation of Debt Maturity and Asset Liquidity (L-2) Limit runs Contain fire-sale externality 6. Regulation of the Shadow Banking System (in L-8) Collapse of ABS market very damaging in this crisis Any institution that acquires ABS and funds them with short-term debt Avoid margin spirals Impose similar capital requirement on a type of credit exposure/asset (Geanakoplos, 2010)
Other Macro-Prudential Instruments There are a number of other macro-prudential instruments to be adopted In the time dimension (in L-3) In the cross-section dimension (in L-4, L-5) Also a number of different taxonomies (by BIS, IMF, others) We will discuss issues related to the effectiveness of some of these instruments, which have been already used in many EMs
Cross-border Bank Resolution Problem: LCFI operate globally but resolution regimes are guided by local legal framework Living wills would be very useful tool IMF recently proposed a framework of enhanced coordination to deal with problem Ideally countries with major financial centers would agree to the enhanced framework in the near future (in L-6)
C. Framework for Macro-Prudential Regulation and Supervision Macro-Prudential Sup&Reg is much harder in practice than in principle If somebody has to do it, is central banks This may complicate monetary policy and require broader objectives; models (in L-7) Different countries are setting up different institutional architectures to carry out this important function (in L-6)
Framework for Macro-Prudential Regulation and Supervision U.S. and E.U. have already established their systems, led by a Financial Stability Oversight Council (FSOC) and the European Systemic Risk Board (ESRB); others in EMs Both gather main regulators/supervisors with central bank officials In case of U.S., Office of Financial Research (OFR) has authority to request info from FI
AThe E.U. new financial architecture ESRB in charge of macro-prudential policy recommendations 51 Macro-prudential oversight European Systemic Risk Board ECB National central banks European Supervisory Authorities European Commission Risk warnings National Supervisors (non-voting) Representative of EU finance ministries (non-voting) Macro-prudential recommendations Micro-prudential supervision European System of Financial Supervision European Banking Authority European Insurance and Occupational Pensions Authority European Securities and Markets Authority National supervisors (including supervisory colleges) EU-wide technical supervisory standards Coordination of supervisors (also in crises)
Rules versus Discretion Rules are desirable, especially to provide support in application during upswing and visà-vis powerful LCFI However, constrained discretion is probably the maximum one will be able to achieve To be successful, will need to be transparent and accountable
Final Thoughts Transparency can go a long way towards reducing systemic risk Tightening of micro-prudential is likely to push activities to non-bank financial intermediaries: macro-prudential policy needs flexibility and authority to regulate and supervise them International harmonization and cooperation (including info sharing) are critical
Thank you