A New Capital Regulation For
Large Financial Institutions
Oliver Hart
Harvard University
Luigi Zingales
Motivation
• If there is one lesson to be learned from the
2008 financial crisis, it is that large financial
institutions (LFIs) are too big to fail.
• Why?
– Economic rationale
• Fear of Armageddon
• Possibility of creating value (ex post)
– Political rationale
• Time inconsistency • Asymmetric loss
Motivation - 2
• A regime in which LFIs won’t be allowed to go fail can impose large costs on society.
– large future taxpayer losses to cover bail-outs. – moral hazard
• We already see signs of problems:
– Large banks, which before the crisis could borrow at 29 basis points below small banks, can borrow at 78 basis points below today.
– For the 18 bank holding companies with more than $100 billion in assets the 49bps advantage corresponds to a $34.1 billion subsidy a year.
• Given that society will bear the ex post costs of LFI
failures there is a strong argument for regulating them ex ante.
How to regulate?
• Many suggestions have been made:
– limit the size of LFIs;
– restrict the kinds of activities an LFI can undertake; – encourage LFIs to issue convertible debt
– write living wills;
– constrain executive pay;
– require contingent capital insurance, etc.
• Rather than micro-managing an LFI, we propose an early warning system that will alert the regulator to the fact that an LFI is in trouble.
• Regulator can intervene before the damage spreads to other institutions and social costs are incurred.
=>We propose a market-based mechanism that achieves this goal.
Idea
• We distinguish between
– “systemically relevant” obligations, e.g., short-term interbank borrowing
– “non-systemically relevant” ones, e.g., long-term debt.
• Long-term debt mostly held by mutual funds
and pension funds, which can absorb losses on
this debt in the same way that they can absorb
the losses on equity investments.
• Our approach is to protect the systemic
obligations in all circumstances but leave open
the possibility that the non-systemic
Our Mechanism
• We would require banks to hold two layers of capital below fragile, systemic obligations. First, as currently, a layer of
equity. The second layer would consist of long-term debt that is junior to the systemic obligations.
• Our mechanism mimics the way margin calls function.
– Describe
• LFIs will post enough collateral (equity) to ensure that their debt is paid in full.
• If the fluctuation in the value of the underlying assets puts junior debt at risk, LFI equityholders will be faced with a margin call and they must either inject new capital or face intervention by the regulator.
Some important differences from standard margin calls: LFI assets are not easily valued, and creditors are dispersed and cannot easily act.
So how do we know when the second cushion of long-term debt is in trouble? One can exercise political pressure on a credit rating agency or a regulator. But it’s hard to influence a market.
Therefore we use price of CDS on LFI junior debt as a trigger, and regulator to coordinate action.
• Trigger mechanism: CDS price. A CDS is a contract that promises to exchange a bond with an amount of cash equal to the bond’s notional value in the event of default. The price of this contract in basis points is the insurance premium paid every year on a notional amount of $100 of debt. By arbitrage the price satisfies
where π is the (risk neutral) probability of default and the recovery rate is the proportion of the value of the debt recovered in the event of default.
(1 recovery rate) 10000 CDS p
• The trigger is activated if the CDS price rises and stays above a threshold for an extended period of time. During this period the LFI can raise equity to
bring the CDS price back down. If this effort fails and the CDS price stays above the threshold for a
predetermined period of time ( say its average over the preceding month exceeds 100 basis points),the regulator intervenes.
• If the regulator intervenes, she first inspects the firm—in effect, carries out a stress test– and :
• If she decides the debt is not at risk, she declares
the firm adequately capitalized and injects some
government funds that are pari passu with
existing financial debt.
• If she decides the debt is at risk, she replaces the
CEO with a receiver ( trustee).
• The trustee eliminates all the debt except for the systemic obligations and runs the new “debt-lite” firm until he can find a cash buyer
(alternatively, he may raise cash by recapitalizing the firm and carrying out an IPO).
• Any cash raised is used to pay off creditors partially—however, they receive a haircut of at least 20%. Shareholders are wiped out . Any remaining funds go to the taxpayer.
• The stress test is important to eliminate the
incentive to carry out bear raids.
• The haircut is important to ensure that the CDS
price is informative about the risk of default on
the junior debt.
1) The Model
• Our model focuses on the agency benefits of debt. We assume that the LFI manager can “steal” a
fraction of the cash flow available after having paid down the debt.
• Idea: managers can pay themselves large bonuses as long as the firm does not become insolvent
afterwards.
• In the basic model we do not distinguish between systemically relevant and non-systemically relevant debt. Relax later.
• A two-period model with the following structure—it is as much a model of GM as Bank of America or Goldman Sachs.
p1 p2 (1 – p1) (1 – p2) (1 – p3) V1 V2 V3 V4 p3 1 2 0
Assumptions
• The firm’s capital structure consists of a
choice of long-term debt D due at time 2.
• Capital structure is set in a value maximizing
way at time zero
• At time 1, the LFI manager can modify the
capital structure by issuing equity only if he
has shareholders’ approval.
• At time 2, the company pays out the cash flow
V and terminates.
• The market is risk neutral, and the interest rate
is zero.
In the absence of any debt, the market value of the LFI
(which we label VU, i.e., value of the unlevered firm) would be
If we introduce a debt D such that V4 < D < V3, then the market value of the debt VD at issue will be
and the total value of the levered LFI (VL) will be
]. ) 1 )( 1 ( ) 1 ( ) 1 ( )[ 1 ( p1p2V1 p1 p2 V2 p1 p3V3 p1 p3 V4 VU , ) 1 )( 1 ( ] ) 1 ( ) 1 ( [p1p2 p1 p2 p1 p3 D p1 p3 V4 V D . ) 1 )( 1 ( ] ) 1 ( ) 1 ( [p1p2 p1 p2 p1 p3 D p1 p3 V4 V V L U
The Unregulated Outcome
• Since there is a benefit, but not a cost, of debt, the
value of a LFI is monotonically increasing in the level of debt.
• An unregulated value-maximizing LFI will pick a debt level that would lead to bankruptcy with
probability one.
• A regulator could impose a debt level less than or equal to V4.
• This would eliminate bankruptcy risks, but impose a high cost for the LFI.
The Regulated Outcome
• Consider a time-zero debt level D such that V4 < D <
V3.
• At time 1, if the realization is good -> debt not at risk • If the realization is bad, then debt becomes risky =>
LFI receives a margin call, i.e., it is forced to raise more equity.
The Regulated Outcome -2
By diluting the entire value of existing equity, LFI can raise
3
(1
)(
3).
p
V
y
D
Hence feasibility requires
,
)
)(
1
(
3 3V
y
D
y
p
which implies that for a debt level D to be made riskless through a margin call it must satisfy
).
)(
1
(
3 4 3 4p
V
V
V
D
The Regulated Outcome -3
y p D D y V p p D V p p D V p p V L (1)[ 1 2( 1 ) 1(1 2)( 2 )(1 1) 3( 3 )] (1 1).
)
1
(
1 4 1D
p
V
p
V
V
L
U
• LFI valueSubstituting the value of y, we obtain (1)
• Since (1) is increasing in D, it will be optimal for the LFI to set D at the maximum level compatible with the financing constraint.
• Substituting in (1) and rearranging we obtain: (2) VˆL VU V4 p1p3(1)(V3 V4).
• Above supposes that states of world are verifiable. We propose a mechanism that works when states of world are not verifiable.
• Trigger mechanism: CDS price. A CDS is a contract that promises to exchange a bond with an amount of cash equal to the bond’s notional value in the event of default. The price of this contract in basis points is the insurance premium paid every year on a notional
amount of $100 of debt. By arbitrage the price satisfies
where is the (risk neutral) probability of default and
the recovery rate is the proportion of the value of the debt recovered in the event of default.
(1 recovery rate) 10000 CDS p
• The trigger is activated if the CDS rises and stays above a threshold—in the model, zero-- for an
extended period of time. During this period the LFI can raise equity to bring the CDS price back down. If it does not, the regulator intervenes.
• If the regulator intervenes, she first inspects the firm—in effect, carries out a stress test-- and
– If debt is not at risk, she declares the firm adequately capitalized and injects some
government funds that are pari passu with existing financial debt.
– If debt is at risk, she replaces the CEO with a
receiver ( trustee), who recapitalizes and sells the firm for cash, ensuring in the process that
shareholders are wiped out and creditors receive a haircut—say 20%.
Timing
Figure 1: Timing
First shock LFI decides Market price of CDS Regulator Second is realized whether to issue observed decides shock is
realized at equity whether realized
at date 1 to intervene at date 2
Key Result
Proof:
A) Suppose lower branch and LFI raises less than D – V4 in equity => it cannot be a rational expectations equilibrium for the regulator not to intervene: there is a positive
probability that the debt will not be paid at date 2, and the CDS price will reflect this.
Proposition 1: Assume Then the
equilibrium price of a CDS, pCDS will be greater than zero if and only if the lower branch of the tree is followed and the LFI raises equity with value less than D – V4 at date 1.
). )( 1 ( 3 4 3 4 p V V V D
• Suppose instead that market expects regulator to intervene.
• The regulator will find that the LFI is under-capitalized and so he will reorganize, imposing a creditors’ haircut => the CDS price will be positive.
• Thus the unique rational expectations equilibrium is for the CDS price to be positive and for the regulator to
B) Lower branch and LFI raises equity equal to D – V4. If the regulator intervenes he will find that the debt is not at risk, and so he will do nothing. The debt is also not at risk if the regulator does not intervene. Thus the unique rational expectations equilibrium in this case is for the CDS price to be zero and for the regulator not to intervene.
C) Upper branch of the tree is followed. Then the debt is not at risk, and so the unique rational expectations
equilibrium is one where the CDS price is zero and the regulator does not intervene.
Systemic vs. non-systemic debt
• Suppose that systemic debt can be issued at a
lower interest rate than non-systemic debt.
• If our CDS mechanism works perfectly, the
regulator should not fear 100% systemic debt
• If concerned about “out of equilibrium” sequence:
– manager does not, issue equity at time 1;– regulator is also unable to issue equity; – regulator is unable to sell the company.
• If occurs, the firm will default on its systemic
debt, possibly leading to a public bail-out.
4
How can this be avoided?
• (a) limit the fraction of total debt that can be
systemic;
• (b) make the systemic debt senior.
• In the model the most systemic debt that can
always be paid back at date 2 is , and so the
fraction of systemic debt should be limited to
4 V 4 4 3(1 )( 3 4) V V p V V
• Requiring that an LFI issue non-systemic junior long-term debt has another benefit.
• For CDS prices to provide useful information, the
underlying instrument should face the risk of default, at least out of equilibrium.
• Junior long-term financial debt plays this role.
• In theory, it is irrelevant how much junior long-term debt there is, as long as there is some.
• In practice, the amount is important because it determines the thickness of the market for the security underlying the CDS.
Why the CDS?
• Equity not good because
– Affected by the upside
• Other debt-like instruments (bonds, yield spreads) good as long as
– Liquid
– Not easy to manipulate – Easily observable
• CDS is where price discovery first occurs
– It leads the stock market (Acharya and Johnson, 2007), the bond market (Blanco et al, 2005) and even the credit rating agencies (Hull et al, 2004).
• CDS should be traded in a regulated market and properly collateralized
Layer of junior debt
• The junior long-term debt cushion has a double function:
• 1) It provides a security that can support the CDS
• 2) It provides an extra layer of protection for the systemic obligations
• Minimum amount of long-term debt should be mandated by regulation
Injection of government
funds
• The injection of government funds is designed to
– Make it politically costly to say that the LFI debt is not at risk – Protect systemically relevant contracts (which are senior) from
the regulator’s mistakes
• Political cost maximized by making the government claim junior to financial debt
• But we want to reduce lobbying pressure from claimholders -> debt senior
Endogenizing LFI activities
• Investment has a cost of i and return R
with probability π and r otherwise.
• Realization of this investment opportunity
is perfectly correlated with the value of the
underlying assets.
• This introduces an additional agency cost:
– manager captures a fraction of the upside ofany investment (in the form of stealing), he suffers no downside cost.
Proposition 3:
Under the CDS trigger mechanism, no
negative NPV investments will be
undertaken.
• Manager better off if
– Investment is positive NPV (he can steal a fraction of it;
– or new equity is issued (he can steal a fraction of it).
• In second case shareholders do not
approve
Our rule eliminates all the
agency costs of debt.
1) It eliminates incentives to undertake
negative NPV investments for risk-shifting
reasons.
2) It eliminates debt overhang problem by
forcing equityholders to issue equity
when debt becomes risky.
3) It eliminates any discretion in the decision
to issue equity, removing any signal
Would This Rule Have Worked?
(Bps of premium to insure against default)
Financial Institution 8/15/2007 12/31/2007 3/14/2008 9/29/2008 BoA 11 29 93 124 CITI 15 62 225 462 JPMORGAN 19 32 141 103 WACHOVIA 14 73 229 527 WAMU 44 422 1,181 3,305 WELLSFARGO 23 45 113 113 BEAR STEARNS 113 224 1,264 118 GOLDMAN 28 78 262 715 LEHMAN 38 100 572 1,128 MERRILL 29 159 410 666 MORGAN 31 129 403 1,748 AIG 31 59 289 821
•
False vs. True Positives
"Failed" institution Date of Average CDS Average CDS Default 6 months 9 months
before before BEAR STEARNS 3/14/2008 121 10 LEHMAN 9/15/2008 288 106 WAMU 9/25/2008 957 430 WACHOVIA 9/30/2008 176 45 MERRILL 9/15/2008 282 177 AIG 9/16/2008 234 70 CITI 9/30/2008 162 44
"Surviving" Institutions False Positive Date with a Trigger at 100 40 BoA 9/22/2008 1/22/2008 WELLSFARGO 9/18/2008 11/23/2007 JPMORGAN 9/29/2008 2/15/2008 GOLDMAN 2/14/2008 8/20/2007 MORGAN 11/13/2007 8/22/2007
The Main Alternatives
Coco bonds ( Squam Lake Report)
• Debt that converts into equity when a trigger is set off.
• Advantage: This does not require any resolution authority
• Disadvantages: 1) Which trigger?
– Market price of equity -> manager can talk down value of the bank to obtain equity on the cheap
– Accounting numbers -> possibility of manipulation – Political decision -> lobbying, influence
Coco bonds -2
2) They do not enhance protection of systemic
obligations, only delay bankruptcy
– Our mechanism forces equity issues, boosting the protection offered to systemic claims
Bail In (Debt for equity swap)
• equity -> warrants
• preferred & sub debt -> new equity
• senior unsecured debt -> 15% new equity
(85% no change)
• No impact on customer positions, repo,
swaps or insured deposits
• Management is removed
• What triggers it?
• Huge political return from delaying pulling
the trigger.
How does mechanism compare with
the Dodd-Frank Bill?
• Resolution authority useful step but
– Not clear what the rules of impairment are – What triggers intervention?
• Too late
• Too clumsy
• Our mechanism could be implemented in the
context of Dodd-Frank
• Possible private response to Government
Regulation
Does It Help to Avoid Systemic Crisis?
• 2 reasons why an LFI failure has systemic effects:
1) Losses on the credit extended to the insolvent LFI can make other LFIs insolvent.
– Our mechanism eliminates this problem since no LFI will become insolvent.
2) The failure of an LFI can force assets’
liquidation leading to downward spiral in asset prices
– Our mechanism does not force any asset liquidation, thus avoiding a downward spiral in assets prices.
Other Advantages
1) Easy to apply across different institutions
(banks, hedge funds, insurance companies).
2) Except for the new resolution and trigger
mechanism, not very far from existing capital
requirements.
3) Easy to implement in an international setting.
4) The mechanism encourages early action: banks
must issue equity well before they are close to
default. A crisis is nipped in the bud.
A Recent Application:
How Not to Be Systemically
Relevant
• Fed can establish a lower CDS threshold
saying that if you
– stay below
– have the required amount of junior long-term debt
you are deemed non systemic
• As soon as you violate one, you become
fully regulated
Conclusions
• The too-big-to-fail problem arises from a combination of
– an economic problem : cost of default on systemic obligations is very large
– a political economy problem: time inconsistency induces the government/regulator to sacrifice the long-term effect of
permitting default to avoid the short-term costs
• Our mechanism addresses both these problems.
• It is similar to existing and proposed capital requirements:
– two layers of protections for systemic obligations: equity capital and junior long-term debt.
Conclusions -2
• It differs in
– trigger mechanism (based on CDS) – resolution mechanism.
• Very importantly, our mechanism encourages early action: banks must issue equity well before they are close to default.
• Credit default swaps have been demonized as one of the main causes of the current crisis. It would be only fitting if they were part of the solution.
