170 likes | 351 Views
Discussion by Jeffrey Zwiebel. Should Derivatives be Senior Patrick Bolton and Martin Oehmke. TexPoint fonts used in EMF. Read the TexPoint manual before you delete this box.: A A A A. Summary:.
E N D
Discussion by Jeffrey Zwiebel Should Derivatives be SeniorPatrick Bolton and Martin Oehmke UBC Winter Finance Conference – 3/5/2011 TexPoint fonts used in EMF. Read the TexPoint manual before you delete this box.: AAAA
Summary: • Derivatives and repos effectively enjoy seniority in bankruptcy as collateral can be seized right away, unlike other claims (they are exempt from the automatic stay on collections in bankruptcy) • Does this exemption make sense? Is this a good law or not? • Paper develops a model to analyze these questions • Very sensible and clear model that includes many of the considerations that are discussed • Results generally indicate that this seniority is unwarranted, and does not promote efficiency, and rather can leadto excessive use of derivatives
Three effects of Derivatives • Benefit of cash management preventing inefficient liquidation • Cost of exogenously given hedging cost using derivatives • Asset substitution effect. Ex-post transfer from debtholders to shareholders if derivative can be made senior to debt.
Benefit of Derivatives • The benefit is: Provide valuable risk management: • Simple Bolton-Sharfstein style contracting friction of nonverifiable cash flows yielding inefficient liquidation and inefficient ex-ante investment: • Financing F needed for an investment, raised with debt of face value R • Cash in period 1 from an investment is either CL or CH, but nonverifiable. Cash in Period 2 is C2 and can always be diverted. • If CL < R is paid back in P1, creditor liquidates, but this is inefficient. No renegotiation is allowed here.
Benefit of Derivative Continued • A derivative is a contract on random variable Z with outcome ZL or ZH with the realization of Z correlated with firm cash flows • Key point: Cash from derivatives contract is verifiable, and can therefore be used to hedge project risk and pay creditor • Hence the derivative allow a firm to verifiably transfer cash (that can be committed for payments) from good to bad states on average, yielding less inefficient liquidation
Cost of Derivative • Cost of Derivative: Use of derivatives is assumed to involve a hedging cost based on notional amount of the derivative • This is a pure deadweight cost • ½(X) for notional value of X • Social optimum determined by trading off this hedging cost with the risk management benefit
Seniority and Transfers • Asset Substitution: If derivatives can be issued senior to existing debtholders, there is an ex post transfer from debtholder to shareholders • Seniority effectively follows from the exemption for derivatives to the automatic stay under bankruptcy, together with the pledging of collateral to derivatives • Paper argues that under current law, it is not possible to commit not to undertake such asset substitution (more on this later) • Of course, debtholders anticipate this ex-ante, and demand a higher interest rate • This in turn leads to more default and less financings if derivate hedge is imperfect, which is a real social loss • If instead derivatives were junior to debtholders (or equivalently, are without collateral), then there is no asset substitution effect
Results • Results follow from these 3 effects. Among these results: • Lower required face value for debt when debt is senior vs. when derivatives are senior • Junior derivatives yields less inefficient failures to finance and less inefficient liquidation subject to financing than senior derivatives • A constraints that restricts derivative collateral to a partial level will yield results in between junior and senior derivative • Asset substitution may lead to the use of derivatives beyond the point where they provide any risk sharing benefits
Comments: • Very nice intuitive story that gets to the heart of what many have in mind with derivatives and seniority • Senior derivatives given a “fair shot” here, insofar as they are endowed in the model with a valuable use that comes from a natural contracting friction • The use of derivatives raises welfare • Key point is derivatives need not be senior to serve risk management role • And effective seniority requires higher debt which yields more inefficient liquidation
A few quick questions • Paper includes repos into the same discussion as derivatives. But is this the same issue? Repos are surely not serving the same risk management role as derivatives may be serving • Similarly, the paper discusses netting provisions for derivatives as similar to exemption for automatic stay in conferring effective seniority. But don’t many other creditors effectively get similar netting treatment? • Why no renegotiation? Paper claims unimportant. Not immediately clear.
Law vs. Contracts • A key question discussed in paper is whether can contract around legal treatment of seniority • Shareholders would like to be able to commit when issuing debt not to endow future derivative counterparties with seniority through collateral • Paper argues may not be able to commit to doing so, insofar as this is ex-post optimal • And recall, do not want to restrict derivatives, which are beneficial, but only collateralization that make them senior
Contracts Continued • Not obvious why clause in debt could not prohibit collateral for derivatives, or require that derivatives be treated as junior claimants to debt in bankruptcy • Why has this not been done? • Not possible. Hard to define all derivatives in debt contract • Not desired for some reason outside the model • Extent of potential expropriation of debtholders through effective seniority of derivatives was not well-appreciated in practice before, but we will see such restrictions going forward
Deadweight Loss with Derivatives • What is the source of this hedging cost ½(X)? • On one hand seems natural, that counterparty may face some costs of this risk that are passed on to firm • This is of course a common assumption in the literature • Debtholders and shareholder however face very similar risks in the model, without such a cost • Model has a partial equilibrium feel when we consider extra costs for holding derivatives and not other securities without a clear story
A Different Story for Derivatives for Financial Firms • Possibly not for risk management at all, but for speculation • Paper acknowledges something like this briefly under certain parameters when asset substitution is extreme • Consider a particular type of speculation: • Suppose financial firms use derivatives (and leverage) to create distribution with extreme left tail risk • Most of the time yields constant gain • In a rare state, lose a lot • This is likely to be easy for a financial firm to do, in a manner that is hard for outsiders to observe
Tail Risk - Continued • Why? Managerial career concerns • That is, focus on Managerial-Shareholder conflict rather than Shareholder-Debtholder conflict • Would be in managers interest if • Manager cared about relative performancein some nonlinear manner (say from career concerns and firing), and • Either market is unaware of risky position, or the market can infer it but not observe it • In such a setting, managers might not care much about the tail event (they get fired anyways), hence bankruptcy rules might not affect behavior much • However, suppose that there is an externality from financial crisis, which can be worsened if derivatives are not settled • This is a very different story that could perhaps justify the exemption of the automatic stay for derivatives
Conclusion • Clean and compelling model of an important issue • Results are intuitive and believable • Highlights and clarifies several critical considerations with regard to current legal treatment of derivatives under bankruptcy • Focuses attention on a number of interesting new questions