|
|
| (9 intermediate revisions by the same user not shown) |
| Line 1: |
Line 1: |
| [[File:Crossing Threshold Hope.jpg|400px|thumb|right|an equally forlorn attempt to influence matters beyond one's control]] | | #redirect[[Cross Default - ISDA Provision]] |
| This is a page about the general, generally stupid, concept of {{t|cross default}}.
| |
| ==Before we start==
| |
| ===As a standard term in [[master trading agreement|master trading agreements]]===
| |
| For specific provisions see:
| |
| *{{isdaprov|Cross Default}} ({{tag|ISDA}})
| |
| *{{gtmaprov|Cross Default}} ([[GTMA]])
| |
| *{{efetprov|Cross Default}} ([[EFET]])
| |
| *{{fbfprof|Cross Default}} ([[FBF]])
| |
| *'''[[Cross Default - GMSLA Provision|Stock lending and repo have no cross default]]''': Neither the {{gmsla}} nor the {{gmra}} have, as standard, either a [[cross default]] or a [[default under specified transaction]] provision. ''[[Cross Default - GMSLA Provision|Unless some bright spark thinks it is a good idea to negotiate one in]].''
| |
| | |
| ===Compare and contrast===
| |
| *'''[[Default under specified transaction]]''': {{t|DUST}} is ''like'' [[cross default]], but just between you and me, on [[derivative]] transactions and without a {{isdaprov|Threshold Amount}};
| |
| *'''[[Cross acceleration]]''': like [[cross default]], only for a kinder, gentler world where people wait for [[indebtedness]] to be actually [[accelerated]] before closing out their exposures.
| |
| | |
| ==History==
| |
| ===[[Cross default]] in the [[loan]] market===
| |
| [[Cross default]] developed in the loan market. If a [[lender]] advanced a large sum to a [[borrower]] with only periodic interest or principal repayments, there would be long periods — months; quarters; even ''years'' — where the borrower was not scheduled to make any payments to the lender at all.
| |
| | |
| Now a borrower that is not due to pay anything, can hardly [[Failure to pay|fail to pay]].
| |
| | |
| This presented our lender with a risk: if, in the meantime, the borrower failed to pay under a loan from ''another'' lender, ''our'' lender would be in a difficult spot: it has good reason to think the borrower is in trouble, but the borrower hasn’t missed any payments. (How could it? ''None were due''.) Waiting for the next payment to see if the borrower will pay won’t do. Our borrower wants to accelerate its loan ''now'' — while the going is still tolerably good.
| |
| | |
| Whence came the notion of a [[cross default]]: ''If you default under a loan you have borrowed from someone else, you default under your [[loan]] with me.''
| |
| | |
| But this is a drastic measure. It means the borrower and its various lenders are in a Mexican stand-off: The lenders will all tend to be trigger happy: they will want to accelerate before some other blighter does. Therefore some thresholds were put around it: The size of the loan being defaulted on would need to be material enough to threaten the borrower’s very solvency.
| |
| | |
| Note the key vulnerabilities that [[cross default]] clause is designed to protect against:
| |
| *'''Material indebtedness''': Our lender has significant credit exposure to the borrower;
| |
| *'''Infrequent payments''': Our lender is owed infrequent payment obligations and cannot necessarily rely on a [[failure to pay]].
| |
| *'''Material default''': The borrower has taken on ''other'' indebtedness in a size big enough to threaten its own viability.
| |
| | |
| ==={{isdaprov|Cross default}} in the {{isdama}}===
| |
| In their infinite wisdom (or jest), the framers of the [[1987 ISDA Interest Rate and Currency Exchange Agreement]] ([[cro-magnon man]] to the {{2002ma}}’s [[metropolitan hipster]]) thought it wise to include a [[cross default]], perhaps because, in those pioneering days, [[Credit Support Annex|credit support annexes]] weren’t run-of-the-mill, and may not even have been invented.
| |
| | |
| Subsequent generations of [[Mediocre lawyer|derivative lawyers]], being the creatures of habit they are, especially when sequestered into an [[ISDA working group]], never thought to take it out, and even our artisanal coffee-swilling {{2002ma}} boasts a tedious {{isdaprov|Cross Default}} provision, an embarrassing relic of its bogan parentage. It’s like that tattoo you got when you were a drunk, but physically attractive, 19 year-old.
| |
| | |
| You see the thing is, for a derivative [[master agreement]], [[cross default]] is a complete nonsense.
| |
| | |
| A counterparty to an {{isdama}}, particularly one with a zero-threshold daily {{isdaprov|CSA}} and many {{isdaprov|transaction}}s under it, suffers none of those weaknesses it is designed for:
| |
| *'''Little indebtedness''': An {{isdama}} is not a contract of [[indebtedness]], and any [[mark-to-market]] exposure that may ''resemble'' indebtedness is zeroed daily by means of a collateral call;
| |
| *'''Frequent payments''': particularly where there are many transactions, or where the net mark-to-market position is shifting, there are payment obligations flowing every day, ''and if there are not that means there is no net indebtedness at all to the counterparty''.
| |
| | |
| Additionally, regulated [[credit institution]]s have (or should have) enormous concerns about giving away cross default, because it can affect their liquidity buffer calculations.
| |
| | |
| Yet still we persist in our sophistry.
| |
| ===Then the [[lawyer|lawyers]] and [[credit officer|credit officers]] start fiddling with things===
| |
| Cross default is a bad enough idea in a [[derivatives]] [[master agreement]] in the first place, before [[risk managers]] start having a go at it. Misguided things they can do include the following:
| |
| *Widening it to include default under agreements which aren’t in the nature of [[indebtedness]]: for example, [[derivatives]], or even “any payment obligation”.
| |
| :*This is problematic because of the accretive nature of the threshold: A single technical or operational failure may mean one is technically in default on payments which, if aggregated, could quickly exceed even a large threshold (especially in a heavily traded derivative master agreement).
| |
| *Not, in the case of banks, excluding [[deposit|retail deposits]], where operational failure or even governmental action (like a moratorium or currency controls) could lead to technical default on a large amount of indebtedness. (Bank deposits are a form of indebtedness, and will almost certainly be a significant source of indebtedness for any trading bank).
| |
| *Adding in [[grace period]]s or other preconditions, excuses, permission to skip PE class and so on, before a party may invoke a [[cross default]];
| |
| *Arguing the toss about [[threshold amount]]s (should it be shareholders funds or cash? or both? lower or higher of? Is my threshold higher than yours? Is it too big? Is it too small? Does my {{isdaprov|Threshold Amount}} look big in this? Honestly it is so tedious).
| |
| | |
| ==Introduction==
| |
| A cross default provision in an agreement allows a [[non-defaulting party]], on a [[default]] by the other party under any separate contract it may have entered for [[borrowed money]], to [[close out]] the agreement containing the cross default provision. Compare this with:
| |
| *a [[cross acceleration]] provision, where the lender of the [[borrowed money]] must actually have taken steps to accelerate the [[borrowed money]] as a result of the default before the default becomes available as a termination right under the first agreement; and
| |
| * [[default under specified transaction]] which references default under financial contracts which do '''not''' represent indebtedness, but only as between the two counterparties to the present contract.
| |
| | |
| Cross default is potentially a very damaging clause, as this picture to the right amply illustrates. Or would do, if there were a picture to the right. To the extent it doesn't:
| |
| | |
| ===Cross default===
| |
| A cross default right effectively imports into the [[ISDA]] all the default termination rights under any {{isdaprov|Specified Indebtedness}} owed by a party:
| |
| *It dramatically (and indeterminately) widens the definition of {{isdaprov|Event of Default}}.
| |
| *It entitles a [[Counterparty]] to [[cross accelerate|accelerate]] the {{tag|ISDA}} whether or not the {{isdaprov|Specified Indebtedness}} itself has been accelerated.
| |
| *Depending on the market value of the {{isdaprov|transaction}}s under the ISDA it may cause an immediate capital outflow (though is less likely to in these days of compulsory variation margin).
| |
| | |
| ==={{isdaprov|Specified Indebtedness}}===
| |
| {{isdaprov|Specified Indebtedness}} means, generally, any [[borrowed money|borrowings]] that, in aggregate, exceed a designated {{isdaprov|Threshold Amount}}. Because of the aggregation right, even comparatively trivial agreements can trigger the provision where they are relatively homogenous and affected by the same local circumstances (for example, [[retail deposit]]s). A low {{isdaprov|Threshold Amount}}, therefore, presents three challenges:
| |
| *It allows a more varied (and difficult to monitor) range of potential termination rights, because a greater number of agreements will qualify as {{isdaprov|Specified Indebtedness}}.
| |
| *It “lowers the bar” so failures to comply with comparatively trivial financial commitments could be aggregated to trigger the {{isdaprov|Cross Default}}.
| |
| *By not excluding [[bank deposit]]s, it raises the possibility of being triggered by localised events unrelated to a bank counterparty’s creditworthiness (for example, political action in a single jurisdiction which affects the bank's ability to pay on its local deposits)
| |
| *Note that [[repo]] is not considered {{isdaprov|Specified Indebtedness}}: see [[borrowed money]]. But don’t let your inner anal retentive amending the definition in your {{isdaprov|Schedule}} so that it is (even though [[repo]] is more properly dealt with by {{isdaprov|DUST}}).
| |
| | |
| ===[[Derivatives]] as {{isdaprov|Specified Indebtedness}}===
| |
| Be wary of including derivatives in the definition of {{isdaprov|Specified Indebtedness}}, no matter how hight the {{isdaprov|Threshold Amount}} (we would say ''never'' do it, but realistically this sort of thing is controlled by “wise senior heads” in the [[credit department]] whose minds will be well beyond the calming influence of most jobbing negotiators, so you may well be stuck with it).
| |
| | |
| The {{isdaprov|Cross Default}} language aggregates up all individual defaults, so even though a single ISDA would be unlikely to have a ''net'' out-of-the-money [[MTM]] of anything like 3% of shareholders’ funds, a large number of individual transactions if aggregated may, particularly if you’re selective about which transactions you’re counting — which the cross default language entitles you to be.
| |
| | |
| Thus, where you have a large number of small failures, you can still have a big problem. This is why we don’t include deposits: operational failure or regulatory action in one jurisdiction can create an immediate problem.
| |
| | |
| The same could well be true for derivatives. Individual net [[MTM]]s under derivative [[ISDA Master Agreement|Master Agreement]]s can be very large. We have a lot of Master Agreements (18000+).
| |
| | |
| Say we have an operational failure (triggering a regulatory announcement, therefore public) or a government action in a given jurisdiction preventing us from making payments on all derivatives in that jurisdiction. We could have technical events of default on a large number of agreements at once – unlikely to be triggered, but for a cross default, that doesn’t matter.
| |
| | |
| The net MTM across all those agreements may well not be significant. But an opportunistic counterparty could tot up all the negative mark to markets, ignore the positive ones, and reach a large number very quickly.
| |
| | |
| Cross Default is a banking concept intended to reference borrowed money - indebtedness etc - and it really doesn’t make economic sense to apply it to derivatives – the fact that there’s a cross default in derivatives documentation at all is something of a historical accident. There are good points made below about the difficulty of calculating it and knowing what to apply it to ([[MTM]]? {{isdaprov|Termination Amount}}? Payments due on any day?) – bear in mind these values are not nearly as deterministic as amounts due wrt borrowed money: on a failure of a derivative contract the valuation of the termination amount (off which {{isdaprov|Cross Default}} would calculate) is extremely contentious. The market is still in dispute with Lehman, for example.
| |
| | |
| ===Credit Mitigation===
| |
| Cross Default is intended to be a tool for mitigating credit exposure. It should be set at a level which reflects a material credit concern in the context of the entire enterprise. By convention, the market generally imposes a Threshold Amount equating to between 2 and 3 percent of shareholders’ funds.
| |
| | |
| ===Credit Support Annex===
| |
| There are other ways of mitigating credit exposure (such as a zero threshold {{csa}}). If a Counterparty's positive [[mark-to-market|exposure]] to {{Bank}} will be fully collateralised on a daily basis, meaning its overall exposure to {{Bank}} at any time will be intra-day movement in the net derivatives positions (a failure to post collateral itself is grounds for immediate termination).
| |
| | |
| ===Contagion risk===
| |
| It is important to maintain minimum standards which are reflective of genuine credit concerns against the bank so as to limit a “[[snowball effect|snowball]]” effect: were we to allow a £50mm {{isdaprov|Threshold Amount}}, we would potentially be open to a large number of derivative counterparties simultaneously (and opportunistically) closing out [[out-of-the-money]] derivatives positions, which in itself could have massive liquidity and capital implications.
| |