News in Practice

Finance documentation: holding the covenants when the contract moves

One exclusion added to an accounting definition, three lines, on a Tuesday. The seven consequences it produces elsewhere, only one of which will be checked.

For finance teams, in firms and in legal departments. How an amendment that is correct and accepted by everyone produces seven shifts that review does not catch.

Version 4 of the facility agreement. Borrower's counsel asks that the definition of Net Financial Indebtedness exclude lease liabilities arising from leases capitalised on the balance sheet. The request is standard, it is reasoned, and it is accepted by the syndicate the same day. Three lines added to clause 1. From that amendment, seven things have moved elsewhere in the documentation, and only one of them will be checked.

First shift: the leverage ratio

The Leverage Ratio is defined as the ratio of Net Financial Indebtedness to EBITDA. Excluding lease liabilities from the numerator mechanically reduces the ratio. By how much depends on the borrower's property portfolio, and for a group operating a retail network the variation is not marginal.

This is the only one of the seven that is systematically seen, because it is immediate and because finance teams calculate it. It is seen, and it is often the very object of the request: that is precisely why the borrower makes it.

Second shift: the covenant threshold

The agreement sets a maximum leverage covenant, say four times, tested quarterly. That threshold was negotiated against the business plan, that is to say against a ratio calculated on the definition in force at the time of negotiation.

With the definition changed, the threshold no longer represents the same constraint. A covenant at four times, on a ratio whose numerator has just been reduced, is a looser covenant, sometimes appreciably so. The question is not whether that is acceptable, which is a matter for negotiation. The question is whether the lender who accepted the amendment to clause 1 on the Tuesday understood that it had just loosened its covenant.

In practice, the answer depends entirely on whether someone holding the map of dependencies in their head was in the room that day. That is not a competence; it is an availability.

The question is not whether the loosening is acceptable. It is whether the person who accepted the amendment knew they were loosening.

Third shift: the margin ratchet

Pricing is indexed to the Leverage Ratio through a grid: above three and a half times, one margin; between three and three and a half, another; below three, a third. The mechanism is standard.

Changing the numerator potentially moves the borrower from one band to another, without any economic improvement having occurred. The lender concedes a margin reduction in exchange for nothing. Over a seven-year facility and a significant amount, that ceases to be a drafting detail.

This shift is the most frequently missed of the seven, because the margin grid sits at clause 8 or 9, is often drafted by someone other than the person maintaining the definitions, and is perceived as an economic rather than a legal provision.

Fourth shift: events of default

Breach of a financial covenant constitutes an event of default, generally with a remedy period and an equity cure mechanism. The trigger for that whole apparatus is the leverage threshold, itself anchored to the amended definition.

The shift here is second-order and of first importance: it changes the moment at which the lender may act. On a deteriorating credit, a few months' delay in triggering an event of default changes recovery value.

Fifth shift: the intercreditor agreement

This is the hardest of the seven to follow, because it leaves the document. The intercreditor agreement has its own definitions, sometimes incorporated by reference to the facility agreement, sometimes reproduced. Where they are reproduced, they do not follow.

The documentation then carries two definitions of Net Financial Indebtedness that no longer coincide, one in the facility agreement and one in the intercreditor, each correct within its own document. The mismatch does not surface during the normal years. It surfaces on enforcement, that is to say at the worst possible moment.

One further point on this fifth shift deserves noting: it outlives the transaction. Three years later, an amendment will be negotiated by different teams who will read both agreements without knowing which one carries the original definition.

Sixth shift: the compliance certificate

The agreement requires the borrower to deliver, at each test date, a compliance certificate confirming observance of the covenants and setting out the ratio calculations. That certificate follows a scheduled form, generally schedule 7, drafted at version 2 and never reopened.

The form contains calculation lines reproducing the original definition of Net Financial Indebtedness, item by item. It contains no exclusion line for lease liabilities, since that exclusion did not exist when the form was drafted.

The borrower will therefore deliver, quarter after quarter, a certificate that conforms to the contractual form and does not conform to the contractual definition. It will be right to follow the schedule and wrong as against clause 1. The mismatch is benign while the ratio is comfortable. It becomes a point of attack the day it is not, and it then covers three years of certificates.

This sixth shift illustrates a general property of schedules: they are drafted early, rarely reread, and they often contain the operational translation of the definitions one later amends.

Seventh shift: the comfort letter

The parent company has given a comfort letter whose undertaking is indexed to the financial ratios of the agreement. It was drafted at version 2, by a different team, and has not been reopened since.

It refers to the original definition, reproduced in an annex. The group therefore stands committed on a ratio calculated differently from the one the agreement tests. The mismatch has no effect while things go well, and determines everything the day the letter is called.

Why review does not catch them

One might object that a careful review of the documentation ought to suffice. It does not, for a reason bearing on the nature of the exercise rather than on the reviewer's diligence.

A review proceeds document by document, in the order of the clauses. It confirms that each clause is correct, which they all are. The seven shifts are not clause defects: they are defects in the relationship between distant clauses, and they appear only when the document is read in an order nobody follows, that of its dependencies.

To spot them one would have to start from the amended definition and traverse everything that depends on it, across two agreements and two schedules. That is feasible once, on an amendment identified as sensitive. It is not feasible across the forty amendments a six-week negotiation produces.

Seven shifts, one checked

To recap. The amendment is correct. It was requested for a legitimate reason, accepted by competent professionals, and properly drafted. Of the seven shifts it produces, one is seen because it is the very object of the request, one is sometimes seen if someone happens to think of it, and five pass, two of which leave the document and one of which sits in a schedule.

None of these seven is visible from where the amendment is made. They are spread across sixty pages, three agreements and two schedules. Spotting them requires holding the complete map of the documentation's dependencies, which a person does well on a deal they have run from the start, and badly on one they have picked up.

And finance documentation, unlike a sale agreement, does not stop at closing. It lives on for years, through amendments, waivers, drawdowns and incoming lenders, each of which alters the system without anyone revisiting the whole. Four years in, knowing what the documentation actually says has become a research exercise, undertaken when an event requires it, and therefore under pressure.

What a system has to hold here

The useful shift does not consist in drafting faster. It consists in the documentation being held as a system whose state can be consulted.

Definitions tracked across the whole set of agreements and not merely within a single document, so that the version reproduced in the intercreditor is linked to its source.

An amendment that surfaces what it affects, including the margin grid at clause 9 and the equity cure mechanism nobody opens.

The consolidated state of the covenants at a given date, consultable rather than reconstructed.

Vocabulary is a subject in its own right here. The terms of finance documentation remain largely in English in French practice, they carry precise meaning, and a system that normalises them destroys the very thing it was meant to preserve. Holding documentation means holding its language.

What the lawyer keeps in hand

Accepting or refusing the exclusion of lease liabilities is a negotiation, and that negotiation depends on what the market accepts at that moment, on the borrower's credit quality, on the balance of power in the syndicate. None of that follows from the documentation.

What does follow from the documentation is the list of the seven shifts. Counsel who accepts the amendment knowing all seven is not doing the same work as counsel who sees one.