Making (Non)Sense of The Lebanese Fiscal Agency Agreement

Mark Weidemaier & Mitu Gulati

After trying but failing to locate the fiscal agency agreements underlying Lebanese bond issues, we finally managed to get our hands on this one. We had hoped that the FAA would clarify the respective legal positions of the Lebanese government and its investors. Nope. Our review of the FAA leaves us scratching our heads. The original contract dates from 1999—this is the 3d amended version from 2010—and was one of the first post-Brady-era bonds issued under New York law to include a collective action clause. We were eager to see it and had even heard it had been carefully designed to minimize the risk of holdouts in the event of a restructuring. Certainly the government has reason to fear holdouts, such as London-based hedge fund Ashmore.

This may be the weirdest CAC ever. Taken as a whole, the FAA also includes just about the weakest set of anti-holdout tools we have seen. The Lebanese government may have to get creative to restructure.

Let’s start with the CAC. For background, Ashmore is rumored to hold over 25% in aggregate principal amount of multiple Lebanese bond issues (here). That’s enough to block a restructuring vote in most first-generation CACs (i.e., those that first took hold in the NY market around 2003). Lebanon’s CACs are even older; it adopted them at a time when CACs virtually never appeared in NY-law bonds.

For reasons not obvious to us, Lebanon’s lawyers were New York specialists but operated out of London. For reasons that are also not entirely clear, they designated New York law to govern but then bolted on modification provisions (the CAC) derived from the template then in use in the English law market. The end result is confounding.

Grossly simplified, most CACs say something like this: We can modify the bonds with the affirmative vote of holders of X% in aggregate principal amount of the affected bonds, except that modifications A, B, or C require the affirmative vote of holders of X+Y% in aggregate principal amount of the affected bonds. Again, this is grossly simplified. For instance, there will typically be provisions governing how votes will be taken at a meeting (including quorum requirements) and votes taken by written resolution. The important point is that the bond includes general power to modify at a relatively low voting threshold but imposes a higher threshold for modifications to key terms, including payment terms. The term “CAC” refers to that higher voting threshold for modifying payment terms. But if the issuer can’t muster enough support to trigger the CAC, it still has options. Most obviously, if it can persuade enough creditors to trigger the lower modification threshold, it can use the Exit Consent strategy. This involves a bond exchange in which participating investors vote to modify the bonds in ways that make non-exchanged bonds unattractive to hold (but don’t trigger the CAC’s higher voting threshold).

The problem with Lebanon’s FAA is that the CAC is weak but the bond does not appear to allow for the use of exit consents:

Most of the details governing modification are found in Schedule I of the Lebanese FAA. (Also see paragraphs 23-24 of the FAA itself and par. 15 of Schedule A.)

Schedule I’s paragraph 6 sets quorum requirements for noteholder meetings (at which a CAC vote might be taken). Most relevant here, for meetings to consider an Extraordinary Resolution, a quorum consists of holders of over 50% in aggregate principal amount of the notes “for the time being outstanding.” However, if the Extraordinary Resolution will modify the bonds in one of the ways listed in the proviso to paragraph 19, the quorum requirement rises to 75%.

Paragraph 19 includes the most important kinds of modifications, including to payment terms. The "proviso" referred to above likely refers to the last part of the paragraph, which covers modifications to payment and other important terms. But the first part of par. 19 empowers bondholders to make a much wider set of modifications by "Extraordinary Resolution" and defines the term so broadly that it arguably includes any modification to the bonds  (e.g., par. 19(a), which empowers noteholders to “sanction any proposal … for any modification … in respect of the rights of the Holders of Notes against the Republic…”)

Note that, thus far, we have been talking only about quorum requirements. To recap, the quorum is 50% of outstanding debt for most Extraordinary Resolutions but rises to 75% for Extraordinary Resolutions affecting important terms. But we’ve said nothing about the actual voting threshold for approving modifications. The threshold applicable to meetings can be found in paragraph 1(c), which defines Extraordinary Resolution to mean a resolution approved by holders of at least 75% in aggregate principal amount of notes represented at the meeting (assuming a quorum). For example, if holders of 90% of the outstanding debt attend the meeting, and 75% vote in favor, the Extraordinary Resolution would pass with the affirmative support of holders of only 67.5% of the entire outstanding debt.

Paragraph 21 sets the voting threshold for written resolutions, equal to holders of 75% in aggregate principal amount of notes “who for the time being are entitled to receive notice of a meeting in accordance with the provisions herein contained.” (Presumably this includes just about everybody?)

The 75% voting threshold applicable to Extraordinary Resolutions (and written resolutions) is the only voting threshold we can find. Normally, the documents would establish a lower threshold for modifications that don’t affect key terms. And various parts of the Lebanese documentation suggest that such a threshold exists. For example, par. 15(a) of Schedule A (“Terms and Conditions of the Notes”) references the 75% voting threshold, implies that this elevated threshold applies only to a subset of modifications, and otherwise advises the reader to consult the FAA. But again, we don't see any other voting threshold in the FAA.

Likewise, the documents suggest that the voting threshold of 75% applies only to a subset of modifications affecting important terms (i.e., those listed in par. 15(a) of Schedule A and in the proviso to par. 19 of Schedule I). But again, Schedule I (in par. 19) seems to include basically any modification in the definition of “Extraordinary Resolution,” and the FAA seems to specify a 75% voting threshold for all Extraordinary Resolutions. The only difference is that "important" Extraordinary Resolutions are subject to a higher meeting quorum.

What is going on here? Where is the voting threshold for relatively unimportant modifications, of the sort that would normally be targeted via exit consents? Surely the drafters did not mean to use the quorum requirement (50%, for Extraordinary Resolutions that don’t implicate key terms, versus 75% for those that do) as an indirect way to set a lower voting threshold? That would be bizarre, as the difference wouldn't matter unless the meeting were poorly-attended.

Even if we can identify the lower voting threshold, when does it apply? Common sense (and, implicitly, provisions like par. 15(a) of the Terms and Conditions) suggest it should apply to any modification not subject to the elevated quorum requirement. But that sensible intuition is undermined by the fact that almost every modification seems to constitute an Extraordinary Resolution subject to a 75% vote.

In any event, if the FAA is interpreted to always require the approval of holders of 75% in aggregate principal amount represented at a meeting, then restructuring will prove very difficult. No matter the quorum requirement, Ashmore will have a blocking stake large enough to prevent amendment, whether of payment terms under the CAC or of other terms via exit consent. The lack of clarity also significantly weakens Lebanon’s position in restructuring negotiations.

Uncertainty over voting thresholds is not the only thing that weakens the government’s position. There’s also the question of who can vote on any restructuring proposal. That’s what is called the disenfranchisement provision. It is important in the Lebanese context because the Lebanese Central Bank and a bunch of local commercial banks under the influence of the government are rumored to hold a boatload of Lebanese debt. The provision, however, is absent from both Schedule A (the Terms and Conditions) and Schedule I (the part about voting).

If one keeps digging, one eventually locates it in the definition of “outstanding” in the FAA’s Definitions section. The provision disenfranchises notes held by the Republic, any department, ministry or agency of the Republic, and any corporation, trust or financial institution or other entity “owned or controlled” by the government. Alas, for the Lebanese government, this is a pretty broad disenfranchisement provision. It seems to rule out having the Central Bank vote in favor of a restructuring and might be interpreted to disenfranchise even some commercial banks.

Curiously, the voting provisions of Schedule I, and especially the 75% voting threshold in par. 1(c), don’t require modifications to win the support of 75% of the “outstanding” debt. They require only approval of 75% of in aggregate principal amount of the “Notes,” and this term isn’t limited to “outstanding” Notes. Now, the quorum requirements (Sch. I, par. 6&7) are computed as a proportion of “outstanding” notes, but the voting threshold itself arguably isn’t so limited. Anyway, the definition of "outstanding" disenfranchises government-controlled bonds "for the purposes of the right to attend and vote at any meeting..."

In short, this FAA leaves a whole lot up in the air. It presents the government with the risk not only that a CAC vote will fail but that the government will not be able to fall back on the use of exit consents. This result isn’t inevitable. But these documents seem to make it more likely than would usually be the case. Perhaps we are missing something. But the one thing we are sure of is that these documents do a good job of obscuring important information.

If Ashmore can block any vote to modify the bonds, there are two easy options remaining for the government, and neither is good. First, it can offer financial terms so sweet that even dyspeptic holdout creditors will accept. Second, it can close an exchange offer with willing investors and then batten down the hatches, removing attachable assets from foreign jurisdictions and hoping to withstand the coming litigation.

But maybe there is another way. We’ll have more to say on that later.

Comments

[Chris Smith]

The poorly written FAA demonstrates why countries must always enter these agreements aware of the possibility of default. One of the biggest mistakes in Lebanon’s FAA is the disenfranchisement provision. The central banks and domestic commercial banks would likely be some of the friendliest debtholders, so disenfranchising them serves to strengthen more aggressive creditors.

Faced with powerful and aggressive debtholders, Lebanon may have no choice but to “batten down the hatches” and face litigation. Because this is Lebanon’s first default, it has a strong argument that it is not a “uniquely recalcitrant debtor” and, therefore, can hopefully avoid being raked over the proverbial coals.

[Chuck M]

I am not sure what excuse there is for such muddled language in a key contract provision. It suggests that even investors don't pay close attention to these clauses when deciding to buy.

Maybe I'm misunderstanding, but I think Lebanon may have a good chance to argue that they're allowed to vote on major changes since the word "outstanding" is omitted. Of course, an attempt to do this will lead to protracted and expensive litigation Lebanon may want to avoid.

[Charlie Fendrych]

Given there really doesn't seem to be any explicit guidance, I wonder if 15(a)'s language "duly passed" simply refers to a majority vote of those present at quorum. I'm skeptical because this seems like a really imprecise way of dictating voting requirements, but I would also generally interpret an unqualified vote as majority rules. Since pretty much anything important requires the heightened quorum # and 75% vote maybe a show of hands is all that a non-Extraordinary Resolution takes.

Given the FAA's disjointed nature, its hard to pick at portions without reading the whole document so I'm probably missing something. Especially since previous posts suggest nothing happens for no reason/really gets "copied and pasted."