Brazil goes off on a CACs tangent

BRAZIL’S $1 BILLION BOND issue at the end of April took the markets by storm. The timing and the pricing were nigh-on perfect: the most difficult part of the deal was deciding how many bonds each of the 430 participants in the over-$7 billion order book was going to get. So it was with some surprise that Euromoney picked up the telephone the following morning only to hear a string of fluent invective aimed at Brazil and its advisers. “Testicular weakness” was one of the more memorable phrases used.

The caller was from the official sector, but was not part of some zealous minority at the IMF. In fact, many private-sector observers took a similar view.

The problem was simple: why had Brazil issued bonds with 85% collective action clauses, or CACs? Everybody had assumed, before Brazil came to market, that not only were CACs here to stay but that the market had standardized on a 75% threshold of bondholders needed to change the payment terms on any bond. After all, that’s what Mexico, the trailblazer, had done; that’s what Uruguay had put into its own exchange offer; and that’s what the G7 countries had said they would do in their own issuance.

What’s more, it was blindingly obvious from the success of the deal that Brazil had no need whatsoever to increase the threshold from 75% to 85%. The extra 10 percentage points in CAC threshold made no difference at all to the level at which the bond priced when it came to market: Brazil didn’t save any money by making it that much more difficult to renegotiate its new, CAC-laden debt.

A number of theories started winging their way around the market. The first had to do with the lead managers. UBS Warburg and Merrill Lynch are fine at lead managing bond issues, but aren’t known to house experts on the finer points of CAC issuance. Shops with in-house experts, such as JPMorgan and Citigroup, would probably have had more confidence in telling Brazil that bondholders’ bark was worse than their bite. What’s more, UBS Warburg’s chief Latin America economist, Michael Gavin, who does know a lot about the subject, is – properly – very close to the buy side and has hosted meetings of the Emerging Markets Creditors Association.

EMCA reacted loudly and negatively to Mexico’s 75% CACs, raising the possibility that they would behave in a similar manner were Brazil to follow suit. Mexico can afford largely to ignore EMCA’s wailings, because dedicated emerging-market investors make up a very small part of Mexico’s investor base. But Brazil needs as many investors as possible on its side, since it is far from out of the woods yet. If 85% CACs would make EMCA happier, Brazil probably saw little harm in adopting them. It would certainly help on the goodwill front, and while it wouldn’t make any difference on this particular bond issue, it might, at the margin, make a difference in the future.

Marcelo Delmar, head of Latin debt capital markets at UBS Warburg, says that “Brazil saw very clearly that 85% was what the investment community wanted, and Brazil embraced the recommendation that the investment community had put to them”.

Conspiracy theorists, however, identified an individual whom they blamed for the elevated level of Brazil’s CACs: Eli Whitney Debevoise II, the partner at law firm Arnold & Porter in Washington DC who advised Brazil on the issue. Debevoise is a veteran in the world of emerging-market debt, but as one friend says of him: “Whitney is a pretty consensus-oriented guy: his advice to Brazil has always been better safe than sorry.” The thing that excited some observers, however, was that as well as advising Brazil, Debevoise was also advising both EMCA and EMTA, the Emerging Market Traders Association.

EMTA’s executive director, Michael Chamberlin, had been first among equals in drafting what he calls the “marketable CACs”: the model clauses for would-be CAC issuers that were put out jointly by half a dozen trade associations and then promptly ignored when Mexico actually took the plunge. The marketable CACs had a threshold of 85% with a 10% veto, making them much closer to Brazil’s bond than to Mexico’s.

Indeed, immediately after the Brazil issue came out, a few EMTA members found themselves in receipt of an “investor-oriented scorecard,” also drafted by Chamberlin, in which he graded the three issuers of CACs on a scale of one to five. Uruguay got 3.57, Mexico got 4.00, and Brazil, thanks to its 85% threshold, got an impressive 4.33. (The marketable CACs got 4.82, since the threshold was 85%, rather than 95%; EMCA’s own covenants, which pre-date the marketable CACs, got a full 5.00.)

The Chamberlin scorecard does look a little bit as if it has been reverse-engineered from a pre-existing idea of where the different countries should place. Mexico and Brazil, for instance, score four out of five for subscribing to the IMF’s Special Data Dissemination Standard, even though they don’t covenant that subscription in their bond documentation as EMCA and EMTA would like them to. Uruguay, on the other hand, which does have data dissemination clauses in its new bonds, still manages to score lower than Mexico and Brazil on “Reporting/Disclosure”.

Meanwhile, Uruguay gets harshly penalized for having a trustee rather than a fiscal agent – something many bondholders actually welcome. On the other hand it receives no credit for closing the loophole that allows such countries as Brazil and Mexico, if they’re feeling particularly nasty, to use exit consents to brutally amend the payment terms on their CAC bonds.

“I think the Uruguay deal is quite clever, and it goes beyond Mexico in providing investors protection against abuse,” says a sell-side analyst who wasn’t involved in the deal. But the trade associations clearly don’t see it that way, and it is they who seem to have persuaded Brazil that, on the one hand, Uruguay is irrelevant as a precedent, and, on the other, that even if it were a precedent, it would be a bad one to follow.

A vogue for minimalism What that means in practice is that future CACs are almost certain to take the minimalist approach to documentation, rather than Uruguay’s maximalist approach. Mexico and Brazil simply added in a couple of short extra clauses, and the deal was done; Uruguay, on the other hand, had a very great deal of new and interesting material in its bonds. The market seems to have decided that new and interesting material is, at the margin, a bad thing: it just means more sitting down and explaining clauses to would-be investors. Even if the new clauses are investor-friendly, the extra cost of having to explain them is usually greater than the benefit of including them.

Uruguay, however, was a bit of a special case: it needed to go on extensive roadshows, anyway, to explain its whole debt exchange and convince investors of the sustainability of its new amortization schedule.

“In all of these deals, when there’s something new, the investor takes a second look,” says Citigroup vice-chairman Bill Rhodes. “But Citi and Uruguay did a lot of roadshows, and all of the work that was done on roadshows made the difference.”

For countries that don’t want to do that much work on their bond issues, however, Uruguay is not going to be a useful precedent. “Uruguay is very, very small,” says Pimco fund manager Mohammed El-Erian, who is one of the largest investors in emerging-market debt in the world, and was one of the founding members of EMCA. “Mexico, Brazil, Russia, South Africa, Korea: they set precedents. Those are the guys who cause systemic ripples.”

On the other hand, notes Rhodes: “Uruguay now has the largest sovereign debt issue under New York law with CACs.” It’s also surely the country most likely to use them, so Uruguay runs a good chance of being the test case for a CAC-based restructuring.

But even if Brazil didn’t follow Uruguay’s lead in terms of CAC documentation, it also broke with Mexico on more than just the CAC threshold. In a development that might be more significant than it first seems, Brazil was the first CAC issuer that didn’t use Cleary, Gottlieb, Steen & Hamilton as its law firm.

Most observers seem to be in agreement that had Cleary represented Brazil, the country would have come with 75% CACs. The fact that Brazil’s lawyer is also working for EMCA raises at the very least the appearance of a potential conflict.

Debevoise says that there isn’t one: he has never done any work on CACs with EMCA or EMTA. The trade organizations only hired him to work on a specific case involving litigation by Argentine bondholders, and he has never talked to them about Brazil or about CACs. Even so, the appearance of a conflict remains.

Debevoise, interestingly, does leave the door open for Brazil moving to 75% CACs in future issues. “This may evolve,” he says: “Beny Parnes [the deputy governor of Brazil’s central bank] was clear that Brazil sees this as an evolutionary process. If the market seems to converge on a lower percentage, maybe Brazil will go there.”

There are certainly good reasons to go for 75% rather than 85%. The whole reason why institutions such as the US Treasury were pushing hard for CACs in the first place was that they would make it easier for countries to restructure their bonds in a crisis situation. But in fact it’s far from clear that it’s easier to restructure bonds with 85% CACs than it is to enter into a present-day restructuring of non-CAC bonds such as that which Uruguay has recently closed.

Uruguay told its bond investors that it would certainly go ahead with its bond exchange if more than 90% of its bonds were tendered, and would certainly not go ahead with the exchange if fewer than 80% of the bondholders agreed to it. If the vote fell somewhere in between the two, the matter would be left to Uruguay’s discretion, and would presumably depend on the nature of the hold-outs.

Aggregation’s easier

In the end, Uruguay managed to get 90% approval, and the exchange went ahead. But the 90% was measured in aggregate, across all of Uruguay’s outstanding bonds: there’s a good chance that at least one bond voted less than 85% in favour. “Getting between 80% and 90% in aggregate [as Uruguay did] is a lot easier than getting 85% approval on each bond [as Brazil requires under its new CAC]” says one analyst. If more countries follow Brazil’s lead and go with 85% CACs, they might actually be making it harder, rather than easier, to restructure their debt, thereby defeating the whole purpose of introducing CACs in the first place. Of course, there are other reasons why CACs might still be a good idea. If Brazil does get 85% of each of its bonds to agree to an amendment of terms, then it doesn’t need to worry about litigation from hold-outs: the minority who vote against will have the new terms crammed down on them. On the other hand, Brazil will now be much more open to blackmail from individual bondholders during the negotiation process: anybody with a 15% holding of a single small bond can now veto a restructuring. (Funds that specialize in accumulating such holdings during a time of crisis call themselves “orphanages”.) In the case of an exchange such as Uruguay’s, such veto power could only come from a holding of 20% of all the country’s outstanding foreign bonds in aggregate.

In the immediate wake of the Brazilian issue, there was some speculation that Brazil had changed the tenor of the whole CAC debate – that, as Pimco’s El-Erian put it, “it will be very difficult for someone new to come out with 75% CACs”.

But El-Erian was proved wrong almost immediately, when, in a very peculiar development, South Africa did just that. The reason that South Africa’s bond was peculiar was that it was issued in London and denominated in euros: there was no reason at all why it should be issued under New York law. All of South Africa’s previous euro-denominated bonds had been issued under London law, which has always included CACs, so there would seem to be no reason why the country should choose New York law for this issue. Clearly, South Africa was trying to make a point; what, exactly, the point was, however, was harder to discern.

While some read the South African bond as a vote in favour of 75% thresholds (and implicitly against the El-Erian view of the world), the general consensus was simply that South Africa wanted to join the ranks of the new CAC vanguard as quickly as possible, and didn’t want to wait until it got around to issuing in dollars.

Even so, South Africa added another important data point to the list of bonds with 75% CACs, and so far only Brazil has been an outlier in that respect. Brazil, too, was something of a special case: having had no access at all to international capital markets for more than a year, it could afford to take no chances with this bond. “If Brazil had not been coming back from a year-long absence, perhaps they would have approached it differently,” says one of the country’s advisers on the issue. “The stakes were enormous.”

Brazil had to be as cautious as possible in deciding what level of CACs to go with, and not only because it absolutely had to have a hugely successful deal to mark its reintroduction to the capital markets.

One of the downsides of moving to CACs is that countries can no longer do small reopenings of old bond issues if they can’t, for some reason, do a big benchmark deal. Since the old bond issues don’t have CACs, and since all new bond issues will have CACs, the old bonds can’t be reopened.

Nearly all new issuance from the likes of Brazil and Mexico, then, will be in the form of large bonds, vulnerable to swings in market sentiment. Brazil therefore needed to ensure that it had the whole market on its side, not only now, but at any foreseeable point in the future when emerging-market debt might look a lot less attractive.

But in making what might have been a sensible decision, Brazil also ran the risk of setting what looked to be a dangerous precedent: that there can and even should be some kind of correlation between credit quality and CAC threshold.

Many observers now have good reason to think that while investment-grade countries such as Canada, Mexico and South Africa can issue with whatever clauses they like, speculative-grade issuers such as Brazil have to make concessions to investors and include 85% CACs. Uruguay, on this reading, does not count as a relevant precedent because its bonds were issued as part of a coercive exchange offer in which bondholders either accepted the 75% CACs or resigned themselves to being defaulted on.

So the debate continues: is CAC documentation going to standardize at the 75% level, or will a two-tier system emerge?

It will take further issuance, from such countries as Peru, Colombia and Venezuela, before the answer to that question becomes clear.