Mozambique’s Guarantees on the Tuna Bonds: Can They be Repudiated?

Mark Weidemaier & Mitu Gulati

There have recently been headline articles in the press about three loans made to state-owned security companies in Mozambique (see here, here and here) and guaranteed by the government. The reason for the attention to these loans – made originally between 2013 and 2014 by Credit Suisse and the Russian bank VTB – is that US federal prosecutors are pursuing charges against a number of bankers from Credit Suisse and government officials from the Mozambique finance ministry. (Somehow the VTB folks seem to have escaped so far.) To simplify, these individuals were allegedly involved in siphoning off funds ostensibly intended to support Mozambique’s fishing industry and enhanced security in its territorial waters. Concretely, the loan was supposed to be used for new boats: some to catch fish (hence the moniker “tuna bonds”) and others to bolster the coast guard (“maritime surveillance”).

Instead, much of the money seems to have disappeared. The loans went into default; few tuna were caught. For contemporaneous reporting, see here, here, and here.

We have been thinking about debt repudiation of late. And Tracy Alloway of Bloomberg (and formerly of FT Alphaville) specifically got us thinking about the Mozambique tuna bonds on a recent podcast for Bloomberg’s Odd Lots (Tracy is a spectacular host).  Prompted in part by Tracy, we wondered--now that the corruption on the part of the agents for the banks and agents within the Mozambique finance ministry has been revealed—whether the government can repudiate the loans on the grounds that they were infused with illegality.

One of the three loans is worth treating separately from the others. This loan was made specifically for tuna boats. It involved an $850m bond for a company called Ematum—allegedly a sham—which has since been converted from a state-guaranteed bond to a sovereign Eurobond. For the other two loans, the repudiation question—since the borrower companies seem to have no assets—is whether the state can withdraw its guarantee on account of the corruption. There is a good argument that the answer is “yes.” Contract law in many key legal jurisdictions makes contracts infected by corruption and bribery voidable.

Some years ago, one of us analyzed this question in an article with Lee Buchheit, where we analyzed the question of “corrupt debts” (here – at pp 1234-39). We quoted this illustrative language from a 1960 New York Court of Appeals case: “Consistent with public morality and settled public policy, we hold that a party will be denied recovery even on a contract valid on its face, if it appears that he has resorted to gravely immoral and illegal conduct in accomplishing its performance.” Jeff King, in his new book on Odious Debts (here – at pp 119-23), has a section on sovereign obligations infected by corruption and makes much the same point under English and a number of other laws. And Jason Yackee tackles the corruption defense for sovereigns in the BIT context here. Bottom line: There is a pretty good defense here.

One objection to this argument—which we mentioned in our previous post on Puerto Rico’s attempt to avoid $6 billion in debt (here)—is that courts sometimes impute the wrongdoing of past government officials to the current government. For instance, in Republic of Iraq v. ABB et al., 768 F.3d 145 (2d Cir. 2014), the post-Saddam government of Iraq tried to disclaim responsibility for the previous government’s involvement in corrupt transactions. The Second Circuit, however, refused to draw a distinction: “where a plaintiff in Iraq’s position bears fault, it does not escape the consequence of its wrongdoing on the basis of a change in leadership.”

We are not inclined to read too much into this reasoning, which may be limited by its context—Iraq was trying to recover damages under RICO, not avoid a contractual obligations—and by its specific facts—including the fact that the corruption was “coordinated pursuant to the policies of an entire government.”

Finally, it’s worth considering the $850m Ematum bond, which in 2016 was “sanitized” into a sovereign bond of Mozambique (its first ever eurobond, sadly). That bond was apparently restructured into another eurobond plus a so-called “value recovery” instrument giving investors some of Mozambique’s future gas revenues (here). Government officials in Mozambique proclaimed this restructuring to be a good deal. But was it really? If this were still a guaranteed instrument, the government would have a credible argument in favor of withdrawing the guarantee. If successful, withdrawal would have left investors with essentially valueless claims. But by converting the guaranteed debt of a shell company into a direct sovereign obligations, the government likely sanitized the debt from challenge.

We wonder why the Mozambique government did this sanitization exercise. Did officials think it would help improve the government’s perceived creditworthiness? Or did they hope to deflect attention from the initial corruption surrounding the loan? (If the latter, it didn’t work.) As we understand matters, investors in the two other dodgy loans wanted a similar deal (indeed, there were discussions on the subject at the end of 2018). Apparently, the holders of the “sanitized” government bond objected on the basis that the guarantees enjoyed by holders of the other two bonds were illegal (see here). Hmm.  (Also, the IMF was already irate about the Mozambique government have failed to adequately disclose to it their debt shenanigans).

Comments

[Natalie Pita]

I think one of the key issues to consider here is whether the corruption was one-sided or plagued both parties. In the Puerto Rico post, a lot of people were willing to put the burden on investors, who could have researched the constitutional limitations themselves and who were also a highly sophisticated party. If they were aware of the corruption or also at fault, I think this becomes a similar situation. However, if this was only on the part of the government, I think it becomes an issue of odious debt, and we have a case similar to that or Iraq with one administration trying to wipe the mistakes of another in a way that simple is not allowed by sovereign debt practice.

[Maximilian Hess]

I believe the ProIndicus and MAM loans are still syndicated loans, not bonds, whereas Ematum was a bond (with a guarantee) from the get-go. so one key distinction is that those who may have bought into the Ematum bond or the restructured Eurobond on the secondary market have ‘clean hands’ whereas VTB and Credit Suisse were the lead syndicators in the or MAM and ProIndicus issuances respectively. So the bond investors, on the basis they were buying into an exchange listed instrument not taking part in a corrupt syndicated loan, argue their debt should be recognized whereas the loans should not. I would think that arguments about the relative strength of different loan structures under such circumstances must have come up during the Brady bond discussions.

[Isabelle Sawhney]

As a preliminary note, I find it suspect that extremely sophisticated parties––Credit Suisse and Russia’s VTB Bank––gave $2B in loans to a Mozambique-owned company when this country is among the poorest countries in the world. How much of a guarantee is it really to have the state back the loan? How much could these investors truly expect to gain back their investment? So I am not sympathetic for these banks.

What is interesting is the odious debt argument. At first I thought that if the argument didn’t work for Iraq’s debt under Saddam Hussein, then the argument just won’t work at all. But what might make the difference in this case is the fact that the banks paid the money directly to the contractor in the United Arab Emirates. And it is known that the money was for kickbacks. So there is no real argument that you can make that the money was for the Mozambique people (whether their benefit or detriment).

Finally, sanitizing only a portion of the bonds doesn’t quite make sense to me. If they wanted to legitimatize their debts or decrease scandal, then it seems incomplete to leave over $1B as “dirty” debts. It also doesn’t seem like that effective of a method either: when the $850M bonds are due in about 15 years, Mozambique is still unlikely to be able to meet the obligation, and if their gas revenues do not cover the cost, litigation will probably ensue and the shady origins of these loans will be readily apparent.

[Federico Espinosa Fernandez]

Mozambique and Puerto Rico are in similar positions, but each of them alleges different arguments and factual considerations. While Puerto Rico tries to refuse the payment of some loans for considering that they are odious debt, Mozambique asserts that it is not liable under a corrupt debt. Thus, the question here if those loans are equally binding.

Even though the Common Law provides principles and rules that would apply to these cases (especially given by Agency Law), I am not sure whether those rules apply to sovereign debts, which are commonly governed by the local rules of the borrower’s lending system (?).

In any case, beyond the legal considerations, it is clear that these types of conducts from countries can produce uncertainty and mistrust from the investors. Thus, Mozambique and Puerto Rico will likely face negative consequences, such as exclusion from future transactions, legal sanctions, and loss of reputation, impacting severely to future generations.

[Shane E.]

In an odd sort of way, the Puerto Rico situation is simpler than the Mozambique one. The Puerto Rican bonds were issued in spite of a *constitutional* ban, whereas the tuna bonds aren't inherently problematic- if they were used correctly they may be exactly the kind of thing we want to encourage in "developing” economies. Do we want to treat the odiousness of Mozambique's debt on a "who knew what when?" scale or do we want to make a broader rule about misappropriated funds and the ability to repudiate?

Surely, both situations reveal that the international debt market needs discipline- perhaps we can rely on ex-post criminal enforcement of fraud and corruption rules to take care of the bad actors in government. But as long as governments are paying back lenders (and downstream holders) for illegal or “clearly should be considered” illegal debt there’s no incentive for the underwriters banks (and dare I say: lawyers) to improve their practices. The solution seems to me to be increased liability for those actors in the form of repudiation and compensation for secondary market holders. It doesn’t seem all that different from say, liability for selling (or organizing the creation of) a derivative instrument based on bad loans.

[Hailey Klabo]

I'm struck by the role reputation plays for Puerto Rico and Mozambique. As discussed in both pieces, as well as by Nick in a comment to the Puerto Rico piece, and by Isabelle here, the ephemeral value of reputation necessarily comes into the conversation. While Puerto Rico seems to be risking its reputation with its attempt to void $6 bill., Mozambique seems to be attempting to save its reputation through this sanitation exercise. It seems likely that neither country will take a hit (or see a bump) in their reputation--at the end of the day, the investors want to make money and will take the risks they see fit to take without much mind to the history of the government, especially given the comparatively small amount of debt at issue here (It's still a large amount of money, Mitu!). I agree with Isabelle, however, that this seems to be for show, given the remaining "dirty" debt. Nevertheless, good faith actions are good faith actions and that has to count for something.