26 April 2018

Odious Debts: Lessons of the Hunger Bond Penalty


Together with Mitu Gulati from Duke University School of Law, Professor Ugo Panizza has published a paper called “The Hausmann-Gorky Effect”, which looks at so-called Venezuelan “hunger bonds”. They find that the public branding of buyers of Venezuelan bonds by Ricardo Hausmann has led to a drop in the price of the Hunger bond. More about methodology and findings with Professor Panizza.

What’s the story behind the paper?

Venezuela doesn’t have enough foreign currency to import medicines and basic goods and there is severe malnutrition among the population. Yet, until recently, the government has been regularly paying its foreign debt. Ricardo Hausmann, who is a Venezuelan economist and not a friend of the current government, wrote an article on Project Syndicate on 26 May 2017 saying that investors who buy Venezuelan bonds are basically financing an odious regime. The background is that Venezuela is part of the Emerging Markets Bond (EMBI) Index by JP Morgan, which is a very important index as investors benchmark their activity in emerging markets on it. Hausmann says that the financing to the Venezuelan government should be reduced, so the article (titled “The Hunger Bonds”) was a call to JP Morgan to remove Venezuela from the EMBI Index. What Hausmann didn’t know at this point was that two days before, one branch of Goldman Sachs had bought some Venezuelan bonds (issued by the state-owned oil company PDVSA) which had officially been issued by the state-owned oil company PDVSA in 2014 and set to expire in 2022. The tricky thing about those particular bonds was that the government had never put them on the market, but simply gave them to the Central Bank in 2014. When Goldman bought the bonds in 2017, it claimed that it had bought them on the secondary market and that they had been around for a while, which wasn’t actually true. They bought them from the Central Bank, which meant that de facto they provided direct financing to the Venezuelan government. There is even proof for this as two days after the purchase by Goldman Sachs (around USD 800 million), the reserves went up by about USD 750 million, meaning there was a near-perfect match between the Goldman purchase and the increase in the reserves. When Hausmann was told about this purchase, he went on TV and denounced it. There was a big backlash, Marco Rubio tweeted about it, it was all over the news and people started protesting outside the office of Goldman Sachs in New York.

While the article by Hausmann was more a criticism of all Venezuelan borrowing, the term “hunger bonds” was attached to this particular bond bought by Goldman. This bond was already trading at a deep discount compared to other Venezuelan bonds, but after the Hausmann article, its price went down even more and it has been trading at a discount with respect to comparable Venezuelan bonds ever since then. We decided to write this paper because we wanted to understand better what had happened to the price of this particular bond.

Regarding your methodology, you used an interesting measure. Could you tell us more about it?

In the econometric analysis, we were asking the following: Can we see in more detail if the interest in this specific bond is linked to its price? To measure this interest, we computed an index that is based on Google searches for the term “hunger bonds”. We went on Google Analytics, typed in “hunger bonds” and Google gave us an index showing the relative popularity of this term. The higher the number, the more people were searching. Since Google Analytics only provides an overview of weekly searches, we had to look at the behaviour on a weekly basis, though.

What we actually were able to show is that when there were many searches for this term, the price of the hunger bond would decrease. There is a link between the saliency of Hausmann’s article and the price of the bond.

We then did some event studies by looking at other types of shocks and their influence on the price; however, we still found that the link between the search for “hunger bonds” in Google and the price of the bond is robust when controlling for all sorts of other variables.

What is the “Hausmann-Gorky effect”?

In the legal world, and also a little bit in economics, there has been a discussion going on for a very long time on the concept of odious debt. Let’s suppose you have the Apartheid regime in South Africa borrowing on international capital markets. Once the regime is gone, should the new government be responsible for the debt of the Apartheid regime, which was money collected to actually oppress the people? There was an article written nearly 100 years ago by a legal scholar called Alexander Sack, who said that this debt is odious and therefore should not be paid. However, the normal legal principle in international law is that all legal obligations of a state are transmitted to successor states, which includes borrowing. If this were not the case, a state would never engage in legal contracts because its successor could refuse to recognise them. This idea of odious debt was never resolved. What we show in this paper is that there are other mechanisms (such as publicly denouncing the immorality of odious debt) to increase the cost of borrowing, which do not depend upon an international agreement. This is the Hausmann part of the effect. We call it the Hausmann-Gorky effect because interestingly enough something similar happened about a century ago. At the time of the Tsarist regime in Russia, the Russian writer Maxim Gorky wrote an article in the French press, saying that lending money to the Tsarist regime was immoral and should therefore not be done. The effect was similar to that of the Hausmann’s article as for some time the cost of borrowing for Russia increased. So we discuss in the paper whether we could have a public registry of such immoral/illegal actions with the goal of keeping the cost of borrowing for odious regimes high for a long time.

You have also conducted interviews with investors as to why they stopped buying this particular Venezuelan bond. What do you conclude from these interviews?

Investors said that the reason for them not to buy this bond anymore wasn’t so much its immorality, or the plight of the Venezuelan people, as it was the fact that they would make too little money compared to the reputational cost, including for example people demonstrating outside their offices or compliance departments questioning the legality of the operation. This is a fully economical approach motivated by self-interest, which is fine with our story: if you can make it costly to buy this bond for investors, it will ultimately increase the cost of borrowing for the regime, thus achieving the goal of cutting off odious regimes from the international capital markets.



Full citation of the paper:
Gulati, Mitu, and Ugo Panizza. “The Hausmann-Gorky Effect.” International Economics Department Working Paper HEIDWP02-2018, Graduate Institute of International and Development Studies, Geneva, 2018.

Interview by Nadia Myohl, Master student in International Economics.