The Power and the Money

Constitutional hardball in Peru

So, yesterday the President of Peru dissolved Congress and called for new legislative elections. The Congress called that illegal, and deposed the President. What’s going on?

Constitutionally, we have a bit of a mess. The reason, however, isn’t really that the Constitution is unclear. (In fact, the Peruvian constitution is a very nice document.) It is that all sides are playing hardball, sticking to the letter of the rules but without any informal norms to keep things on the rails.

Short version: Peru has a mixed Presidential-Parliamentary system with a President and a Prime Minister. The President, however, retains executive authority. Think France, not Germany.

The Congress, controlled by the Fujimorista opposition, tried to pack the Peruvian supreme court. The President said this was against the rules, and told the Prime Minister to send a “confidence vote” against the move directly to the floor. (Article 133.)

The first problem: does the confidence vote take priority over all other matters? Article 133 doesn’t explicitly say that confidence votes should take priority, but that appears to have been the intention of the 1993 authors of the document. (Weirdly, no one is asking them.)  The Congressional leadership, in a chaotic session, decided to vote on court packing before the confidence vote. (Chaotic means chaotic: fistfights, shouting, physically stopping the Prime Minister from taking the floor by locking the door. )  

The President took Congress’s decision to postpone the confidence vote to be the equivalent of having his government lose a confidence vote. According to him, this was the second vote lost during the current presidential term, and so he dissolved Congress and called new elections under Article 134 of the Constitution.

But this brings us to the second problem. Article 134 gives the president the power to call new congressional elections should the administration lose two confidence votes. But has President Martín Vizcarra lost any confidence votes? In point of fact, he has not lost any confidence votes since assuming office. 

But Vizcarra claims that since he is only finishing disgraced ex-President Pedro Kuczynski’s term, a lost confidence vote on education policy during Kuczynski’s term should count as a loss for him. Add in Congress’s failure to hold the second confidence vote, and he has the two failed votes he needs to call new elections. 

Well, once Vizcarra declared Congress dissolved, Congress counter-declared that the President was breaking the law and suspended him from office for a year. Mercedes Araoz then became president. Article 114 allows Congress to declare the president “temporarily incapacitated.” Once again, though, that is supposed to be an equivalent of the 25th Amendment, something to be used when the president is, well, temporarily incapacitated. Article 117 concerns removing the president, by trial.

But back in the bad days of Alberto Fujimori, in 2001, Congress did remove him from office after he fled the country. Fujimori tried to resign, a la Nixon, but Congress insisted on holding the vote. I don’t think anyone intended to establish a precedent, but they did, so now Congress has suspended the President for a year and elevated his Vice-president in his place.

I should mention here that the opposition tried to remove Vizcarra permanently, but they didn’t have the votes. I don’t know if that was because legislators couldn’t stomach the outright unconstitutionality and preferred to have the fig leaf of Article 114 or if it was for some more Machiavellian reason.

Politically, my sympathies are with the President. It’s clear that the opposition has been playing constitutional hardball. They haven’t been breaking rules and nor has the government; Peru isn’t Bolivia or Honduras, let alone Nicaragua or Venezuela. But they have been playing to the letter of the law in an extremely destructive way.

I don’t suppose it matters. Right now you have Peru with two people claiming to be the legitimate president and no clear-cut constitutional case. Nonetheless, this was not how the framers of Peru’s constitution expected things to go. They did expect that the confidence vote would get priority and they did not expect the Congress to be able to remove the President with a simple vote, even temporarily.  

This won’t lead to civil war or a military coup (the military backs Vizcarra). But it is sobering and it is ugly. And closer to home, it should remind us of where constitutional hardball could lead here in the United States. I am glad that Mitch McConnell recognizes that if the house votes to impeach then the Senate has to hold a trial, but he could have gone the other way. He did in the Merrick Garland case. Republicans have elsewhere, and Democrats are tempted to follow suit.

And that will, as Peru shows, inevitably end in tears. The only question is how many tears.

Nobody looks good in the Peruvian debt saga

 Gramercy Funds Management is a good operation. Its founder and boss, Robert Koenigsberger, was on the side of the angels in the Argentine debt dispute. It also did good work with Nicaragua, masterminding a 2007 exchange that wrote off 95.5% of $1.4 billion in debt. (Data from here.) Moreover, Koenigsberger is a smart fellow, a graduate in Latin American history from UCSD. It is the kind of firm I find difficult to dislike, let alone villify.

But it is in the wrong on Peru, where it is suing the country for $1.6 billion in defaulted bonds under the U.S.-Peru Trade Promotion Agreement.

I wrote about the history of the bond default here. Short version: Peru issued a bunch of bonds to compensate landowners as part of a land reform in 1968-69. In the 1980s it defaulted on those bonds, even though inflation was rapidly reducing them to nothing.

The Peruvian Constitutional Tribunal, for some reason, decided that letting inflation eat up the value of the defaulted bonds would be an unconstitutional seizure of property. Note, please, that American courts have rejected that notion. In the 1970s, unexpected inflation ate up the value of debts by more than a factor of two ... good luck getting an American court circa 1980 to tell you that the Fifth Amendment means that you should get double back from a debtor in default to whom you loaned cash in 1970. 

But anyway, the Constitutional Tribunal decided that creditors in Peru deserved more protection than creditors in the United States. 

Between 2006 and 2008, Gramercy bought the bonds cheap. (Background testimony here.) In a twist that makes my Luddite self happy, they had to actually go to Peru and buy up 9,773 individual paper bonds. I find that amazingly cool; it further increases my respect for Gramercy.

They are still wrong. But I like them.

Anyway, the Peruvian government stalled and stonewalled on settling the outstanding debt. In part, this was because Gramercy insisted that the right way to value them was by running them through the Peruvian CPI. Note, however, that Peru underwent a hyperinflation over that period: calculating the CPI for a hyperinflating country involves a lot of guesswork; it would be economic malpractice to simply run the value through the inflation index unless that index had been checked and recalibrated over multiple different base years to correct for the inevitable distortions. 

The Peruvian government understandably said no. The Constitutional Tribunal then stepped in again and proposed a very sensible way to value the bonds that involved neither a specious CPI calculation nor hiring a group of economic historian to carefully recalibrate historical price indices. (Obviously, I would have preferred the latter!) It suggested rather that the value of the bonds be converted into U.S. dollars, which would then be compounded at the contemporary rate on one-year U.S. Treasury obligations.  

OK, fine. In his testimony, Mr. Koenigsberger complains about that valuation method, but it is way more sound than running it through the Peruvian CPI. But his complaints are not groundless, because the Peruvian government used a bogus exchange rate to convert the nominal value of the bonds from gold soles into dollars. Frex, at a time when the black market rate of the gold sol was around 44 per dollar, the government used a rate of 1,029 per dollar.

So he has a point.

But the claim of $1.6 billion is too big. It involves making some heroic assumptions about the accuracy of the CPI under a long and extremely volatile time period. 

What would a fair value be?

Well, on page 5 of the arbitration filing, Gramercy implies that the bonds in its possession had a face value of approximately one billion gold soles around 1972.  The exchange rate in that year was around 43 per dollar. Using a 6% interest rate would value the bonds at $283 million in 2015; alternative methods (discussed here) would put it between   $190 and $321 million. If you assume that the bonds were not dated 1972 but rather 1969, the valuation range would increase to $222−$386 milion.

That is a lot more than Peru wants to pay, but it is a lot less than $1.6 billion. 

I were the Peruvian government, I would offer to settle for around $200 million. I have no idea what Gramercy paid for the bonds. Mr. Koenigsberger says on page 11 that Gramercy bought them at a discount, but we do not know at a discount off what. Face value? CPI-adjusted face value? Something else? I suspect, however, that the company likely paid rather less than $200 million for the bonds and should be willing to settle. 

Or they could ask Fat Crybaby to make a deal for them. He does deals, right! Gangsta lean straight from Queens, only not lean, not at all. 

Vultures attack Peru!

 In 1968, a left-wing military government ascended to power in Peru. It expropriated a lot of things, not least a bunch of American investments. The Americans had the power of the United States behind them: as I explain here, they were more than fairly compensated. Domestic Peruvian landowners, however, were of no interest to Richard Nixon. They were forced to accept Agrarian Reform Bonds denominated in the local currency and paying between 4% and 6%  interest over 20-30 years. (See page 5.) 

Unfortunately for everyone involved, the Peruvian government stopped paying interest on the bonds around 1982. (See page 4.) Why they defaulted, I could not tell you: the country was entering a period of hyperinflation that reduced the effective value of those bonds to zero. If the Belaúnde and García administrations had just kept on paying, the debt would have been inflated to nothing and the current government would have been spared a major headache. 

But default they did, and thus the current problem. In 2001, the Constitutional Tribunal declared that the government had to repay the bonds with some adjustment for inflation. To do otherwise violated Article 70 of the Peruvian constitution: “No one shall be deprived of his property, except, exclusively, on grounds of national security or public need determined by law, and upon cash payment of the appraised value, which must include compensation for potential damages.”

The decision was not a surprise to the Peruvian government. In 2000 the Fujimori administration (then on its last legs) issued a decree which said that the government would honor the unpaid debt by converting it into U.S. dollars at the exchange rate prevailing at the date of issue and then applying a 7.5% interest rate, compounded annually. So far so good ... except the government did not move to pay cash. Rather, it moved to issue interest free debt for the face-value of the bonds due in 2030. In other words, the compensation offered was fair, but wouldn’t be paid for three decades. (Page 16.)

The creditors were not satisfied. When the new Toledo administration proved unwilling to alter the terms of the Fujimori offer they went back to court. In 2004 the Tribunal declared that bondholders could accept the government’s terms, but only if they wanted to. So it went back into the government’s lap.

The Toledo administration ran out the clock and handed things over to incoming President Alan García in 2006. (Yes, that would be the same García who helped create the problem when he was president in 1985-90.) President García did nothing, handing off the problem to the newly-elected Humala administration in 2011.

Three months after President Humala’s inauguration, the bondholders sued again. In 2013, they got a decision. The Tribunal once again ruled that the government had to pay. But the Tribunal also said that the government didn’t have to simply run the value of the debt through the Peruvian CPI, because that would lead to number so high that it would break the Peruvian fisc. (Page 18.) It suggested rather that the value of the bonds be converted into U.S. dollars, which would then be compounded at the contemporary rate on one-year U.S. Treasury obligations.  

Basically, the method gives the bondholders what they would have earned had they plonked their money into one-year T-bills back in 1969 and kept renewing them. Maybe not fair, but not patently unfair either: the rates on T-bills were higher than the 4%-6% coupons on the original Peruvian bonds until around 2001. (See below.)

How much money are we talking about? In 2006, the total unpaid principal came to about 2.5 billion old soles. (They were actually called “gold soles,” but old soles is more straightforward.) Adjusting for inflation the simple way (which, remember, the Tribunal said not to do) would bump that up to 10.0 billion new soles, or US$3.5 billion, plus unpaid interest. The investors claim that they are owed $5.1 billion with interest using this method. (Page 7.) 

Using the suggested Peruvian method (compounding at contemporary T-bill rates, see page 2), the 2006 value of the debt would be around 1.9 billion new soles, or US$568 million, assuming that it all dates back to 1969. (I have interpolated for the missing years. Data from the Fed here.) Unpaid interest included. 

Alternatively, the government could just convert the debt to dollars at the exchange rate at the date of issue and compensate investors at the interest rate they were promised. At 6%, that would come to 1.7 billion new soles, or $531 million. (The actual amount would be lower, since the actual coupons varied between 4 and 6 percent.) The result would also be similar to assuming that the investors would have plonked their money into long-term T-bills: the 20-year Treasury bond yielded 6.32% in 1969. At that interest rate you get a 2006 value around $595 million. 

So far, so good; it may not be what the bondholders want, but the proposed method seems to give fair and reasonable values. So why are vulture funds involved? What is all the yelling about?

Well, the government proposed that the value of the bonds be converted to U.S. dollars not at the 1969 official exchange rate of 38.7 per dollar, or the open market rate of 43.5 (which the above calculations use) but the somewhat absurd rate of 1,029 per dollar. (Page 8.) 

Under that calculation the debt would have been worth in 2006 (for comparability) 78 million new soles ... or only $24 million. 

My. That is some difference! Add to that insult the injury that the government intends to compensate different bondholders in different ways and still has not made it clear whether they will be paid in cash, bonds with a market value around the equivalent amount in cash, or some other method. Plus other stuff. But that exchange rate calculation is where the Peruvians are giving the bondholders a giant middle finger and the real source of the trouble. 

So now the vulture funds intend to attack Peru, first in Peruvian courts, then using the stipulations in its Free Trade Agreement with the United States. If that fails, they will presumably do the full Argentina. I cannot say if they will be successful, but Peru (unlike Argentina) needs outside funding. 

I understand why Peru is wary. The national debt is a little under $38 billion, with about $18 billion of that owed to foreigners. Interest on the foreign debt amounts to about $1 billion per year. Why add to that? 

But still. There is something a little gratuitous about this, even for me. The $5 billion that the bondholders want is too much, but something on the order of $500-$600 million is entirely reasonable and entirely affordable. Lima should pay.

A review of the Empire Trap misses the big finding, with special attention to how Richard Nixon slapped around Peru

 Negative reviews are not disheartening. People get things wrong. Interpretations differ. These are discussions worth having. 

But it is disheartening when a reviewer you respect appears to have just plain missed one of the primary empirical findings of your work. I do not mean that the reviewer thinks my data is incorrect. I mean that the reviewer did not note that I provided any data at all!

Here is Elizabeth Cobbs Hoffman’s review of The Empire Trap. It is a well-written review that makes very cogent criticisms. I liked it, even if it is less than positive!

But there was one place in the review where I slapped my face with my palm and started to cry, “Where, dear Lord, did I go wrong in my writing?” 

Here is where I did the self-face-slapping. (It sounds better in Spanish: autobofetada.) In the 1970s, the Peruvian military government nationalized a swathe of American investments. The incoming Nixon administration put the screws in, yanking aid and making other dire threats. Regarding my account of the episode, she writes:

The military junta in control of Peru in the early seventies, for instance, nationalized a number of industries, from oil to copper to sugar. President Nixon, the State Department, and the CIA all trembled that retaliation might “lead to reprisals against other U.S. investments, alienate the Peruvian people and . . . perhaps push Peru further toward . . . relations with the Soviet bloc” (p. 372). American investors were effectively held hostage, lest the local government switch sides in the Cold War. In Peru as in countless other countries, what followed was essentially a long, drawn-out process of rescuing the hostages. Maurer calls these negotiations “imperial intervention,” though it’s hard to see how the horse-trading that achieved partial compensation for dispossessed investors was terribly different from any other bruising business transaction. Both sides sought the best deal possible — and peaceably came back to the table when they saw ways to advance their interests. 

I added the italics, because one of the big empirical findings of the book was that over the entire 20th Century the United States managed to extract rather more than market value for aggrieved American investors. Seriously! That was one of the big findings. If the Red Army was not involved, then the U.S. government helped American investors beat foreign governments again and again and again, across administrations and across continents. I tried to highlight it over and over throughout the book, including the first chapter. Right here on page 2:

The Empire Trap is about the shift from politicized confrontations like the imbroglio of 1900 to legalized disputes like the more orderly affair of 2007. It advances four basic findings. First, American government intervention on behalf of U.S. foreign investors was astoundingly successful at extracting compensation through the 1980s. [Italics in the original.]

I am not bothered that someone might disagree with the finding; I am saddened that someone would miss it altogether.

Specifically, the U.S. managed to obtain full compensation in Peru. My primary evidence comes from secret communications between the expropriated companies and the American negotiators. The companies listed the minimum valuation that they would accept as compensation. These numbers were not negotiated by the Nixon Administration: the companies laid them out. 

And the companies all in fact got more than those numbers! My argument largely stops there. But it is possible to check the result against market benchmarks for the two largest investments in Peru: the Cerro copper mining enterprise and the IPC oil company. (They together received 57% of the total compensation Peru paid.) 

The copper company, Cerro Corporation, was publicly traded. The market value of its Peruvian investments was $58 million in 1971. One might argue that fear of the Peruvian “Revolutionary Government of the Armed Forces” had pushed down that value, but Cerro’s p/e ratio matched that of other copper producers, including those geared towards domestic American production. Cerro told the Nixon Administration that it wanted $65 million; it got $68 million. (It publicly asked for $133 million, but it would, no?)

The International Petroleum Company (IPC) was not publicly traded, but it did provide data that lets us see if it received fair compensation. It privately told the Nixon Administration that it wanted $20 million. It publicly told Peru that it wanted $85 million, a palpably ridiculous number. Now, on paper Peru made like Michael Corleone.

The Peruvian government, however, paid compensation to the U.S. government, not the companies directly. The U.S. government diverted $22 million from the settlement to IPC. In practice, then, the Peruvian government agreed to overpay other American claimants (in the knowledge that the funds would be diverted to IPC) in order to be able to pretend that it paid nothing for the oilfields and refineries. 

Was $22 million a fair amount? Yes. Over the previous decade, IPC earned around $1.9 million per year at a time when U.S. oil companies traded at p/e ratios rather below five. Unless you believe that an investment in a country ruled by the Revolutionary Government of the Armed Forces was less than half as risky as an investment in a country ruled by Richard Nixon, then IPC got more than twice a market price.

(Hmm. Richard Nixon? Another day, another day.)

I am fine with saying that I am wrong in calling Nixon’s strong-arming of the Peruvians “imperialism.” I am also fine with saying that others have argued that covert intervention was not used to defend American investments ... although I believe that such a conclusion is incorrect, I am also well-aware that I may be placing too much weight upon a few internal communications and President Nixon’s taped offhand comments. And if you show me how, I am fine with saying that my numbers are wrong!

But please acknowledge that I brought some numbers to the table! (And not just for Peru; for every single expropriation of American natural-resource FDI from the 1890s through the 1980s.) Those numbers may well be wrong. But they took several years to gather! So even if the numbers are wrong, please acknowledge that the numbers are there? Or at least somebody tell me how my writing went so off the rails that a great historian could miss that I presented data to back my claims.

If you want to know more, a working paper about the Peruvian episode can be found here.

Thoughts and opinions are very welcome.

Does Russian expansion matter beyond the border regions of Russia?

Back in April 2012, we discussed the idea of the “1945 line” in comments. In October of the same year, we discussed it again in reference to the Falkland Islands. Key quote:

“Any planning for a war [with the U.K.] would need very careful diplomatic preparation these days, much more than in 1982. Back then, Argentina was following a long line of actions that held that the “1945 line” did not hold for European overseas territories. Morocco could invade the Spanish Sahara, India could fight a war with Portugal for Goa, Indonesia could march into East Timor and Dutch New Guinea. Given this history, it was not (completely) crazy for the Argentine generals to think that the world would accept their victory. (And to be honest, I suspect that they were right: if they had won, las Malvinas hubieran sido argentinas.)  But that is not the case today, even though plenty of countries support the Argentine claim [to the Malvinas].”

In short, there has been a remarkably effective taboo on unilateral annexations unless they involved European overseas territory. There is only one real exception: Israel after 1967. (Wikipedia adds Sikkim, which India annexed in 1975, but Sikkim had not been an independent state.)

How do we know that the taboo is effective? A case can be made that the taboo limited the extent of interstate wars in Latin America. For example, El Salvador and Honduras fought the Soccer War in 1969 over the status of Salvadoran settlers in Honduras. Since annexing the area settled by Salvadorans was not a possible outcome given the taboo, the OAS (and behind it, the United States) had the leverage to force Salvadoran troops to withdraw. In another example, Peru and Ecuador fought the Cenepa War in 1995. Neither side could reasonably expect to gain or lose any territory outside the disputed region, which made Peru more willing to accept a limited Ecuadorean victory even when Peru had military superiority. (See pages 2 and 3 at the link.) In short, one might believe that the taboo kept both wars from spiralling into protracted total conflicts.

OK ... but the taboo did nothing at all to stop Israel from annexing some border regions. Nor did it prevent the Eritrean-Ethiopian War from turning into a protracted total conflict. Similarly, Paul Collier made arguments that Africa might in theory be better off if governments feared that their states might be forcibly absorbed by neighboring states — but he rejected the implications of that idea by arguing that the lack of interstate conquest was overdetermined: the international taboo was just one of many obstacles to a would-be empire builder in modern Africa. (See pp. 184-85 of Wars, Guns, and Votes.) Considering those examples, perhaps other factors were at work in keeping Latin American wars from spiralling out of control.

The debate matters. If the taboo is key, then Putin’s actions matter beyond near-abroad of the Russian Federation. Successful expansion opens the door to similar actions all over the planet.

But if the taboo on conquest is just a reflection of some other deeper cause, then current Russian actions have implications only for future disputes that involve Russia. Interstate relations between other countries won’t change. David Weman is correct that this will be the only successful non-colonial annexation outside Israel, but maybe that does not matter.

So which is it? Does precedent matter? Will we see territorial aggrandizement that would not have happened absent current Russian actions?

The Pacific Alliance is not a big deal

I am in Colombia — if things go well I will be spending more time here — and the news is full of the Pacific Alliance summit in Cali.

The Alliance has a very clever logo, combining the flags of the different countries.

But it is nothing more than a commitment to liberalize restricitions on the movement of goods, capital, and (eventually, maybe) people between Mexico, Chile, Colombia, and Peru. There is no secretariat, no courts, no bureaucracy. Its founding document is called simply the “Framework Agreement.” If you click the link, you can read it: it is just a list of objectives, with a few statements about intergovernmental cooperation.

In Cali this week, the leaders agreed to eliminate tariffs on 90% of products between them. That is nice, but the countries do not trade a lot between each other and barriers (except in agriculture) are already pretty low. (The link goes to WTO trade profiles for the four current members. You can specific tariff profiles for the four countries here.) Moreover, with no common external tariff, the countries will need to develop rules-of-origin. In addition, with no common bureaucracy or legislature, non-tariff barriers will continue to loom large. And with no dispute resolution mechanism, the countries will continue to rely on the WTO for enforcement.

In other words, there is not a whole lot of there there. If you are an exporter from one of the members looking at another, then it might matter (depending on which products are included in the 90% to be liberalized) but the macro impact will be tiny. If they liberalize labor flows, then I will change my tune, but call me skeptical.

The breathless declarations from the FT, the Miami Herald, and the Economist are really not warranted. In fact, at the end of the article the Economist called the Pacific Alliance “a brilliant piece of diplomatic marketing.” That sounds about right.

Brazilian magnaminity explained

  A while back, we discussed Brazil's decision to grant Paraguay a greater share in the rents from their joint hydroelectric project.  There were a number of excellent hard-headed reasons for the Brazilian government to do this, but we've just discovered another:  persuading the Peruvian government that it is worth getting into bed with Brasilia. 

Brazil is contemplating five hydroelectric projects in Peru that would cost about $15 billion and generate 6 gigawatts of power, 80 percent of which would go to Brazil.  That represents about 4½ percent of the country's current installed capacity.  The expansion will involve Electrobras purchasing Peruvian energy companies, something the Peruvian government won't accept lightly unless it trusts that it can negotiate with Brasilia.   After all, as things stand now, Brazil approaches its neighbors much as Teddy Roosevelt did:  it speaks softly, but it carries a big stick.  That makes people nervous.

By showing grace towards Paraguay, the Brazilian government showed Peruvians that it would be a reliable partner and take their interests into account, rather than using its power to take up all of the surplus.  We'll see if that attitude lasts (power corrupts and all that) but for now Brazil is managing the sudden and surprising emergence of its informal empire in South America with grace, aplomb, and plausible deniability.