The Power and the Money

Correction: Colombia does not have birthright citizenship and seems to be trying to avoid letting Venezuelans get it

 In the Western Hemisphere, birthright citizenship is close to universal. I thought that this included the Republic of Colombia, as did the author of this article in the Harvard Human Rights Journal. 

But it turns out that Colombia is an exception. But it is not clear-cut. Here is Article 96 of the 1991 Constitution:

Article 96. Colombian nationals are:

1. By birth: 

(a) Natural-born Colombians, under one of two conditions: that the father or mother were natural-born or nationalized Colombians or that, being children of foreigners, one of their parents were domiciled in the Republic at the moment of birth.

(2) The children of a Colombian father or mother who were born overseas and later domiciled in Colombian territory or registered in a consular office of the Republic.

So what is the definition of the word “domiciled”?

Colombian jurisprudence has held that “domiciled” does not mean “resident.” A 1993 law made it more specific: “Residence is understood as residence in Colombia accompanied with the intention to remain in the national territory according to the relevant stipulations of the civil code.”

Or, if that wasn’t clear: if you don’t have legal permission to reside in Colombia, then your Colombian-born children don’t automatically get Colombian citizenship.

Specifically, if you want your kids to be natural-born Colombians, then you need to have a type R resident visa; a type M work, marriage, partnership or business visa or a type TP3, TP4, TP5, TP7, TP9, TP10, or TP15 temporary visa. Now, in case you were wondering, it turns out that refugees qualify for a TP9 visa. (The link explains all the TP visa types.)

But here is the rub. Decree 1288 granted some sort of permanent status to 442,462 Venezuelan refugees, on top of the 380,000 who had already gotten legal status. Holders of the new status can work in Venezuela and access public services. 

The Venezuelans do not get TP-9 visas. Rather, according to Article 2 of the implementing regulation, they get a two-year renewable something “issued in order to authorize the stay of Venezuela nationals who find themselves in the national territory without the intent to settle permanently, which is why it is not equivalent to a visa.”

This strikes me as a bad idea. I understand why the Colombian authorities want to pretend that the Venezuelans will all go home soon. But that is a bad bet. It is hard to imagine that granting TP-9 visas instead of the temporary whatever-it-is would engender that big an outcry. But it would at least insure that you avoid creating an underclass should things not work out and the millions of Venezuelans wind up staying in your country.

Unless there is some legal requirement at work, Colombia really should change this policy. As insurance for their own future. Because some eggs cannot be unscrambled, and the great Venezuelan exodus is beginning to look like one of them. 

Observatorio argentino 38: Who needs checks and balances?

It is astounding to realize just how much power the President of the Argentine Republic possesses. 

This came home to me while I was checking over how President Macri recently tightened the immigration laws. Context: Argentina has a rather liberal immigration regime. (I am surprised even more Venezuelans have not shown up. I guess the limiting factor is that they need a job offer to get a work visa, although there are no limits on such offers.)

Last year, the Macri administration wanted to tighten the law. (See Observatorio Argentino 32.)  The changes were small, but they involved specific clauses in the legislation itself. You can find the original 2004 text here. So they could not be made via simple executive order.

In other presidential republics, the only option would be to go through Congress. Get the law changed. How hard that is depends on the country. 

An American president needs to be a hella politician. They need to convince a legislator to introduce the legislation; they then need to cajole the Speaker and Majority Leader to bring it to a vote. Even then, under the new 21st Century norms, they need at least 60 senators to vote yes. 

A Mexican president is a little more powerful, but only a little. Under the original wording of Article 71 of the constitution, they could introduce bills straight to the relevant committee. But by itself that gave them only a little more power than an American president. In 2012, however, a constitutional amendment changed Article 71 to give the president the ability to introduce two bills every session that go straight to the floor. Each house then gets 30 days to vote on them; if they don’t hold a vote, then the bill becomes the first item on the agenda in the next session. So congressional leaders can still delay things; and note that if they delay both initiatives the President cannot introduce any in the next session. So even with the additional post-2012 powers, they need to be good legislative politicians if they want to get anything done. 

A Colombian president is more powerful still. She or he can introduce bills and demand that Congress vote on them within 30 days, with no limits. But Colombian presidents are not legislators. They have no direct control over the amendment process. And, of course, Congress can just say no. 

But Argentina? Sweet Mary. Under a 1994 amendment to Article 99 of the constitution, the President can decree pretty much anything. The article is nuts. Read section 3 for yourself:

The Executive Power shall in no event issue provisions of legislative nature, in which case they shall be absolutely and irreparably null and void. Only when due to exceptional circumstances the ordinary procedures foreseen by this Constitution for the enactment of laws are impossible to be followed, and when rules do not refer to criminal issues, taxation, electoral matters, or the system of political parties, he shall issue decrees on grounds of necessity and urgency, which shall be decided by a general agreement of ministers who shall countersign them together with the Chief of Cabinet.

Within the term of ten days, the Chief of Cabinet shall personally submit the decision to the consideration of the Joint Standing Committee of Congress, which shall be composed according to the proportion of the political representation of the parties in each House. Within the term of ten days, this committee shall submit its report to the plenary meeting of each House for its specific consideration and it shall be immediately discussed by both Houses. A special law enacted with the absolute majority of all the members of each House shall regulate the procedure and scope of Congress’s participation.

Yes, that is as crazy as it sounds. Exceptional circumstances are not defined, and every president since 1994 has interpreted the clause to mean that they can make legislative changes to anything they want whenever they want. In theory, the joint standing committee has ten days to vote and then Congress can reject the decree. In practice, the committee often stonewalls or deadlocks and then the decree just goes ahead and becomes law, poof, just like that. 

How the hell did that happen? The 1853 drafters of the constitution did not set out to create an elected dictatorship. But they got an unelected dictatorship in 1930, when the military seized control. The Supreme Court remained in session, though, and it validated the decrees handed down by the junta. In 1947, the Court tried to block a presidential decree ... at which point President Juan Perón impeached all but one of the justices. You won’t be shocked to find that the restaffed Court ruled in Ziella v. Smiriglio:

The decree-laws issued by the de facto government are valued by reason of their origin and since they have the force of law they continue in force even though they have not been ratified by Congress as long as they are not repealed in the only manner in which they may be so, that it is to say, by other laws.

Still, only 15 decrees were issued between 1930 and 1982. When democracy returned for good in 1982, the new civilian presidents were loth to give up the power. Raúl Alfonsín used the power ten more times, mostly for small-bore stuff; his big decree replaced the peso with the austral. 

But Alfonsín was succeeded by Carlos Menem, a Trump before Trump, a Berlusconi before Berlusconi. Menem liked executive power. He started by increasing the size of the Supreme Court from five to nine. He then issued a decree that seized most bank accounts and replaced them with government bonds. People sued. 

But the packed court ruled that President Menem was entirely within his rights as long as Congress did not explicitly repeal the decree with another law. American legal scholars will be amused to know that the Argentine justices cited Home Bldg. & Loan Association v. Blaisdell (1934) as justification for a decision ratifying exactly the sort of thing that Hughes, Cardozo and Brandeis were trying to prevent.

The 1994 constitutional amendment creating the new Article 99 was an attempt to limit the decree power. As such it failed. Menem would go on to issue over 300 decrees. (Note again that before Menem there had been only 25 presidential decrees in the history of the Republic.) Congress would not even bother to create the joint standing committee until 2006. And it remains toothless. 

Consider the immigration reform that started this discussion. What happened after President Macri issued it? Well, the joint standing committee has only 16 members but it could not get everyone to attend. It voted 6-5 to support the decree on February 22, 2017 ... but because there was no majority its recommendation was never reported out. The decree continued as law. In October, a federal judge ruled that it was in fact law even though the whole Congress had never voted. (In this he had ample precedent, although Argentine judges are not formally bound by precedent the way they are in the Anglosphere.) He also ruled that the deportation of criminals was somehow not a criminal matter and thus the reform was kosher under Article 99. 

No matter how you slice it, it is unbelievable that a democratic president has the power to make law unless Congress actively says no. Giving Donald Trump the power of a Mexican or Colombian president would change little; GOP initiatives are hard to pass because they are unpopular. Allowing the President to force votes on them would not help. But giving Donald Trump the power of an Argentine president? Well, that is scary.

And that may be one of the prosaic reasons why Argentina is not particularly well governed. They really ought to do something about that Article 99, I think. 

Colombian peso update, part 2

In 2015, I predicted that Colombia would not chop three zeros off the peso until the May 2018 presidential election. The last time around, Congress balked at the cost of replacing the bills. In 2016, though, the Colombian central bank issued new bills that replaced the three zeros with the word “mil.” On February 5th, the head of the central bank, Juan José Echavarría, pronounced himself in favor of redenominating the currency since the new bill design would make it cheap: just let new bills without the word “mil” circulate alongside old bills that have it. 

Congress took up the discussion on February 8th. 

So why bother? Well, first, redenomination is now free. (Although the private sector will have to pay some cost for new software.) That is a big reduction from the $123 million it would have cost to recall and reprint or remint everything. Second, the finance minister, Mauricio Cárdenas, wants Colombia to have a currency close in value to the prestigious hard monies in the world. This is supposed to have good psychological effects; hey, we have a serious currency now! Finally, getting rid of some zeros just makes things easier. No need to write down long strings of numbers or talk about “millones de millones.” 

Of course, introducing the new redenominated bills piecemeal takes away the benefit of forcing people to turn in unregistered stashes of cash. It isn’t that doing so would uncover nefarious doings and bring crimelords to justice; they use dollars. Rather, forcing people to exchange old bills for new would provide useful data as to the size of the underground economy.

But that looks to be off the table. What is left amounts to prestige. 

Nothing wrong with prestige! I would vote to redenominate were I a Colombian congressperson. But still bet Congress does nothing unless the new president takes the up mantle of giving Colombia a currency that can beat anyone else in Latin America. Absent that, why burn the political capital? 

Where to now in Colombia: the law and the polls

This post, unlike the previous one, is a backgrounder.

Plebiscites in Colombia are governed by Title IV, Chapter 1 of the Constitution, aka Articles 103-106. Essentially, it allows the President to call a referendum with the approval of the Senate. The chapter does say, “The decision of the people shall be binding,” but it vague on everything else.

And so, the Constitutional Court was called to issue an opinion as to the referendum law.  It said that a no would only prevent the President from implementing this accord. Another one could be negotiated or Congress could overturn the referenda and pass the accord via legislation. Similarly, with the approval of the Senate and his cabinet, President Santos could amend the accord and send it back out for a vote. 

There is still a mystery about the polls. Early polls in June recorded 53-point margins. But that soon changed. By mid-August, polls were all over the map. Cifras y Conceptos and the Centro Nacional de Consultoría indicated “yes”; Ipsos and Datexco projected “no.” In September, however, the polls were back to showing support, although one tracking poll showed support declining in the middle of the month. (Page 3.) That said, by the end all the polls projected a victory.

One facil explanation is low turnout. And it is true, turnout was shockingly low at 37%. But that only pushes back the mystery: the polls expected 67% turnout. At the end of the day, it is still a mystery. More data analysis is needed ...

How Colombia voted

The Colombia referendum result was a huge shock. We still don’t know why the polls were wrong and wrong by no small margin. But that’s a second-order question. The first-order question is: why did Colombians vote against the peace pact?

Jim Robinson, an expert on Colombia and a friend of mine, has some cogent things to say:

At minute 0:45, he states that the places where the FARC guerrillas have been active the longest voted against the peace treaty. Given that those are the places that would benefit the most from peace, he says, it’s telling as to how much distrust there must be towards the Marxist forces.

But is he right? Did the areas with most experience of the FARC vote against the pact? A former executive-education student of mine, David Ibáñez, is an economist with the Colombian Ministry of Commerce, Industry and Tourism. He recently sent me data indicating that Robinson is wrong. The areas most afflicted by the fighting voted no. Rather, it was the center of the country, where the government long ago battled the FARC into impotence, what voted against the pact.

Lets start with these two maps:

The map on left shows the election results, by province. Green means yes, orange means no. The map on the right shows the intensity of political violence: the redder, the more deaths in political violence. There is a remarkable amount of overlap between the two maps ... with the most afflicted areas voting for peace. 

More anecdotally, but tellingly, locales known for particularly bloody battles and massacres all voted yes by large margins:

Caloto	73%
Cajibio	71%
Miraflores	85%
Silvia	73%
Barbacoas	73%
Tumaco	71%
San Vicente del Caguan	62%
Apartado	52%
Mitú	77%
Valle del Guamez	86%
Macarena	73%
Puerto Asis	57%
Turbo	56%
Toribio	84%
Bojaya	96%

The correlation holds up at a more fine-grained level, using the proportional amount of internal displacement as an indicator of the intensity of the violence:

The correlation is even more strong between rich jurisdictions and poor ones, measured by the unsatisfied basic needs index. Poor jurisdictions in Colombia voted yes rather dramatically:

It is a reasonable working hypothesis that poor areas were more vulnerable to FARC violence, even if they had not experienced it in the recent past. 

Finally, most of the remaining variation can be soaked up by Colombia’s unfortunate political polarization. (That is something very depressing for an American to report; I like to hope that our country’s recent descent into maddening political tribalism isn’t a universal phenomenon.) Consider these two scatterplots. The plot on the left compares the size of the yes vote with the share that President Santos won in the last election. (Santos negotiated the pact.) The plot of the right compares the size of the yes vote and the size of the vote share for former-President Uribe (who vociferously opposes the pact) with the yes vote. The correlation is striking.

Uribe really has been vociferous in his opposition. The linked story (which you should read!) does not quite communicate just how vociferous he has been. He did not just argue that a vote for “the Santos deal” was a vote for impunity (which he did), or paint it as a general referendum on the Santos administration (which he did), or that it was insane to allow the FARC to become a political party (which he did), but also that the pact would change the very definition of family ... this in a country where the Constitutional Court only ruled for marriage equality this April in a very contentious decision. 

This data brings home why referenda are generally a terrible idea. In this case, people in wealthier areas, where the FARC had long been defeated and excluded, could turn the referendum into a proxy vote on the Santos administration or gay marriage. Others could indulge their understandable revulsion at forgiving the FARC. I understand the feeling, but it is not one conducive to ending civil wars.* Perversely, the referendum failed because the FARC was near defeat ... in the places where the guerrillas were still a threat, people held their noses and voted for peace. 

Thoughts? There is clearly an article here ...

A terrible surprise from Colombia

More later. But right now, this is unexpected very bad news. For those of you who do not want to click the link, the Colombia peace deal to end a half-century of civil war just failed in a 50.2-49.8 percent vote. The polls predicted an overwhelming victory. May God save Colombia.

The above photo is from two years ago, at a hotel in La Guajira which only existed because of the peace process. I think it will survive the failure. In fact, one reason the referendum may have failed is that peace had already arrived in many locations (but not all!) and that may have made voters complacent.   

There is a good case to be made that referenda work well as legitimizing a decision to undertake a peace process, but are a terrible way to try to finalize peace deals once made. That case just became quite a bit stronger.

There is also a case that the polling industry may be in trouble.

Is Ecuador’s resource nationalism hurting it in the oil price crash?

 In comments, reader Cad points out that Ecuador’s oil policy has a downside: the same fixed-price policy that lets the country take all of a price windfall exposes the government to large losses in the event of a price crash. 

Under the best-known service contracts, the Ecuadorean government pays private companies a fixed price per barrel. It then resells the oil at the market price. In other words, it takes all the upside from price rises. (This new policy was the cause of several of the country’s big expropriation disputes.) A subset of service contracts allow private companies to help Ecuador’s state-owned companies (Petroamazonas and Petroecuador) extract oil, also at a per barrel fee.

Finding out Ecuadorean operating costs has proven remarkably hard, although there is some rough evidence that costs have recently risen rather dramatically. What we do know, however, is how much money crude oil extraction has generated for the Ecuadorean state. That revenue, in real terms, has fallen 55% since 2013. On a per barrel basis, revenue for the state has fallen 56%.

Of course, prices have also fallen. Ecuadorean exports fetched $95.63 per barrel in 2013, $84.14 in 2014, and $42.00 in 2015. (Those are weighted averages.) That is a ... 56% decline, which is the same as the per barrel decline in state revenue. 

So not much evidence that resource nationalism has hurt revenues from those data.

What about international comparisons? Sadly, the data for Colombia only runs through 2014, but the Mexico data is available through 2015. In Mexico, oil revenues declined 55% in 2013-15. Revenues per barrel, however, declined by only 50%. So Mexico did slightly better than Ecuador at squeezing the upstream oil sector. (These figures do not count revenues from excise and value-added taxes on refined products.) 

In short, Ecuador’s oil policy has not led the country to do worse than would be expected given the crash in oil prices ... but it is possible that an alternative policy would have kept government revenues from falling quite as far. 

Now, Mexico’s oil tax policy is not particularly recommended. It is fiendishly complex, but it essentially boils down to a tax on Pemex’s gross revenues. The result is a company starved of resources and declining production. (Down 10% on 2013, whereas Ecuador is up 3%.) It is not clear that Mexico is an example to emulate.

But at least from these data it appears that Cad was incorrect when he wrote, “Mashi Correa’s greedy and short-sighted gamble was a disaster and now the people of Ecuador stand to suffer.” Rather, thus far his oil policy seems to be holding up remarkably well both in comparison to the price crash and to at least one benchmark comparison. Colombia may have done better; when we have the numbers, I will analyze them.

Colombian peso update

 Colombia did not redenominate its currency in 2014. The Santos administration says that the plan to chop three zeros off the money continues to be in play.

Short version of what happened: Congress resisted the cost and President Santos decided not to use his considerable legislative power to push the bill through. In February 2015 the country introduced a 100,000-peso bill, temporarily reviving the debate. But for the moment, nothing is in store. Given that Santos has a lot on his plate, I doubt it will happen for several years ... I would be surprised if it happens before the 2018 election.

Hope that helped!

The near-war with Colombia and the sad state of the Venezuelan military

 The Bolivarian Republic had a “Georgia moment” in 2008. I’m referring to the Russo-Georgian War of that year, when the Russians discovered just how severe were their training and mobilization problems. The result was a sustained modernization program. The Russian armed forces are not a par with the United States, of course, but they have improved markedly. The Russian Federation will not become a peer competitor with the United States because of its long-term economic and demographic problems, not because its military institutions are irredeemably broken.

In 2008, Venezuela almost went to war with Colombia. On March 2, President Chávez ordered ten battalions to the Colombian border. 48 hours later the U.S. embassy reported that “less than 12 light tanks or armored fighting vehicles have been seen preparing for movement at Carora in the western state of Lara and at Fuerte Mara in the border state of Zulia.” The embassy went on to say,  “A lack of training, rehearsal and failure to invest in unglamorous military equipment like trailers and cargo planes has slowed the Venezuelan military's mobilization. This may explain the lack of patriotic visuals, from this otherwise propaganda savvy leader, of troops marching to the front.”

Six days later, not much had happened. Five heavy equipment transports, capable of carrying five main battle tanks or two light tanks, finally made it to a military base in Valencia. There was also the first evidence of desultory helicopter movements. The military was apparently buying fuel on the spot from commercial vendors. The U.S. believed that the AK-103 assault rifles purchased from Russia had been issued to Army units, so that was something. But generally, the mobilization was a complete disaster.

Chávez was a political genius and rarely averse to recognizing reality ... so on March 9th he pivoted to peacemaker after it became clear that only a third of his troops had made it to the border. (And those were operating beyond their logistical tail.) The deterioration in capabilities was a little ironic considering that the logistical and aviation branches had remained loyal during the 2002 putsch.

How did President Chávez react to the Georgia moment? Not by modernizing. Rather, in December 2009 a new law gave the President the authority to declare military districts that could be used to undercut recalcitrant state governors; it also further pushed resources towards the militia and allowed foreigners to serve in the military. (Our 2009 post on the subject is here.) Presumably the relaxation on the citizenship requirement was made to allow Cubans to take up posts in the service.

The result has not, shall we say, been an increase in operational capabilities. 

In other words, Guyana can breathe easy. So can Aruba, for that matter. Seriously, stop worrying. This stuff stopped around the same time that Caracas realized its military was useless for interstate war.

The Philippines move right!

 Carlos Yu and I have a long-term plan to write a book on the Philippines under American rule. But it won’t be done until 2018 at the earliest.

One thing that we discovered while writing our book on the Panama Canal was that Panamanians drove on the left until April 15, 1943. The reasons for the changeover are obvious; less so is why Panama started out on the left. The influx from the British West Indies is often cited. (Colombia did not start to regularize its traffic codes until Panama had already seceded but appears to have always driven on the right.) Argentina drove on the left between 1889 and 1945. The first traffic code, from 1872, mandated driving on the right but the country switched to the left in emulation of the British-built railroads. (You can still see this in the Buenos Aires subway.) It moved back to the right on June 10, 1945.

But neither of us had any idea that the Philippines also drove on the left until March 10, 1945! That is more than slightly amazing. The vast majority of cars were imported from the United States. So how did the rule survive so long? Even more amazing, the reason for the shift given in the text of the executive order implies that imported cars had been physically reconfigured to right-hand-drive before the switch:

EXECUTIVE ORDER NO. 34: AMENDING TRAFFIC REGULATIONS FOR VEHICLES

WHEREAS it is deemed advisable to change the existing regulations providing for the driving of vehicles on the left side of the road so that the vehicle traffic in the Philippines shall conform with the practice of driving on the right side of the road obtaining in most countries of the world;

WHEREAS such a change would bring about certain economic advantages to the people of the Philippines in that it would reduce the price of motor vehicles imported into the Philippines from the United States;

WHEREAS during the present emergency the great majority of the motor vehicles of the United States Army used on the roadways in the Philippines are right-hand driven and the drivers thereof are accustomed to driving on the right side of the road; and

WHEREAS the present affords the most propitious opportunity to effect this desired change;

NOW, THEREFORE, by virtue of the powers vested in me by the Constitution and the laws of the Commonwealth of the Philippines, particularly Commonwealth Act No. 671, I, Sergio Osmeña, President of the Commonwealth of the Philippines, do hereby order;

1. That section 60 of the Revised Motor Vehicle Law, Act No. 3992, be amended, as it is hereby amended, to read as follows:

“SEC. 60. Drive on right side of road.—Unless a different course of action is required in the interest of the safety and security of life, person, or property, or because of unreasonable difficulty of operation in compliance herewith, every person operating a motor vehicle or guiding an animal drawn vehicle on a highway shall pass to the right when meeting persons or vehicles coming toward him, and to the left when overtaking persons or vehicles going the same direction, and, when turning to the left, in going from one highway into another, every vehicle shall be conducted to the right of the center of the intersection of the highway.”

2. That all other sections or parts of the said Revised Motor Vehicle Law and amendments thereof shall be considered as amended or modified as they are hereby amended or modified accordingly so as to conform with the provisions of the Executive Order; and all other provisions of law inconsistent with or contrary to the provisions of this Executive Order shall be considered repealed as they are hereby repealed.

Done at the City of Manila, this tenth day of March, in the year of Our Lord, nineteen hundred and forty-five, and of the Commonwealth of the Philippines, the tenth.

SERGIO OSMEÑA
President of the Philippines

Unless most American imports had been reconfigured before sale in the Philippines, then there is no reason why switching the side you drive on would make imports cheaper. Today the Bahamas drives on the left but enforces no such rule; neither does the U.S. Virgin Islands. Was there a mechanics lobby group in the Philippine Islands that insisted on reconfiguring imports before 1945? 

There has got to be a story here.

We are back! Well, half back, anyway

We apologize for the radio silence! Doug Muir is swamped with repairing Kosovo. Noel Maurer was working in a coal mine.

Really, those hills behind his left shoulder are piles of coal.

Anyway, your regularly-scheduled blogging will be returning.

In other news, we report that at least one Jamaica-born-and-raised Jamaican-American gives a qualified thumbs-up to Mayor Ford’s accent. Meanwhile, we await Randy McDonald to explain the Ford phenomenon to us and continue to beg for input from Alex Harrowell on British political polarization or lack thereof. Enjoy!

Is American democracy doomed?

Juan Linz has gotten a lot of well-deserved attention in the blogosphere recently. Linz wrote that presidential regimes tend to break down because of irreconciliable conflicts between the executive and the legislature.

Many observers, not least Matt Yglesias, have interpreted Linz as stating that the problem is that presidential system allow for gridlock between the executive and the legislature, especially when the parties are ideologically coherent and commited. Kind of like, uh, the modern United States. So are we doomed now that our parties have become European-style organizations?

Well, maybe. The cross-country data is mildly supportive of the presidentialism = bad hypotesis, but is by no means a slam-dunk. Moreover, while presidential regimes are more likely to break down there is really rather little evidence that the reason was gridlock between the executive and the legislature.

The closest parallel to the current American situation is Chile in the early 1970s — but it is not very close. In the 1969 election, the People’s Union (Unidad Popular) got 60 out of 150 seats in the lower house. The Christian Democrats got 57 and the National Party received 33.  The next year, Salvador Allende (the People’s Union candidate) won a plurality of the presidential vote, throwing the election to Congress, where he won with Christian Democratic support. (The Nixon Administration tried to prevent his election but failed.) Allende started a program of widespread nationalizations using existing legal authority under a 1932 law that allowed for the temporary requisition of basic goods when in the public interest. (The law also allowed the executive to put companies into temporary receivership should they be closed by a strike.) In October 1972, the Supreme Court ordered the President to obey lower-court orders overturning some of the nationalizations; Allende refused.

In March 1973, the opposition won the Congressional elections. The Christian Democrats aligned with the National Party to form the Democratic Confederation, which took 55.6% of the popular vote and 87 out of 150 seats. (The People’s Union got the rest.) Congress first attempted to amend the constitution, which failed. It then tried to impeach Allende, but could not get the requisite ⅔ majority. In May, the Supreme Court again reprimanded the President. At this point, Congress went on record asking for a coup. The coup happened on September 11, and its leaders proceeded to disband Congress and jail or “disappear” many of the legislators who had called on the military.

In short, the Chilean crisis was caused by a President who overstepped his authority and disobeyed the Supreme Court and resolved by a military that believed itself above the law. Suffice it to say that this does not characterize America’s current situation.

What about Argentina? Well, Argentina has had a lot of coups since the first one in 1930. In 1930, President Hipólito Yrigoyen’s party had comfortable majorities in both houses. The 1943 coup was against a dictatorship. The 1955 coup overthrew the elected Juan Perón — but Perón’s party controlled Congress by overwhelming margins. As in 1930, gridlock was not the issue. In 1962, the miltary overthrew President Arturo Frondizi: in that case, the prompt was Frondizi’s loss of several governorships to the Peronists in the ’62 elections. Again, gridlock was not the issue. The 1966 coup kinda sorta maybe if you squint could be gridlock-related — but really had more to do with President Arturo Illia’s confrontation with American oil companies. And 1976? That one was against President Isabel Perón’s supposed inability to defeat Communist guerrillas. Her coalition controlled Congress; once again, gridlock was not the issue.

What about Brazil? Well, the Revolution of 1930 took down an oligarchical government with an extremely limited electorate. Not really comparable. The 1964 coup knocked out a left-wing government with American support. It is true that Congress did not support President João Goulart. It is also true, however, that for his first three years Goulart presided over a parliamentary system in which power resided with Prime Minister Tancredo Neves. In 1963, a referendum restored President Goulart’s executive power: the result was paralysis. In that sense, the ’64 democratic breakdown is consistent with the hypothesis that presidentialism causes gridlock causes democratic failure. The problem is that the parliamentary system tried beforehand was doing no better at resolving the country’s political conflicts.

Colombia? Well, the 1953 coup happened after President Laureano Gómez seized dictatorial power. Gómez, in turn, won the 1950 election because the opposition boycotted it. Again, gridlock not the problem.

Let’s turn then to Mexico. Uh … no, let’s not. The 1910 revolution got started against a dictatorship. Porfirio Díaz government was many things, but gridlocked was not one of them. (In fact, Congressional votes were all unanimous after 1893.)

I could go on around the other countries of the Western Hemisphere, but the pattern is similar. Divided government is not the cause of most democratic breakdowns in Latin America. Chile 1973 is the closest, but that is one event, with little resemblance to the government-by-crisis currently affecting the United States.

So breathe easy America! We will get through this. How, you ask? Well, if the GOP keeps this up, it will lose a lot of elections, even with our gerrymandered House of Representatives ... and there is a lot of ruin in the United States. The worst-case scenario on the debt ceiling is pretty bad, and I would certainly a 5-10% fall in GDP and the corresponding human suffering a failure of the United States. But in the unlikely event that the Republicans take us there, they will not be in a position to take us there again.

The possible pernicious effect of the Colombian-American alliance

As recounted earlier, the United States and Colombia have a very close relationship. It is somewhat assymetrical, of course, not least in the sense that American public opinion barely thinks about it while it is a major issue in Colombia. That said, it is not (yet) a controversial issue in Colombia: the special relationship enjoys broad and deep support.

But should the alliance enjoy such broad and deep support? It got Colombia a free trade agreement with the U.S., which is almost certainly a good thing given the country’s economic structure. And it got Colombia a lot of American assistance against both organized crime and the FARC. So far, so good.

Suresh Naidu and Oeindrila Dube decided to ask if there might be any downsides.

They put together a list of all Colombian military bases from this site. (See map.) Then they interacted the presence of those bases in a county with changes in American military aid to Colombia.

If American aid finds its way to the paramilitaries via the Colombian military, then increases in aid should increase paramilitary attacks in the counties with a permanent military presence and not in other counties. (It is important to note that the dataset involves only regular military bases, not the National Police.)

They found a large effect. It was robust to myriad control variables. And it got stronger when they used global American military aid as an instrument for aid to Colombia. (After all, one might think that paramilitary activity causes greater U.S. aid rather than the other way around. On the other hand, greater aid to Colombia is correlated with greater U.S. aid overall, but there is no way that Colombian paramilitary activity determines overall U.S. aid levels.) FARC and ELN attacks were not affected by American aid.

You can see the basic correlation in this chart. The blue line shows U.S. aid; the red line shows the effect of aid on paramilitary activity in counties with military bases relative to everywhere else:

In addition, U.S. aid appears to have caused voter turnout to fall in the counties with military bases ... and turnout fell more the higher the initial level of political competition.

Upshot: “The estimates imply substantial effects: a 1% increase in US military assistance increases paramilitary attacks by 1.5% more in base municipalities, and lowers turnout for mayoral elections by 0.2% and 0.12% more in militarily and electorally contested regions, respectively. These results are consistent with the idea that the influx of foreign military aid enhances the capacity of paramilitary groups, both to carry out political attacks, and to intimidate voters, which reduces political participation.”

In other words, there is evidence (from a declassified document at George Washington University’s invaluable National Security Archive) that the Colombian military abets the paramilitaries and channels U.S. aid to them. Dube and Naidu provide evidence that this matters for paramilitary activity. They also provide evidence that it weakens the functioning of democratic institutions in the affected counties.

It is a very powerful finding: American aid may directly weaken Colombian democracy.

Now, to be fair, there may be many upsides to the U.S.-Colombia military relationship that they do not measure.

I have asked here for suggestions before and usually come up empty, save for Alex Harrowell’s useful questions about the Colombian coal industry. (I am also very grateful to Alex for pointing out when my charts are confusing or my posts badly written.) But I will ask again: if you wanted to see if there were any upsides to U.S. aid, what would you look for?

Colombia is not joining NATO, but it is an American ally

OK, I got some emails about Colombian membership in NATO. The rumors are true: President Santos did float a trial balloon about joining the organization. NATO, unsurprisingly, said no. NATO would have said no regardless, but it was useful that the North Atlantic Treaty states that new parties to the organization have to be in Europe. Article 10:

The Parties may, by unanimous agreement, invite any other European State in a position to further the principles of this Treaty and to contribute to the security of the North Atlantic area to accede to this Treaty. Any State so invited may become a Party to the Treaty by depositing its instrument of accession with the Government of the United States of America.

The country will sign a cooperation agreement with NATO, but it is still not clear what that means. NATO says, however, that all cooperation agreements are the same, and you can find out what the organization says about said cooperation here.

NATO member or not, Colombia and the U.S. are close allies. We have a free trade agreement. Small numbers of U.S. troops operate in-country. The Colombian supreme court overturned an agreement giving the U.S. access to Colombian bases (see page 39 of this CRS report) but the U.S. still builds facilities there. (The text of the agreement can be found here.)

And the U.S. is surprisingly popular. Here is Latinobarómetro’s 2009 results on the question, “Does the U.S. treat your country with respect?”

SIDE OBSERVATION #1: It is interesting that Venezuelans place themselves in a category with Uruguay and Brazil, and not with their Bolivarian comrades in Bolivia, Ecuador and Argentina. My hypothesis is that it reflects Venezuelan polarization: all Venezuelans know that relations between the U.S. and Venezuela are bad, but many more Venezuelans take the American side than do their compatriots in the other big Bolivarian countries.

Here are the results to a general question on the state of bilateral relations:

Colombians do not see their relationship with the U.S. to be as good as do members of America’s near-abroad (like the D.R. and Panama) but they do see it as very good.

SIDE OBSERVATION #2: The Venezuelan results are consistent with my hypothesis: everyone knows that relations with the U.S. are terrible, but many of them blame their own government. 

SIDE OBSERVATION #3: A large subset of Mexicans are uniquely suspiscious of the United States, and uniquely prickly about national sovereignty. In fact, the U.S. treats Mexico with at least as much respect as Colombia (likely more) and relations are at least as good as with Colombia (likely better). But for understandable historical reasons, Mexicans don’t see it that way.

Finally, the Colombian results appear to hold over time. It is not just a phenomenon of the Obama-Santos relationship.

In short, even though Colombia will not be joining NATO, it has a uniquely close relationship with the United States. Moreover, that special relationship is supported by Colombian public opinion, and thus likely to endure.

None of this is to say that the U.S.-Colombian relationship is necessarily good for Colombia. The jury is out on that one, although if there is demand I will present some evidence that it just might be bad.

How Congress works ... in Colombia-with-an-“o”

In comments, Tzintzuntzan asks about the procedural rules in the Colombian congress. What are the rules by which the legislature makes the rules? Who controls them?

This is something often overlooked when Americans examine foreign countries. It’s an odd oversight, since procedural rules define American politics more than most places. In the Senate, the need to get 60% to pass anything is a procedural rule. In the House, the Speaker’s near dictatorial control over what bills get to the floor is a procedural rule. And don’t get me started on the baroque rules governing the conference committees that reconcile bills passed by both houses. In short, American politics is entirely incomprehensible if you don’t understand the procedural rules.

The same applies (albeit generally to a lesser extent) to other countries. If you don’t understand the rules by which the legislature makes the rules, then you can’t understand their politics.

Fortunately, Royce Carroll (Rice) and Mónica Pachón (Uniandes) have studied the issue for Colombia. Who controls the legislative agenda? Where are the vetoes?

The U.S. president, as Barack Obama knows all too well, cannot introduce legislation directly. Now, it is usually easy for a President to get a sympathetic legislator to introduce a bill, but he or she cannot force Congress to vote on it. Speaker Boehner or a coalition of forty Senators can kill anything before it even comes up for a vote.

Not so in Colombia! Here, the President can directly introduce legislation. Article 200 of the constitution (English here, Spanish here, and note that in context “gobierno” is used in the British sense and ought to be translated in American English as “administration”) lays out the authority:

1. Help draft the laws, present bills through ministers, exercise the right of objecting to them, and approve them in accordance with the Constitution.
2. Convoke the Congress to special sessions.
3. Present the national development and public investment plan, in accordance with the provisions in Article 150.
4. Send the budget bill of revenues and expenditures to the Chamber of Representatives.

This article has been interpreted very broadly.

In addition, the President can force votes on his or her bills within 30 days. In fact, they can force a vote within 15 days by invoking a joint session of Congress. Better yet, they can skip committees and insure that the House and Senate debate the same initial version of the bill.

But there is a snag. Bills can be amended on the floor. So the President needs to pay close attention to the legislative process in both chambers. If the House and Senate amend bills differently, then a conference committee will determine its final shape ... and the President of the Republic has no direct authority over the committee.

So who does? The answer: each chamber’s three-person Board of Directors. They are elected to one-year terms, by secret ballot. Each Board consists of one president and two vice-presidents, with (by tradition) the president coming from the majority coalition and the veeps from the largest minority parties. These guys serve pick the conference committee, and these are the guys that the President needs on board to get his or her bills through the Congress. In fact, they pick all committees, so getting any bills through Congress requires their assent.

The board presidents are, to get back to Tzin’s question, the equivalent of the U.S. Speaker. The one-year limit, however, insures that the Colombian congress produces no Willie Browns. Me, I think that is a bad thing, but mileage may vary.

What is the upshot? First, presidents in Colombia have a good-but-not-perfect legislative track record. (See page 17.) President Pastrana got 85% of his major bills passed. President Uribe got 79% passed in his first term, and 57% in his second. Presidential bills generally, however, take a lot longer than 30 days to pass: the average is actually 224.

Congress, however, maintains a lot of influence over the process. Carroll and Pachón calculated the proportion of characters that changed between the first version of a bill and the last. For presidential initiatives, 54% of the bill (among passed laws) changed throughout the process. For legislative initiatives, however, the proportion was only 26%. (See page 20.)

The unanswered question, however, is why presidential bills change more than Congressional ones. There are three hypotheses.

Hypothesis 1: The conference committees run the show. They change presidential bills more than ones introduced by their fellow congresspeople.

Hypothesis 1 is probably wrong, or at least incomplete. If it were correct, the changes to presidential bills should mostly happen in the conference committee. Conference committees, however, change only 14% of presidential bills. Most changes occur on the floor.

Hypothesis 2: The only kinds of congressional bills that can pass are uncontroversial motherhood issues.

Hypothesis 2 is also probably wrong. It is true that congressional bills usually fail: none passed under Pastrana and only 36% in Uribe’s first term. But congresspeople introduced such significant (and successful) pieces of legislation as the Violence Against Women Act, the Congressional Ethics Act, and a law allowing the confiscation of property from convicted criminals.

Hypothesis 3: The Board of Directors exercise control early on over congressional bills, but though means other than the conference committee. (Of course, the threat of an unfriendly conference committee might lead the executive to roll over on amendments earlier in the process.)

This one seems to fit the facts. The Board of Directors (in Spanish, the “mesa directiva”) are very powerful. Their power is limited, however, by the fact that they serve only one-year terms.

The upshot is that the President of Colombia can work even with the kinds of fragmented Congresses produced by Colombian electoral law as long as he can get a friendly Board of Directors. He can even work with the Senate. Barack Obama should be so lucky.

Tzin, does this answer your question?

But this is not to say that all is well with Colombian politics. More on that if there is interest.

Senates: Colombia

Finally, I will have a good thing to say about a senate!

This blog is on record trashing the senates of America, Argentina, Bolivia, Brazil, Canada, and the Philippines. It has been slightly kinder to the one of Mexico, but only slightly. So, finally, here is a senate that I can support.

Under the Constitution of 1991, Colombia has two houses. The House of Representatives is elected by district. Each department gets two representatives, plus an additional one for every 250,000 people. The representatives, however, are not elected from geographic districts. Rather, they are elected by proportional representation at the level of the department. For example, Bogotá divvies up its 18 representatives according to the percentage of the vote taken by each party in Bogotá. La Guajira, then, does the same thing with its two representatives. (The geographic division of the 164 representatives can be found here.)

The 100 members of the Senate (actually 102, since two Senators represent indigenous communities) are also elected by proportional representation, but according to their percentage of the national vote. Voters vote separately for the two chambers.

Now there may be problems. First, proportional representation might not be the best way to organize a legislature under a presidential system — but given the train-wreck that is the American legislature, I would hesitate to make big claims about that. Moreover, Colombia is a country where it is vital that small political minorities receive direct representation.

Second, the actual system of proportional representation used is one of the worst (see page 6 here) advantaging small parties. Basically, they divide the number of votes cast by the number of seats. That becomes the magic number. Every party that gets above that number gets a seat. If it gets more than twice the magic number it gets two. More than three times the magic number, then it gets three. Any fractions over the magic number are thrown away. Then the smaller parties get any remaining seats in order of their vote tallies. Bigger parties get slightly fewer seats than they should and some tiny parties can sneak in. This is a problem for the smaller House districts, but does not really apply to the Senate, since the magic number there is a relatively small 1% of the total vote.

Finally, the benefit of a two-house legislature is unclear. There is a case that checks-and-balances are good, but multiple veto points can also empower minorities.

None of these three objections strike me as compelling.

Colombian politics has a lot of problems, of course. (Is there interest in the recent politics of Bogotá?) And in a perfect world I would not pick the Colombian system for the United States. (Although I would take it over our own crazier system in a heartbeat!) But similar to the way Mexico has made its Senate more functional by using proportional representation to dilute its undemocratic features, Colombia has made its Senate more functional by using proportional representation to make it more democratic.

But I will say that the problems of Colombian politics owe little to the constitutional design of the Senate. (These political scientists agree: their reforms mostly involve changes to electoral law, not constitutional design.) Moreover, in some ways the Senate is better designed than the House. So, finally, an upper house in the Americas that I can support.

At least sort of. The idea that legislators can vote by secret ballot strikes me as nuts. But that problem applies to both Colombian chambers, not just the Senate.

I am a very new observer of Colombian politics. Thoughts?

Should Colombia chop three zeros off the peso?

UPDATE (June 13): The Santos administration just announced that it intends to redenominate the currency by the end of 2014.

President Juan Manuel Santos has long wanted to chop three zeros off the Colombian peso, which trades around 1,800 to the dollar. Except for Paraguay, the next weakest in Latin America are the colón and the Chilean peso, which bounce around 500. (Wikipedia has a list of the smallest-value currency units here.) In October 2011, the proposal failed in the Senate by a vote of 41-15. In September 2012, the President reopened the idea, but by February 2013 it was clear that it was not going to happen in this legislative session. The idea remains current, however, and I will not be surprised if it comes back after the peace talks with the FARC are sorted out.

Redenominating the currency is not crazy: France got rid of two zeros in 1960; Israel revalued by a factor of 1,000 in 1986 as did Russia in 1998; Peru hacked off six zeros when the sol replaced the inti in 1991; the next year Argentina cut four zeros off when the peso replaced the austral; Mexico sliced off three in 1993; Turkey hacked off six in 2005; Mozambique redenominated by three orders of magnitude in 2006; Ghana introduced a new cedi at 10,000-to-1 in 2007; and Venezuela chopped off three in 2008 with the introduction of the bolívar fuerte.

The direct cost of introducing a redenominated currency would not be insignificant. In 2010, the Colombian government estimated it around $123 million. (In a sign of the problem with high-denomination currencies, that number turned into $123 thousand on an English-language website!) Printing new bills is cheap: 84% of the cost would come from minting new coins. There would also be an additional $32 million spent in advertising and explaining the change, for a total cost of $155 million.

(Side note: in Ghana, coins were basically out of circulation by the time the government proposed redenominating the cedi. Since coins had to be reintroduced anyway, the monetary cost of redenominating was tiny.)

So why bother? The issue is understudied, but the economics literature (such as it is) focuses on credibility. Few currencies (I am tempted to say none) are denominated in hundreths or thousandths of a euro for long-standing historical reasons. Rather, it comes about either because of a burst of triple-digit inflation (say, Mexico in the 1980s), outright hyperinflation (say, Argentina) or a long period of relatively moderate double-digit inflation (like Colombia; see the below chart.)

So the big economic reason to redenominate, then, is to reiterate that the inflationary period is over, kaputsky, finis. Over here, the Banco Central de Venezuela states four reasons that boil down to making mental math easier and one which says “Leave behind the consequences the history of inflations of the past had on the currency.”

Layna Mosley at UNC-Chapel Hill tried to figure out why countries redenominate. She hypothesized that credibility might be a reason, of course, but she also thought that conservative governments would be also be more likely to redenominate for reasons of national pride. Her evidence was supportive, but not compelling.

Colombia, however, mostly fits her model. (You can read the Colombian authorities’ reasons for wanting to change.) In addition to standard discussions about credibility and easier math, the central bankers mentioned the feel-good pride effect from having a serious currency valued at around half a dollar (or euro or real) instead of some teeny tiny amount. “There is possibly a positive psychological impact since it reduces the gap between local and foreign currency denominations.”

That fits the Mosley model. The final reason, however, does not:

“A countrywide redenomination exercise would also mean all citizens had to exchange their currency, and this could potentially shed light on the country's vast shadow economy. It is thought that drug cartels which operate in Colombia trade mainly with greenbacks. However, officials do not have any definitive figures that prove the extent to which pesos are used in drug trade, and a wide-scale currency exchange could help in this respect.”

And that, ultimately, is why I think Colombia should redenominate.

Sic transit gloria Bogotá

Bogotá is famous for its TransMilenio system. (I continue to confuse Bogotanos by referring it to it as the Metrobus, which is the name of Mexico City’s imitation.) TransMilenio is bus rapid transit (BRT): elevated stations with entry turnstyles and dedicated bus lanes. TransMilenio beats its Mexico City counterpart in having two lanes on some routes, allowing for express buses. I was last here in 2010; since then, the system has been extended east-west through downtown.  

But it still, well, sucks. It’s overcrowded; always, all the time, sardine-overcrowded. And since the traffic lights don’t give it priority, it isn’t all that fast. It beats the smaller buses, and sometimes it beats a taxi, but it is not pleasant. Thus, angry riots in 2012.

The city is proposing a light rail system to supplement the TransMilenio. A surface tram would be cheaper than an underground metro, and trams have more capacity than buses. Sounds great! There are even private consortiums ready to cough up the money. So what’s not to like?

Simply put: light rail can’t handle the demand.

First, a bit about Bogotá’s geography. The city is sharply limited to the east by a large mountain range. The “Borde Oriental” sprawls north from the old downtown along the edge of the mountain range. Employment and (wealthy) housing has crowded along the Borde. (The Borde is traditionally held to end around 72nd Street, but has been creeping further north.) The city has then sprawled out to the west from the Borde. Imagine Chicago, only with a mountain range instead of a lake. Roads are numbered from the beginning of the mountains; each subsequent road runs is a little under 400 feet west of the previous, or about  50% longer than a Manhattan city block. Since the border of the mountain range is not really straight, Seventh Road is the easternmost road that runs continuously from downtown to the prosperous northern suburbs.

(This link gives a good overview, although the map inexplicably rotates the maps 90-degrees counterclockwise.)

Not surprisingly, then, one of the two proposed light rail routes runs north up Seventh Road from downtown all the way to 193rd Street. North of 119th Street it would transfer to an existing right-of-way, but south of there it would be right down the middle of Seventh Road. It is supposed to move 263,000 people per day at an average speed of 25 miles per hour. Trains will run every 2½ minutes.

Or not.

Each train will have a capacity of 760 people, cramming in ‘em at TransMilenio levels of six people per square meter of floorspace. (Try it: that is crowded. In fact, by the standards of everywhere save the Manhattan-bound F-train at 8:30am, that is overcrowded.) To meet the travel projections, peak demand will be 23,188 people per hour. At 760 people per train, that’s 31 trains per hour ... or one every 118 seconds. 118 is rather less than the projected headway of 150 seconds.

The traffic light at 72nd Street right now cycles at 120 seconds for cross-traffic. Keeping that would cut the number of trains per hour by a factor of three. Otherwise trains would back up at traffic lights and arrive in bunches ... the way buses always seem to. Moreover, cars often block-the-box: you would need to ban left turns, but even that might not suffice.

In fact, nowhere has at-grade light rail running at 25 mph with 150-second headways. Melbourne (unlike everywhere in the United States, including Philly and S.F.) kept its tram network after WW2. The most heavily used line runs at 10 miles per hour with a seven-minute headway. Not modern enough for you? Well, Paris is building a boatload of trams in its outer boroughs. They average 13 mph with a four-minute headway. (All data in the above is courtesy of Uniandes.)

Now, trams are a bit better than more TransMilenio lines. TransMilenio was always considered a second-best option, even by Mayor Enrique Peñalosa, who greenlighted it because the Metro was unaffordable at the time. But unless you can be assured that the private contractors will be picking up all of the capital cost, then it is far from clear that a light rail beats TransMilenio in cost-benefit terms.

A metro, though, clearly does beat both TransMilenio and the light rail. And design contracts were just signed.

But. The first problem is that a metro has a giant capital cost, and both the city of Bogotá and the national government are capital constrained. The second problem is that the city of is not, shall we say, well governed. Combine them and the privately-funded light rail line is going to be a hard thing to kill.

Which would not be terrible ... if you were sure that no public money would go into the light rail. Uh huh.

How important are property rights?

It is a bit of a truism that property rights are key to economic development. An efficient system, runs the logic, needs to be transparent, transferable, excludable, and enforceable. In a bit more detail:

1. Transparent means that it is cheap and easy to find out who owns what.
2. Transferable means that it is cheap and easy to sell or transfer property rights from one entity to another.
3. Excludable means that if you own the rights to a piece of property, then the law allows you to exclude others from enjoying those rights.
4. Enforceable means what it says.

Without those characteristics, runs the logic, you get underinvestment and confusion and commons-tragedies and all that. And the logic has some empirical backing. For example, in 1986 the Buenos Aires provincial government offered a group of landowners compensation for land that had been seized by a group of squatters. Some landowners took the offer, and the province then transferred title to the squatters. Others did not. On the assumption that the squatters had no way to know whether the landowner from whom they had seized the land would take the provincial offer, Sebastian Galiani and Ernesto Schargrodsky found that the squatters who received titles invested more in their homes and the education of their children.

Here comes the New York Times with an article about terrible land titles in Greece:

“In this age of satellite imagery, digital records and the instantaneous exchange of information, most of Greece’s land transaction records are still handwritten in ledgers, logged in by last names. No lot numbers. No clarity on boundaries or zoning. No obvious way to tell whether two people, or 10, have registered ownership of the same property.”

The article goes on to say that even the former Yugoslav states have better land rights.

One reasonable conclusion would be that bad property rights are a drag on economic recovery. “Many experts cite the lack of a proper land registry as one of the biggest impediments to progress. It scares off foreign investors; makes it hard for the state to privatize its assets, as it has promised to do in exchange for bailout money; and makes it virtually impossible to collect property taxes.”

But another reasonable conclusion would be that the evidence about the importance of clear property rights is wrong ... or at least oversimplified.

There is no getting around that fact that Greece became a rich country despite terrible property rights. The above chart tracks GDP per worker in 2005 dollars (adjusted for purchasing power) for four countries from 1950 to 2011. (The data through 2010 is from the Penn World Tables 7.1; the 2011 data comes from the OECD.) In 1950, Greek workers produced as much as Colombian ones; they were significantly less productive than in Mexico. Over the next three decades, Greece far outdistanced its Latin American counterparts. The current depression has not eliminated that gap.

Greece, of course, had many advantages that Colombia and Mexico did not, but poor property rights did not stop the country from taking advantage of them.

It has to make you wonder. Despite the micro evidence from places like Argentina, are efficient property rights really as important as we think they are?

Why I am in Colombia: Spreading the word on the resource curse, or lack thereof

I am in Colombia for many reasons. One is to start a big research project. Another is because I like the country. But the main one was to attend a conference on mining put on by the Universidad de los Andes.

El Tiempo reported on the conference. Here is a translation of part of their report:

“In his presentation, Noel Maurer, a professor at Harvard Business School, said ‘there is no curse’ from the resources created by mining discoveries. For him, there is no evidence from any country where either democracy or development has severely deteriorated after an increase in income generated by mining activity.”

That is worded a little more strongly than I would like (you can find some evidence of a resource curse in Nigeria and Zambia, and it is unlikely that the 1997 crisis would have unseated Suharto at a time of high mineral prices) but it does about sum up my opinion. To be fair to El Tiempo, they added my caveat in this tweet.

Lina Holguin of Oxfam-Québec tweeted in response to my caveat, “Congo, Chad and Niger...????” I would answer by saying that while all three countries are poor and unstable, there is no evidence that they would be less poor or less unstable without natural resources ... with an additional caveat that natural resources may have increased violence in particular regions of Congo. Possibly.

El Tiempo also tweeted a few other quotes from yours truly. First: “The decrease in Venezuelan democracy started before the oil boom.” Second, a somewhat misquoted, “No resource boom has affected countries rich in resources.” (I am mildly unhappy with that one: my argument was only about democratic stability.) Third: “Colombia ought to avoid wasting the current mining boom.”

While my discussion was about democracy, that last quote is correct: lots of countries have blown resource booms, leaving them no better off than they entered. Colombia should avoid that. Of course, that is something easier said than done!

Finally, let me end by presenting the (weak) evidence for a resource curse in Zambia.The red line measures the country’s Polity score, a commonly-accepted measure of democracy. The blue line measures the government’s real direct income from the copper industry; the green line measures the percentage of government revenues from copper.

Zambia’s polity score dropped precipitously between 1968 and 1972, while copper revenues boomed. The country then democratized in 1991, after a period of prolonged low copper prices. The argument would be that copper revenues made it easier for Kaunda to establish a one-party state, while the period of low prices was essential to see the fall of that state.

The rub, of course, is that lots of countries in Africa followed the same trend around the same time. (See the below chart.) But you cannot rule out a curse for Zambia.

Chris Blattman is a genius

For those of you who have not, start reading Chris Blattman’s blog. It is the best blog on development economics (and a whole lot more).

Five minutes ago I started to procrastinate. As I do, I turned to his blog and found this post. Its title is exactly what I just told a Colombian reporter when asked what the government should do to ease local dislocations created by large-scale mining!

My answer? Give people cash.

What is nice about Chris’s work is that he has experimental evidence that giving poor people cash works to improve their earnings potential, and not just their current earnings. In an experiment, the Ugandan government gave randomly-selected young people a no-strings-attached grant of $400; roughly the average annual income. And what happened?

“Most start new skilled trades like metalworking or tailoring. They increase their employment hours about 17%. Those new hours are spent in high-return activities, and so earnings rise nearly 50%, especially women’s.”

So there you have it. Want to help mitigate the adverse effects of mining on poor people? Give them cash.

What is interesting about the Colombian coal industry?

In an effort at crowdsourcing, I am asking the question: What do you want to know about the Colombian coal industry? What do you want to know about coal in general? What do you want to know about mining in Colombia in general?

There is a case in the works; possibly a paper. I will be in-country in May. Before I start, I would very much like to know what people consider to be the big interesting questions.

Imperialism can pay

There's a debate out there about imperialism.  Did it pay?  The consensus answer is:  no, it did not.  The British spent far more in defense than they earned from imperial investments; in fact, by the 1880s, the profitability of imperial investments had fallen below domestic and non-imperial ones.  The French believed that their empire after WW2 helped them maintain their balance-of-payments; the numbers don't support that idea either.  The big counterexamples involve the Nazi conquest of France, and the Soviet occupation of Eastern Europe after 1945 ... but after 1970 or so the balance on the Soviet empire hinges on how much you believe control of the Warsaw Pact armies was worth to Moscow.

But there is a case of imperialism paying, and paying quite well:  Panama in 1903. 

Here's the simple story.  A French company, see, had tried and failed to build a canal across Panama.  In 1904, as per the terms of its concession, the assets of that French company would revert to the government of Colombia, which could then sell them to (say) the United States, which wanted to build a canal.  Those assets included a whole lot of big ditches, and a working railroad, and were worth around $40 million — $21.3 billion as an equivalent share of national income today.  So when the U.S. comes sniffing around to strike up a canal agreement, the Colombians rather intelligently decide to stall. 

The shareholders in that French company, though, want a deal quickly.  So does the United States, run by, you know, Teddy Roosevelt.  (Any resemblance to a current postulant for the Oval Office is entirely coincidental.)  Of course, the U.S. isn't a dictatorship, Congress matters, so it kinda helped that the shareholders of that French company had made rather large donations to the GOP and hired all the best lobbyists of the time, including a Brooklynite named William Cromwell.   

So one of the shareholders of the French company cuts a guy named Manuel Amador a personal check for $100,000 in a New York City hotel room ($53 million) to make sure that a revolt gets organized on schedule.  It does, and coincidentally enough American ships are right offshore to make sure that Colombia can't intervene.  The new government then appoints the French dude who cut the personal check to be its foreign minister, and he negotiates a canal treaty with the United States.  When the people in Panama City see the agreement they balk, but their foreign minister (and, uh, the U.S. secretary of state) warn them that if they don't sign then the half-life of their new republic is going to be significantly less than its two-month history, and poof, Panama signs.  Without a Spanish translation.

So far, so sordid, but honestly, that tells you nothing.  After all, there was competition between the Panama and Nicaragua routes.  I don't care that there was skullduggery in the dealings, and I don't care that Colombia got screwed.  What I want to know is whether the U.S. got a better deal because it used military force than it would have gotten in the absence of force.

Fortunately, we have indeed got six existing deals struck with either Colombia or Nicaragua which did not involve force or the threat of force.  (Well, one of them involved the implicit threat of force.)  All of them incorporated the fear of competition from another route.  And all of them gave the owner of the Isthmus a hella lot more squeeze than the deal Panama actually received.

Here's a table showing the net present value of how much less the U.S. would have received under various other agreements compared to the one it foisted on Panama in 1903:

	% of U.S. GDP	2007 dollar equivalent	% of Canal cost	% of Panama GDP
Hay-Herrán treaty	0.07%	$10.1 bn	4.3%	117%
Concha memorandum	0.10%	$13.4 bn	5.7%	155%
MCC-Nicargua contract	0.10%	$13.6 bn	5.8%	157%
French-Colombia concession	0.10%	$13.8 bn	5.8%	159%
Senate counteroffer	0.17%	$22.9 bn	9.7%	265%
Nicaragua-U.S. treaty	0.23%	$31.7 bn	13.4%	366%

So, yes, sometimes imperialism does pay.