Mainstream opinion holds that Chávismo socialism drove the nation of Venezuela to its current collapse. As with all mainstream opinion, there is a Leftist counter-narrative, in this case that Venezuela actually grew in leaps and bounds under Chávez and Maduro, and the recession only turned into a full-blown crisis under the weight of US sanctions. The piece that started the counter argument is Mark Weisbrot and Jeffrey Sachs’ “Economic Sanctions as Collective Punishment: The Case of Venezuela,” which claims that US sanctions have directly caused 40,000 deaths in Venezuela. In the wake of this publication I was dragged to three different leftist events, every one of which was opened by explicitly quoting this 40,000 figure. This narrative seems to steadily sunk into the water, and this study has been cited everywhere from the Independent to the Washington Post and even Fox News. Let’s take a closer look
Have You Heard of CEPR Before?
The paper in question came from the Center for Economic and Policy Research, which despite the objective, neutral sounding name is a very much a committed leftist think tank - one which does not subject its publications to peer review. Its staff do not strike me as obviously impartial truth seekers. The Director of International Programs, Deborah James, and the Director of International Policy, Alexander Main, were respectively the Executive Director and a Staff Analyst at the Venezuela Information Office, a think tank funded by the Venezuelan government and filed under the Foreign Agents Registration Act.
Mark Weisbrot himself, the Co-Director of CEPR and principal author of the paper in question, has been a vocal fan of Chávez from the start and has a kind of dubious track record as an economist. He’s been predicting the course of the Venezuelan economy for some time now and has been almost hilariously consistently wrong, authoring gems like: “Why Chávez was Re-Elected” and “Sorry, Venezuela haters: this economy is not the Greece of Latin America.”
Lest you think I’m being unfair to CEPR, I don’t really think they’re really much more biased than your average think tank, which after all are usually promoting some kind of narrative. But their claims on the performance of the Venezuelan government should be taken with the grain of salt that multiple CEPR employees were once professionally paid by the Venezuelan government to say it was performing well.
The co-author from outside of the organization, Jeffrey Sachs, is a whole different story. He’s won an impressive series of awards and accolades and worked everywhere from the World Health Organization to the World Bank to the UN. His work has received its fair share of criticism, but his resume still contrasts conspicuously with the work of everyone else in CEPR.
Anyway, let’s give “Economic Sanctions as Collective Punishment: The Case of Venezuela” a fair treatment. The paper claims that following US sanctions in 2017 three critical changes took place: 1) Venezuela lost access to international finance, 2) Venezuelan oil production went down and 3) the mortality rate shot way up. The first round of sanctions in August barred Venezuela from using markets denominated in US Dollars, thus preventing the Maduro regime from restructuring its country’s debt, one of the few tools available for tempering an advanced recession. Oil production soon began to fall at three times its previous rate.
The following sanctions in January turned the pressure up even higher by halting all United States’ imports of Venezuelan oil, equivalent to roughly a third of Venezuela’s total exports. Venezuelan overseas assets were also frozen, including the gas station CITGO, some international credits and billions in gold in the Bank of England, under the slightly dubious logic that Maduro wasn’t the legitimate leader of Venezuela, and therefore couldn’t legitimately access Venezuelan bank accounts. So far, so good, sounds accurate.
Weisbrot and Sachs supply the above graph to substantiate their theory about oil production by using Colombia as a counterfactual. Here we can see that while Colombia and Venezuela followed similar production paths normally, US sanctions clearly show a stark divergence where Venezuelan oil began to plummet.
How do We Know Sanctions Killed People?
Without wading too far into any statistics, Weisbrot and Sachs cite a figure from the National Survey of Living Conditions (ENCOVI) which says that Venezuelan mortality rose by 31% between 2017 and 2018 due to reduced food and medicine imports. Our authors extrapolate from there to claim that the sanctions must have caused this spike in mortality, equivalent to a staggering 40,000 lives lost. Note that this is not the conclusion of the Survey itself, but rather Weisbrot and Sachs' own interpretation, and that the footnote says: “The ENCOVI 2018 survey has not been made public; the mortality statistic cited here is from the UN Report. United Nations (2019).” The UN Report in question isn’t obviously available anywhere either. So a survey happened somewhere, we don’t know anything about the methods or the sample size, but maybe it was cited in a UN report, which was then cited in CEPR, which chose to connect this essentially unsourced claim to the US government killing tens of thousands of people.
I am not an economist. This is what I’ve noticed from my distinctly amateur reading of this publication and it seems a little sketchy. I consider Weisbrot something of a hack, and I wouldn’t even be reviewing this if it wasn’t for his co-author’s more impressive track record. But while I know Jeffrey Sachs is probably brilliant in his own regard, he seems to be committing to a narrative here in direct absence of hard evidence. In interviews on this paper I hear him make patently false statements, such as the claim that hyperinflation began after US sanctions, which even his own paper directly contradicts. But don’t hear it from me, hear it from literally everybody else who’s weighed in on the issue:
And So It Begins
The rebuttals came rolling in fast and hard only a month after the paper was published. We’ve heard from a think tank that employs Venezuelan government lobbyists, now let’s hear from an economist employed by the Venezuelan government’s opposition. Yes, really. Ricardo Hausmann was the former Venezuelan Minister of Planning in the 90s and is the Founder and Director of the Growth Lab under the Center of International Development at Harvard. He is also literally an advisor to Maduro’s main political opponent, Juan Guaidó, so while his credentials are very impressive, his work is not a clearly objective source.
That said, his “Don’t Blame Washington for Venezuela’s Woes: A Rebuttal,” co-authored with fellow Harvard CID economist Frank Mucci, is probably the most shared response to Weisbrot and Sachs’ work, so let’s give it a fair shake. Hausmann claims that while oil production did fall around sanctions, you can’t draw a straight line from one to the other, because Venezuela simply shifted most of its exports from the US to friendly Russia, China and India. He pushes back against Weisbrot and Sachs’ graph on Venezuelan and Colombian oil production with a graph of his own:
Hausmann’s point is that if you scroll out to a broader timeline, there was never much convergence between the trends to begin with. He also rightfully points out that comparing Venezuela to Colombia is a bit of an apples-to-oranges study, since Colombia lacks some of Venezuela’s confounding factors, like the administrative overhauls in 2003 and 2017 (this theory expanded more here). He suggests that falling oil production is mostly in line with the course Venezuela had been heading on for some time, possibly jumping down a bit in 2017 following the blackouts and general collapse of electrical infrastructure.
Also, scroll back up and take a look at the Y axes in the original Weisbrot graph. They’re quite clearly for very different quantities of oil but re-scaled to make the gap between the countries look much smaller than it is. Hausmann calls Weisbrot out for this, but sort of does the same thing to a lesser extent, so I guess it’s not a mortal sin in graphcraft. Note that while this graph, and the others in the paper, were built on data from OPEC and the EIA, Hausmann also cites “own calculations,” and doesn’t list these calculations anywhere. This isn’t actually an unusual practice for economics papers, and I’m sure that a decorated Harvard economist is capable of doing these calculations. That said, given his glaring bias in this situation it would have shown academic integrity to be forthcoming about his methods.
In response to Weisbrot and Sach’s claim that food and medical imports dropped following sanctions, Hausmann throws up this sleek new graph produced by the Harvard Center for International Development:
As you can see, medical imports started their drop quite a bit before 2017. Man that trendline looks grim.
The Home Team Defends Their Title
Weisbrot and Sachs fire back with the snarkily titled: “Economists Use ‘Fuzzy Graphs’ to Challenge Data on the Human Cost of Trump Sanctions in Venezuela” They take shots at an economist named Miguel Ana Santos who insulted their graphs on Twitter, and defend their choice to alter the Y axes as normal statistical practice. Notably Weisbrot and Sachs do not dispute the above CID graph showing medicine imports falling long before 2017, but they do take issue with the fact that both responses failed to directly address CEPR’s claims about the second round of sanctions in January. Fair enough.
Reinforcements Arrive
Perhaps Santos felt deeply offended, for he proceeded to band together a plucky team with Dany Bahar, Sebastian Bustos and Jose Morales to author “Impact of the 2017 Sanctions on Venezuela: Revisiting the evidence” for the Brookings Institute. Do the authors in question have any really obvious bias? Well . . . yes. A year ago when I first stumbled upon this paper I’m embarrassed to say that I concluded I had found the One True Neutral Take and shared it with all my friends as such. But actually Miguel Santos and Jose Morales both worked on the economic policy team for Henrique Capriles, who faced off against both Chávez and Maduro in presidential elections.
It also turns out that all four of the authors work in varying capacities for the Harvard Center for International Development Growth Lab, which Hausmann founded and chairs. This is all a little weird and incestuous and it might be fair to be skeptical of their work, which backs up the claims of a guy who’s actually their colleague.
Santos et al. open by building on a point that Hausmann touched upon, that it’s silly to act like US sanctions cut Venezuela off from a chance at restructuring their bonds when the financial markets made that decision long ago. Venezuelan government debt was already considered such a risky investment that Venezuela was required to pay almost 29% on returns for bonds, an astronomically high rate about 8 times higher than the rest of Latin America.
If you narrowed this graph down only to 2017-2018, all you would see is that huge spike. However, with the view stretched out across a wider timeline, it’s clear that sovereign risk rate had been out of control long before a US sanction ever touched a Venezuelan market; in fact, the risk rate itself was unmoved for months following the imposition of sanctions
Santos et al also disagree that sanctions directly reduced oil production, and recreate CEPR’s graph starting in 2010, showing that Weisbrot and Sachs conveniently cut off the data where it stopped lining up with their chosen narrative. You can see that the synchronicity between the two countries was more of a blip, and the broader trendline shows a long Venezuelan collapse next to a jolting Colombian market. As for Weisbrot and Sachs' central claim about the effect of sanctions on food, medicine and mortality rates, well, they have a few graphs for that:
Santos et al assemble all these graphs together to make the point that no, the crisis really didn’t start in 2017 but actually way earlier. Weisbrot and Sachs frame the situation as though falling food and medicine imports and rising mortality rates appeared out of nowhere in 2017, ignoring the sharp three year downward trend prior to sanctions. Santos et al, like me, also raise their eyebrows at Weisbrot and Sachs' evidence for 40,000 deaths resting on a survey that cannot be found and a UN report that also cannot be found.
Looking through the sources for these graphs, our old nemesis “Own Calculations” strikes again, along with the calculations of a mysterious man named Douglas Barrio, who turns out to also work at the Harvard Center for International Development. Bustos-Yildirim refers to the dual team of Sebastian Bustos, one of the authors of this paper, and Muhammad Yildirim, who also works at - well, you can already guess by this point. The easiest criticism of the above graphs is that (except for daily calories) they cut off in 2017, so they don’t really measure the rates after the sanctions. I thought I might try to chart the updated trends myself, but looking through Santos et al’s main sources, their own CID Atlas, the World Bank and UN Comtrade, unless I’m hopelessly lost it doesn’t look like there actually are updated numbers. Maybe there won’t be until the dust has settled a bit.
In fairness Santos et al acknowledge that they can really only show good data for the past, but they point out that Weisbrot and Sachs’ are making the wildly unrealistic assumption that these trendlines would have suddenly improved in absence of US sanctions. Weisbrot and Sachs themselves make no attempt to dispute the basic truth of these trends; you’ll recognize “Figure 6” as the graph from Hausmann’s original paper that the CEPR response didn’t even try to refute. The two authors even explicitly wrote in their original study that “Real GDP had already declined by about 24.7 percent from 2013 through 2016, and consumer price inflation for January to August 2017 was probably somewhere between 758 percent and 1,350 percent at an annual rate.” Furthermore, because Weisbrot and Sachs are attributing every death they believe happened after 2017 to the United States, they are essentially assuming the counterfactual that no deaths would have taken place without sanctions, which is patently absurd based on their own description of the economy.
Do Two Wrongs Make a Right?
My Leftist friend called Hausmann and company a “regime change conspiracy,” which is a bit much, but not totally off the mark. It is a little weird that both of the main publications rebutting Weisbrot, presented by seemingly independent institutions, are actually all written by a bunch of economists who work together and help make each other’s graphs, and also sometimes work for the Venezuelan opposition party. To ground this in reality though, it’s basically a crack team of some of the most highly respected and highly accomplished economists out there vs. the CEPR economists who are consistently wrong about everything. Should the man who served as the Chief Economist for the Inter-American Development Bank and Chair of the IMF World Bank Development Committee really be considered equally credible as the guy who wrote: “Sorry, Venezuela haters: this economy is not the Greece of Latin America”?
That said, I find it as frustrating as everyone else that all the people involved in this discussion seem to be either clearly schilling for the Venezuelan government or clearly schilling for the opposition. So let’s close with a source that actually is relatively neutral - or at least respected by people on both sides of the aisle.
Can We Synthesize These Two Stories?
When Weisbrot and Sachs first wrote their rebuttal, to corroborate their attack on Santos they cited Francisco Rodriguez, the former head of the Economic and Financial Advisory for the Venezuelan National Assembly under Chávez. It’s an interesting choice for them to name drop Rodriguez, because he himself has taken such issue with the credibility of Weisbrot’s work in the past that he authored the scathingly titled: “How to Not Defend the Revolution: Mark Weisbrot and the Misinterpretation of Venezuelan Evidence.”
Rodriguez has no direct connection to the Center for International Development, but he actually did co-edit a book on Venezuela with Ricardo Hausmann. At this point I’m starting to believe that the circle of people doing in-depth econ research on Venezuela is so small that they basically all know each other and have worked together. Prior collaborations with Hausmann aside, Weisbrot and Sachs seem to consider Rodriguez a legitimate source, citing him not only in the rebuttal but also twice in their original paper. Wikipedia calls him “one of the world’s foremost leading experts on the Venezuelan economy.”
Rodriguez added his own voice to the conversation with the preliminary Crude Realities: Understanding Venezuela’s Collapse, which agrees that the overwhelming blame for the current state of collapse lies with Chávez and Maduro, but suggests that sanctions are worsening the situation. He pushes back against Santos et al’s claim that Venezuela’s sovereign risk rate was too high for investors by pointing out that Goldman Sachs had purchased Venezuelan debt shortly before sanctions. Santos et al actually acknowledged the Goldman Sachs deal, and I think everyone here is looking at the same data, just from a different angle. Only one bank over an entire year offered to buy any Venezuelan bonds, and at terrible rates: the CID economists concluded that there was low to no chance at bond restructuring; Rodriguez concluded there was low but not zero chance.
Rodriguez followed up with “Sanctions and the Venezuelan Economy: What the Data Say,” which compares oil production not just between Colombia and Venezuela, but a series of other countries as well, along with a “synthetic control group” averaged out to replicate Venezuela’s own natural production rate.
Yup, looks like oil production fell after sanctions. Rodriguez also measures and dismisses Hausmann and Morales’ theory that 2017 administrative overhaul made a big difference, but agrees that much of the oil sales were indeed shifted off of America onto other allies.
Rodriguez goes on to conclude that any real world effect on Venezuelan revenue was cancelled out because rising oil prices offset the falling production. Given that these sanctions were implemented with the specific goal of hurting Venezuelan oil sales, it’s impressive it took this much methodological rigor to tease out a modest effect with no practical implications. Rodriguez suggests that falling oil revenues could still possibly hurt the economy in the future, but he echoes the now growing chorus of voices that there is no evidence sanctions have caused more people to die.
I think this is actually a fairly rigorous, good faith attempt to reconcile the differing conclusions of previous research. Rodriguez laboriously details his methods, earning him first prize for transparency out of everyone who’s weighed in so far. For his hard work this paper has been mentioned in the popular press exactly once, in Nicholas Kristof’s New York Times’ article “Kids in Venezuela are Dying. Are We Responsible?” which cites this study as “evidence” for the title claim, despite the abstract specifically saying: “it is incorrect to infer any effect of sanctions from the increase in mortality rates in 2018.”
Conclusions
There is good evidence that sanctions have impacted oil production, but the effects are modest and haven’t measurably changed Venezuelan outcomes. It is certain that sanctions cut Venezuela off from many international investors, but it's a little unclear how much of a difference this made given the context of their long failed economy and highly risky bonds.
There is no evidence that sanctions killed 40,000 people. Over the course of the crisis that many people certainly have died, and many more, but there’s no data anywhere showing that sanctions changed mortality rates. It was actually overkill going through all these rebuttals because at the end of the day Weisbrot and Sachs just didn’t provide any sources.
Weisbrot has in the past conceded how weak these original claims were, admitting to the Independent that he “could not prove those excess deaths were the result of sanctions.” Apparently forgetting this admission, last year he doubled down and claimed on Twitter that “the world needs more journalists willing to report on how US sanctions are killing people in Venezuela.”
Jeffrey Sachs has at least taken a somewhat mollified stance:
“Let me be clear: Nobody knows. This was a very basic, simple calculation based on estimates of universities in Venezuela that mortality had increased by a certain proportion after the sanctions. I don't want anyone to think that there is precision in these numbers. What is certain, though, staring us in the face, is that there is a humanitarian catastrophe, deliberately caused by the United States”
So close, yet so far. Sachs does not appear to have ever collaborated with CEPR again.
None of this can fully disprove the claim that US sanctions will continue to act as a lid preventing a recovery from ever taking place, though we should be clear that judging by the three year course of the economy pre-sanctions, such a recovery would be quite unlikely. There is also the possibility that sanctions, coupled with other forms of US support for the opposition, are driving up civil tension and violence. This second possibility is proposed by Rodriguez and Sachs, who collaborated to argue for peaceful power sharing in Venezuela here and here (I told you these guys all knew each other and worked together).
For these hypotheticals I do personally oppose the sanctions. Lest you think I’m a regime change schill too, let me clarify that I actually oppose sanctions in most cases, not just this one. I think often sanctions serve as an economic cudgel, a way for the US to bully other countries and throw its weight around without being accountable to anyone else. But we can hold that general principle without making up actual nonsense to white wash the failures of a domestic regime, which is the path CEPR went down.
This sanctions-are-the-real-culprit shtick is part of a broader narrative that actually Venezuela was doing just fine under socialism til the pesky United States got involved. This claim that Chávez and Maduro secretly did well for their country isn’t incoherent, just so long as you cherry pick the right metrics. Chávez oversaw a dramatic rise in literacy, vaccinations and poverty reduction, Maduro built three million homes for Venezuela’s poor. All that is fine and great, but it should really also be put in the context that the entire economy collapsed under their watch and now people are starving. I worry that the counter narrative focuses on Chávez and Maduro’s careers right up until things stopped going well, then starts paying attention again when the United States gets involved. In between are three years representing tens of thousands of lives lost and millions forced to flee their homes as refugees. The least I ask is that we look honestly at the entire timeline.
コメント