Wall Street Journal gets it wrong (UPDATED)
There is an alarming Wall Street Journal article doing the rounds right now, and I don’t think the numbers add up. Written by Matthew Rosenberg, it focuses on the movement of allegedly illicit money through Kabul International Airport. Rosenberg states that $3.6b in cash is being flown out over the course of a year. The main point is made by unnamed officials who speculate:
“…some of the cash, if not most, is siphoned from Western aid projects and U.S., European and NATO contracts to provide security, supplies and reconstruction work for coalition forces in Afghanistan.”
As you’d expect, this revelation is lighting up the chattering classes as evidence that the coalition efforts are merely funding the Taliban and the drug lords. Other recent articles along these lines have directly led to the creation of a new Pentagon task force to stop procurement spending from entering Taliban hands.
But before anyone pulls the panic cord, we need to take a closer look at these numbers, because they just don’t make sense.
For example, PDT has been tracking donor spending on behalf of DFID for the last few years. All the major donors cooperated, including the US, and provided us with the details on what was spent and on who. Our field staff then drilled down to see how much of that actually entered the Afghan economy. (It’s a fascinating project, I encourage you to read the latest report if you are interested in aid flows and aid effectiveness.) What they found is that in 2008-9 only 38% of ODA spending remained in Afghanistan, totaling $788m. While the project only looked at the largest donors, the total amount of those not examined was inconsequential.
Looking at military spending, the story isn’t much different. The WSJ article states that NATO spent $14b last year alone. That is hugely misleading, though. The vast majority of military spending doesn’t go anywhere near Afghanistan. It is spent on German tank manufacturers, Russian transport companies, soldiers’ salaries, medical facilities stateside, ammunition, field rations packaged in Alabama, satellites made in Seattle, and bottled water flown in from Dubai. PDT has spent more time than anyone tracking this spending and trying to shift as many of these contracts as possible to Afghan vendors. It’s hard going, as you can imagine, because there aren’t many entrepreneurs in Kabul who can build an M1 Abrams main battle tank. There are many, however, who can provide fruits and vegetables or build roads. If you add up all the NATO and US military spending on these items, it’s considerable, but it isn’t measured in the billions.
Adding it all up, there is simply no way that $3b a year of aid and military spending is entering the hands of Afghan businessmen, even assuming they had 100% profit margins. The numbers just don’t add up to anything near that amount.
So assuming the article is correct that $10m a day is leaving Kabul, where is that money coming from? I think there are a two likely explanations.
First, Rosenberg talks about the hawalas system, an informal banking network used in the Islamic world. He believes that the Afghan businesses are using this system to stash their profits in Dubai. What is far more likely is that these are hawalas loans being repaid. This informal network, like the formal global banking system, moves money back AND forth. One of the major obstacles faced by Afghan businesses is a lack of access to credit. If you are the proud owner of Abdul’s Bricklayers Emporium, and you win a $1m contract from USAID to provide materials for the construction of a new school, you won’t get that money until 90 days after the bricks are delivered. But it is going to cost you $800k to buy the raw materials, fire the kilns, hire the extra labour, and ship the final product to the building site. So you turn to a hawalas to lend you the operating capital (at 30% interest). That money has to be physically shipped to Kabul and comes in on those same planes that are sending other bags of money back to Dubai.
Second, Rosenberg makes the mistake of referring to Afghanistan GDP as $13.5b (pointing out that the movement of cash out of Kabul is about a quarter of this figure). But the real GDP in Afghanistan, which includes the heroin trade, is far higher. How much bigger? Well, no one has an accurate estimate, but here is a fascinating piece of information: In a 2007 report by the US Army War College the authors state:
“…less than 20 percent of the $3 billion in opium profits actually goes to impoverished farmers, while more than 80 percent goes into the pockets of Afghan’s opium traffickers and kingpins and their political connections.”
That $2.4b in opium profits has to go somewhere, and I propose to you dear readers, that much of it is being flown through KIA.
So, to sum up, the Wall Street Journal is wrong. There is no way that $10m a day in donor and military money is being siphoned off and shipped out of Afghanistan. And if that amount of money is indeed being transferred, it is far more likely to be the hawalas system working as it is supposed to (a good thing) and/or the opium economy working as it is not supposed to (a bad thing).
UPDATE: This is turning into a classic example of how bad data can go viral. According to Josh Rogin at Foreign Policy, in Congress the foreign ops subcommittee chairwoman Nita Lowey has reacted to this misleading WSJ story by announcing “I do not intend to appropriate one more dime for assistance to Afghanistan until I have confidence that U.S. taxpayer money is not being abused to line the pockets of corrupt Afghan government officials, drug lords, and terrorists,”
Now let me be clear here, there is corruption in Afghanistan, mountains of it. But the WSJ story is misrepresenting the scale, probably by a factor of 10. So before Congress decides to freeze assistance, given what is at stake, let’s hope they invest some time and energy to dig into the hard data and do not just rely on unverified stories like this one.
UPDATE 2.0: This thing won’t die and no one seems to be trying to fix it. The WSJ continues to reference its own (unreferenced) data on allegedly stolen money flying out of Kabul, and respected aid bloggers are referencing them. Killing bad data is liking stomping cockroaches.