When US troops leave Afghanistan, will US aid dollars be withdrawn as well? The Obama administration says no. But House Republicans are beginning to push back, bolstered by damning testimony from the Special Inspector General for Afghan Reconstruction (SIGAR), the watchdog tasked with overseeing US aid to Afghanistan, arguing that American aid is being wasted on a corrupt regime.
For the past year plus, my colleagues Charles Kenny, Sarah Dykstra and I at CGD have been wading through geo-coded data sources on aid, violence, public opinion, and development progress in Afghanistan, examining where US aid is going, and what it has achieved. The full report on our number-crunching is still forthcoming, but even a cursory look at the data (see maps below) reveals a stark contrast with the assumptions underlying the discussion on Capitol Hill -- especially on the advantages of collaborating with the Afghan government when trying to spend aid money in Afghanistan.
But in a hearing of the House Subcommittee on National Security earlier this month, Rep. Jason Chaffetz (R-UT) went on the attack against exactly this idea.
"USAID has embarked on an effort to significantly boost the amount of funding that goes directly to foreign governments and non-US organizations… [which] all too often result[s] in funneling grants directly to unaccountable and often corrupt foreign governments."
Chaffetz then called SIGAR's John Sopko to make his case. As befits an auditor, Sopko's testimony erred on the side of risk aversion:
"The Afghan government does not appear to have the capacity to manage the amount of funding envisioned in the international community's pledges of direct assistance. Funds provided through direct assistance are... especially risky, given the pervasiveness of corruption in Afghanistan."
This sounds like legitimate cause for concern. Sopko noted that the Afghan National Army has failed to produce receipts for hundred of millions of dollars in fuel purchases, for instance. But that's not USAID's territory. If we focus just on the civilian, development component of American aid to Afghanistan -- a few billion of the roughly $100 billion we spend on the war each year -- the dynamics are actually very different.
Here are two simple facts Congress should keep in mind before embracing SIGAR's conclusions.
1. "On-budget" aid does not mean writing a blank check to Karzai.
First, let's clear up one huge misconception about what "on budget" aid means. If you listen to the noises coming from Capitol Hill, you might come away with the impression that the United States places blind faith in a government known to be very, very corrupt. (USAID backs up a Brinks truck with palettes full of cash to the loading deck of the Afghan Ministry of Finance, drives away, and hopes for the best.) This is not how on-budget aid actually works.
Instead, USAID typically relies on something called “host country contracting”. The Afghan government negotiates contracts with companies and NGOs (often American companies and NGOs) to implement projects, and USAID pays these contractors directly. Key point: for many on-budget programs, money never passes through Afghan government coffers. For programs where assistance is given directly to the Afghan government, it’s paid on a cost-reimbursement basis, upon receipt of invoices. There are no large, up-front chunks of cash being given to Hamid Karzai and company.
Take, for example, USAID's biggest "on budget" aid program in Afghanistan: the Ministry of Public Health's (MoPH) Basic Package of Health Services, designed to provide the most essential health services to underserved rural areas. While this program is on the Afghan government's budget, it's really an innovative collaboration between many players. The MoPH sets policy and coordinates at the national level. In each province, an NGO is chosen to coordinate services at the clinic level. An independent team of experts from Johns Hopkins Bloomberg School of Public Health monitors progress, and USAID reimburses implementing NGOs.
2. Aid through the Afghan government receives more scrutiny, and has more development impact, than when America goes it alone.
On the military side, the idea of a "transition" from US and NATO leadership to Afghan National Army leadership is often little more than a verbal fig leaf on a hasty withdrawal. But the opposite is true on the civilian side.
The two most celebrated, and by all accounts successful, American aid programs in Afghanistan have been programs run through Afghan government Ministries. In health, the BPHS program described above has -- according to researchers from Johns Hopkins -- led to enormous gains in access to care despite the ongoing war.
In rural development, the government's National Solidarity Program (NSP) -- supported by the World Bank, United States, and other donors, and implemented by international NGOs -- has distributed hundreds of millions of dollars for local infrastructure projects, overseen by democratically elected local councils. NSP has been subject to a large-scale randomized impact evaluation by a team from MIT and the World Bank, who found significant improvements in local governance and economic conditions as a direct result of "on budget" aid. This is a level of scrutiny almost no "off budget" aid programs can claim. As a result, NSP was held up as a model for future development work in Afghanistan by the Senate Foreign Relations Committee under John Kerry.
From a development perspective, there’s a strong case to be made that on-budget aid programs such as NSP are much better targeted than off-budget American aid. The maps below from our forthcoming CGD working paper illustrate this point.
Mapping the priorities of off- and on-budget aid:
American off-budget aid is oriented toward winning hearts and minds in unstable areas. The top left panel shows the pattern of civilian deaths from terrorist attacks across districts – as defined broadly by the National Counterterrorism Center’s Worldwide Incidents Tracking System, to include most insurgent attacks. As seen in the bottom left panel, the pattern of violence predicts quite well the allocation of US aid quick close. (Darker blue districts represent more American aid per capita, combining spending from both USAID and the Pentagon, the vast majority of which is off-budget.)
In contrast, funding through on-budget aid programs like NSP is more closely linked to development needs in more stable areas. The top right panel shows province-level poverty estimates from the 2007 National Risk and Vulnerability Assessment. The bottom right panel shows grant disbursements as of 2007 by NSP, as reported by Beath, Christia, and Enikolopov. Roughly speaking, blue districts (off-budget aid) track red districts (violence), while green districts (on-budget aid) track poverty (black areas). This pattern suggests the shift from off- to on-budget aid may improve the allocation of resources from a development perspective.
Bottom line, effective development aid and working with the Afghan government go hand in hand.
It's in America's interest that Afghanistan have a functional state. The United States can't hope to contribute to that goal if it insists on bypassing the Afghan state with its development aid.